SOVEREIGN CLIMATE AND DISASTER RISK POOLING World Bank Technical Contribution to the G20 SOVEREIGN CLIMATE AND DISASTER RISK POOLING World Bank Technical Contribution to the G20 © 2017 International Bank for Reconstruction and Development / International Development Association or The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its 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 the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ACKNOWLEDGMENTS This report has been prepared by the World Bank, Salvador Perez Maldonado, Associate Director drawing on inputs from a group of internationally General of Insurance, Ministry of Finance and renowned experts comprising David Bresch, Dario Public Credit, Mexico, for sharing information on Luna, and Simon Young. Mexico’s Natural Disaster Fund (FONDEN). The team also thanks participants in the stakeholder The World Bank team was led by Olivier Mahul consultation meetings held in Berlin on February and included Bianca Adam, Benedikt Signer, 9 and March 21, 2017. Simeon Abel, Roberto Dario Bacchini, Richard Poulter, and Emily White, all from the Disaster Internal reviews and inputs from Luis Alton, Risk Financing and Insurance Program, a joint Michael Bennett, Martin Buehler, Samantha partnership between the World Bank’s Finance Cook, Julie Dana, Francis Ghesquiere, Eugene and Markets Global Practice and the Global Gurenko, Stephane Hallegatte, Hideaki Hamada, Facility for Disaster Reduction and Recovery. Niels Holms-Nielsen, Nicholas Jones, and John Pollner greatly enhanced the report. The report The team thanks Isaac Anthony, CEO of the was edited by Anne Himmelfarb. Design and Caribbean Catastrophe Risk Insurance Facility layout were by Studio Grafik. (CCRIF SPC), for sharing information on CCRIF and CCRIF for Central American countries (CCRIF- The report was prepared during the period CA); Mohamed Beavogui, General Director of October 2016–April 2017 and was sponsored African Risk Capacity (ARC), and Dolika Banda, by the Federal Ministry for Cooperation and CEO of ARC Ltd., for sharing information on ARC; Development of Germany (BMZ). ABBREVIATIONS ACP-EU African, Caribbean, Pacific–European Union APEC Asia-Pacific Economic Cooperation ARC African Risk Capacity ARC Ltd. African Risk Capacity Insurance Company Limited ARV Africa RiskView ASEAN Association of South East Asia Nations BMZ Federal Ministry for Economic Cooperation and Development CARICOM Caribbean Community CAT bond catastrophe bond CAT-DDO Catastrophe Deferred Draw-Down Option CCRIF Caribbean Catastrophe Risk Insurance Facility (CCRIF SPC since 2014) CCRIF-CA Caribbean Catastrophe Risk Insurance Facility for Central American countries COSEFIN Council of Finance Ministers of Central American countries plus the Dominican Republic and Panama DFID United Kingdom Department for International Development DPF Development Policy Financing EAP East Asia and Pacific ECA Economics of Climate Adaptation FONDEN Natural Disaster Fund (Fondo de Desastres Naturales) GDP gross domestic product GFDRR Global Facility for Disaster Reduction and Recovery HSNP Hunger Safety Net Programme IBRD International Bank for Reconstruction and Development IDA International Development Association IMF International Monetary Fund IRMS Integral Risk Management Strategy L4D Licensing for Development program MDB multilateral development bank NPV net present value OECS Organization of Eastern Caribbean States PCRAFI Pacific Catastrophe Risk Assessment and Financing Initiative PEF Pandemic Emergency Financing Facility PSNP Productive Safety Net Programme SDG Sustainable Development Goal SPC Secretariat of the Pacific Community TCIP Turkish Catastrophe Insurance Pool V20 Vulnerable Twenty Group SOVEREIGN CATASTROPHE RISK POOLS TABLE OF CONTENTS Overview 3 Introduction 8 PART 1. Managing the financial impact of climate and disaster risks 10 1.1. The drivers of disaster impacts: A look into the future 11 1.2. The many impacts of disasters: From public finances to poverty reduction 14 Fiscal and economic impact 14 Development and poverty impacts 14 1.3. The elements of an effective approach to disaster preparedness and crisis response 16 Coordinated plan for post-disaster action agreed in advance 16 Fast, evidence-based decision-making process 17 Pre-planned financing to ensure that the plan can be implemented 17 Growing interest 18 Going beyond climate and disaster risks: A comprehensive approach to risk management and crisis response 19 1.4. Financial protection: From post-disaster emergency borrowing to proactive risk management 20 1.5. Selecting the appropriate financial protection instruments 24 1.6. Insurance: A powerful tool to be used with caution 28 PART 2. Lessons learned from experiences of sovereign catastrophe risk pools 30 2.1. Global landscape of sovereign catastrophe risk pools for developing countries 33 Existing sovereign catastrophe risk pools 33 New initiatives for sovereign catastrophe risk pools 39 2.2. Efficiency of catastrophe risk pools 40 Political ownership 41 Financial efficiency 43 Operational efficiency 48 2.3. Sustainability of catastrophe risk pools 50 Who owns the risk? 50 Technical expertise 50 Risk-based pricing 51 Premium financing 51 Regional ownership 56 2.4. Increasing the impact of sovereign catastrophe risk pools 57 Public financial management of natural disasters 57 Shock-responsive social protection 58 Critical infrastructure 59 Crowding in the private sector 59 PART 3. Recommendations moving forward 60 3.1. Catastrophe risk pools can enhance the financing of climate and disaster risks 61 3.2. Catastrophe risk pools do not solve all climate and disaster risk financing problems 63 3.3. Recommendations for G20 members 64 SOVEREIGN CATASTROPHE RISK POOLS 1 Glossary 66 Annex 1. Brief description of existing sovereign catastrophe risk pools 69 Caribbean Catastrophe Risk Insurance Facility (including CCRIF-CA) 69 African Risk Capacity 70 Pacific Catastrophe Risk Assessment and Financing Initiative (PCRAFI) 71 Sovereign catastrophe risk transfer: The experience of Mexico 72 Annex 2. Sovereign catastrophe risk pools: Annual portfolios 74 References 76 BOXES Box 1. Turn down the heat: Climate change impacts around the world 11 Box 2. The economics of climate adaptation 13 Box 3. A comprehensive disaster risk management framework 21 Box 4. Index insurance: A few definitions 22 Box 5. Turkish Catastrophe Insurance Pool 23 Box 6. The principles of effective financial protection 26 Box 7. Quantifying the benefits of financial protection 27 Box 8. Insuring the world: Risk pooling and reduction in capital requirements for low- and middle-income countries 32 Box 9. Toward a regional approach for disaster risk finance in Asia 38 Box 10. Exploratory work on new catastrophe risk pools 40 Box 11. Leveraging donor funds in risk pools 43 Box 12. Diversification benefits and their limits: An illustrative case 45 Box 13. Gathering further diversification benefits through risk swaps between regional pools? 46 Box 14. CCRIF-CA options for risk transfer 47 Box 15. Concessional financing of premium 53 Box 16. Quantitative comparison of seed capital versus premium subsidies 54 Box 17. PCRAFI Facility 56 FIGURES Figure 1. Loss events worldwide, by hazard type (1970–2015) 12 Figure 2. Welfare impacts from climate shocks 15 Figure 3. The elements of an effective approach to disaster preparedness and crisis response 16 Figure 4. Sovereign disaster risk layering 24 Figure 5. Map of countries participating in a sovereign catastrophe risk pool (as of December 2016) 34 Figure 6. Catastrophe risk insurance premium decomposition 41 Figure 7. Diversification benefits modeled for the PCRAFI program: Simple aggregates versus pooled risk for 1-in-250-year return period losses 44 Figure 8. Flow of funds 58 TABLES Table 1. Summary overview of existing regional sovereign catastrophe risk pools (as of December 2016) 5 Table 2. Costs of different instruments for financing post-disaster expenditure 29 Table 3. Detailed overview of existing regional sovereign catastrophe risk pools (as of December 2016) 35 Table 4. FONDEN risk transfer program, 2006–2015 (US$ million) 73 2 SOVEREIGN CATASTROPHE RISK POOLS OVERVIEW More than 1 billion people have lifted themselves A growing number of governments are moving out of poverty in the past 15 years, but climate toward a proactive (and more cost-effective) and disaster risks threaten these achievements. approach to financial planning to protect national Global economic losses from disasters are now budgets, as well as the lives and livelihoods of reaching an average of more than US$300 billion their citizens, from the impacts of disasters. a year. A recent World Bank report finds that the This approach helps governments consider impacts of disasters on well-being are equivalent disaster and climate shocks as part of their to a US$520 billion drop in consumption (60 fiscal risk management strategies. In addition, it percent more than the asset losses usually complements other elements of a comprehensive reported) and force some 26 million people disaster risk management strategy, ranging into poverty every year (Hallegatte et al., 2017). from investments in risk reduction to improved And countries face increasingly complex threats preparedness and resilient reconstruction. that often compound the negative impacts of disaster and climate shocks—from migration Financial protection involves planning ahead caused by fragility and conflict situations, to to better manage the cost of disasters, the risk of pandemics. It is estimated that 93 ensure predictable and timely access to much- percent of people facing extreme poverty today needed resources, and ultimately mitigate are living in countries that are politically fragile long-term fiscal impacts. By combining various or environmentally vulnerable, and in many cases financial instruments—such as contingency both. The United Nations’ humanitarian appeal for budget, contingent loans and grants, and risk 2017, for example, stands at a record US$22.2 transfer solutions—financial protection allows billion, to help almost 93 million people affected governments to manage the full range of disaster by conflicts and natural disasters. impacts. Different instruments help address different risks (ranging from recurrent to more Climate change exacerbates some of these rare events) and different funding needs (ranging risks by increasing the frequency and intensity from short-term emergency relief to recovery and of extreme weather events. In addition, economic reconstruction). growth and rapid urbanization increase exposure. Building resilience is therefore crucial to safeguard Financial protection is also an important topic poverty reduction efforts and promote sustainable on the global agenda, for example under the and inclusive development— particularly for the Sendai Framework for Disaster Risk Reduction, poor and vulnerable who are the least able to the Humanitarian Financing Agenda, and the G7 cope with and adapt to increasing risks. InsuResilience Initiative. The G7 InsuResilience Initiative, sponsored by the German presidency Current post-disaster response financing, of the G7 with the goal of expanding climate risk including donor assistance and commercial insurance coverage to an additional 400 million insurance, covers only a fraction of disaster poor and vulnerable people in developing countries losses, creating a protection gap. On average by 2020, has already taken significant steps only about 30 percent of catastrophe losses toward expanding existing disaster and climate have been covered by insurance over the past risk insurance programs and creating new ones. 10 years. That means that about 70 percent of catastrophe losses have been borne directly by In the immediate aftermath of a disaster, being individuals, firms and governments (SwissRe, able to rapidly access financial resources is 2016). Donor assistance is struggling to keep up crucial to save lives and livelihoods. Quick- with growing needs. In 2015 for example, almost disbursing financial protection instruments, half of the of the UN’s humanitarian appeals were such as contingent credit and insurance, can left unmet (UN, 2016). reduce humanitarian impacts and save money by SOVEREIGN CATASTROPHE RISK POOLS 3 enabling rapid crisis response and relief efforts. provides indemnity coverage, where payouts are In Ethiopia, for example, every US$1 secured based on actual losses on public infrastructure. ahead of time for early drought response can Catastrophe risk pools could also be used to save up to US$5 in future costs (Wiseman and aggregate insurance of public infrastructure, or Hess 2007). to manage the contingent liability from shock- responsive social protection schemes more cost- Catastrophe risk pools are emerging as a cost- effectively. Some countries in South East Asia effective vehicle to help countries access are also exploring risk pools as a more effective rapid financing for disaster response. They approach to reserves as standby financing. allow countries to (i) pool risks in a diversified portfolio; (ii) retain some risk through joint A decade of experience has shown that political reserves/capital; and (iii) transfer excess risk to commitment, sound operational design, and the reinsurance and capital markets. By putting financial sustainability are at the foundation of a price tag on risk, risk pools also increase the successful risk pools. When those foundations value of risk information and create incentives to are in place, risk pools can in turn generate invest in risk reduction. Their emergence over the positive externalities that further enhance their last decade provides governments with access to impact, by fostering political, operational, and a new set of instruments to enhance the financial financial effectiveness. management of climate and disaster risks. The establishment of sovereign catastrophe Over the past 10 years, 26 countries in three risk pools relies on a combination of highly regions—Africa, the Pacific, and the Caribbean specialized technical assistance, significant and Central America—have joined sovereign financial support, and strong political catastrophe risk pools. They have purchased commitment. The long-term sustainability of parametric catastrophe risk insurance for an sovereign catastrophe risk pools depends on aggregate coverage of US$870 million and an their ability to generate regular and large enough aggregate premium volume of US$56.6 million premium income, possibly with financial support (2016/17), backed by more than 30 reinsurance from donor partners; broaden the set of financial companies. The three pools have so far made instruments offered beyond parametric insurance; payouts totaling just over US$105 million (table maintain strong political commitment; and link O.1 provides an overview of existing risk pools). financial instruments to pre-agreed post-disaster programs, such as shock-responsive social Parametric insurance solutions allow for rapid protection programs or critical infrastructure payouts in the event of a disaster, providing recovery programs, to ensure that funds can be liquidity within a couple of weeks to finance efficiently channeled to support targeted post- rapid response. For example, having purchased disaster responses. insurance through the Pacific Catastrophe Risk Assessment and Financing Initiative (PCRAFI), Catastrophe risk pools have significantly the government of Vanuatu received a payout of relied on donor partners for their technical and almost $2 million just seven days after Tropical financing capacity. All sovereign catastrophe Cyclone Pam made landfall in March 2015. risk pools have benefited from donor support to This amount was eight times the government’s start operations and to remain sustainable during emergency provision and was critical for funding their first years. Donor financing has at various a number of urgent priorities, including flying stages covered start-up costs, capitalization, nurses to the most affected areas and providing and sometimes (partial) premium financing. lifesaving assistance. Existing sovereign catastrophe risk pools have also required many years of sustained technical Beyond parametric insurance, other financial assistance from credible third parties; the instruments can be structured and offered World Bank Group has assisted the Caribbean by risk pools. For example, Mexico’s disaster Catastrophe Risk Insurance Facility (CCRIF) and fund, which acts as a national level risk pool, PCRAFI, and the World Food Program has assisted 4 SOVEREIGN CATASTROPHE RISK POOLS African Risk Capacity (ARC). Risk pools have also the enabling environment and incentives for relied on the technical expertise and capacity of systematic adoption of disaster risk finance and the private insurance and reinsurance industry. insurance solutions, including catastrophe risk pools. Regional catastrophe risk pools require a regional partner organization to facilitate the political The private sector has contributed to making and policy dialogue and coordination between catastrophe risk pools cost-effective. The private participating governments. Given the level of insurance industry has been heavily involved in cross-country coordination required to establish the preparation and implementation of sovereign and manage such a pool, regional political bodies catastrophe risk pools. It provides not only risk are essential to facilitate the process. Sovereign capital but also technical expertise to inform the pools have relied on their respective regional design of effective risk pools. political organization at various levels. This report focuses on sovereign climate and Sovereign catastrophe risk pools—and disaster disaster risk pools as a mechanism to enhance risk financing solutions more generally—require financial protection of national and subnational that participating countries be committed to governments. It discusses for G20 consideration implementing necessary policy reforms. In how those pooling mechanisms could be further this context, humanitarian and development expanded and replicated, how the cooperation donors have a role to play in creating incentives between G20 countries and developing countries (both at the country level and within donor could be further strengthened to integrate disaster organizations) for investments in pre-agreed risk and climate risk into broader financial protection management and risk financing solutions, and for strategies for vulnerable people, and how to set reducing reliance on post-disaster humanitarian expectations about the role of catastrophe risk assistance. Innovative concessional financing pools as a meaningful, cost-effective instrument instruments may be necessary to create to that end. TABLE 1. SUMMARY OVERVIEW OF EXISTING REGIONAL SOVEREIGN CATASTROPHE RISK POOLS (AS OF DECEMBER 2016) CCRIF (Caribbean) CCRIF-CA ARC PCRAFI (Central America) Perils Earthquake, tropical Earthquake, tropical Drought, tropical Earthquake, tropical cyclone, extreme cyclone, extreme cyclone, flood cyclone, extreme rainfall rainfall rainfall Initial capital Multi-donor grants Multi-donor grants Interest free loan Multi-donor grants via World Bank via World Bank from 2 partners via World Bank Participating 20 eligible; 16 have 6 eligible; 1 has 32 signatories; 8 15 eligible; 6 have countries participated,14 have purchased coverage have participated, 6 participated, 5 have purchased coverage in 2016/17 purchased coverage in 2016 in 2016/17 Avg premium income US$20 million US$1.5 million US$22 million US$2 million Cumulative payouts US$67.3 million US$0.7 million US$34 million US$3.2 million Avg aggregate US$622 million US$28 million US$50 million US$45 million coverage Source of premiums IDA credits, CDB IDA credit National budgets, Grants, national credits, grants grants budgets, IDA credits Reserves US$117 million US$1.3 million US$98.5 million US$6 million Note: IDA = International Development Association; CDB = Caribbean Development Bank. SOVEREIGN SOVEREIGN CATASTROPHE CATASTROPHE RISK RISK POOLS POOLS 5 RECOMMENDATIONS FOR G20 MEMBERS The G20 could promote a set of priority action Activities under this action area could include areas designed to reduce the protection gap in vulnerable developing countries. These actions ŠŠ Knowledge exchange to learn from would advance financial protection against experience and consolidate good practices climate and disaster risks, in part by encouraging in disaster response planning the scale-up of catastrophe risk pools at the ŠŠ Technical assistance and investments supranational, national, and subnational levels. to develop shock-responsive scalability Specifically, the G20 could promote activities that mechanisms for existing social safety nets support the following priority action areas: to protect the poor and vulnerable ŠŠ Technical assistance and investments ƒƒ Facilitate the adoption of financial protection to identify, prioritize, and protect critical strategies that include a mix of financial infrastructure at risk, both ex ante (by instruments against disaster and climate mainstreaming disaster risk reduction risks, such as budgetary instruments, in investment planning) and ex post (by contingent credit, and catastrophe risk transfer developing pre-agreed financial plans for to increase the ownership, impact, and cost- post-disaster reconstruction) efficiency of disaster response financing. ƒƒ Promote institutional and legal frameworks Activities under this action area could include that enable the implementation of financial protection strategies. This includes creating ŠŠ Technical assistance to support the the legal base that enables governments to development of financial protection establish disaster risk management funds, strategies, including diagnostic reviews of pay insurance premium and manage insurance countries’ approach to financial protection, proceeds, and join supranational financial and identification of policy options for entities such as catastrophe risk pools. strengthened financial resilience ŠŠ Technical assistance and investments to Activities under this action area could include support the implementation of national financial protection strategies, including ŠŠ Knowledge exchange among countries to for specific line ministries or sectors learn from experience in public financial management of climate and disaster risks ƒƒ Support the development of pre-agreed ŠŠ Technical assistance to incorporate disaster response plans backed by financial climate and disaster risks into public protection strategies to help poor and finance frameworks vulnerable households and protect key lifeline infrastructure. Such plans can help raise awareness of the benefits of risk reduction and financial protection by engaging a wide range of stakeholders, including members of civil society. 6 SOVEREIGN CATASTROPHE RISK POOLS ƒƒ Develop new concessional financing for The Global Partnership could bring together catastrophe risk transfer instruments to relevant partners from developing and developed incentivize vulnerable developing countries countries, international organizations, the private to develop and adopt sustainable financial sector, and civil society. To achieve maximum protection strategies. impact, the Global Partnership would leverage the comparative advantages of all partners and build Activities under this action area could include on the work of existing platforms and initiatives. In particular, it would leverage the technical ŠŠ Cofinancing of capitalization and operating expertise and capacity of the private insurance costs of catastrophe risk pools and reinsurance industry. ŠŠ Cofinancing of premiums for insurance solutions (designed to incentivize The G20 could develop a work program countries to progressively increase their structured around the four priority action areas contributions over time) identified above to specify how countries would support specific activities. Such efforts would not To achieve the overarching objective of reducing only promote financial protection and help close the protection gap in vulnerable developing the protection gap, but would also support the countries, and to catalyze action around these broader disaster and climate resilience agenda. priority areas and activities, the G20 could promote the creation of a Global Partnership for Climate and Disaster Risk Finance and Insurance Solutions. SOVEREIGN SOVEREIGN CATASTROPHE CATASTROPHE RISK RISK POOLS POOLS 7 INTRODUCTION Climate change and extreme weather events noted that risk pooling does not reduce the threaten global efforts on poverty reduction and underlying risk (which should be reduced through sustainable development. In this context, more appropriate risk mitigation measures), but rather and more governments are making financial allows for improved spreading of risk, leading to resilience a priority to protect national budgets as potentially significant reductions in the cost of well as the lives and livelihoods of the poorest and risk, particularly for severe events . Risk pools most vulnerable . How to better manage climate therefore contribute to more effective and more and disaster risks is an increasingly important affordable financial risk management. topic on the global agenda, for example under the Sendai Framework for Disaster Risk Reduction, The principle of risk pooling is extensively used the Humanitarian Financing Agenda, and the G7 for managing catastrophe risks. The insurance InsuResilience Initiative. industry has developed catastrophe risk insurance pools as a way to offer affordable coverage Financial protection, including climate and disaster to homeowners and enterprises in developed risk financing and insurance, can help countries countries and, more recently, in developing increase their financial resilience against disaster countries (e.g., the Turkish Catastrophe and climate risks in several ways: it allows them Insurance Pool established in 1999).In countries to secure access to rapid financing to address where subnational government entities (such emergency response and early recovery needs; to as provinces or states) have substantial power smooth the budget volatility associated with post- and responsibility in the financial response to disaster expenditures; and to incentivize disaster disasters, subnational risk pools can increase risk reduction. the financial resilience of local governments (e.g., Mexico’s FONDEN). This approach has recently The insurance industry has a long record of been used to establish sovereign catastrophe dealing with the risks associated with natural risk pools, that is, pools where the contributors hazards, and the international community has and beneficiaries are sovereign governments. learned valuable lessons from this experience. Sovereign catastrophe risk pools include the But the specific challenges faced by low-income Caribbean Catastrophe Risk Insurance Facility populations and the new challenges posed by (CCRIF, originally established for Caribbean climate change call for development partners and countries and now expanded to Central American industry to think beyond the financial instruments countries); the Pacific Catastrophe Risk Insurance currently available. By applying proven financial Facility (the PCRAFI Facility); and the African Risk risk management tools and insurance systems to Capacity (ARC). public policy, countries can unlock new solutions to improve their financial resilience. When The G7 InsuResilience Initiative, sponsored by applied to rules-based programs such as shock- the German presidency of the G7 with the goal of responsive social protection systems that rapidly expanding climate risk insurance coverage to an increase assistance following a disaster shock, additional 400 million poor and vulnerable people insurance can reach large numbers of people. in developing countries by 2020, has already Risk pooling is a fundamental principle of risk taken significant steps toward expanding existing management and insurance: by combining and disaster and climate risk insurance programs spreading the risks faced by a large number of and creating new ones. To date, 26 countries in contributors into a single portfolio, pools ensure Africa, the Pacific, and the Caribbean and Central that each contributor’s share of the portfolio America have purchased sovereign catastrophe is less risky than its initial share. It should be risk coverage. However, climate and disaster 8 SOVEREIGN CATASTROPHE RISK POOLS risk insurance coverage for vulnerable people in and poverty impacts. It presents key principles developing countries is far from comprehensive, for effective risk management: a coordinated and indeed remains low. A key question for G20 plan for post-disaster action agreed in advance; consideration is how to strengthen cooperation a fast, evidence-based decision-making process; between G20 countries and developing countries and pre-planned financing to ensure that the plan to build solutions that integrate disaster and can be implemented. Finally, it describes the climate risk into broader financial protection menu of financial instruments and approaches strategies for vulnerable people, and how to set (including risk pooling) available to governments, expectations about the role of catastrophe risk and presents an approach to defining the optimal pools as a meaningful, cost-effective instrument mix of financial instruments, given a government’s to that end. specific needs and priorities. Risk pools can play an important role in moving the Part 2 describes the sovereign catastrophe management of disaster and climate shocks away risk pools now operating, examines potential from uncertain, ad hoc humanitarian assistance gaps in coverage in those pools, and identifies and making it part of planned development. possible opportunities for new sovereign pools. The challenge for the international community It discusses the efficiency of pools from a is to provide the right set of incentives. Both political, operational, and financial perspective international partners and potential recipient and reviews pools’ sustainability and impact. The governments have a responsibility to plan and purpose of this discussion is not to evaluate the program financing in advance. effectiveness of the different instruments offered by existing sovereign pools, but rather to identify This report focuses on sovereign climate and key success and failure factors and opportunities disaster risk pools as a mechanism to enhance for further improvement. financial instruments available to national and subnational governments. Taking into account lessons learned from existing sovereign climate and disaster risk pooling and Part 1 presents the broader framework for transfer mechanisms, part 3 recommends a financial resilience that policy makers in G20 set of priority action areas that G20 member countries should consider and the role of risk countries could consider to strengthen the existing pools within this framework. It briefly describes landscape of climate and disaster risk finance and the main drivers of risks and their development insurance solutions, including risk pooling. SOVEREIGN CATASTROPHE RISK POOLS 9 PART 1. Managing the financial impact of climate and disaster risks 10 SOVEREIGN CATASTROPHE RISK POOLS 1.1. THE DRIVERS OF DISASTER IMPACTS: A LOOK INTO THE FUTURE Economic growth, demographic trends, and construction standards are not applied. For rapid—often unplanned—urbanization are among instance, it is estimated that over half of Africa’s the main drivers of disaster losses worldwide. urban population (61.7 percent) lives in slums When people and economic assets concentrate (UN-HABITAT 2013). in cities located in hazard-prone areas, exposure to disaster risk rises, pushing potential losses This increasing urbanization is likely to continue upward. Half of the world’s population now lives apace and has implications for future trends in in cities, compared to 39 percent in 1980. In disaster risks. The urban population is expected to most developing countries, rapid urbanization is reach two-thirds of the global population by 2050, leading to the creation of informal settlements with most of that growth concentrated in middle that are not reached by public services and where and low-income countries (UNDESA, 2014). BOX 1 - TURN DOWN THE HEAT: CLIMATE CHANGE IMPACTS AROUND THE WORLD EASTERN EUROPE AND SOUTH ASIA  MIDDLE EAST AND CENTRAL ASIA Inconsistences in the monsoon season and NORTH AFRICA The impact of climate change unusual heat extremes will affect crops. will vary region to region. Melting Reduced snow melt from the Himalayas will Extreme heat will spread glaciers and warming temperatures decrease the flow of water into the Indus, across more of the will shift the growing season and Ganges and Brahmaputra basins. Together, they land for longer periods the flow of glacier-fed rivers further threaten to leave hundreds of millions of people of time, making some into spring in Central Asia, while without enough water, food, or access to reliable regions unlivable and in the Balkans, worsening drought energy. Bangladesh and the Indian cities of reducing growing areas conditions will put crops at risk. Kolkata and Mumbai will be confronted with for agriculture. Rising temperatures will put increased flooding, intense cyclones, sea-level intense pressure on crops rise, and warming temperatures. and already scarce water resources, potentially increasing migration and the risk of conflict. LATIN AMERICA AND THE CARIBBEAN Longer droughts, extreme weather, and increasing ocean acidification will SOUTH EAST ASIA affect the region. In the SUB-SAHARAN AFRICA A sea-level rise of 30 cm, possible tropical Andes, rising Food security will be the overarching challenge, by 2040 if business as usual temperatures will reduce with dangers from droughts, flooding, and shifts continues, would cause massive glacier ice and the in rainfall. With 1.5°C-2°C warming, farmers flooding in cities and inundate low- meltwater that some 50 will loose 40-80% of cropland from drought and lying cropland with saltwater. In million people in the low- aridity by the 2030-40s. In a 4°C warmer world, Vietnam’s Mekong Delta this could land farms and cities rely around the 2080s, annual precipitation may reduce rice production by 11%. At on. Rising temperatures decrease by up to 30% in southern Africa, while the same time, storm intensity is will affect food security. East Africa will see more rainfall. likely to increase. SOVEREIGN CATASTROPHE RISK POOLS 11 Climate change compounds this increase in intense or frequent hydrometeorological hazards exposure, exacerbating potential impacts of and climate extremes is a potential driver of future disasters. In some regions, changes in climate losses (see Figure 1). The world’s largest coastal patterns are expected to cause an increase in the cities, for example, could experience losses of frequency and intensity of climatic disasters, in US$1 trillion by mid-century (Hallegatte et al. 2013). the form of rising temperatures (e.g., heat waves), changing precipitation patterns (e.g., heavier Clearly, climate change threatens the objective of rains leading to flash floods), and sea storms eradicating poverty. Climate is involved in many (IPCC 2013) (see Figure 1). if not most of the shocks that keep households poor, or cause them to fall into poverty: these Over the past 30 years, over 70 percent of include natural disasters such as floods (which total economic losses from disasters have cause asset loss and affect human capital), been attributable to hydrometeorological and health shocks such as malaria (which result in climatological events (storms, floods, droughts, health expenditures and lost labor income or and extreme temperatures). The threat of more worse), and crop losses and food price shocks FIGURE 1. LOSSES AND LOSS EVENTS WORLDWIDE, BY HAZARD TYPE (1970–2015) FIGURE 1. LOSSES AND LOSS EVENTS WORLDWIDE, BY HAZARD TYPE (1970–2015) 140 6 9 120 100 80 10 5 7 60 1 3 4 8 2 40 20 0 1970 1975 1980 1985 1990 1995 2000 2005 2010 2016 Earthquake/tsunami Weather-related catastrophes Man-made disasters 1 1992: Hurricane Andrew 6 2005: Hurricanes Katrina, Rita, Wilma 2 1994: Northridge earthquake 7 2008: Hurricanes Ike, Gustav 3 1999: Winter Storm Lothar 8 2010: Chile, New Zealand earthquakes 4 2001: 9/11 attacks 9 2011: Japan, New Zealand earthquakes, Thailand Flood 5 2004: Hurricanes Ivan, Charley, Frances 10 2012: Hurricane Sandy Source: Swiss Re 2016 12 SOVEREIGN CATASTROPHE RISK POOLS (due to drought or crop disease). A recent World Countries face increasingly complex threats that Bank report estimates that climate impacts often compound the negative impacts of disaster could push an additional 100 million people and climate shocks, from migration caused by into poverty by 2030 unless there is action to fragility and conflict situations to the risk of reduce extreme poverty, provide access to basic pandemics. For instance, it is estimated that 93 services, strengthen resilience, and increase percent of people facing extreme poverty today adaptive capacity (Hallegatte et al. 2016). are living in countries that are politically fragile or environmentally vulnerable, or in many cases both. While combating climate change will require investments in mitigation, inclusive, climate- informed development—development that invests in cost-effective adaptation measures and builds the resilience of economies and people—can prevent most (though not all) of climate change’s impacts on poverty (see Box 2). BOX 2. THE ECONOMICS OF CLIMATE ADAPTATION Decision makers understand that adaptation million and US$26 billion) annually as a result measures can make societies more resilient of existing weather patterns. Climate change to adverse weather and the impacts of climate could worsen this picture significantly: under an change. Hence finance ministers, mayors, extreme climate change scenario, annual losses private sector leaders, and others have sought from flood, drought, heat waves, and tropical to quantify the potential weather- and climate- storms could increase by US$33 billion in the related damage to their economies and societies decades to come. over coming decades. But prevention and mitigation measures can The Economics of Climate Adaptation (ECA) help manage such weather and climate risks. methodology is a tool that can help decision Adaptation measures, from strengthened flood makers understand the impact of weather events defenses and improved building codes to beach and climate change on their economies, and nourishment and roof cover retrofits, can avert identify actions that will minimize impact at the between 15 and 80 percent of the total risk. lowest cost to society. For risks that cannot be mitigated, risk financing Assuming that current development trajectories strategies can provide protection by capping continued to 2030, the ECA methodology was losses and smoothing the costs of climate events applied to some 20 regions worldwide. The to individuals, corporations, and governments. results showed that national and local economies Effective risk reduction notably lowers the costs could lose between 1 and 20 percent of gross of such arrangements. domestic product (GDP) (or between US$47 Source: Bresch and Müller 2014. SOVEREIGN CATASTROPHE RISK POOLS 13 1.2. THE MANY IMPACTS OF DISASTERS: FROM PUBLIC FINANCES TO POVERTY REDUCTION In 2015, the share of the global population living long-term macroeconomic impacts that affect in extreme poverty was estimated to have fallen the government’s budget. When record floods below 10 percent for the first time. More than 1 inundated large swaths of Thailand in the fall of billion people have lifted themselves out of poverty 2011, for example, total damage and loss was in the past 15 years alone (World Bank 2016). estimated at some US$46.5 billion, more than 13 However, climate and disaster risks threaten percent of that year’s GDP. But the financial impact these achievements. Global economic losses continued even after the water receded. According from disasters are now reaching an average of to estimates, the floods reduced real GDP growth US$250 billion to US$300 billion a year (UNISDR in 2011 by 1.1 percent from preflood projections, 2015). With climate change exacerbating some reduced the current account to US$11.9 billion disaster risks and economic growth and rapid from a projected US$20.6 billion, and caused a urbanization increasing exposure, building the loss in tax revenue of 3.7 percent from estimated resilience of societies is ever more crucial to preflood revenues (World Bank and Government protect poverty reduction efforts and promote of Thailand 2012). sustainable and inclusive development. When governments are ill-prepared to respond to fiscal In addition, there is some evidence that disasters shocks, development programs and budgets often can have an impact on a country’s credit rating. suffer. When productive assets are destroyed and Recent Standard & Poor’s analysis finds that savings wiped out by extreme events, it is much tropical cyclones have the potential to lead to harder for the poor to get back on their feet and downgrades of up to two notches. Post-disaster recover, let alone prosper. financing costs might therefore substantially increase after a disaster, especially for countries FISCAL AND ECONOMIC IMPACT with low ratings (S&P Global 2015b). And climate change can further aggravate the situation (S&P Governments play a central role in emergency relief, Global 2015a). recovery, and reconstruction in the aftermath of climate and disaster shocks. During and directly DEVELOPMENT AND POVERTY IMPACTS after a disaster, the government and international partners provide emergency relief to the affected Statistics on macroeconomic and fiscal impacts population, ranging from the distribution of food do not reflect the direct and indirect impacts that aid in areas affected by drought, to the drainage disaster losses can have on the poorest members of water in flood zones. Such disaster response of society, nor their long-term impacts on human expenditures require immediate access to and economic development. Often, the absolute liquidity and swift mobilization of funds to mitigate economic losses of the poorest households are the negative impact of disasters on people and small relative to those of the wealthy, and thus assets. If public infrastructure is damaged (and the consequences for the poorest communities not insured), the government will also need to pay are marginalized in the analysis. This is not just a for significant reconstruction costs. monitoring issue but one that has implications for the selection and design of projects to mitigate Disasters (especially those that affect large and/ risks, as more weight is often given to impacts or industrial areas of a country) can also have that can be expressed fairly easily in monetary 14 SOVEREIGN CATASTROPHE RISK POOLS FIGURE 2. WELFARE IMPACTS FROM CLIMATE SHOCKS EDUCATION MALNUTRITION POVERTY TRAPS The impacts of disasters on Even temporary malnutrition in The poor are at risk of getting education are two-fold. On one early childhood (as early as in locked into poverty traps. A hand, poor households often utero) can permanently stunt disaster can destroy their take children out of school in growth and lower cognitive productive assets, or wipe out the aftermath of a disaster to abilities, with consequences their savings, making it harder for help with manual labor or simply potentially lasting a lifetime. poor households to recover, let because they cannot afford school Stunted children are likely to grow alone prospe r. In the absence of anymore. up to be less productive adults insurance, lending opportunities, and will earn less. or social security institutions, even In El Salvador, for example, school the prospect of a disaster can attendance fell by 7 percent and Children who had suffered from have devastating consequences. children from affected households stunting during the 1982–84 were three times more likely to drought in Zimbabwe, for example, For instance, in anticipation of future work after the 2001 earthquake. were still doing worse in school disasters a household may decide And after Hurricane Mitch in than non-affected peers some 16 to minimize savings (such as in- Nicaragua, child labor increased years after the disaster. kind savings in the form of physical as much as 58 percent in the assets), because they risk being affected areas. And there is clear evidence that destroyed in the event of a disaster. early childhood nutrition plays a This results in limited financial On the other hand, children may not huge role in future development buffers that would otherwise allow be able to go to school because prospects. Studies on nutrition households to smooth consumption the building collapsed or was supplementation programs in in the event of transitory shocks damaged during a disaster. In Guatemala, for example, have such as droughts or floods—and China for example, the Wenchuan found that children who received may thus have further poverty earthquake destroyed more than supplements until the age of three implications. 7,000 schools. earned as much as 46 percent more than others as adults. terms, such as aggregate damages, rather than earning potential. Within households, the impacts impacts on well-being.1 of disasters are also not spread equally; women and girls often bear the brunt of both direct A recent World Bank report finds that when impacts impacts (mortality rates as a result of disasters on well-being are accounted for, natural disasters are higher among women) and indirect impacts are equivalent to a US$520 billion drop in (negative impacts on nutrition and school consumption in the global economy (or 60 percent performance are disproportionately high among more than the asset losses usually reported), and girls) (Baez, de la Fuente, and Santos 2010). that they force some 26 million people into poverty every year (Hallegatte et al. 2017). In this context, it is ever more urgent both to address key drivers of risk through mitigation Looking at the impacts on households and and adaptation measures and to improve communities, it is clear that the poor are the preparedness and response capacity. Section ones that bear the brunt of disasters and climate 1.3 provides an overview of the elements of an shocks (figure 2). Through their impact on human effective approach to disaster preparedness capital (in particular, nutrition, education, and and crisis response, based on pre-agreed post- health), disasters can severely affect household’s disaster plans and pre-planned financing. 1. When projects to reduce disaster risk are assessed on the basis of the value of damages that can be avoided, analyses favor projects that will protect or support richer areas or people. Imagine two flood protection projects with similar costs. The first would cover a wealthy neighborhood in a capital city. Because of the density of high-value assets, it would avert on average US$10 million a year in losses. The second project would target poorer areas in a second-tier city and prevent just US$5 million a year in losses. A traditional analysis would unambiguously select the first project. But a US$5 million loss may matter more to poor people than a US$10 million loss to richer people. If the second project benefits very poor people, it may generate greater benefits for well-being. And because well-being is the ultimate goal of public policy, the second project may be more attractive (Hallegatte et al. 2017). SOVEREIGN CATASTROPHE RISK POOLS 15 1.3. THE ELEMENTS OF AN EFFECTIVE APPROACH TO DISASTER PREPAREDNESS AND CRISIS RESPONSE Preventing losses and alleviating the impacts An effective approach to managing the impacts of of disasters requires a comprehensive disaster shocks should be based on the principles approach to disaster risk management—one described below and summarized in Figure 3. based on reducing and managing conditions of hazard, exposure, and vulnerability, but also COORDINATED PLAN FOR POST- on coordinated and pre-agreed post-disaster DISASTER ACTION AGREED IN ADVANCE plans backed by effective financial protection measures. Not all disasters and crises can be The first element of a coordinated approach prevented, and governments must be ready to disaster response is a single, solid disaster to manage the impacts of any residual risk. To response plan agreed in advance. Without such manage and mitigate the impacts of increasingly a plan, responsibilities may be ill-defined, work complex threats, governments must move away steps may be either duplicated or omitted, and the from reliance on traditional humanitarian support exploitation of economies of scale in logistics may financed with funds raised after an event and be lost. To avoid such inefficiencies, all stakeholders toward a system that emphasizes preparedness need to work together before disasters strike based on national response systems. to establish a credible plan with a clear decision FIGURE 3. THE ELEMENTS OF AN EFFECTIVE APPROACH TO DISASTER PREPAREDNESS AND CRISIS RESPONSE COORDINATED PLAN FOR POST-DISASTER ACTION AGREED IN ADVANCE • Consists of a single, credible plan for disaster response • Defines explicit responsibilities and liabilities of all stakeholders (who or what will be protected, against what, and who will pay for what) • Establishes clear decision process • Clarifies what risks the national/local government will take on, and what risks have to be shared with households and firms, as well as the role of international partners FAST, EVIDENCE-BASED DECISION-MAKING PROCESS • Identifies ahead of time objective and transparent rules to guide decision making • Requires investing in early warning systems and better data/information (ground data on loss of human life, building damage, area average index data on damage and losses, parametric indexes), including the human and technological capacity to collect data in a timely manner • Define rules and triggers that result in pre-agreed interventions to promote decisive, timely action PRE-PLANNED FINANCING FOR EARLY ACTION • Ensures that funds are available quickly when—and only when—they are required • Binds partners to pre-agreed objectives, decision processes, and implementation modalities • Promotes greater discipline, transparency, and predictability in post-disaster spending • Ensures rapid mobilization of funds, reducing humanitarian costs and potentially saving money Source: Based on Clarke and Dercon 2016. 16 SOVEREIGN CATASTROPHE RISK POOLS process and explicit responsibilities and liability. parametric indexes. No rule is perfect, so there The relevant stakeholders must decide who or should be some discretionary backup system to what will be protected, against what, and who will deal with situations in which the rules fail. pay for what. By removing any ambiguity about responsibility, such a plan clarifies what risks the PRE-PLANNED FINANCING TO national or local government will take on, and what ENSURE THAT THE PLAN CAN BE risks have to be shared with households and firms. IMPLEMENTED It is also necessary to plan in advance how post- Finally, a coordinated approach to disaster disaster financial resources will be included in response involves financial planning. Pre-planned the national budget, how they will be disbursed financing not only ensures that the money is and executed, and how they will reach end available when it is needed; it also serves as the beneficiaries, especially the poor and vulnerable. glue that holds all the pieces of the plan together Delivery mechanisms are just as important as and makes it credible. Specifically, it ensures the resources they will channel, and selecting that funds are available quickly when—and only the right modality will ensure that funds reach when—they are required by the plan, and it binds beneficiaries rapidly and efficiently. Applying the various partners to pre-agreed objectives, insurance principles to enable the scale-up decision processes, and implementation of existing social safety nets is increasingly modalities so that the plan is strong enough to considered an effective way to reduce the negative withstand the whirlwind of highly charged post- impacts of shocks on the most vulnerable. Using disaster politics. Sections 1.4 and 1.5 describe data-driven, rules-based triggers ensures rapid, in greater detail the elements of a pre-planned transparent, and accountable responses in the approach to financing disaster response, also immediate aftermath of a disaster, or at the early known as financial protection. stages of slow-onset disasters such as droughts. This strategic approach to financing disaster FAST, EVIDENCE-BASED DECISION- response offers a number of benefits, including MAKING PROCESS greater discipline, transparency, and predictability in post-disaster spending (see Box 6). Pre-planned Effective disaster response must avoid costly financing also helps ensure the rapid mobilization delays. Thus countries must not only invest in of funds to support relief efforts, which is crucial early warning systems and better information, to limit humanitarian impacts. This rapid response but also identify well in advance objective and can also save money. For example, well-targeted transparent rules to guide decision making. Clear early interventions in slow-onset disasters such rules and triggers tied to pre-agreed actions will as droughts cost a fraction of emergency aid after promote decisive, timely action and limit the a famine develops (Clarke and Hill 2013). number of decisions that stakeholders must make when a disaster strikes. Experience shows that seemingly minor improvements in the way disasters are planned for The data driving these decisions need to be and financed have significant positive impacts on resistant to manipulation and strike the right people’s lives. Below are just a few examples that balance among cost, speed, and accuracy. highlight the benefits of a coordinated, pre-agreed Any data that could trigger action will depend approach to disaster response backed by pre- on pre-disaster investments in design of the planned financing. These examples suggest how data collection system, and in the human and such an approach helps national governments technological capacity to implement the system. manage the impact of disasters on their budget, Three types of data are particularly useful for while also reducing the negative consequences of triggering post-disaster action: individual damage disasters for individuals. and losses (affecting people and buildings), area average index data on damage and losses, and SOVEREIGN CATASTROPHE RISK POOLS 17 In Kenya, donors, nongovernmental organizations In Ethiopia, a major drought in 2011 did not (NGOs), the World Bank, and the government result in major loss of life (unlike in Somalia). are working together on the Hunger Safety Net One important reason was that the government, Programme, which uses a system based on with donor support, had set up the Productive insurance principles to provide pastoralists Safety Net Programme, which was designed to be with additional cash transfers in the event of scaled up to absorb more funding and reach more a shock. Such approaches are proving useful people during a crisis. In 2011 it expanded to in (re)designing disaster response programs, support 9.6 million people. By providing support even in cases that do not involve commercial to households at the early signs of droughts, it insurance contracts. The Kenya scheme has allows them to maintain consumption, reduce simple triggers and rules about when, where, malnutrition, and keep children in school. By and how additional support will be provided to ensuring that beneficiaries receive financial vulnerable pastoralists. It pays a cash transfer to support before the lean season begins, the a predefined group when the rains fail and the program is estimated to reduce drought impacts harvest is bad, so that pastoralists can afford to by as much as 25 percent. buy inputs and food for their families—that is, so they can invest in their cows, goats, and camels GROWING INTEREST without worry that the next drought will ruin them. Donors and government cofinance the program by Recognizing the benefits of this proactive drawing on specific contingency funds, and costly approach to managing the financial impacts need assessment and delays are avoided. of disasters, a growing number of donors, development partners, and international financial In Mexico, the country’s Natural Disaster Fund institutions are supporting financial protection (FONDEN, or Fondo de Desastres Naturales), solutions in developing countries. operates a rules-based system to reconstruct public infrastructure such as roads, hospitals, and In addition, financial protection has become an schools after a disaster hits.2 In this collaboration increasingly important topic in high-level global among the federal government, state governments, policy initiatives. This attention is serving to and the private sector, all participants have increase awareness of the agenda and is driving agreed to an objective procedure to determine the political commitment, investments, and new degree of damage, which is implemented by an partnerships. The World Bank supports many independent third party and audited by all parties. of these initiatives as a neutral broker, bringing The result is clarity before a disaster about who together stakeholders to invest in technical advice will pay for what. FONDEN also offers incentives and knowledge that support the implementation for investments in risk reduction by rewarding of policy reforms and financial instruments. them. Financial markets are used to lock in this rules-based approach. By facilitating faster A number of examples show how the post-2015 reconstruction of infrastructure assets, FONDEN development agenda has embraced disaster has contributed to increasing post-disaster local risk management and financial protection as key economic activity by 2–4 percent on average (De elements for building resilience and securing Janvry, del Valle, and Sadoulet, 2016). development gains: In India, the government and farmers share the ƒƒ The Sendai Framework for Disaster Risk cost of crop insurance. This approach allows Reduction, adopted by UN member states in cheaper input credit because the banks can now 2015, guides global efforts to prevent new trust that farmers will be able to repay even if their and reduce existing disaster risk through harvests fail. Meanwhile, farmers are protected 2030 and highlights financial protection as a and able to invest more in their farms. key element of resilience. 2. FONDEN also covers low-income housing. It provides building materials that amount to US$250 for minor damage, US$1,200 for partial damage, and US$6,000 for total damage. 18 SOVEREIGN CATASTROPHE RISK POOLS ƒƒ The Addis Ababa Action Agenda, adopted ƒƒ The Association of South East Asia Nations in July 2015, lays out the level of ambition (ASEAN) and the ASEAN+3 (Japan, Republic for financing the Sustainable Development of Korea, and China) have discussed Goals (SDGs), which replaced the Millennium opportunities for regional disaster risk Development Goals in September 2015. financing for several years. In 2015, ASEAN Climate and disaster resilience are announced the establishment of the ASEAN mainstreamed across the SDGs and their Disaster Risk Financing and Insurance associated targets, ensuring that global Programme, which aims to further explore development priorities over the next 15 national and regional disaster risk financing years will integrate climate and disaster risk solutions. management considerations. GOING BEYOND CLIMATE AND ƒƒ The benefits of financial protection as part of DISASTER RISKS: A COMPREHENSIVE the agenda of the G20 Mexican presidency APPROACH TO RISK MANAGEMENT were first discussed by the G20 finance AND CRISIS RESPONSE ministers in 2012. Most recently, the 2015 G7 German presidency sponsored InsuResilience, Financial protection against climate and disaster an initiative to expand climate risk insurance risks is increasingly seen as a model with wider coverage to an additional 400 million poor relevance, one that can be adapted to, and offer and vulnerable people in developing nations lessons for, efforts to manage the financial by 2020. impacts of other shocks and crises, including pandemics and crises related to fragility and ƒƒ The 2015 Asia-Pacific Economic Cooperation forced displacement. (APEC) ministerial statement highlighted innovative and regional disaster risk financing The Pandemic Emergency Financing Facility (PEF), and insurance mechanisms as key tools for example, is a global financing facility designed to increase financial resilience as part of to channel funds swiftly to governments, broader disaster and fiscal risk management multilateral agencies, NGOs, and others seeking frameworks. The statement emphasized the to respond to dangerous epidemic outbreaks need to explore (with World Bank support) before they turn into pandemics. Developed by the possibility of developing regional risk the World Bank in partnership with the World pools and other risk financing mechanisms Health Organization, the PEF will include both an for interested APEC economies. A dedicated insurance component and a cash component to APEC Working Group on Regional Disaster Risk provide more flexible funding. Financing Solutions has been established to take this forward. Risk management and crisis response are becoming key elements of an approach to ƒƒ The Vulnerable Twenty (V20) ministers of development focused on global public goods. finance, representing a group of countries that Given the rapid expansion of the financial solutions share a vulnerability to climate and disaster available to governments it is necessary to risks, is also exploring joint risk financing package available instruments and mechanisms solutions—including regional options for in a comprehensive and coherent offering that cuts Asia—to protect against the financial shocks across sectors and focuses on helping countries of climate risks. manage the full range of risks they face. In this spirit, the World Bank Group recently announced its Global Crisis Response Platform (GCRP), an umbrella to organize existing crisis management tools, so as to improve effectiveness, strengthen complementarity, and fill gaps . SOVEREIGN CATASTROPHE RISK POOLS 19 1.4. FINANCIAL PROTECTION: FROM POST-DISASTER EMERGENCY BORROWING TO PROACTIVE RISK MANAGEMENT When disaster strikes, governments often face fiscal impacts, governments can adopt a acute financing needs. If national resources are strategic approach built on pre-planned financing limited and financing options were not agreed mechanisms. This approach to financial protection upon and set up in advance, governments are left complements other elements of a comprehensive with a limited set of options in the aftermath of disaster risk management strategy, ranging a disaster: from investments in risk reduction to improved preparedness and resilient reconstruction. It ƒƒ They can appeal for international assistance works by helping governments proactively manage and wait for it to materialize, which takes the residual risk that cannot be fully mitigated time and implies significant uncertainty, as (because doing so is not feasible or not cost commitments do not always match emergency effective). appeals. Financial protection involves planning ahead— ƒƒ They can use budget reallocation to divert before the disaster actually happens—and funds from other government programs, setting resources aside to finance disaster with potential negative impacts on long-term response activities. Governments can choose development programs. Since opportunity from a menu of financial instruments and costs of post-disaster budget reallocations can mechanisms that help address different risks be high, particularly if funds are not replaced (ranging from recurrent to more rare events) and and reprogrammed, this approach has clear different funding needs (ranging from short-term disadvantages. In addition, the process often emergency relief to recovery and reconstruction). takes time, as it requires an analysis of Based on experiences over the past 15 years available resources as well as a certain level across more than 60 countries, the menu of ex of consensus across affected line ministries. ante financing options available to governments Nevertheless, this is a fairly standard way of includes the following: accessing the immediate liquidity needed to finance early response, especially following ƒƒ In many countries, contingency/reserve funds small disasters. are used to finance relief, rehabilitation, reconstruction, and prevention activities ƒƒ They can issue debt, though this is not always for national emergencies. Sovereign funds an option or may be very costly. For countries specifically dedicated to disaster response where debt sustainability is a concern, or if exist in Colombia, Costa Rica, India, markets are not liquid and/or costs are high, Indonesia, the Lao People’s Democratic this approach can be prohibitively expensive. Republic, the Marshall Islands, Mexico, the Unexpected borrowing could also derail debt Philippines, and Vietnam, among others. In management strategies and/or progress the Philippines, the National Disaster Risk toward debt
targets. Reduction and Management Fund finances a range of disaster-related expenditures, but To better manage the cost of disasters, ensure it is not able to disburse rapidly in response predictable and timely access to much-needed to a crisis. For that reason, the government resources, and ultimately mitigate long-term created the Quick Response Fund for 20 SOVEREIGN CATASTROPHE RISK POOLS BOX 3. A COMPREHENSIVE DISASTER RISK MANAGEMENT FRAMEWORK Financial protection is an integral PILLAR 1 Risk assessment and part of a comprehensive disaster risk RISK IDENTIFICATION risk communication management framework. Structural and nonstructural PILLAR 2 To sustainably reduce the impact of measures—infrastructure, RISK REDUCTION land-use planning, regulations disasters on people, livelihoods and national budgets governments should PILLAR 3 always consider ways to identify and Early warning systems, PREPAREDNESS contingency planning reduce the underlying drivers of risk. Financial protection complements risk reduction by helping governments Assessing and reducing PILLAR 4 contingent liabilities, financial address risks that cannot be mitigated FINANCIAL PROTECTION planning for disaster response (residual risks). It helps shift the paradigm of risk management towards PILLAR 5 Resilient recovery and a more proactive approach focused RESILIENT RECOVERY reconstruction policies on planning financial responses in advance, rather than relying on fund- raising efforts after disasters. emergency response. In Mexico, FONDEN was The World Bank’s contingent instrument for created as a budgetary tool to rapidly allocate natural disasters, the Development Policy federal funds for emergency response and Loan with Catastrophe Deferred Draw-Down rehabilitation of public infrastructure affected Option (CAT-DDO), allows countries eligible by disasters.3 to borrow from the International Bank for Reconstruction and Development to secure A number of other countries are working to immediate access to budget support of up establish similar funds. In Kenya, for example, to US$500 million, or 0.25 percent of GDP the government is in the final stages of (whichever is lower), following declaration of operationalizing a national contingency fund a national emergency. Since the introduction dedicated to drought emergencies. of the instrument in 2008, CAT-DDOs have been used in 10 countries for an aggregate ƒƒ Contingent loans are financial instruments amount of US$2.3 billion. These loans also designed to give countries access to liquidity provide a platform for policy reform and immediately following an exogenous shock, strengthening of national risk management such as a terms-of-trade shock, financial capacity. Going forward, the CAT-DDO will be shock, or natural disaster. They are typically adapted to address pandemic risks and will offered by multilateral development banks and be made available to low-income countries international financial institutions (including eligible for International Development the World Bank, the Asian Development Bank, Association (IDA) financing. the Inter-American Development Bank, and the International Monetary Fund). ƒƒ Market-based risk transfer solutions are used in every sector of the economy and have 3. FONDEN is complemented by FOPREDEN (Fondo para la Prevención de Desastres Naturales), which finances prevention activities (mainly studies) for subnational governments. SOVEREIGN CATASTROPHE RISK POOLS 21 growing relevance in development due to drought, tropical cyclone wind speeds, etc. Box increased exposure to risks that result in 4 has more detail on index-based products. economic loss. A broad menu of underlying instruments—derivative contracts, insurance ƒƒ Catastrophe risk pools in particular are contracts, and catastrophe bonds (CAT emerging as a promising vehicle to help bonds)—can be used to transfer the risk of countries access cost-effective risk transfer specific meteorological or geological events solutions. They facilitate (i) diversifying risk (droughts, hurricanes, earthquakes, and across multiple countries with different risk floods) to actors in the market (insurance profiles; (ii) establishing joint reserves (joint companies, reinsurance companies, banks, surplus capital) to self-insure a part of the risk; and investors) who are willing to accept them. (iii) transferring excess risk to the reinsurance These market-based risk transfer products use and capital markets; (iv) sharing operational scientific information and actuarial modeling costs, such as program development and day- to estimate losses that would be sustained to-day back-office operations; and (v) building due to a specific event and to “price” the risk. up a better foundation of risk information. Examples of sovereign/supranational pools Disaster risk transfer solutions can rely on a include CCRIF, PCRAFI, and ARC. Part 2 parametric trigger, where payments are triggered presents a detailed discussion of the role of by the performance of a prespecified index sovereign catastrophe risk pools in the financial such as levels of rainfall, length and intensity of management of climate and disaster risks. BOX 4. INDEX INSURANCE: A FEW DEFINITIONS Unlike traditional indemnity-based insurance payouts, lower administrative costs, and reduced products that require the assessment of individual moral hazard and adverse selection. For example, losses following an insurable event, index-based at the micro level it allows domestic insurance (including parametric) insurance policies make companies to offer farmers simple and transparent payouts based on a predetermined trigger, such solutions for transferring weather risks such as as crop yield estimates, in a given geographical drought, excess rainfall, or low temperatures. area. Other triggers could be based on the location or intensity of natural hazards such as But index insurance is not without its challenges, wind speed, rainfall levels, or ground acceleration notably basis risk. Implicit in all index insurance, this from earthquakes. is the risk that the index measurement will not match individual losses. For example, an insured individual The particular index used can be tailored to the or asset may experience a loss from a disaster availability of data: when only hazard data are that is not fully captured by the parametric index, available, a parametric index (which pays out and hence does not lead to a payout. Alternatively, on a given hazard event) can be used; but when a payout could be triggered without any damage exposure data are available, it is more appropriate and losses incurred. Improved accuracy of hazard to use a modeled loss index (which pays out in line data collection systems, increased openness and with loss modeled using actual exposure data and centralization of historical data, and better-quality parameters such as wind speed from the actual risk assessments could reduce basis risks, enabling event). Parametric coverage demands improved a more efficient and effective use of parametric accuracy of hazard risk data collection systems insurance. Capacity building and education are key to because of the heavy reliance on objective ensure understanding of basis risk. For governments measurement of weather and hazard parameters. considering parametric insurance, it is crucial to Index insurance offers several advantages over carry out a cost-benefit analysis of different potential traditional or indemnity insurance, such as quicker indexes with different levels of basis risk. 22 SOVEREIGN CATASTROPHE RISK POOLS Beyond sovereign catastrophe risk pools that can assist governments in accessing rapid (but limited) post-disaster financing, governments can also promote the development of domestic catastrophe risk pools to transfer disaster risks from households and small and medium enterprises to the private insurance and reinsurance markets. Since the establishment of the Turkish Catastrophe Insurance Pool (TCIP) in 1999 (see Box 5), very few developing and emerging countries have implemented such pools. (Indonesia and Romania are among the few exceptions in emerging economies.) The role of the government is essential in such endeavors, which must align incentives and generate a sustainable demand for catastrophe risk insurance, especially where insurance literacy and penetration are low. BOX 5. TURKISH CATASTROPHE INSURANCE POOL With a majority of the population living in 2. Limit the government’s financial exposure to earthquake-prone areas, the persistent potential natural disasters for large-scale natural disasters has become a 3. Build long-term catastrophe reserves to fiscal and social issue for the Turkish government. finance future earthquake losses These pressures led to the establishment of the 4. Encourage risk reduction and mitigation Turkish Catastrophe Insurance Pool in 1999. Aside practices in residential construction from fiscal exposure, the main rationale for the creation of TCIP was a very low level of catastrophe The program provides earthquake insurance insurance penetration among households. TCIP coverage to approximately 4.5 million Turkish was established as a public sector insurance entity homeowners (corresponding to approximately 30 providing catastrophe risk insurance for Turkish percent of the insurable housing stock). homeowners. A genuine public-private partnership, TCIP has no public employees. All of its business The creation of TCIP has greatly increased public functions—from sales to reinsurance to claims awareness of earthquake risk due to a wide, management—are subcontracted to the private ongoing public information campaign dedicated to insurance industry. The government’s role is limited earthquake insurance. For example, the concept of to (i) enforcing compulsory earthquake insurance earthquake risk management and insurance has for all urban dwellings and (ii) providing contingent recently been introduced in school textbooks. The liquidity support in the remote event that TCIP’s program also provides incentives for local builders financial capacity is insufficient to pay all claims in to comply with the construction code because full. The program has four principal objectives: TCIP does not provide insurance coverage for any buildings that do not carry valid construction and 1. Provide compulsory earthquake insurance occupancy permits. coverage at affordable but actuarially sound rates for all registered urban dwellings SOVEREIGN CATASTROPHE RISK POOLS 23 1.5. SELECTING THE APPROPRIATE FINANCIAL PROTECTION INSTRUMENTS Not all instruments serve the same purpose, cover against severe events, but using it to and governments can take a strategic approach protect against low-intensity and recurring to financial protection by combining instruments events may be inefficient and costly. For such with different characteristics. disasters, a dedicated contingency fund that retains this lowest layer of risk may be a more More specifically, depending on the frequency and appropriate solution (Figure 4 provides a graphic severity of risks to be managed, governments representation of this risk-layering approach). can combine (or layer) financing instruments Since climate change may over time affect a that address different needs and have different country’s risk profile by potentially increasing cost implications. Such an approach prioritizes the frequency and intensity of such hazards, cheaper sources of funding, ensuring that the the combination of financial instruments used most expensive instruments are used only to address disaster impacts will also need to in exceptional circumstances. For example, evolve to account for changes in risk and other sovereign insurance may provide cost-effective considerations beyond pure financial aspects. . FIGURE 4. SOVEREIGN DISASTER RISK LAYERING SOVEREIGN RISK TRANSFER Low frequency/ High severity • Insurance (including through risk pools) • Derivatives INSURANCE OF PUBLIC ASSETS • CAT bonds CONTINGENT FINANCING POST-CRISIS FINANCING • IBRD DPL with a CAT-DDO • Emergency lending operations • Hybrid IPF with a contingent financing component • MDB financing • IDB deferred draw-down option (DDO) • Bilateral financing • IDB Contingent Credit Facility for Natural Disasters • IMF Rapid Credit Facility (RCF) HAZARD (CCF) • IMF Rapid Financing Instrument • IMF Flexible Credit Line (RFI) CONTINGENT EMERGENCY RESPONSE COMPONENTS • IDA Immediate Response Mechanism (IRM) • Contingent emergency response components (CERCs) High frequency/ BUDGETARY INSTRUMENTS Low severity • Sovereign reserve funds • Contingency budget • Government reserves • Budget reallocation Short-term liquidity Long-term financing needs TIME Note: IBRD = International Bank for Reconstruction and Development; IDA = International Development Association; DPF = Development Policy Financing; CAT DDO = Catastrophe Draw-Down Option; IMF = International Monetary Fund; MDB = multilateral development bank. 24 SOVEREIGN CATASTROPHE RISK POOLS Combining instruments also enables governments ƒƒ Cost of capital: How cost-effective is the to take into account the evolving need for funds— instrument in accessing financial resources initially for emergency response and eventually after a disaster, either in absolute terms (i.e., for long-term reconstruction. For example, a how much does US$1 of disaster response government could purchase (exante) quick- cost?) or relative to other instruments available? disbursing risk transfer instruments to ensure immediate liquidity in the aftermath of severe ƒƒ Timeliness: Can the selected instrument events , but raise the much larger sums required make funding available at the right time? to finance reconstruction efforts through (ex post) budget reallocations or by issuing bonds. ƒƒ Discipline: How well can the instrument support post-disaster spending discipline, Policy makers considering the most effective accountability, and transparency? Does the approach to financial protection need to evaluate instrument support risk-based pricing? various potential instruments. An evaluation framework is helpful for this purpose because ƒƒ Ownership: How well can the instrument it allows public officials to assess (i) whether a clarify risk ownership? Is the entity that pays given instrument is the right tool to achieve their the cost of the instrument (e.g., premium) objectives, and (ii) how well the selected instrument also the entity that bears the risk? is implemented and ways to make it more efficient. ƒƒ Risk reduction: Does the instrument incentivize A possible framework for evaluating risk financing investments in risk reduction and preparedness? solutions draws on the key principles of effective financial protection, listed here and described ƒƒ Risk information: Can the chosen instrument further in Box 6: help countries understand and price their risk? SOVEREIGN CATASTROPHE RISK POOLS 25 BOX 6. THE PRINCIPLES OF EFFECTIVE FINANCIAL PROTECTION Effective disaster risk financing strategies are marked by several mutually reinforcing characteristics that promote disciplined, timely, and appropriately priced access to finance: Appropriate Risk Information Ownership ƒƒ Cost of capital. Access to capital (i.e., public and/or private financial resources) is necessary Discipline for effective emergency response and Financial reconstruction as well as for investment in risk Protection reduction and prevention. Yet different sources Cost of of funds come with different costs. Through Capital optimal use of instruments such as reserves, Timeliness contingent credit, risk transfer solutions, and post-disaster credit, disaster risk financing policies can secure access to disaster financing for governments, businesses, and households ƒƒ Ownership of risk. Clarity about who owns the before an event strikes and ensure timely and risk —that is, about the contingent liability cost-effective financial resources for recovery of the national government, subnational and reconstruction. governments, donors, the private sector, ƒƒ Timeliness. Different levels of funding need to and households—promotes better decision be available at the appropriate time to cover making. In the absence of clear rules relief, response, and reconstruction efforts. regarding what share of costs for response Rapid mobilization of funds to support relief and reconstruction the national government efforts is crucial to limit humanitarian impacts. will assume, businesses and households This rapid response can also save money. have little incentive to invest in risk reduction For example, well-targeted early interventions or purchase catastrophe risk insurance, and in slow-onset disasters such as drought cost delays can occur in post-disaster response a fraction of emergency aid after a famine and recovery. Clearly established rules for develops. While immediate liquidity is crucial the amount and timing of payouts under to support relief and early recovery operations, social protection programs allow vulnerable the government has more time to mobilize households to plan and budget effectively. the majority of resources for reconstruction. Businesses and households also need access ƒƒ Risk information. Lack of knowledge about to timely financing through catastrophe risk exposure to risk can lead to suboptimal insurance and/or post-disaster credit. investment decisions by public and private actors. Better information on which ƒƒ Discipline. Disaster risk financing helps populations and assets face potential disaster governments, businesses, and households impacts can help overcome behavioral biases, plan in advance of a disaster and agree ex such as the reluctance of businesses and ante on rules and processes for securing funds households to buy catastrophe risk insurance. through their budget (budget mobilization) and It may also help monitoring and evaluating the spending this money (budget execution). This effectiveness of risk finance and insurance approach fosters discipline, transparency, instruments. Finally, putting a price on risk can and accountability in post-disaster spending. help elevate decision making to the finance Discipline is also important for a government ministry level and provide further incentives for to credibly commit when it will or will not act, ex ante risk reduction and adaptation efforts. thus facilitating ownership of risk. Source: World Bank Group 2014. 26 SOVEREIGN CATASTROPHE RISK POOLS BOX 7. QUANTIFYING THE BENEFITS OF FINANCIAL PROTECTION Cost-benefit analysis provides governments with a in Ethiopia through an expansion of the Productive quantitative tool to assess (i) whether a specific Safety Net Programme (Clarke and Coll-Black et al. financial instrument would offer effective financial 2016). The analysis compared three strategies, protection against the risks faced by the country; each involving different combinations of federal (ii) how a specific instrument would complement contingency budget, emergency budget reallocation, an existing portfolio/approach; and (iii) whether insurance, and humanitarian response. the price of the proposed instrument is cost- effective compared to other options. In the case of Ethiopia, the analysis showed that financing disaster response costs through The World Bank recently published a framework a combination of budget reallocation, reserves, for evaluating the economic costs and benefits and insurance could on average reduce the of disaster risk financing instruments, which can cost of response by a quarter, as compared to be used to inform the design of an appropriate waiting for humanitarian aid to be mobilized. and cost-effective approach to the financing of The cost savings from including insurance as disaster response (Clarke and Mahul et al. 2016). part of a disaster risk financing strategy become The framework was used to analyze different risk far greater for more severe droughts, which are financing approaches to support drought response expected to occur less frequently. Evaluating an instrument against these indicators achieve its objectives through contingent helps governments take informed decisions on financing. If a government has sufficient reserves whether a given instrument serves their purpose. for short-term response but is looking for the The objectives, and hence the most appropriate cheapest possible source of longer-term financing tool, will not be the same across countries, for reconstruction, ex post borrowing may be the and each government will have to find its own most attractive option. preferred combination of sources of funding. For instance, a government that prioritizes speed Box 7 presents an approach to support decision and predictability of post-disaster funding may be making through the quantitative analysis of the willing to pay a potentially higher price for these costs and benefits of different combinations of benefits in the form of an insurance premium. On financial instruments. the other hand, a government without sufficient resources to respond even to more frequent and less severe events might more cost-effectively SOVEREIGN CATASTROPHE RISK POOLS 27 1.6 INSURANCE: A POWERFUL TOOL TO BE USED WITH CAUTION Market-based risk transfer and insurance solutions Insurance, like some other ex ante financing play an important role in the mix of financial instruments, allows for rapid response financing. instruments available. But while insurance can It secures access to immediate (but limited) be a powerful tool for risk management, it is also liquidity in the aftermath of an insured disaster an expensive one for governments that otherwise event. While there is not a clear consensus on have access to sufficient sovereign financing. the effect of delay on post-disaster costs, there is agreement that this effect occurs, and the range Insurance can effectively reduce the volatility of seems to be between 2:1 to 5:1 depending on disaster impacts on government accounts and the peril and the location. For example, evidence therefore promote budget stability. In addition, it from Ethiopia shows that every US$1 secured in provides rapid liquidity without requiring access contingency financing for timely and predictable to credit. This is a significant benefit to some disbursement for emergencies can save up to governments, particularly small states that lack US$5 over the long term.4 sufficient capacity to build reserves and are restricted in their access to credit due to already Taking these considerations into account, a high debt-to-GDP ratios. government can combine market-based risk transfer with other instruments to ensure that An important consideration when assessing risk cheaper sources of money are used first, with transfer and insurance is the comparative cost the most expensive instruments used only in of the instruments. Different sources of funding exceptional circumstances. For example, while come with different costs. insurance can provide cover against severe events, it is likely not appropriate to protect Table 1 provides an indicative cost multiplier for against low-intensity events that recur regularly. different financial risk instruments. This multiplier For such events, the government could consider is defined as the ratio between the cost of the setting up a dedicated contingency fund or access financial product (such as the premium of an pre-arranged donor resources to retain this lowest insurance product, or the expected net present layer of risk. value of the cost of a contingent debt facility) and the expected payout over its lifetime. A ratio of Part 2 of this paper considers how risk transfer 2 indicates that the overall cost of the financial can be implemented most effectively and product is likely to be twice the amount of the efficiently through catastrophe risk pools. expected payout made over a long period of time. Note that these multipliers are only indicative and aim to illustrate the cost comparison of financial products. The speed at which funds can be obtained is also determined by the legal and administrative processes that drive their use (Ghesquiere and Mahul 2010). 4. See Wiseman and Hess (2007). Cabot Venton et al. (2012) suggest even higher figures for Ethiopia and double the cost of a late response in Kenya compared with an early response. 28 SOVEREIGN CATASTROPHE RISK POOLS TABLE 2. COSTS OF DIFFERENT INSTRUMENTS FOR FINANCING POST-DISASTER EXPENDITURE Instruments Indicative Disbursement Amount of funds cost (months) potentially multiplier available Ex post financing Donor support (humanitarian relief) 0–1 1–6 Uncertain Donor support (recovery and reconstruction) 0–2 4–9 Uncertain Budget reallocations 1–2 0–9 Small Domestic credit (bond issue) 1–2 3–9 Medium External credit (e.g., emergency loans, bond 1–2 3–6 Large issue) Ex ante financing Budget contingencies 1–2 0–2 Small Reserves 1–2 0–1 Small Contingent credit 1–2 0–1 Medium Parametric insurance 1.3 and up 0–2 Large Alternative Risk Transfer (for example CAT 1.5 and up 1–6 Large bonds, weather derivatives) Traditional (indemnity-based) insurance 1.5 and up 2–12 Large Source: Based on Ghesquiere and Mahul 2010. SOVEREIGN CATASTROPHE RISK POOLS 29 PART 2. Lessons learned from experiences of sovereign catastrophe risk pools 30 SOVEREIGN CATASTROPHE RISK POOLS For governments defining cost-effective combinations of financial instruments, risk transfer solutions can be implemented in various forms, including direct access to the reinsurance and capital markets or indirect access through a dedicated vehicle such as a catastrophe risk pool. Catastrophe risk pools create a platform that allows governments to (i) take a collective and standard approach to quantitative analysis and modeling; (ii) improve information sharing; (iii) coordinate response; (iv) lower costs of coverage (through pooling of diverse exposures, retention of some risk, and transfer of excess risks to the capital and reinsurance market); and (v) strengthen regional/ subregional cooperation and policy dialogue. In the Caribbean, for example, the World Bank Group has supported 16 countries that decided to join efforts through the Caribbean Catastrophe Risk Insurance Facility, which pools hurricane, earthquake, and excess rainfall risks. The purpose of this regional approach was to enable countries to access rapid (but limited) response financing —with payouts disbursed within two or three weeks after a disaster. The sovereign mechanism CCRIF set up in 2007 in the Caribbean was the first of its kind. Since then, the concept has been replicated in two other regions, in the Pacific Risk pools also entail certain risks and costs that through the PCRAFI Facility, and in Africa through should not be underestimated, including the time the ARC. The government of Mexico has also and (technical and political) effort to establish drawn on some of the principles of catastrophe such pools, the cost of insurance (premium), risk pools in its subnational disaster response and the political risk of an insurance policy not pool, which relies on a standardized approach to triggering a payout after several years. disaster risk modeling and assessment. Globally, a number of countries and their development Box 8 illustrates the potential benefits of risk partners are exploring how risk pools, or similar pooling with countries from Latin America and the financing structures, could help them manage the Caribbean, East Asia and Pacific, South Asia, and financial impacts of disasters. Europe and Central Asia. The risk-pooling benefit is captured here as a reduction of the capital There are a number of reasons why national requirements to sustain a 1-in-200-year loss, and subnational governments may seek to that is, the amount of capital that should be held collaborate and pool their catastrophe risks. The to sustain a loss with an annual probability of benefit of diversification of losses (and ultimately occurrence of 0.5 percent. This parameter directly its tendency to lower premiums) is the principal influences the price of insurance (premium): the driver. Diversification of losses through pooling more (less) capital a pool needs to hold, the of risk is the underlying basis of the global more (less) expensive the premium is. Using data (re)insurance markets. Other drivers include provided by the risk modeling firm AIR Worldwide, economies of scale (in shared administrative the analysis shows that the 1-in-200-year level costs and larger transactions with the of loss in the aggregate portfolio is about 45 reinsurance markets) and political cooperation percent lower than the sum of the 1-in-200-year and solidarity in managing disasters. level of regional losses. SOVEREIGN CATASTROPHE RISK POOLS 31 BOX 8. INSURING THE WORLD: RISK POOLING AND REDUCTION IN CAPITAL REQUIREMENTS FOR LOW- AND MIDDLE-INCOME COUNTRIES An analysis was undertaken to illustrate the ƒƒ Separate regional values for low- and middle- hypothetical impacts of pooling low- and middle- income countries in Europe and Central Asia, income countries’ economic losses on a global East Asia and the Pacific, South Asia, and scale. The benefit of risk pooling manifests as Latin America and the Caribbean.a (data for a reduction in the amount of capital required to Africa were not available for this analysis) manage the occurrence of severe but infrequent ƒƒ Values for selected regional collaborations loss events. For the purposes of illustration, a 1-in-200-year return period level of loss (standing The figure below shows the difference in capital for the level of a severe but infrequent loss that requirements for pooled risks versus nonpooled would need to be managed) is examined as a risks. This reduction in capital requirement simple aggregate (i.e., countries managing severe generated by risk pooling leads to lower premiums. losses separately) versus a pooled risk (i.e., a common fund for financing severe losses). The following scenarios are considered: 1400 1200 1000 US$ Billions 800 600 400 200 0 LAC + ECA LAC + ECA LAC + EAP LAC + EAP EAP + SA EAP + SA LAC + EAP LAC + EAP (SUMMED (POOLED) (SUMMED (POOLED) (SUMMED (POOLED) + SA + ECA + SA + ECA NOT NOT POOLED) NOT (SUMMED (POOLED) POOLED) POOLED) NOT POOLED) South Asia (SA) Low and Europe and Central Asia (ECA) Middle Income Countries (LMICs) Low and Middle Income Countries (LMICs) Latin America and the Caribbean (LAC) East Asia and the Paci c (EAP) Low and Middle Income Countries (LMICs) Low and Middle Income Countries (LMICs) Source: World Bank based on results provided by AIR Worldwide. Note: ECA = Europe and Central Asia; EAP = East Asia and the Pacific: LAC = Latin America and the Caribbean; SA = South Asia. All losses presented here represent AIR Worldwide’s upper-bound estimates of economic losses for low- and middle-income countries on an aggregate basis. The income categories and regional country classifications adopted are as defined by the World Bank (see note a). While AIR models losses in over 90 countries, assumptions were made regarding losses from perils and countries that AIR does not currently model; assumptions were also made concerning the nature of the scaling factor between insured/insurable and economic losses. These assumptions are based on AIR’s current understanding of historical losses and relative hazard levels. Numbers presented should therefore be interpreted as estimates. a. World Bank income classifications were used, with high-income countries removed from the analysis. Where data were not available for in-scope countries, AIR Worldwide used an extrapolation methodology to make estimates. See World Bank, “World Bank Country and Lending Groups,” https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups. 32 SOVEREIGN CATASTROPHE RISK POOLS 2.1. GLOBAL LANDSCAPE OF SOVEREIGN CATASTROPHE RISK POOLS FOR DEVELOPING COUNTRIES Sovereign catastrophe risk pools can provide a PCRAFI began facilitating insurance transactions mechanism for governments to access rapid for a number of Pacific island states starting in liquidity post-disaster in a cost-efficient and 2013, and in November 2016 issued the first mutually supportive way. They enhance financial policies from the newly established catastrophe preparedness against climate and disaster risk pool, the PCRAFI Facility. risks by (i) pooling risks into one single, more diversified, less risky portfolio; (ii) retaining some As of December 2016, the three programs risks through joint reserves/capital; and (iii) provide an aggregate coverage limit of US$870 accessing the reinsurance and capital markets million (80 percent from CCRIF-Caribbean). While when it is the most cost-effective. However, risk these existing sovereign risk schemes share pools do require significant technical and political many common features, they are different in support and time to be established, and even detail. Table 2 provides a comparative matrix longer time to become sustainable. summarizing the schemes’ key features as well as a detailed description of their evolution and Three sovereign catastrophe risk pools currently current status. Further information on each exist, covering 26 countries in three regions— program and its respective annual portfolio is the Caribbean and Latin America, Africa, and provided in annex 1 and annex 2.5 the Pacific. All are heavily supported by donor partners. Pool members represent the vast majority of developing countries that have purchased sovereign catastrophe risk insurance. Opportunities for one or more supranational pools may arise in Asia, and large countries across the world are exploring national level solutions, following the example of Mexico. EXISTING SOVEREIGN CATASTROPHE RISK POOLS The three existing sovereign catastrophe risk pools could potentially cover more than 50 countries in Africa, the Pacific, and the Caribbean and Central America, for multiple perils. Many current members are among the world’s most vulnerable nations (see Figure 5). Since its launch in 2007, CCRIF has been restructured as a segregated portfolio company to be able to expanded to gradually include countries of Central America (as of December 2016 only Nicaragua has joined). ARC was launched by the African Union in 2012 and issued its first insurance contracts in 2014. 5. Annex 2 also includes a description of Mexico’s fund for natural disasters FONDEN. SOVEREIGN CATASTROPHE RISK POOLS 33 FIGURE 5. MAP OF COUNTRIES PARTICIPATING IN A SOVEREIGN CATASTROPHE RISK POOL (AS OF DECEMBER 2016) CCRIF ARC PCRAFI Marshall Islands Anguilla Dominica St. Kitts Burkina Faso Niger & Nevis Samoa Mali Senegal Antigua Grenada Saint Lucia & Barbuda Tonga Mauritania The Gambia Barbados Haiti Trinidad & Tobago Cook Islands Belize Jamaica Turks & Vanuatu Caicos Islands Cayman St. Vincent Nicaragua Islands & the Grenadines 34 SOVEREIGN CATASTROPHE RISK POOLS TABLE 3. DETAILED OVERVIEW OF EXISTING REGIONAL SOVEREIGN CATASTROPHE RISK POOLS (AS OF DECEMBER 2016) Scheme CCRIF (Caribbean) CCRIF (Central ARC PCRAFI America) Form of Modeled loss Modelled loss Modelled loss Modelled loss insurance parametric parametric parametric parametric Perils Earthquake, Earthquake, Drought, Tropical Earthquake (ground covered tropical cyclone Tropical Cyclone Cyclone (wind + shaking + tsunami), (wind + surge), (wind + surge), surge), Flood under Tropical cyclone extreme rainfall Extreme Rainfall development (wind + surge) Modeling EQ/TC - built for EQ/TC - built for In-house (license AIR Worldwide and licensed by and licensed by owned by ARC model CCRIF, available CCRIF, available Agency), TC and FL to participants for to participants for will use licensed noncommercial non-commercial feed for hazard use. XSR - in-house use. XSR - in-house data Date of first 2007 2015 2014 2013 policies Initial Multi-donor grants Multi-donor grants Development Multi-donor grant capitalisation via World Bank via World Bank capital (interest- via WB free loan) from 2 partners Ownership Purpose trust CCRIF Purpose Mutual insurance Foundation trust company formed at direction of ARC Conference of the Parties Operational Segregated Cell in CCRIF SPC Class 2 captive Captive insurance entity portfolio company, insurer company multiple cells Domicile Cayman Islands Cayman Islands Bermuda Cook Islands Governance Board of 5 Management Board of 7 Board of 5 directors, 2 Committee for CA Directors, Directors appointed appointed by cell, under CCRIF appointed by the by Council of Caribbean SPC board members Members Development Bank, 2 by CARICOM, and 1 by other 4 Directors Operational CEO and COO on Operated by CCRIF CEO and small Operated by Pacific staffing staff, remainder SPC technical/ Catastrophe outsourced to operations support Risk Insurance service providers team, remainder Company (PCRIC). outsourced to CEO, remainder service providers outsourced to service providers SOVEREIGN CATASTROPHE RISK POOLS 35 Scheme CCRIF (Caribbean) CCRIF (Central ARC PCRAFI America) Number of 20 are eligible for 6 are eligible, 1 32 have signed 15 are eligible, participants coverage, 16 have has purchased a ARC Treaty, 8 have 6 countries have participated, 14 policy participated, 6 are participated, 5 purchased policies participating in countries are in 2016 2016/17 participating in 2016/17 Average US$20 million US$1.5 million US$22 million US$2 million premium income 2016 US$27.7 million US$1.5 million US$25 million US$2.3 million premium (2015/16 policy income year) Cumulative US$67.3 million US$0.7 million US$34 million (to US$3.2 million payouts close of 2015/16 since policy year) inception Average US$622 million US$28 million US$150 million US$45 million aggregate coverage 2016/17 US$697 million US$28 million US$100 million US$45 million aggregate (2016/17 policy coverage year) limit Source of Initial IDA credits IDA credit for sole National budgets, Grants (first 3 premiums for 4 countries for current participant grants (1 country) years), national 3.5 years premium. (3-5 years budget, IDA credits CDB credits for 0.5 premium) years premium for 8 countries, full grant of premium each year for 1 country Payout Initial estimate in Initial estimate in Payout calculated Payouts made process 3-5 days, payout 3-5 days, payout within 10 days within 10 business made after 14 days made after 14 days of end of risk days. (partial payouts (partial payouts period (for have been made have been made drought), 7 days sooner). Self- sooner). Self- for TC/FL. Self- certification of loss certification of loss certification of loss required. required. required. Certified contingency plan also required before payout is made. 36 SOVEREIGN CATASTROPHE RISK POOLS Scheme CCRIF (Caribbean) CCRIF (Central ARC PCRAFI America) Reinsurance Panel of traditional Traditional Traditional Panel of 5 summary reinsurers and reinsurers, reinsurance reinsurers capital market separate agreement with 24 element, most placement for CA participants, multi- recently via World cell peril Bank CAT Bond Portion of 25% 66% 41% 90% (to decrease Agg. Limit significantly once reinsured the facility is fully (2016/17) capitalized in 2017) Capital/ US$117 million US$1.3 million US$ 98.5 million US$6 million (to Reserves (Statutory Capital & increase to US$25 (2016) Surplus as of end- million in 2017) 2015) Associated Two products n.a. Licensing for n.a. meso co-developed by Development or micro CCRIF, one meso initiative allows schemes (inactive) and one for use of ARC micro (active), both model to underpin utilize CCRIF model commercial to some extent, no transactions. risk taken by CCRIF Revenue to L4D to date Trust to support ARC, ARC Ltd. could take some risk Note: EQ = earthquake; TC = tropical cyclone; XSR = excess rainfall; FL = flood; L4D = Licensing for Development. SOVEREIGN CATASTROPHE RISK POOLS 37 BOX 9. TOWARD A REGIONAL APPROACH FOR DISASTER RISK FINANCE IN ASIA According to a review undertaken by the World of risk pooling. A model similar to that of CCRIF, Bank in partnership with the Rockefeller where countries enter into an insurance contract Foundation, a regional approach to disaster with the facility and pay a premium for access to risk financing could improve Asian countries’ rapid liquidity as post-disaster bridge financing, resilience if structured to accommodate the could be considered. particular conditions of the region. Asia presents particular challenges of significant heterogeneity The risk transfer platform could function as a in peril exposure, and also in both the geographic clearinghouse for transferring sovereign disaster and economic size of countries. A proposal for risk in Asia to the international markets. It would a platform that could accommodate a range of allow large economies to approach the market country priorities is shown below. Such a platform directly and smaller economies to approach the could make floods a priority, given the prevalence market as a group, as the Pacific countries have of this peril across the region; but it should also done. Standardized contracts could be used offer solutions to protect against less frequent as well as a standardized process for readying but more severe shocks, such as earthquakes countries for transacting. An approach for and tropical cyclones. In addition, it would need to collectives of subnational entities (such as cities) serve countries focused on livelihoods assistance to this platform could also be considered. as well as those focused on reconstruction of homes and infrastructure. It should build on the The technical assistance facility would be the extensive national level work that has already home of public goods such as catastrophe risk been undertaken on financial management of models that would support the above components. disasters across the region. It could also assist countries with their national strategies for financial protection, and specifically The joint disaster insurance fund would be best with mechanisms for disbursing funds in country to suited for smaller economies, with uncorrelated better reach affected households and businesses. but similar risk, looking to gain from the benefits A REGIONAL PLATFORM FOR DISASTER RISK FINANCING FOR ASIAN COUNTRIES INTERNATIONAL (RE)INSURANCE/CAPITAL MARKETS Platform for Risk Transfer to International Markets Technical Assistance Joint Disaster Insurance Fund Facility National Disaster Risk Finance Strategy Asian Countries by Increasing GDP Source: World Bank and Rockefeller Foundation 2016. 38 SOVEREIGN CATASTROPHE RISK POOLS NEW INITIATIVES FOR SOVEREIGN forward, assessed, and ultimately implemented CATASTROPHE RISK POOLS by participating countries. Preliminary technical conversations on the subject of sovereign Asia: With its high exposure to flood, tropical catastrophe risk pooling are also underway for cyclone, earthquake, and drought events, Asia Cambodia, Lao PDR, and Myanmar (Box 10). remains perhaps the most conspicuous coverage India, Pakistan, and the Philippines are exploring gap with regard to regional sovereign disaster subnational structures for disaster risk financing, risk financing programs. Asian countries face with advanced technical work underway in the particular challenges in establishing an efficient Philippines on the possibility of creating a pooling regional sovereign disaster risk financing pool due mechanism for Local Government Units (Box 10). to the heterogeneity of perils they confront and their vastly different geographic and economic Latin America: Chile, Colombia, Mexico, and sizes (World Bank and Rockefeller Foundation Peru are exploring a multicountry (parametric) 2016). See Box 9. catastrophe bond against earthquake under the Pacific Alliance. Still, although no Asian facility has been established to date, some activity has occurred at Subnational catastrophe risk pools: Another the national level, and a number of Asian countries significant gap is the establishment of subnational are already individually using ex ante disaster risk structures for pooling and managing the financial financing mechanisms to manage the impacts impacts of disaster risk. As the FONDEN case of disasters. In addition, regional platforms for study shows (See Annex 1), such structures can collaboration on disaster risk financing more add significantly to the financial resilience of generally now exist. In 2015, the Association countries where subnational government entities of Southeast Asian Nations announced the (such as provinces or states) have substantial establishment of the ASEAN Disaster Risk power and responsibility in the financial response Financing and Insurance Programme, a regional to disasters. India, Pakistan, and the Philippines platform through which options for managing are exploring subnational structures for disaster the financial impacts of disaster could be put risk financing. SOVEREIGN CATASTROPHE RISK POOLS 39 BOX 10. EXPLORATORY WORK ON NEW CATASTROPHE RISK POOLS Although no regional risk pool currently exists flood loss estimation as the basis for a potential for Asia, technical work is already underway to future rapid-response financing instrument. The explore the feasibility of risk-pooling solutions at work will comprise the development of hazard, the sovereign level for Cambodia, Lao PDR, and loss, and exposure data collection and analysis. Myanmar, and at the subnational level for the Philippines. The development of a subnational risk pool for Local Government Units in the Philippines is For Cambodia, Lao PDR, and Myanmar, technical further advanced; international market-standard assistance work is underway with the World Bank, catastrophe risk models for tropical cyclone with financial support from the Global Facility for and earthquake are complete and ready to form Disaster Reduction and Recovery (GFDRR) and the basis of a modeled loss trigger. This work the government of Japan. The goal is to develop is led by the Philippine Department of Finance options for a catastrophe risk pool, especially and the public insurance company GSIS, with for floods. While rainfall-induced flooding has technical support from the World Bank and been included as part of parametric tropical financial assistance from the UK Department cyclone triggers (e.g., by PCRAFI) and as part of for International Development (DFID) through a standalone excess-rainfall product (by CCRIF), GFDRR. The development work for the parametric capturing flood impacts using the indirect hazard insurance contracts themselves is also complete, measure of excess rainfall has its limitations. with the intent to transform a bundle of parametric The work for Cambodia, Lao PDR, and Myanmar insurance contracts at the Local Government seeks to improve on existing parametric triggers Unit level into a derivative product passing 100 and use innovative methods for near-real-time percent of the risk onto the international markets. 2.2. EFFICIENCY OF CATASTROPHE RISK POOLS To assess the efficiency of implementation and Financial efficiency and operational efficiency aim operations of a catastrophe risk pool, some key ultimately to reduce the cost of insurance (that is, indicators can be considered: the insurance premium in excess of the average annual loss paid) while ensuring the financial ƒƒ Political ownership: Do risk pools support viability of the catastrophe risk pool. Financial political ownership by one country or efficiency helps reduce the cost of capital, and collaboration by multiple countries for operational efficiency helps reduce the operating increased attention to building resilience? costs (i.e., the cost of establishing and running the pool) and the uncertainly loading (i.e., the ƒƒ Financial efficiency: How financially efficient is quantity and quality of risk information available it to transfer risk jointly through a catastrophe to design and price an insurance product). It is risk pool as compared to individually? How important to note that catastrophe risk pools financially efficient is it to retain some risk cannot reduce the underlying risk measured by through joint reserves? the annual expected loss; only risk mitigation measures can reduce the annual expected loss. ƒƒ Operational efficiency: How efficient is See Figure 6. the operation of a catastrophe risk pool as compared to individual risk transfer? 40 SOVEREIGN CATASTROPHE RISK POOLS FIGURE 6. CATASTROPHE RISK INSURANCE PREMIUM DECOMPOSITION WITHOUT risk pooling WEAK risk information COST OF INFORMATION Standardized AFTER risk pooling information systems IMPROVED risk information COST OF INFORMATION COST OF CAPITAL (reserves and cost of risk transfer) Diversified portfolio structure Joint reserves retain first losses COST OF CAPITAL (reserves and cost of risk transfer) OPERATING COSTS Economies of scale benefits OPERATING COSTS Underlying risk ANNUAL EXPECTED LOSS is unchanged ANNUAL EXPECTED LOSS Source: World Bank Disaster Risk Financing and Insurance Program (2015). Note: The size of the boxes is not meant to be proportional to each component’s actual contribution to the insurance premium. POLITICAL OWNERSHIP has not triggered any claim for several years. In low-income countries where public resources Ownership and discipline are scarce, these challenges can be even more Catastrophe risk pools allow sovereign states problematic. Multiyear insurance contracts and to work together for their mutual benefit before premium discounts (e.g., in case of no claims) and after disasters, and demonstrate solidarity may help mitigate this problem. through a fair and transparent mechanism. However, sovereign states can face financial The payment of an insurance premium is often and political costs for participating in those poorly perceived even by government officials, catastrophe risk pools. Politicians often who see it as competing directly with the (limited) have difficulty justifying investments in risk resources available for preparedness and management that require governments to pay risk reduction. In reality, both investments are for something that does not demonstrate an worthwhile, and a cost-benefit or other financial immediate benefit. Investment in insurance has a analysis should allow the government to allocate value proposition that extends over a long period efficiently their limited resources. of time, while budgetary and political cycles have short-term time frames. Moreover, uncertainty While countries may value solidarity with one about the trajectory, size, and scale of disaster another, they are reluctant to cross-subsidize shocks means that governments frequently face payments of premiums. For this reason, the questions about whether the insurance will be premium paid by each country should reflect “needed.” These factors make it difficult for only its own risk exposure and coverage. The politicians to justify insurance investments in use of parametric triggers helps avoid this risk the first place; continued investment becomes of cross-subsidization, but even when insurance even more difficult when an insurance policy is properly priced, problems may arise. Countries SOVEREIGN CATASTROPHE RISK POOLS 41 that have contributed for several years without by the Mexican Ministry of the Interior. In the event any payout may question the insurance product of a declaration of disaster, states submit funding and its pricing when they see payouts made to requests based on damage assessments, and other participants. a prescribed process is followed to determine access to funds from the central FONDEN Capital ownership of the facility is also an structure. Under the FONDEN rules, the FONDEN important element in the overall political financing mechanism finances 100 percent of the ownership. All existing risk pools were built reconstruction costs for federal assets and 50 on a large base of seed capital—capital that percent for state and municipal assets. If states helps reduce premiums, but also contributes do not purchase insurance for their reconstructed to increasing ownership of the risk pool by assets, they are penalized under FONDEN by a participating countries. reduction in the percentage of reconstruction costs deemed eligible for funding. For participating countries (and supporting donor partners), risk pools can offer a platform Leveraging donor funding for political dialogue on the financing of climate Catastrophe risk pools allow entry points for and disaster risks and collective action against development partners to support financial them—dialogue that may be even deeper resilience in targeted ways across a group of because the participating countries own the pool countries. For example, when the Multi-Donor and are ultimately responsible for its success Trust Fund established to support CCRIF was (or failure). Sovereign catastrophe risk pools can approved, three donors put forward around also be a vehicle allowing member countries to US$30 million in grant funding, after an initial negotiate with donor partners and secure funding US$2 million grant from Japan for the preparation that would not be available to individual countries, work.6 This amount subsequently grew close to as donor partners are often keen to promote and US$70 million. The use of donor contributions for support regional initiatives. CCRIF has provided further leverage on funds by giving the facility the space to build a substantial In addition, catastrophe risk pools can serve to asset base—now supporting over US$115 promote financial and operational discipline by million of capital—and by protecting premium members. Pools should apply a systematic rules- income from insurance payouts and expense based approach to the request for and release erosion in the early years. CCRIF has benefited of funds for post-disaster activities. They can from investment income from the original donor also require that countries consider in advance funding and from the asset base these funds how any proceeds released through the pool will have allowed it to develop. In the Pacific, Japan be deployed. Such discipline can sometimes be provided premium subsidies and grants for difficult to maintain. The governance structure of technical assistance to support the piloting of the the pools is therefore critical to ensure that the PCRAFI insurance program during its first three pool is managed according to sound operational years. Now that a successful proof of concept has and financial principles and to avoid any political been established, the evolution of the pilot into interference that may not be consistent with a stand-alone facility has prompted involvement the long-term sustainability of the pool (such by Germany, Japan, the United Kingdom, and as setting premiums below sound rating or the United States, with contributions of US$40 requesting payouts for noninsured events). million in funding toward the establishment and capitalization of the facility and additional FONDEN provides an example of how a technical assistance to the countries. In the collaborative approach to financing can instill case of ARC, multiple donors have contributed to financial and operational discipline across multiple funding the development of the program and the entities. FONDEN comprises a system of federal ongoing capacity building and client support work and state-level reconstruction financing managed undertaken by ARC Agency. The UK and German 6. The donors were DFID, the Agence Française du Développement, and the Canadian International Development Agency. 42 SOVEREIGN CATASTROPHE RISK POOLS BOX 11. LEVERAGING DONOR FUNDS IN RISK POOLS Risk pools offer a number of mechanisms through Operating costs. Operating costs, including which donors can apply effective financial support: management costs and reinsurance costs, can be reimbursed for a given amount of time to allow the Premium subsidies. Direct subsidization of premium income to be used to build up reserves premiums is facilitated when countries collaborate and capital. This approach was taken by CCRIF. on risk pooling, allowing donors to support pilot programs and to create a proof of concept to Seed capital. Insurance becomes more cost- test demand from countries. Some participating efficient per dollar of coverage when a higher countries in all three sovereign catastrophe risk attachment point is set, that is, for lower- pools have benefited from (partial) premium frequency/higher-severity events (e.g., major subsidies, including Marshall Islands, Samoa, floods, severe earthquake, or tropical cyclone). Tonga, and Vanuatu during the pilot phase of Since donor support of risk pools can finance PCRAFI; Haiti under CCRIF; and Senegal under joint reserves that cover first loss layers, the ARC. Premium subsidies should be designed to pool can transfer higher layers of risk where this incentivize countries to contribute to the payment is most cost-effective. The principle of applying of the premium over time. risk-bearing capital can also extend beyond the first loss layer, to provide more cost-effective Start-up costs. The early years of risk-pooling alternatives to market-based reinsurance for schemes are the most expensive, and donors may additional layers of loss. This approach has been look to apply funding to initial costs. In the case of used by ARC and PCRAFI. the Pacific, donors funded the large up-front cost of developing catastrophe risk models essential to disaster risk financing. governments have provided interest-free 20-year each country were covered to the same level loans (“development capital”) to ARC Ltd. as the on an individual basis. Diversification occurs founding risk capital7. both across multiple perils and across a geographical area. Diversification does not Pooling initiatives deliver additional efficiencies in reduce the risk (as measured by the annual donor funding by offering donors a range of options expected loss), but does reduce the capital for subsidizing the cost of financial management requirements to cover the full risk spectrum. of disasters, and thereby allowing donors to apply funds in the most effective way possible. These ƒƒ Joint reserves. Establishing joint reserves mechanisms are described in Box 11. allows the pool to retain a fraction of the risks (typically more frequent losses) and to transfer FINANCIAL EFFICIENCY the excess risks to the reinsurance and capital markets when it is most cost-efficient. There are a number of financial benefits to transferring climate and disaster risk through a ƒƒ Larger reinsurance transaction size. Pooling risk pooling mechanism, which ultimately reduces creates larger transactions that are more the cost of insurance: attractive to global reinsurance and capital markets, thus reducing the cost of reinsurance ƒƒ Risk diversification. Covering each country and ultimately the premiums paid by the in a pool requires much less capital than if participating countries. 7. See Annex 1 for a detailed description of ARC, including the different roles of ARC Agency and ARC Ltd. SOVEREIGN CATASTROPHE RISK POOLS 43 FIGURE 7. DIVERSIFICATION BENEFITS MODELED FOR THE PCRAFI PROGRAM: SIMPLE AGGREGATES VERSUS POOLED RISK FOR 1-IN-250-YEAR RETURN PERIOD LOSSES $40 $30 Millions $20 $10 $0 Individual Countries Portfolio Source: World Bank and PCRAFI (2016). Risk diversification countries. This reduction leads to a premium The principle benefit of pooling catastrophe risk reduction in excess of 40 percent. In its first year, is the impact that diversification of losses has ARC Ltd. likewise captured a significant portion on the risk-bearing capital needed to support of the full diversification benefits despite having the risk across the pool. This applies whether only four countries (five growing seasons) and countries opt to retain the risk collectively, or a single peril. ARC Ltd. wrote a total of US$129 transfer it to the international markets. This million in drought coverage, and the 1-in-250-year benefit manifests as a reduced catastrophe load loss to the portfolio was just US$62 million, a (see Figure 7) applied to the premium charged 52 percent reduction in required capital. This for risk transfer, or a reduced amount (and thus reduction in capital requirement could exceed 60 cost) of capital that needs to be held to support a percent with further expansion of the portfolio risk retention mechanism such as a joint reserve to new countries and perils (e.g., flood), thereby facility. It arises from the fact that severe losses further reducing the insurance premium paid by will not be experienced simultaneously across all the participating countries. participating entities. Therefore, the “backstop” required to ensure that the pool or risk transfer There is, however, a limit to the additional product can (with a very high level of certainty) diversification benefits delivered by adding extra meet its obligations to the insureds is lower than units of uncorrelated risk to a pool. Once a critical the sum of supporting capital required if countries mass of uncorrelated units (or units with low consider their possible maximum losses (to the correlation) has been added, the marginal benefits same level of certainty) individually. of adding more become small and may not merit the cost of expansion. Box 12 illustrates such Figure 7 uses the PCRAFI program to illustrate this limitations. In this specific example, 80 percent of effect; the combined portfolio shows a 65 percent the diversification benefits (in terms of premium reduction in the 1-in-250-year return period loss reduction) can be achieved when one-third of the compared to the sum of individual values for potential states/provinces join the pool. 44 SOVEREIGN CATASTROPHE RISK POOLS BOX 12. DIVERSIFICATION BENEFITS AND THEIR LIMITS: AN ILLUSTRATIVE CASE To illustrate how the benefits of risk pooling evolve cyclone risk for this case study is correlated across with pool expansion, an analysis was conducted provinces. There are varying degrees of correlation using data from a large Asian country for tropical between individual provinces, depending on their cyclone risk. The analysis takes estimated respective locations; the order in which provinces premiums for covering the modeled tropical are incorporated therefore affects the premium cyclone losses in each of 60-plus provinces/ benefit trend. In this analysis, provinces were states, and compares the aggregate of these incrementally added to the pool with no ordering against the estimated premium for a pool of with respect to correlation. provinces, as provinces are incrementally added to the pool. PREMIUM REDUCTION (AS A PERCENTAGE OF MAXIMUM PREMIUM REDUCTION) FOR The figure below shows the share of premium INCREMENTAL ADDITION OF PROVINCES reduction (compared to the maximum premium 100% reduction when all provinces join) as the number of provinces joining the pool increases. Twenty 90% provinces joining the pool allows for 80 percent 80% of the maximum premium reduction due to risk diversification. 70% 60% The results show that the premium reduction generated by adding provinces is most significant 50% at the beginning of the pool expansion, tapering 40% off as more and more provinces are included. 30% The rate of occurrence of this effect will vary from pool to pool depending on the correlation between 20% incorporated units, and their respective sizes and 10% risk exposure. However, the trend of decreasing marginal benefits beyond a critical mass of units 0% 10 20 30 40 50 60 is universal. It is important to note that the tropical Number of provinces joining the pool Where pools have reached optimal diversification may not be available in most of the existing benefits, they could potentially achieve additional sovereign catastrophe risk pools). It would also savings by swapping risk via financial instruments require strong political will on the part of pool between regional pools (see Box 13). However, members, who may not support an arrangement assuming that the potential savings are that uses a substantial amount of the capital to significant, such an approach would require not pay claims faced by another pool, even if they are only stable and mature portfolios of risks but also adequately compensated. specific underwriting and pricing expertise (which SOVEREIGN CATASTROPHE RISK POOLS 45 BOX 13. GATHERING FURTHER DIVERSIFICATION BENEFITS THROUGH RISK SWAPS BETWEEN REGIONAL POOLS? While significant diversification benefits are of US$5 million were swapped in an illustrative already being captured by the existing regional risk portfolio from two regional pools. The result pools, each is constrained in diversifying further showed that compared to the technical price of for two main reasons. First, all the existing risk a specific reinsurance transaction, savings on pools serve the client/member states within the technical price are only 1 percent to 2 percent pool, so that maximization of diversification is not for the full portfolio, but that savings of up to 20 a key factor in building a portfolio; rather, client percent of the cost of a top risk layer are possible. demand in terms of coverage selection supersedes other considerations. Second, political and/or However, there may be significant challenges in governance constraints are a barrier to offering such catastrophe risk swaps. First, they could be coverage to countries outside of the region of the technically complex if the portfolios being swapped pool, so that further geographical diversification have different reinsurance periods (e.g., due to may be difficult. seasonality of the insured risks). Second, there are significant political risks to such transactions; One potential mechanism for addressing the countries from one pool might not like to see a second constraint is to swap risk between substantial portion of their capital being used to catastrophe risk pools. Risk could be swapped pay claims of countries from another pool, even if directly between two catastrophe pools, so that they have been adequately compensated (through a pool in one region could take on some portion swapping out an equivalent amount of risk). of risk from a pool in another region (which would Similar political concerns prevented the Central likely bring diversification benefits) and vice versa. American countries from joining the original CCRIF pool; instead, a stand-alone underwriting cell, with Portfolio risk swapping brings some additional its own capital, was formed to accommodate the diversification benefit to both parties. For purposes incoming Central American countries. of analysis, four separate (and matching) tranches Asset management is an area where additional The engagement of an experienced investment financial efficiencies might be garnered, manager to deliver the investment strategy is depending on the risk tolerance of the entity— critical. CCRIF has significant flexibility in this area; which in turn depends on the nature of the its effective shareholder is a passive trust, so the capital and interest of the shareholders/country board has broad scope to manage the investment members of the risk pool. Investment income can portfolio consistent with its own views of risk and be a significant component of overall income for a reward. In contrast, ARC’s capital providers and facility and can contribute to financing operational policyholders are its effective shareholders, and costs over the longer term, thereby contributing to the initial capital providers in particular are highly the pool’s sustainability. Member countries and averse to investment losses (which are a source development partners need to carefully consider of major political risk). Policyholder capital, once what type of risk profile is appropriate for the it accumulates in ARC, will be subject to different pool’s investment strategy, given the desire considerations coming from those policyholders, for financial efficiency, the need for long-term who may be less focused on total return and more sustainability, and the objectives of all parties in on appropriate investments in Africa. establishing the pool. The developed insurance markets provide useful benchmarks for this Joint reserves purpose, including what balance of asset liquidity A catastrophe risk insurance pool can also will deliver the best returns possible without be viewed as a joint reserve mechanism, compromising rapid claims-paying capacity. with contribution levels selected by individual 46 SOVEREIGN CATASTROPHE RISK POOLS BOX 14. CCRIF CENTRAL AMERICA OPTIONS FOR RISK TRANSFER In preparing the expansion of CCRIF to Central states to the international reinsurance market, America, four options for catastrophe risk transfer they could reduce the cost of catastrophe risk were considered for Honduras and Nicaragua as insurance by 27 percent without joint reserves, part of a wider Central America grouping. The table and by 41 percent with joint reserves. As a below presents the estimated reduction in the group, COSEFIN countries could reduce indicative indicative commercial premium across the options commercial premiums paid by close to 45 percent compared to the baseline option (independent should they chose to transfer risk through CCRIF. catastrophe risk transfer). Preliminary analysis This premium reduction could be shared between found that when Honduras and Nicaragua participating COSEFIN countries and CCRIF transferred catastrophe risk with other COSEFIN member countries. Option Premium reduction 1. Central American countries each independently transfer catastrophe risk to - the international reinsurance/capital markets 2. Central American countries jointly transfer catastrophe risk to the 27% international reinsurance/capital markets, without joint reserves 3. Central American countries jointly transfer catastrophe risk to the 41% international reinsurance/capital markets, with joint reserves 4. Central American countries work with CCRIF to jointly transfer catastrophe 44–45% risk to the international reinsurance/capital markets Source: World Bank 2014a. Note: For option 3, initial capital from donors of US$50 million is assumed. Ranges reflect uncertainty over the correlation of tropical cyclone losses between Central American countries. Prototype CCRIF-style policies are assumed, with partial coverage in excess of retention of annual losses equivalent to 1-in-15 years for hurricane and 1-in-25 years for earthquakes. participants and a set of rules to ensure that of losses, passing a further US$140 million to in the long term, each participant will receive the international markets. For its 2015/16 policy payouts relative to the premium (contribution) it year, ARC Ltd. retained about US$22.5 million of has paid. By combining resources, this vehicle risk through its reserves and passed a further can provide a higher amount of available capital US$72.5 million to the international markets. than an individual sovereign contingency fund, and it also brings discipline to the management The capacity of catastrophe risk pools to diversify and protection of the funds, especially protection their risk portfolio and retain a portion of losses against possible political interference, so that the using joint reserves can reduce the insurance funds are there when really needed. premium paid by the participating countries. According to an actuarial analysis conducted for Joint reserves give the pool the flexibility to retain expansion of CCRIF to Central America, where (self-insure) some risks, so that the excess risk several risk transfer options were considered and transferred to markets is a smaller slice of the the impact on premium reduction was estimated, exposure, which in turn makes premium pricing risk pooling benefits translated into premium more cost-effective. All existing sovereign pools savings of an estimated 27 percent for the two have heavily relied on donor support to build up countries analyzed. The analysis found further their initial joint reserves. For example, CCRIF’s that retaining losses through joint reserves annual joint reserve layer is US$25 million, which increased the premium savings to 40 percent or means it is able to retain the first US$25 million more. See Box 14. SOVEREIGN CATASTROPHE RISK POOLS 47 The financial efficiency of capitalizing several segregated portfolio company that operates pools separately can be questioned. It may as a virtual organization; it is supported by a be worthwhile analyzing not only the financial network of third-party service providers covering feasibility but also the political and operational captive management, risk and reinsurance feasibility of establishing a global fund of risk- management, risk modeling, asset management, bearing capital to support sovereign catastrophe and information technology, among other areas. risk pools. Such an approach could, at least Certain costs that would otherwise sit in house conceptually, yield a number of financial benefits. (such as for overhead/IT infrastructure in addition It could pool resources to increase the overall to the provision of the service itself) are pushed retention capacity of the individual pools and out to the third-party service providers and potentially reduce costs through diversification wrapped into their fees. across pools. From an operational point of view, it could also yield cost efficiencies through the The comparison of operating costs for facilities use of a single service provider for investment is difficult, partly because facilities use differing management and fund administration. cost classifications and partly because public information on their operation structure is limited, The feasibility of such an approach would, since these pools operate as private insurance however, depend on the nature of the entities companies. However, the average operating looking to share the common pool and would costs of the sovereign catastrophe risk pools are require a strong convergence of interests across estimated to be around 10 percent of their annual those entities. At present, PCRAFI operates with premium income (when in full operation) and a dedicated captive insurer domiciled in the Cook higher during the first years of operation. Compare Islands; CCRIF is a segregated portfolio company this to average operating costs of 30 percent for domiciled in the Cayman Islands; and ARC traditional insurance companies (not including operates through a limited-by-guarantee (mutual) initial start-up costs incurred during the first years insurance company domiciled in Bermuda. In of operations). Some of the costs that traditional addition, establishing a financial structure to insurance businesses usually account for as serve such entities with risk-bearing capital operating costs (such as product development, poses legal and regulatory challenges that are not capacity building, and communication) are financed trivial. Finally, detailed technical analysis would be through donor-funded technical assistance in the needed to determine a financial structure able to case of sovereign pools. achieve this common pool function without adding substantial additional administrative costs. These operating models raise two questions for the efficiency of catastrophe risk pools: Is OPERATIONAL EFFICIENCY there a way for multiple regional pools to share any back-office functions without compromising Operational efficiency in catastrophe risk pools the sense of regional ownership of the facility? is critical to keep the operational costs low. The And is an outsourcing model—which moves operating sovereign pools have opted for slightly expertise and knowledge out of the facility, and different strategies to achieve this. ARC is in the typically also out of the region—detrimental to process of building staff resources within ARC Ltd. a pool’s sustainability? The following challenges and ARC Agency to support client engagement need to be addressed by facilities considering and operations. ARC utilizes external service consolidating functions across regions: providers for standard financial services such as insurance management, external audit, internal ƒƒ Physically locating a facility within its region, audit, legal counsel, and company secretary. and having a distinct regionally owned entity, CCRIF and PCRAFI, on the other hand, outsource is critical from a political and ownership the majority of back-office operations to trusted perspective, as discussed in the previous third-party service providers in order to keep the section. This limits the potential to consolidate in-house operating costs as lean as possible. operational functions across regions for the For example, CCRIF recently restructured as a purpose of cost saving. 48 SOVEREIGN CATASTROPHE RISK POOLS ƒƒ Catastrophe models, data, and monitoring assignment of government staff to coordinate services may vary greatly from one region to activities in country. ARC provides technical another. support bilaterally as well as through regional and continental workshops and coordinates inputs ƒƒ Given the potential application of catastrophe with nongovernmental stakeholders. ARC’s in- risk models beyond catastrophe risk pools, house risk quantification platform, Africa RiskView there is a strong rationale to develop and (ARV), helps to illustrate and promote discussion retain as much expertise in this area as of sovereign risk profiles and offers strategies to possible within each individual region to build manage the identified risk. the technical capacity required for the long- term resilience agenda. CCRIF’s main capacity-building efforts are currently implemented via the Technical Assistance ƒƒ Country priorities across different regions Programme. Launched in 2009, the program may not be aligned with regard to the risk/ consists of three components: scholarship/ return profile appropriate for the asset base of professional development, regional strategic the facility, or with regard to appetite for risk knowledge building, and support for local disaster retention versus risk cession. risk reduction initiatives. It is designed as an ongoing mechanism offering grant support within Pools as providers of public goods the region for capacity-building initiatives and Catastrophe pools have driven the development of the development and implementation of projects catastrophe risk models and other public goods likely to make risk management more effective. that have demonstrated their utility beyond the risk The program aims to help CCRIF countries pool itself. For example, the development of the deepen their understanding of natural hazards, PCRAFI platform, and the successful placement catastrophe risk, and the potential impacts of of risk, have had substantial positive impacts climate change on the region. within the region. The combination of a technical and financial program has raised the profile of disaster risk financing, and managing the financial impacts of disasters has become a priority not only for disaster risk management entities, but also for ministries of finance. The insurance component of PCRAFI has been complemented by technical assistance work on public financial management—a critical link in the path from receipt of insurance proceeds to deployment of funds on the ground. One tangible result of this has been the development of post-disaster budget execution manuals for participating countries. In addition, the models developed under the PCRAFI initiative have been used for urban planning, as part of a technical assistance project by the Asian Development Bank for selected PCRAFI-covered countries (ADB 2013). ARC provides scope for engagement with member countries for capacity building, not only in risk financing, but also in contingency planning, natural hazards early warning, and risk quantification. Engagement is led by ARC Agency as a public sector entity, and is operationalized via a memorandum of understanding and SOVEREIGN CATASTROPHE RISK POOLS 49 2.3. SUSTAINABILITY OF CATASTROPHE RISK POOLS Beyond efficiency, sustainability is essential to costs and underline the necessity of financial the long-term success of catastrophe risk pools. management through disaster risk financing The field of catastrophe risk financing has evolved instruments. However, even with the full costs rapidly in recent years, as technical advances elucidated, many developing countries, and based on lessons learned have delivered especially low-income countries, do not hold increasing value for money for countries and much disaster and climate risk, so the potential their development and humanitarian partners. climate and disaster losses are not factored Sovereign catastrophe risk pools are recent; into their budget. As a consequence, moving except CCRIF-Caribbean (established in 2007), countries from a largely ex post approach, where they have been operational for less than five years. financing mainly relies on donor partners, to an ex It is therefore premature to draw any conclusions ante model, where the countries themselves are on their sustainability or to seek lessons on expected to pay for some (if not all) of the cost, sustainability for other pools. However, some is a significant challenge. This is an important key challenges for sustainability of sovereign discussion in light of growing humanitarian costs. catastrophe risk pools can be identified, including Demand for humanitarian assistance is growing the ownership of risk, technical expertise, risk- rapidly, and an estimated 20 percent of these based pricing, financing of insurance premium, growing costs go to sudden-onset and recurring and regional ownership of the pool. natural disasters. In 2000, UN OCHA estimated that the global budget for humanitarian action was WHO OWNS THE RISK? US$2 billion. In 2017, this number has risen to US$22.2 billion. If trends continue, humanitarian Generating a sustained demand for disaster costs will rise to US$50 billion by 2030.8 risk financing, and catastrophe risk insurance in particular, is a challenge for all ex ante financial African Risk Capacity (ARC) has recently started instruments that incur costs up front, not just for offering UN agencies (such as WFP) the opportunity catastrophe risk pools. The full costs incurred to purchase “replica coverage”—coverage that by countries after a disaster are not always replicates the insurance policy purchased by the easily visible, as many are not direct costs. ARC Member State in which the agency is active. Instead, they manifest themselves as opportunity Through this approach, humanitarian actors can costs associated with budget reallocation or access market risk capital to cover costs related restructuring of development projects; or they to humanitarian action in specific countries. are longer-term costs associated with the impact of negative economic and social development TECHNICAL EXPERTISE trajectories. The increasing role of humanitarian aid also impacts countries’ perspectives on Because catastrophe risk insurance is a highly the cost/benefit trade-off of ex ante financing technical area, catastrophe risk pools have to rely instruments. A fundamental question that must on technical expertise in insurance production, be answered to justify the up-front cost required including underwriting, product design, and pricing. for instruments like insurance is “who owns the To establish themselves and become operational, risk?” and consequently, “who should pay for it?” existing pools have required significant technical Ex ante financing schemes first demand that risk assistance and capacity building in catastrophe is quantified up front for pricing and structuring, risk modeling and insurance operations. CCRIF which can highlight the magnitude of potential and PCRAFI, for example, have benefited from 8. Estimates are from internal UN OCHA and Financial Tracking Service data provided to the High-Level Panel on Humanitarian Financing. 50 SOVEREIGN CATASTROPHE RISK POOLS significant technical assistance provided by the Risk-based pricing can be facilitated by using World Bank with financial assistance of donor parametric insurance and catastrophe risk partners. Private reinsurers and brokers have also models. Parametric insurance—as implemented helped those pools build their technical expertise. in all the sovereign pools examined in this However, even after several years of operations, report—provides an unusual case where payouts those pools still rely on external expertise, which and insurance premiums are not explicitly linked could impact their sustainability. Going forward, to actual losses incurred on the ground. Instead, pool’s sustainability will depend on their ability the cost of insurance is a function of the pure to develop local expertise on catastrophe risk hazard exposure (such as the frequency and insurance, along with international expertise. intensity of tropical cyclones) and the level of coverage chosen by the country. Catastrophe RISK-BASED PRICING risk models have been developed to improve the assessment of catastrophe risk and the pricing Risk-based pricing allows for the pricing of of parametric insurance products. They allow the catastrophe risk insurance based (only) on its pools to calculate not only the underlying risk underlying risk. This prevents cross-subsidization estimated through the annual expected loss, of premiums from occurring, but also has value as but also the cost of capital to be reflected in the an indicator of risk. The regional pools discussed insurance premium. here all use risk-based pricing to determine the premiums payable for each country contract.9 PREMIUM FINANCING Financial decision makers need the information on risk contained within this type of pricing in order Establishing a long-term commitment to payment to evaluate the costs and benefits of different of premiums by, or on behalf of, countries is types of instrument for managing their financial one of the most serious challenges affecting exposure to disasters, and such information is the sustainability of sovereign catastrophe risk therefore important for the long-term sustainability pools. Insurance premiums present an up-front of disaster risk financing programs. cost, which may not produce a financial return in the near (or even medium) term. As described Even where there is a strong political platform above, governments face public and political for regional collaboration, countries are typically pressure when payments for premiums on high- not willing to cross-subsidize premiums for other profile sovereign insurance do not yield a payout pool participants by paying more for their own in the event of a disaster (for example, Jamaica coverage. This type of “solidarity” mechanism after Hurricane Dean in 2007, the Solomon has been used within a single political unit (e.g., Islands after the 2013 earthquake and tsunami, private insurers offering UK households flood and Malawi after the drought in 2016).11 At the insurance incorporated an element of cross- country level, allocating budget for the payment subsidization between low- and high-risk flooding of premiums is generally not a permanent part areas to keep insurance affordable for riskier of budgetary processes, and the expenditure is households).10. But it may not be viable across still treated as atypical, when it is possible. This borders or across distinct governance units with section discusses potential sources of premium substantially devolved political power (such as in financing, ranging from national budgets to federal-state structures). concessional loans/grants, long-term subsidies, and premium rebates. 9. One could argue that the pool’s pricing methodology represents an element of cross-subsidization in the way it allocates capital/ reserves among the participating countries. 10. In April 2016, a new public-private partnership, Flood Re, was launched in the United Kingdom to manage insurance costs for higher-risk properties, leading to a restructure in the market. 11. Malawi’s policy with ARC did not initially trigger an insurance payout because the type of maize selected during customization of the parametric model differed from the actual crop planted. The insurance policy was “amended” to take account of the correct crop type, and a payout of US$8.1 million under the contract was made. ARC, “Press Release—Malawi to Receive USD 8M Insurance Payout to Support Drought-Affected Families,” November 14, 2016, http://www.africanriskcapacity.org/2016/11/14/press-release- malawi-to-receive-usd-8m-insurance-payout-to-support-drought-affected-families/. SOVEREIGN CATASTROPHE RISK POOLS 51 Designing risk financing products that match premiums in the investment budget (e.g., St. countries’ needs and priorities, and that Lucia), whereas others have budgeted such demonstrate maximum value for money, can help expenditures as a recurrent item (e.g., Tonga’s generate and sustain willingness to pay premiums. PCRAFI insurance premium or Jamaica’s CCRIF In particular, frequent and timely payouts help insurance premium). However, the government demonstrate the value of the proposed insurance budget for the provision of insurance often has product. But it is important to keep in mind to be approved separately and is not included that insurance is most cost-effective for less among the line items with automatic budget frequent and more severe risk, and that insurance appropriation. In the Philippines, for example, premiums increase significantly if more frequent loan repayment and interest payments receive payouts are targeted. Finding sustainable and automatic budget appropriations, while sovereign practical options that allow countries to finance insurance does not. premiums remains a key question for maintaining sovereign risk pools in the long-term. Concessional loans/grants The use of concessional lending instruments to National budget allocation fund insurance premiums has been an indirect If countries decide to participate in sovereign way for international financial institutions (such risk insurance pools, they will need to include as the World Bank) to support premium financing. the required premium payments in their national International Development Association resources budgets. As indicated above, allocating budget in both grant and credit form have been used to for the payment of premiums is generally not fund premiums and facility entrance fees for a a permanent part of budgetary processes and number of countries participating in CCRIF. For the expenditure is still treated as atypical. Care example, for the first three to four years of CCRIF’s must be taken that the inclusion of premium operation, the Organization of Eastern Caribbean payments in budgets is in line with relevant public States (OECS) Catastrophe Risk Insurance Project finance legislation. The government of Indonesia, allowed four countries (Dominica, Grenada, St. for example, had to pass a new government Lucia, and St. Vincent and Grenadines) to use regulation (PP 45/2013) in 2013 to explicitly allow national and regional IDA financing to cover the its Ministry of Finance to purchase insurance with cost of entrance fees and insurance premiums. funds allocated in the national budget. This program made CCRIF participation affordable to OECS countries and allowed them to test the In some countries, only one entity, usually the use of the instrument; two of the four countries national ministry of finance, can purchase had insurance payouts during the project, which insurance. For example, in Colombia the led them to internalize the premium cost in their Ministry of Finance is responsible for purchasing annual budget decision-making processes. Most insurance for all sectors of government, with clear recently, Nicaragua opted to use IDA credits to limitations in the ability to evaluate the specific finance its CCRIF entrance fees and premiums needs of single municipalities and ministries. across a multiyear timeframe. Likewise, four of the five Pacific island countries that joined PCRAFI In countries where there is no legal and institutional used their IDA credit, together with national framework regulating sovereign insurance, the budget contributions, to cofinance their insurance resulting uncertainty complicates the allocation premiums. of responsibility between government entities and leaves the insurance program vulnerable to There are cases where long-term subsidies may political changes. be appropriate given the fiscal position of the country in question and the potential benefits Depending on a country’s specific guidelines, that sovereign insurance (in pool or other form) insurance premiums could be budgeted either can confer, including access to immediate post- as recurrent expenditure or as investment disaster liquidity. Haiti provides such an example: expenditure. In practice, some countries have fiscal constraints and high levels of catastrophe included expenditure on sovereign risk insurance risk prohibit the country from self-funding CCRIF 52 SOVEREIGN CATASTROPHE RISK POOLS insurance premiums to a level that would provide Seed capital injection versus up-front a meaningful volume of cover. Haiti has received premium subsidies premium subsidies since it first joined CCRIF at Some donor partners have been hesitant to its inception, with the last four years of premium provide up-front premium subsidies (arguing costs funded by the Caribbean Development that this may impact the long-term sustainability Bank. The country has received two payments of the risk-pooling mechanism). However, as from CCRIF —one following the devastating discussed above, low-income countries across 2010 earthquake, and dual payments under all three operating sovereign catastrophe risk excess rainfall and tropical cyclone policies in pools have received such subsidies. More 2016 following Hurricane Matthew. The US$23.4 recently, donor partners have preferred to provide million payout from Hurricane Matthew represents catastrophe risk pools with capital injection (seed the largest payment ever made by CCRIF for a capital) either directly as a grant (PCRAFI) or single event. While the 2010 earthquake payout repayable interest-free loan (ARC), or indirectly by was substantially smaller (US$7.7 million), it reimbursing operating costs, including claims and provided important rapid liquidity in the aftermath reinsurance costs (CCRIF). This capital injection of the event. CCRIF’s demonstrated contribution aims to increase the risk retention capacity of the to Haiti’s resilience has ensured that the donor pool and hence reduce the insurance premiums premium subsidies have delivered strong value charged to member countries over a long time for money. period (ideally in perpetuity). Conversely, up-front premium subsidies by definition reduce the cost The use of concessional financing in this way of insurance for the beneficiaries only for the confers several benefits. Most importantly, it duration of the subsidy program. makes participation affordable for countries while still requiring a financial commitment from To illustrate the trade-off between capital their side. The favorable loan terms offered by injection and up-front premium subsidies, Box 16 development partners, such as IDA credit terms, compares from a financial perspective the cost- can be seen as an implicit subsidy mechanism: effectiveness of a hypothetical grant used for the discounted value of the insurance premium either premium subsidies or seed capital. This financed through IDA loans ranges from 20 of the analysis indicates that the most cost-effective face value of the premium, assuming a 10 percent option depends, among other things, on the level discount rate, to 50 percent of the face value of of premium savings that the pool can pass on to the premium, assuming a 5 percent discount its members if using the grant as seed capital. In rate. But countries use the allocation from their the example, the provision of seed capital is more (limited) overall IDA allocation, and therefore need cost-effective than premium subsidies (given the to be confident of the value of the mechanism set of assumptions) if it allows the pool to reduce to justify the (potentially high) opportunity cost of premium savings by 14 percent or more. doing so. See Box 15 for more detail. BOX 15. CONCESSIONAL FINANCING OF PREMIUMS In this example a country borrows to finance its difference between the NPV of the loan and the insurance premium under standard IDA terms premium amount (as a percentage of the premium (1.47 percent annual interest rate or service amount) can be interpreted as an implicit subsidy. charge, six-year grace period, 38-year duration). With a 5 percent social discount rate, IDA financing The net present value (NPV) of the loan is is equivalent to an up-front subsidy of 50 percent. calculated for various social discount rates. It can The implicit premium subsidy increases as the be compared with the premium amount, should social discount rate increases. the country pay its premium in full up front. The SOVEREIGN CATASTROPHE RISK POOLS 53 The level of premium reduction due to the capital reserves quickly enough, and hence would heavily injection depends on several factors, including depend on the reinsurance markets). On the other the current position in the reinsurance cycle and hand, if a catastrophe risk pool is already well the current level of capital held by the catastrophe capitalized, additional seed capital would only pool relative to the total premium volume. Other marginally reduce the cost of insurance. In this factors also affect the overall effectiveness of case, the donor grant may be more cost-effective the grant, such as the availability of reinsurance than premium subsidies. The example should not capacity in the region where the pool operates, be seen as a general statement on how donors and the effect of subsidies on the operating costs should support catastrophe risk pools, but rather of the pool (e.g., premium subsidies may create as an illustration of the financial trade-offs to incentives to increase operating costs). The be considered when donors decide to provide a result depends on the underlying assumptions catastrophe risk pool with financial support. adopted, and could be significantly different under alternative (but still realistic) assumptions. Donors can also use the provision of concessional insurance (either as seed capital or premium This analysis shows, however, that seed capital subsidies) to create (financial) incentives injection and up-front premium subsidies can for climate and disaster risk reduction and be complementary. When a new catastrophe preparedness, including financial preparedness. risk pool is established, it will rely on a capital injection to make it sustainable (since without seed capital the pool may not be able to build up BOX 16. QUANTITATIVE COMPARISON OF SEED CAPITAL VERSUS PREMIUM SUBSIDIES The benefits of this grant can be analyzed by Consider a hypothetical catastrophe risk pool comparing the NPV of the two grant options over with an annual premium volume of US$1 million. a 20-year time horizon, with a social discount Donors are willing to provide a US$6 million grant rate of 10 percent. The initial premium pricing for either seed capital or premium subsidies. has a premium multiple of 1.4 (before premium subsidies or premium savings due to additional Premium subsidies allow the pool to reduce annual capital injection). premiums every year for 20 years.
It is assumed the grant is depleted over the full 20 years. In each case, the NPV consists of the annual This leads to an annual premium subsidy of 30 premium savings to the pool and the investment percent or, equivalently, to a 30 percent increase return on the unused grant amounts. Based on in catastrophe risk insurance coverage. It is also these assumptions, the NPV of the US$6 million assumed that each year the unused portion of the grant is US$3.3 million under the premium subsidy grant generates investment earnings of 2 percent case. The NPV of the grant used as a seed capital compound interest. injection depends on the level of premium savings the pool is able to pass on to its members (through Seed capital enables the pool to retain a proportion retaining a proportion of the risk), as indicated in of risk and pass on premium savings to participant the figure below. countries. It is assumed the capital cannot be depleted over time, generates investment In this illustrative example, where premium earnings of 2 percent compound interest, and is savings due to additional seed capital are greater shared among the participating countries at the (lower) than 14 percent, a higher (lower) NPV will end of the 20-year period. be achieved if the grant is used for seed capital rather than premium subsidies. 54 SOVEREIGN CATASTROPHE RISK POOLS (continued) 4,000,000 NPV of US$6 million grant 3,500,000 3,000,000 2,500,000 2,000,000 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% Premium savings from seed capital Seed capital Premium subsidy Note: Calculations assume a 20-year time horizon and a 10$ discount rate The actual level of premium savings achieved pool to further reduce reinsurance coverage by retaining risk depends on several factors, at higher layers of risk, where cost savings including the current position in the reinsurance may be limited. Conversely, if a pool has a low cycle and the existing level of capital held by the level of capital (likely for a newly established catastrophe pool relative to the total premium pool, for example) then the capital is likely to volume: be used to retain risk in the first losses (or working layer) of the reinsurance program, ƒƒ The current position in the reinsurance cycle. which can result in significant savings. In the In a soft market characterized by increased above example, if the grant were to be US$10 capacity of reinsurers, lower insurance million rather than US$6 million, then a grant premiums, and broader coverage, the level used as premium subsidies is more cost- of savings from retaining risk are likely to effective unless the seed capital can generate be reduced. Conversely, in a hard market, premium savings in excess of 24 percent. characterized by higher premiums and lower available coverage, the savings from a pool In addition to the above factors, there are other holding capital to retain a proportion of the risk considerations that will also impact the cost- are likely to be significantly higher. effectiveness of the grant. A lower discount rate will mean seed capital is more cost-effective ƒƒ The existing level of capital held by the (i.e., a higher level of premium savings can be catastrophe pool relative to the total premium achieved for seed capital). In the example above, volume. If the pool has a high level of capital reducing the discount rate from 10 percent to 5 relative to the premium volume and is already percent reduces the minimum premium reduction retaining a large proportion of the risk, from seed capital (to be more cost-effective than additional capital is likely to be used by the premium subsidies) from 14 percent to 6 percent. Source: World Bank Disaster Risk Financing and Insurance Program (2016). SOVEREIGN CATASTROPHE RISK POOLS 55 BOX 17. PCRAFI FACILITY In 2015, the Pacific Island Countries made a PCRAFI FACILITY decision to establish a regional facility (PCRAFI COUNCIL OF MEMBERS Facility) to ensure the sustainability of the Pacific OF PCRAFI FOUNDATION Catastrophe Risk Insurance Program, and to appoints bring the management of the program into the region. By creating a regional body accountable to the countries themselves, the facility affords BOARD OF DIRECTORS countries greater control and influence over the design of future disaster and climate risk governs solutions. In June 2016, the PCRAFI Facility was established by legal statute in the Cook Islands; CAPTIVE INSURANCE COMPANY it received an insurance license in September OF PCRAFI FOUNDATION 2016. The governance structure is shown in the manage figure adjacent. INSURANCE CEO AUDITOR MANAGER Source: PCRAFI. REGIONAL OWNERSHIP the ministries of finance, to discuss how the pool can better serve their needs (e.g., through As discussed previously, regional ownership the development of new insurance products), as is an important element in the establishment well as how it fits into their disaster risk financing of sovereign vehicles like catastrophe risk strategies and broader disaster risk management pools. Countries’ ownership is also important and climate change agendas. to ensure the pool’s overall (and not merely financial) sustainability. A sense of ownership Regional political bodies, such as the African prompts participating countries, and especially Union, Pacific Island Forum, or Caribbean Community Market, have played a central role in establishing sovereign regional pools; and once established, the pools play a political role in promoting better financial management of disaster and climate risks in their respective regions. For example, the Pacific islands have recently worked with the support of donors and the World Bank to change the pilot PCRAFI insurance program into a stand-alone facility in the form of a captive insurer, domiciled in Cook Islands. The PCRAFI Facility’s strong position within the region, along with regional ownership, help foster the use of financial mechanisms as a way to build resilience in the region (see Box 17). 56 SOVEREIGN CATASTROPHE RISK POOLS 2.4. INCREASING THE IMPACT OF SOVEREIGN CATASTROPHE RISK POOLS Sovereign catastrophe risk pools are an and accounted for as quickly and efficiently as instrument designed to provide immediate possible. Other ways to strengthen post-disaster liquidity to countries after a disaster, serving as public financial management may include (i) bridge financing while additional funds such as reinforcing the legal environment to support the bilateral aid or reconstruction loans are being development of risk financing and insurance mobilized. They should be part of a country’s solutions; (ii) strengthening risk information comprehensive strategy for public financial and risk analytics for evidence-based decision management of natural disasters, which should making; and (iii) improving countries’ capacities also include contingent disaster response plans to effectively allocate, disburse, and monitor to ensure a timely, transparent, efficient, and recovery and reconstruction funds following effective use of the pools’ resources. Shock- disasters through dedicated mechanisms. responsive social protection allows countries to use existing safety net programs to support poor CONTINGENCY PLANS households affected by natural disasters and reduce the negative impacts of disasters on the The use of the insurance proceeds should also be well-being of the poor and vulnerable. Contingency carefully considered. This is particularly important plans for the restoration of lifeline infrastructure for parametric insurance proceeds, which are not are also critical to reduce impacts and promote linked to any specific asset. Contingency plans swift recovery. But governments cannot bear can help define the potential use of the insurance those risks alone, and the private sector has an proceeds and ensure that agreement about their important role to play in this regard. use is reached in advance. PUBLIC FINANCIAL MANAGEMENT OF Contingency planning is an integral part of the NATURAL DISASTERS ARC insurance program. For ARC member states, one of the preconditions for purchasing insurance A country’s framework for the public financial is the up-front development of an Operations management of natural disasters is a core Plan. Developed in country by government officials element for leveraging the impact of sovereign collaborating with partners and supported by catastrophe risk pools. ARC, it is reviewed against detailed standards and guidelines. While primarily designed as a The PCRAFI program has invested heavily tool to ensure that rapid payouts reach the most in building capacity in the public financial vulnerable, government-led contingency planning management of natural disasters, including the has helped bring together disaster response development of post-disaster budget execution actors—including international agencies and manuals detailing what sources of finances are NGOs—to work on early warning, risk reduction, available for post-disaster response and how and disaster preparedness and response to execute these funds to complement national disaster risk management plans. The guidelines In Mexico, the FONDEN program has a direct link are meant to provide a desk reference for Ministry to a social safety net: if the FONDEN program of Finance staff and other stakeholders following is triggered after a disaster, federal and state the proclamation of a state of emergency or the funding is made available to provide households declaration of a state of disaster. These guidelines living under the poverty line with self-construction are designed to ensure that funds for disaster packs that allow them to repair, reconstruct, or relief and recovery are accessed, disbursed, even change the location of their homes. SOVEREIGN CATASTROPHE RISK POOLS 57 SHOCK-RESPONSIVE SOCIAL populations in the event of a drought. The PROTECTION contingent financing window has been supported by a number of donors and, in By helping people to better manage climate risks 2011, US$134 million was drawn down from and recover after disasters, shock-responsive the window and deployed through the program social protection can play a prominent role in to respond to the drought: in this way, over protecting the poorest and most vulnerable and 3 million additional beneficiaries were in building resilience at the household level. supported, and regular program beneficiaries Governments can utilize social protection systems received expanded support. and programs to deploy assistance swiftly to those most in need after a climate shock. By ƒƒ In Uganda, work is focusing on establishing a assuming this responsibility and utilizing social disaster risk financing component of the Third protection in this way, governments provide Northern Uganda Social Action Fund initiative. a form of insurance to those who are exposed This will provide income support and build the and vulnerable to climate change but unable to resilience of poor and vulnerable households access market-based insurance themselves. This in the case of predefined shocks. approach directly contributes to resilience at the household level by smoothing consumption and ƒƒ In Kenya, the Hunger Safety Net Programme supporting livelihoods after a climate shock has Phase 2 provides regular cash transfers to occurred, potentially helping to break the cycle the poorest households in four northern of poverty and vulnerability that disasters often counties. The program, implemented by the perpetuate. National Drought Management Authority, also has a shock-responsive component and Channeling sovereign-level financing to direct is able to quickly scale up temporary cash beneficiaries through social safety nets is one assistance to vulnerable populations following way to increase the impact of disaster risk predefined drought shocks (as discussed financing instruments in general, and sovereign earlier). The development of this component catastrophe risk pools in particular. required technical assistance to analyze the costs associated with different scenarios, ƒƒ In Ethiopia, contingent financing mechanisms depending on the severity of the shock, the have been used at the sovereign level to number and location of people covered, the fund the scale-up of the Productive Safety additional amount in cash transfers, and the Net Programme and reach food-insecure time period of that extended support. FIGURE 8. FLOW OF FUNDS FINANCIAL RECIPIENT ULTIMATE INSTRUMENT INSTITUTION BENEFICIARIES Payout determination Deployment of funds and fund transfer 58 SOVEREIGN CATASTROPHE RISK POOLS CRITICAL INFRASTRUCTURE being implemented through ARC Ltd.’s reinsurance broker. The initiative will expand ARV’s availability The rapid restoration of critical infrastructure, as a tool for underwriting parametric weather including critical roads and bridges and water insurance, providing funding for ARC itself and supply, is often key for efficient post-disaster supporting agricultural activities on the African response. In small state islands, where continent. L4D allows private sector actors with international emergency aid comes mostly by African exposure to improve their understanding of plane, the restoration of airport runways is the impact of weather volatility on crop production, an essential part of the government’s rapid supply chains, and business forecasting. Further, response, along with restoration of key roads and it facilitates companies’ use of parametric bridges. Most maintenance programs do not take insurance products to transfer the risk of adverse into account unexpected costs caused by natural weather to the global risk markets. Lastly, L4D disasters, and this oversight may ultimately delay can contribute to growth in the insurance sector, the restoration of those assets. In addition, even in particular to complement the transformational when funds are available, few countries have agenda set for the agricultural sector in Africa. standby contracts with suppliers and service CCRIF has provided its catastrophe risk data and providers that would allow for rapid restoration. platform to international partners (MicroEnsure, the Munich Climate Insurance Initiative, and CROWDING IN THE PRIVATE SECTOR Munich Re) and local insurers and financial institutions, which are collaborating to offer Crowding in the private sector can help achieve catastrophe risk insurance to low-income scale, develop the local market, and generate individuals and lenders affected by extreme demand. But public-private partnerships need to weather events in the region. The original program, follow clear pro-poor guidelines to be credible. called Livelihood Protection Policy, ran from 2010 to 2014 and was piloted in Grenada, Jamaica, and The Pacific Risk Information System, a platform St. Lucia; using CCRIF weather data, its products that includes an exposure database of over 4 were designed to help protect the livelihoods of million assets located in the Pacific region, and vulnerable low-income individuals (such as small its associated catastrophe risk model have been farmers and day laborers) by providing quick cash used by domestic insurers and brokers to inform payouts following extreme weather events. CCRIF their underwriting and pricing decisions. In Fiji, data were also the basis for a loan portfolio for example, the model was used to inform the hedge for financial institutions with portfolios of provision of catastrophe risk insurance for hotels individuals and micro, small, and medium-sized and resorts. The model has also been used to enterprises, which was designed to promote explore the feasibility of crop insurance in some investment in areas previously considered too Pacific islands. risky for traditional lending. The results from the first phase of the pilot were positive, but limited; ARC developed the Licensing for Development only a small number of policies was issued. (L4D) program to allow the use of Africa RiskView However, the pilot will be continued in a second to cover nonsovereign agricultural risks on a phase for 2017–19 and expanded to target parametric basis. During its pilot phase, L4D is additional countries. SOVEREIGN CATASTROPHE RISK POOLS 59 PART 3. Recommendations moving forward 60 SOVEREIGN CATASTROPHE RISK POOLS 3.1. CATASTROPHE RISK POOLS CAN ENHANCE THE FINANCING OF CLIMATE AND DISASTER RISKS Catastrophe risk pools can help countries by the pool; (iii) enabling access to international shift financing of climate and disaster risks reinsurance and capital markets that might away from a reactive approach that mobilizes otherwise be impossible; (iv) sharing operational resources after a disaster and toward a more costs, such as program development and day-to- cost-effective, proactive approach that plans day back-office operations; and (v) involving the in advance. Timing matters, and early financing private sector in the operational and financial is more cost-effective than late financing. In management of the pools. Ethiopia, for example, every US$1 secured in financial planning for timely and predictable Catastrophe risk pools can provide further risk disbursement for emergencies can save up to management benefits. Beyond the financial US$5 over the long term (Wiseman and Hess benefits of transferring risk off balance sheets, 2007). In addition, by securing financing ahead catastrophe risk pools can (i) create incentives of time, a pre-planned approach to disaster- for risk reduction by putting a price on risk through response reduces reliance on humanitarian aid, the payment of insurance premiums; (ii) enable which is unpredictable and uncertain, and often rapid disbursement of funds that can reduce takes time to materialize. humanitarian impacts and save money through rapid crisis response and relief efforts; and (iii) Catastrophe risk pools can help countries increase transparency and efficiency through the access insurance and capital markets on adoption of clear, pre-agreed triggers/rules for competitive terms. The 26 countries currently the disbursement of funds as identified in a post- covered by sovereign catastrophe risk pools disaster response plan. The process of developing represent the vast majority of low- and middle- such post-disaster plans and identifying related income countries that have purchased sovereign costs can also create incentives to step up disaster risk insurance. (An exception is Mexico, investments in prevention and adaptation to which issued catastrophe bonds on its own). reduce risks in the first place. Small countries with limited financial market experience or infrastructure may struggle to Catastrophe risk pools can serve as a vehicle for access international insurance markets directly. regional policy dialogue. Catastrophe risk pools If one small state tried to purchase insurance have allowed participating countries and donors coverage individually, the premium volume would to improve collaboration on risk management and be too small to make the transaction commercially risk finance. They offer a vehicle to anchor key attractive and viable for international insurance elements of collaboration: companies. But a risk pool’s joint portfolio and ƒƒ Integrated financial planning. Countries can larger premium volume solve this problem and make their participation in a risk pool part of facilitate access to international markets. a more comprehensive disaster risk financing strategy that brings together various financial Catastrophe risk pools can contribute to more instruments. affordable climate and disaster risk insurance solutions. Risk pools can make risk transfer ƒƒ Contingency planning. The different risk pools more cost-effective by (i) diversifying risk established to date have taken different through the participation of multiple countries views on this, but all encourage countries to with different risk profiles; (ii) establishing joint establish contingency plans in advance to reserves to self-insure a part of the risk managed support disaster response. SOVEREIGN CATASTROPHE RISK POOLS 61 ƒƒ Regional ownership. Since risk pools inherently risk pools to direct beneficiaries. To help poor and require regional cooperation to be established vulnerable people better manage climate risks and maintained, they also serve as a vehicle and recover after disasters, governments can to advance regional collaboration on the utilize social protection systems to protect assets climate risk management agenda. and livelihoods and to deploy assistance swiftly to those most in need after a climate shock. By ƒƒ Discussion on climate and disaster risk assuming this responsibility and utilizing social ownership. This occurs through the pricing of protection in this way, governments provide a climate and disaster risks. form of insurance to those who are exposed and vulnerable to climate change but unable to access Catastrophe risk pools help clarify who owns the market-based insurance themselves. This approach risk. Many developing countries still rely on donor directly contributes to resilience at the household assistance—which may come late but at low or no level through consumption smoothing and livelihood financial cost for the recipient country—to help fund support after a climate shock has occurred, relief and response activities following a shock. In potentially helping to break the cycle of poverty and exchange for the insurance coverage, catastrophe vulnerability that disasters often perpetuate. risk pools require participating countries to pay up front an insurance premium that reflects their Catastrophe risk pools can contribute to the actual risk exposure, thereby shifting payment so provision of public goods. The creation of risk it takes place in predictable installments before pools has driven the development of catastrophe disaster strikes. It may be challenging for countries risk models and other public goods that have that previously relied on donor support to start roved valuable in multiple areas. For example, paying for climate and disaster risks with national domestic insurers and brokers have used the resources through an insurance premium. But Pacific Risk Information System (a platform that moving in this direction, even partially, can provide includes an exposure database of over 4 million the right incentives for proactive planning and assets located in the region) and its associated investments in risk reduction. catastrophe risk model to inform their underwriting and pricing decisions. In Fiji, the model was Catastrophe risk pools can help reduce the used to inform the provision of catastrophe risk impacts of disasters on the poorest and most insurance for hotels and resorts. The model has vulnerable. Social safety nets can be used to also been used to explore the feasibility of crop channel sovereign-level financing from catastrophe insurance in some Pacific islands. 62 SOVEREIGN CATASTROPHE RISK POOLS 3.2. CATASTROPHE RISK POOLS ARE NOT THE ONLY SOLUTION FOR THE FINANCING OF CLIMATE AND DISASTER RISKS The long-term financial sustainability of existing insurance is inherently a costly mechanism (it sovereign catastrophe risk pools is still to be pays a third party to bear the risk). Moreover, proven. With the exception of CCRIF-Caribbean, catastrophe risk pools cannot reduce the which started with 16 countries, pools have faced underlying climate and disaster risks faced by challenges in attracting a large number of countries. the countries, which should be reduced through CCRIF-CA has only one country so far, PCRAFI has appropriate risk reduction measures. five countries, and ARC has six countries. Most sovereign catastrophe risk pools have also faced Catastrophe risk pools can address only challenges in securing the annual payment of specific needs. Governments considering the insurance premiums by member countries. There establishment of a risk pool need to have clarity are financial and practical reasons why some on the objectives to be achieved. Catastrophe countries (often low-income countries that are a risk pools in the Caribbean and the Pacific are high priority for donors) struggle to commit to or designed to ensure access immediate (but limited) afford their annual premium payments through funding for rapid post-disaster response. While their national budgets. the potential payouts following an earthquake or a tropical cyclone represent a small fraction of Catastrophe risk pools come with risks— the overall damage and loss caused by a disaster, managing expectations is key. As with any this immediate cash injection can be substantially insurance product, catastrophe risk pools are larger than—sometimes several times over— exposed to the risk that small disasters may the government’s contingency budget (which not trigger a payout because the loss is below is typically the only source of cash available). a pre-agreed threshold (deductible), or the In Africa, ARC encourages using payouts for peril that causes the loss is not included in early action against drought and incentivizes the insurance policy. In addition, catastrophe contingency planning to that effect. pools offering parametric products (which all sovereign catastrophe risk pools now do) are Catastrophe risk pools cannot be the only exposed to basis risk (that is, the risk that the solution for disaster risk financing and should be index measurement does not match the actual complemented by other financial instruments. losses). Such risks should be carefully mitigated Disaster risk insurance products offered by risk in the design of the insurance products as part of pools are efficient ways to provide timely (but a comprehensive financial strategy that includes limited) financing for rapid post-disaster response. instruments beyond insurance. Moreover, such Experience shows that payouts received from risks should be addressed in the dialogue with risk pools are only a small fraction of the actual participating countries, to ensure awareness of disaster loss. Other financial instruments should the benefits and limitations of insurance. be used by the countries to finance the cost of more frequent disasters and the cost of long-term Catastrophe risk pools cannot make insurance reconstruction. cheap. While these vehicles can improve the efficiency and reduce the cost of risk transfer, SOVEREIGN CATASTROPHE RISK POOLS 63 3.3. RECOMMENDATIONS FOR G20 MEMBERS The G20 could promote a set of priority action ŠŠ Technical assistance and investments areas designed to reduce the protection gap by to develop shock-responsive scalability in vulnerable developing countries. These actions mechanisms for existing social safety nets would advance financial protection against to protect the poor and vulnerable climate and disaster risks, in part by encouraging ŠŠ Technical assistance and investments the scale-up of catastrophe risk pools at the to identify, prioritize, and protect critical supranational, national, and subnational levels. infrastructure at risk, both ex-ante (by Specifically, the G20 could promote activities that mainstreaming disaster risk reduction support the following priority action areas: in investment planning) and ex-post (by developing pre-agreed financial plans for ƒƒ Facilitate the adoption of financial protection post-disaster reconstruction) strategies that include a mix of financial instruments against disaster and climate ƒƒ Promote institutional and legal frameworks risks, such as budgetary instruments, that enable the implementation of financial contingent credit, and catastrophe risk transfer protection strategies. This includes creating to increase the ownership, impact, and cost- the legal base that enables governments to efficiency of disaster response financing. establish disaster risk management funds, pay insurance premium and manage insurance Activities under this action area could include proceeds, and join supranational financial entities such as catastrophe risk pools. ŠŠ Technical assistance to support the development of financial protection Activities under this action area could include strategies, including diagnostic reviews of countries’ approach to financial protection, ŠŠ Knowledge exchange among countries to and identification of policy options for learn from experience in public financial strengthened financial resilience management of climate and disaster risks ŠŠ Technical assistance and investments to ŠŠ Technical assistance to incorporate support the implementation of national climate and disaster risks into public financial protection strategies, including finance frameworks for specific line ministries or sectors ƒƒ Develop new concessional financing for ƒƒ Support the development of pre-agreed catastrophe risk transfer instruments to disaster response plans backed by financial incentivize vulnerable developing countries protection strategies to help poor and to develop and adopt sustainable financial vulnerable households and protect key lifeline protection strategies. infrastructure. Such plans can help raise Activities under this action area could include awareness of the benefits of risk reduction and financial protection by engaging a wide range of ŠŠ Cofinancing of capitalization and operating stakeholders, including members of civil society. costs of catastrophe risk pools Activities under this action area could include ŠŠ Cofinancing of premiums for insurance solutions (designed to incentivize ŠŠ Knowledge exchange to learn from countries to progressively increase their experience and consolidate good practices contributions over time) in disaster response planning 64 SOVEREIGN CATASTROPHE RISK POOLS To achieve the overarching objective of reducing comparative advantages of all partners and build the protection gap in vulnerable developing on the work of existing platforms and initiatives. countries, and to catalyze action around these In particular, it would leverage the technical priority areas and activities, the G20 could expertise and capacity of the private insurance promote the creation of a Global Partnership and reinsurance industry. for Climate and Disaster Risk Finance and Insurance Solutions. The G20 could develop a work program structured around the four priority action areas The Global Partnership could bring together identified above to specify how countries would relevant partners from developing and developed support specific activities. Such efforts would not countries, international organizations, the private only promote financial protection and help close sector, and civil society. To achieve maximum the protection gap, but would also support the impact, the Global Partnership would leverage the broader disaster and climate resilience agenda. SOVEREIGN CATASTROPHE RISK POOLS 65 GLOSSARY basis risk. The risk that the index measurement an ex ante instrument that allows borrowers to will not match actual losses. prepare for a natural disaster by securing access to financing before a disaster strikes. budget allocation. An amount of funding set aside to cover specific planned expenditures. contingent liability. A potential future expenditure. In the context of disaster risk management, a In the case of disaster risk, a government’s or budget allocation can be made so that it can be organization’s contingent liability is a random accessed only in the event of a disaster. variable denoting the liability contingent on potential disaster events. capital base. Money contributed by the shareholders who first purchased shares in a disaster risk finance. The financial protection company, plus retained earnings. of populations against disaster events. Disaster risk finance strategies allow national and capital market instrument. Any financial contract local governments, homeowners, businesses, that can be structured to act as reinsurance, agricultural producers, and low-income but with investors, not reinsurers, providing the populations to respond more quickly and protection. Examples are catastrophe bonds and resiliently to disasters. catastrophe swaps. disaster risk management. The systematic catastrophe bond. An insurance-linked security process of using administrative directives, in which payment of interest and/or principal is organizations, and operational skills and suspended or canceled in the event of a specified capacities to implement strategies, policies, and catastrophe such as an earthquake. improved coping capacities in order to lessen the adverse impacts of hazards and the possibility of catastrophe swap. A contract used by investors disaster. to exchange (swap) a fixed payment for a certain portion of the difference between insurance disaster risk reduction. The concept and practice premiums and claims. of reducing disaster risks through systematic efforts to analyze and manage the causal factors contingency fund. A reserve fund designated of disasters, including through reduced exposure for financing disaster losses. Allocations to the to hazards, reduced vulnerability of people and contingency fund can be made through budget property, wise management of land and the allocations of national or local governments, environment, and improved preparedness for international agencies, communities, or a adverse events. combination of these. Funds are made available immediately after a disaster and are disbursed emergency recovery phase. The disaster using clear and simple rules. response phase that follows the emergency relief phase. During recovery, initial relief efforts have contingent credit. A financial tool that provides been completed; typically people have access to governments with immediate access to funds food, water, and temporary shelter, and children following disaster events to enable a more rapid are able to attend school. The recovery phase can and efficient response. This type of financing last several weeks or months, depending on the is typically used to finance losses caused by initial situation of the country. recurrent natural disasters. Contingent credit is 66 SOVEREIGN CATASTROPHE RISK POOLS emergency relief phase. The disaster response natural disaster. A disastrous event leading to phase that begins immediately after a disaster. loss of lives and livelihoods caused by natural During this phase, key objectives include ensuring hazards such as tropical cyclones, earthquakes, food security, shelter, and medical care. The duration floods, and landslides. of the relief phase depends on the initial situation of the country following the disaster event. parametric insurance. A type of insurance that does not indemnify the pure loss but agrees ex ex ante. Latin for “from before.” In the context ante to make a payment upon occurrence of a of disaster events, ex ante instruments are triggering event. The triggering event is often a arranged, and ex ante decisions are made, before catastrophic natural event likely to cause a loss. an event takes place. post-disaster needs assessment. A government- ex post. Latin for “from after.” In the context led exercise that assesses post-disaster needs, of disaster events, ex post instruments are generally in order to provide a platform enabling arranged, and ex post decisions are made, after the international community to assist the affected the event takes place. government in recovery and reconstruction. global humanitarian system. The network of public financial management. The systematic interconnected institutional and operational process designed to ensure that money is spent entities through which humanitarian assistance and accounted for in a clear and transparent is provided when local and national resources fashion. A public financial management system are insufficient to meet the needs of the affected comprises resource generation, resource population. allocation, and expenditure management (resource utilization). humanitarian aid. In general terms, the aid and action designed to save lives, alleviate suffering, reinsurance. A practice in which insurers transfer and maintain and protect human dignity during portions of risk portfolios to other parties in order and after man-made crises and natural disasters. to reduce the likelihood of having to pay a large Such aid may also be used to prevent and obligation resulting from an  insurance  claim—in strengthen preparedness for the occurrence of other words, insurance of insurance. such situations. risk-based pricing. Pricing of an insurance policy indemnity insurance. An insurance policy that to reflect the underlying risk that is transferred pays claims based on the actual economic losses through the insurance contract. incurred by the policyholder. risk pool. An arrangement whereby several index insurance. An insurance policy that pays individuals, companies, or countries jointly insure claims based on an index . Indexes are typically against a certain prespecified risk. chosen to be a good proxy of the economic losses incurred by the policyholder. risk retention instrument. An instrument whereby a party retains the financial responsibility for loss individual loss adjustment. The process by which in the event of a shock. Although risk retention a loss adjuster objectively assesses the actual instruments do not take risk off the balance damage for each insured building or injured person. sheet—the cost of a disaster must still be repaid— they do offer more flexibility in how and when to pay moral hazard. In the context of insurance, the that cost. Contingency funds, budget allocations, problems generated when the insured’s behavior and lines of contingent credit are all risk retention can influence the extent of damage that qualifies instruments, as are budget reallocations, tax for insurance payouts. Examples are carelessness, increases, and post-disaster credit. fraudulent claims, and irresponsibility. SOVEREIGN CATASTROPHE RISK POOLS 67 risk transfer instrument. An instrument such as trigger. The event that must occur before a an insurance contract that passes on the risks particular insurance policy applies to a given loss. associated with a certain event from one party to For example, for weather index insurance, the another. For example, in disaster insurance the trigger is the weather measurement that causes financial risks associated with a disaster event the insurance policy to pay out, such as a certain are passed from the insured to the insurer. amount of cumulative rainfall. shock-responsive social protection. Social underwriting. The process of issuing an protection that has the ability to increase its insurance policy, thereby accepting a liability and caseload and/or its intensity of support in guaranteeing payment in case a loss occurs. response to catastrophic events. targeting. The process of selecting beneficiaries under a social safety net program. 68 SOVEREIGN CATASTROPHE RISK POOLS ANNEX 1. BRIEF DESCRIPTION OF EXISTING SOVEREIGN CATASTROPHE RISK POOLS CARIBBEAN CATASTROPHE RISK excess rainfall policies have a more frequent INSURANCE FACILITY (INCLUDING trigger level, though generally transfer a smaller CCRIF-CA) quantum of risk. CCRIF started as a Cayman-domiciled captive CCRIF is designed as business interruption–type insurer offering parametric earthquake and coverage, providing rapid liquidity to sovereigns to tropical cyclone insurance policies to the 20 meet immediate, unbudgeted needs in the weeks Caribbean Community (CARICOM) member and after a disaster. Given the basis risk inherent associate member states. CCRIF is owned by a in any parametric insurance contract (i.e., the purpose trust and operates to the benefit of the potential mismatch between modeled losses and participating countries. In 2014, it reconstituted actual losses), CCRIF coverage may not be well as a Segregated Portfolio Company, enabling the suited to covering specific infrastructure damage establishment of separate underwriting pools or long-term rebuilding programs; on the other with differentiated capital (cells). CCRIF now also hand, it can provide financial leeway to put in offers an excess rainfall policy and includes a place more cost-effective financing mechanisms cell dedicated to underwriting risk in the COSEFIN for full post-disaster recovery. CCRIF does not put countries (Central America plus the Dominican any formal constraints on use of payout funds, Republic and Panama). although it increasingly monitors their use. CCRIF offers modeled loss–type parametric When it is efficient to do so, CCRIF utilizes the policies, which input a hazard parameter (or international reinsurance markets to leverage parameters) as the sole variable to a catastrophe its own capital (initially provided as grants by risk model, are locked at the start of the bilateral and multilateral donors) to provide insurance period, and convert the hazard variable much greater aggregate coverage than would to loss. CCRIF’s models use a consistent gridded otherwise be possible. The objective nature of the exposure data set across all the perils it covers parametric contracts used for risk transfer, along and attempt to capture the large-scale impacts of with the diversification of the portfolio, allows natural catastrophes on national economies and for attractive pricing from international markets, governments. including capital markets. CCRIF’s coverage is customizable, with pricing CCRIF’s business targets include maintaining based solely on the quantum of risk transferred efficiency in operations, capturing a risk- (measured by expected loss and variability of appropriate return on its capital, committing those losses). Some limitations are put in place to to pay claims even for the most extreme (e.g., constrain the risk transfer transaction such that it 1-in-1,000-year) events, and offering the lowest targets that portion of the risk profile where such possible (yet actuarially sound) premium pricing insurance provides a cost-effective solution. This to its clients. CCRIF has lowered the long-term design generally provides coverage that triggers premium pricing metric as capital has been every 10 years or so, and provides larger payouts accumulated, and has also used short-term for larger events up to a limit at the 1-in-100- to premium discounting (implemented on an equal 1-in-200-year range. Earthquake policies tend to basis across the pool) to maintain participation have less frequent trigger levels (due to the lower and provide best value to its client countries. frequency of damaging earthquake events), while SOVEREIGN CATASTROPHE RISK POOLS 69 From 2007 to 2016, CCRIF has made a total of required to develop (and have certified by a group 21 payouts to 10 member countries amounting of experts and peers) a final implementation plan to US$68 million. All payments have been made for use of the payout funds before the payout is within 14 working days. In the case of Haiti, for made. This plan builds on an operations plan, example, payments of US$7.8 and US$23.4 which must be certified prior to the original million were made immediately after the 2010 purchase of the insurance coverage. While this earthquake and 2016 Hurricane Matthew, approach tends to make payouts from ARC respectively. occur on the time scale of a few weeks (that is, somewhat more slowly than they would AFRICAN RISK CAPACITY be otherwise), it is critical in ensuring that the benefits of early action are fully captured and that ARC comprises two entities: a treaty-based the most vulnerable people in a given situation international organization, the ARC Agency, which are reached and assisted. is a Specialized Agency of the African Union formed at the direction of union ministers of ARC Ltd. is capitalized through interest-free loans finance; and an affiliated insurance company, ARC and operates on mutual principles, such that any Ltd., which is domiciled in Bermuda as a mutual underwriting profit is used to accumulate capital captive. ARC Agency is ultimately under the control to the ultimate benefit of the premium-paying of the Conference of Parties, which appoints a clients. It has also established cost-efficient governing board, with operations undertaken by a access to the international risk markets, providing secretariat. ARC Ltd. has its own board of seven capacity to underpin expansion of ARC’s insurance directors and operates on commercial principles program. This is in addition to its efforts to build as the underwriter of policies to ARC Agency capacity in risk understanding, early warning, and member states. contingency planning, which are critical to its long- term sustainability. ARC Ltd. offers parametric insurance policies for the key climate risks faced by African sovereigns, ARC Ltd. policies are constructed to pay out every namely drought and tropical cyclone; policies for three to five years, commensurate with the high riverine flood are in development. The drought frequency of droughts in many African countries. policy uses rainfall as the variable input parameter, Trigger levels for cyclone and flood policies will be and ARC’s in-house modeling platform, Africa somewhat higher, covering risk in the 1-in-10- to RiskView, converts that rainfall into an affected 1-in-100-year return period range. population estimate (in low-rainfall situations) and then calculates a response cost using Replica insurance coverage, which exactly assumptions about various types of response replicates an insurance policy paid for by an mechanisms (cash transfer and food aid, for ARC member state, is available to humanitarian example). ARV can be used by ARC member actors (international organizations and NGOs). states to model a country’s risk profile (as a Such policies are already customized to reflect basis for making decisions about risk financing) the country’s experience and, to the extent and as an early-warning tool. For cyclones, ARC possible, needs. has leveraged the experience of CCRIF and offers a very similar product to the exposed countries African countries, including those that purchase in the southwest Indian Ocean, including an early insurance from ARC, also have access to technical warning component embedded in ARV (which assistance and capacity building on disaster CCRIF also offers to its clients). risk finance and public financial management of natural disasters provided by the World Bank, ARC was the first pool to institutionalize an with financial support from the ACP-EU (African, incentive mechanism for contingency planning Caribbean, Pacific–European Union) Natural among the participating countries. Countries Disaster Reduction Program. receiving a parametric payout from ARC Ltd. are 70 SOVEREIGN CATASTROPHE RISK POOLS PACIFIC CATASTROPHE RISK Like CCRIF and ARC, the PCRAFI insurance ASSESSMENT AND FINANCING program aims not to cover the full losses incurred INITIATIVE (PCRAFI) but rather to provide rapid, flexible funds within weeks of an event for use as budget support The Pacific catastrophe risk insurance pilot by the affected countries. Given this objective, program was launched under PCRAFI in January a parametric contract was considered suitable, 2013, when the World Bank intermediated a since these triggers lend themselves to rapid loss portfolio of catastrophe swap contracts on the determination after an event—typically within international reinsurance markets that transferred a couple of weeks of an event. Contracts were catastrophe risk from five Pacific island countries. designed to cover catastrophe risk for events with The pilot was supported by the government of a return period of 10 years and above. Recognizing Japan, the Global Facility for Disaster Reduction that insurance should fit within a framework for and Recovery, the World Bank Group, and the disaster risk financing, such that lower-return- Secretariat of the Pacific Community (SPC). period (and other out-of-scope) events not covered The placement built on several years of prior by policies can also be managed, the PCRAFI work under PCRAFI on catastrophe risk models insurance program is accompanied by a technical (developed and run by the risk modeling firm AIR assistance program on disaster risk finance and Worldwide) that use a unique database (the Pacific insurance and on public financial management Risk Information System) for risk transfer to the of natural disasters, under which post-disaster international market. The Pacific Risk Information budget execution manuals have been developed. System is the largest regional database and Since its inception, the PCRAFI insurance program contains information on the hazard, exposure, has made two payouts for an aggregate amount and vulnerability of 15 Pacific island countries, of US$3.2 million, both times within 10 days of including information on over 4 million buildings the disaster. Tonga received a payout of US$1.3 and their attributes. million within 10 days of being struck by Tropical Cyclone Ian in 2014. The funds were mainly used In its first season, the pilot program placed to purchase fuel for boats bringing emergency US$45 million of catastrophe risk for the Marshall goods to the affected islands. Vanuatu received Islands, Samoa, Solomon Islands, Tonga, and a payout of US$1.9 million within seven days of Vanuatu. The pilot was renewed in its original being struck by Tropical Cyclone Pam in 2015; the form for three subsequent seasons; in its fourth funds were mainly used to bring nurses to the season, the newly established PCRAFI facility is affected areas to provide emergency care. The taking a role in the risk transfer. The Pacific Risk program has also progressed from its initial model Information System plays a pivotal role in the of fully subsidized premiums to the current model insurance triggers, which are based on modeled of cofinancing from countries. For the 2016/17 losses. Reported hazard parameters for tropical season, participating countries have taken on full cyclone, earthquake, or tsunami events were responsibility for payment of premiums, some of used to create event footprints in the catastrophe them using IDA loans to finance the payments. risk models, and from this a modeled loss was The willingness of countries to move away from determined. The selection of a modeled loss fully subsidized premiums, and the decision of trigger—unlike the simpler hazard indexes that the Cook Islands to join without any premium had also been considered—allowed the inclusion support, are strong indicators of the program’s of the tsunami peril, and generated a more refined value to countries. loss estimate to reduce basis risk for countries. The portfolio of policies was expanded with the addition of the Cook Islands in the second season, while the Solomon Islands withdrew in the third season. SOVEREIGN CATASTROPHE RISK POOLS 71 SOVEREIGN CATASTROPHE RISK In 2009, the government of Mexico and the TRANSFER: THE EXPERIENCE OF World Bank launched the MultiCat Program,12 a MEXICO catastrophe bond–issuance platform that aimed to facilitate and lower the cost of catastrophe risk The Fund for Natural Disasters was established in transfer for governments and public entities. The 1996 by the Federal Government of Mexico as a thinking was that by using common documentation mechanism to finance the post-disaster recovery and legal and operational frameworks, developing and reconstruction of Mexico’s public assets countries exposed to natural disasters would and low-income housing. FONDEN consists of have more affordable access to capital markets, three primary financial accounts: (i) the FONDEN and investors would be able to pool multiple Program for Reconstruction; (ii) the FONDEN perils and regions in order to achieve better Trust; and (iii) the Revolving Fund. Collectively, portfolio diversification. Although other members these instruments assist the government of of the Pacific Alliance (Chile, Colombia, and Peru) Mexico in its efforts to respond quickly to natural have expressed interest in using this platform, so disasters by providing funding for emergency far only Mexico has used it to issue catastrophe relief, rehabilitation, and reconstruction. These bonds. instruments are continuously changed to enhance their efficiency and effectiveness. FONDEN has issued three CAT bonds over the period 2006–2015, covering hurricane and The FONDEN Program for Reconstruction provides earthquake risks in several high-risk areas of the financial support to rehabilitate and reconstruct country. The last outstanding issue (2012–15) assets destroyed by natural disasters. It focuses expired in early 2016, and an extension was on the reconstruction and restoration of (i) public granted to receive the only claim to the notes infrastructure at the three levels of government in the history of the program (the program had (federal, state, and municipal); (ii) low-incoming triggered a partial payout after Hurricane Patricia housing; and (iii) forestry, protected natural areas, in 2015). The program’s price decreased 27 rivers, and lagoons. percent as the market was becoming more comfortable with the underlying risks. It should be noted that while the FONDEN Program for Reconstruction is not strictly speaking an In addition, FONDEN has purchased traditional insurance mechanism (e.g., Mexican states do not indemnity-based reinsurance to cover public pay an insurance premium), it uses the principles buildings and infrastructure. Since 2010 Mexico of insurance to finance the reconstruction of has transferred contingent obligations of the public assets: a transparent damage reporting FONDEN program to a panel of more than 40 system, clear rules for how funds are disbursed, a reinsurers. The reinsurance contract uses public clear plan for how money is spent, and a credible loss adjustment rules set up by FONDEN, and monitoring system for expenditures. loss adjustment is conducted by private loss adjusters and government staff. This innovative The Federal Budget Law requires that no less combination has allowed for more discipline than 0.4 percent of the annual federal budget and transparency in the evaluation of damage should be allocated for FONDEN and related to public assets and more efficient pricing. (See activities through a dedicated budget line item. table A1.1 for a summary of FONDEN’s risk To further manage the volatility of the FONDEN transfer program). budget and to leverage its resources, FONDEN is allowed to transfer disaster risks through insurance and other risk transfer mechanisms such as catastrophe bonds. 12. Mexico issued its first CAT bond in 2006 with technical assistance from the World Bank. 72 SOVEREIGN CATASTROPHE RISK POOLS TABLE 4. FONDEN RISK TRANSFER PROGRAM, 2006–2015 (US$ MILLION) 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 CAT bondsa Cover (sum insured) 150 150 150 290 290 290 315 315 315 Premiums paid 8.5 8.5 8.5 31.5 31.5 31.5 27.8 27.8 27.8 Claims received 50 Reinsurance Cover (sum insured) 386 418 392 376 85.2 Premiums paid 75.6 79 71.7 72.5 58.7 Claims received 124 Source: Secretaría de Hacienda y Crédito Público. a. In 2006–08, CatMex was a mix of CAT bonds (US$150 million) and alternative reinsurance (US$300 million). During the summer of 2016, the ministers of services in risk modeling (using the catastrophe finance of three Mexican states (Oaxaca, Colima, risk model R-FONDEN developed by the federal and Hidalgo) established a joint committee to government with technical assistance by the risk coordinate the purchase of catastrophe risk modeling firm ERN), as well as in catastrophe insurance. This effort was led and supported by risk assessment, insurance product design, the federal government as a natural expansion and drafting of insurance policies. The IRMS of FONDEN’s technical support to the Mexican standards have brought several benefits to states. states. As a result of accessing the international For example, the state of Oaxaca has found that reinsurance markets together with FONDEN, the its IRMS not only helped it purchase insurance on states were able to increase their insurance better terms, but also provided crucial information coverage by 20 percent and reduce their insurance to strengthen its disaster response capacity and premium by 30 percent. emergency relief strategy. This collaborative effort aims to help the The pool of Mexican states builds on the state governments develop an Integral Risk experience of FONDEN in developing and placing Management Strategy (IRMS) so that they can catastrophe risk insurance. At least two more meet requirements for federal fundin g (such states have expressed interest in doing some as high data quality and a methodological of the preparatory work involved in joining the approach to risk identification, measurement, pool. In the next stage, it is expected that the and management that follows FONDEN rules). participating states will build joint reserves to It provides the state governments with advisory further improve insurance coverage and pricing. SOVEREIGN CATASTROPHE RISK POOLS 73 ANNEX 2. SOVEREIGN CATASTROPHE RISK POOLS: ANNUAL PORTFOLIOS (AS OF DECEMBER 16, 2016) Insurance Participating countries Number of Aggregate Aggregate Aggregate period participating insurance coverage insurance countries premium limit (US$ payouts volume million) (US$ million) (US$ million) Caribbean Catastrophe Risk Insurance Facility (Caribbean cell)–CCRIF 2007/08 Anguilla, Antigua and 16 19.5 494.8 0.9 2008/09 Barbuda, Bahamas, 16 21.8 563.8 6.3 Barbados, Belize, Bermuda, 2009/10 Cayman Islands, Dominica, 16 21.5 601.2 7.8 Grenada, Haiti, Jamaica, 2010/11 St. Kitts and Nevis, St. 16 20.8 618.4 17.2 Lucia, St. Vincent and the 2011/12 Grenadines, Trinidad and 16 20.0 624.4 - Tobago, Turks and Caicos 2012/13 Islands 16 20.3 624.5 - 2013/14 16 19.5 618.8 - 2014/15 16 23.1 656.8 3.4 2015/16 16 31.1 723.9 2.4 2016/17 Anguilla, Antigua and 14 27.7 697.5 29.5 Barbuda, Belize, Barbados, Cayman Islands, Dominica, Grenada, Haiti, Jamaica, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Trinidad and Tobago, Turks and Caicos Islands Caribbean Catastrophe Risk Insurance Facility (Central American cell)–CCRIF-CA 2015/16 Nicaragua 1 1.0 18.0 - 2016/17 Nicaragua 1 1.5 28.2 1.6 74 SOVEREIGN CATASTROPHE RISK POOLS Insurance Participating countries Number of Aggregate Aggregate Aggregate period participating insurance coverage insurance countries premium limit (US$ payouts volume million) (US$ million) (US$ million) African Risk Capacity (ARC) 2014/15 Kenya (2 seasons), 3 17 129 26.1 Mauritania, Niger, Senegal 2015/16 The Gambia, Kenya (2 7 25 179 8.1 seasons), Malawi, Mali, Mauritania, Niger, Senegal 2016/17 Burkina Faso, The Gambia, 6 11.3 100 - Mali, Mauritania, Niger, Senegal Pacific Catastrophe Risk Assessment and Financing Initiative (PCRAFI) 2012/13 Marshall Islands, Samoa, 5 1.5 45 - Solomon Islands, Tonga, Vanuatu 2013/14 Marshall Islands, Samoa, 5 2.2 67 - Solomon Islands, Tonga, Vanuatu 2014/15 Cook Islands, Marshall 5 1.3 43 1.3 Islands, Samoa, Tonga, Vanuatu 2015/16 Cook Islands, Marshall 5 2.3 43 1.9 Islands, Samoa, Tonga, Vanuatu 2016/17 Cook Islands, Marshall 5 2.3 38 - Islands, Samoa, Tonga, Vanuatu SOVEREIGN CATASTROPHE RISK POOLS 75 REFERENCES ADB (Asian Development Bank). 2013. “Regional: S&P Global. 2015b. “Storm Alert: Natural Disasters Strengthening Disaster and Climate Risk Resilience Can Damage Sovereign Creditworthiness.” in Urban Development in the Pacific.” Technical September 10. https://goo.gl/JtZPS4. Assistance Consultant’s Report. Project Number ———. 2015a. “The Heat Is On: How Climate Change 46162. Can Impact Sovereign Ratings.” November 25. Baez, de la Fuente, and Santos. 2010. https://goo.gl/sccYb5. Bresch, D., and L. Müller. 2014. “A Global Overview Swiss Re. 2016. “Natural Catastrophes and Man- of Case Studies with a Focus on Infrastructure: Made Disasters in 2015: Asia Suffers Substantial Economics of Climate Adaptation—Shaping Losses.” Sigma no. 1. March 30. http://media. Climate-Resilient Development.” Swiss Re, Zurich. swissre.com/documents/sigma1_2016_en.pdf. Cabot Venton et al. 2012 UN-HABITAT. 2013. Clarke, Daniel, Sarah Coll-Black, Naomi Cooney, and UNISDR (United Nations Office for Disaster Risk Anna Edwards. 2016. “A Methodology to Assess Reduction). 2015. Making Development Indicative Costs of Risk Financing Strategies Sustainable: The Future of Disaster Risk for Scaling Up Ethiopia’s Productive Safety Net Management. Global Assessment Report on Programme.” Policy Research Working Paper 7719, Disaster Risk Reduction. https://www.unisdr.org/ World Bank, Washington, DC. we/inform/publications/42809. Clarke, Daniel, Olivier Mahul, Richard Poulter, and Wiseman, W., and U. Hess. 2007. “Reforming Tse-Ling Teh. 2016. “Evaluating Sovereign Disaster Humanitarian Finance in Ethiopia: A Model for Risk Financing Strategies: A Framework.” Policy Integrated Risk Financing.” United Nations World Research Working Paper 7721, World Bank, Food Programme Working Paper. Washington, DC. World Bank. 2012. Turn Down the Heat: Why a 4°C Clarke, Daniel J., and S. Dercon. 2016. Dull Disasters? Warmer World Must Be Avoided. Washington, DC: How Planning Ahead Will Make a Difference. Oxford: World Bank. Oxford University Press. ———. 2013. Turn Down the Heat: Climate Extremes, Clarke, Daniel J., and Ruth Vargas Hill. 2013. “Cost- Regional Impacts, and the Case for Resilience. Benefit Analysis of the African Risk Capacity Washington, DC: World Bank. Facility.” Discussion paper, International Food Policy ———. 2014a. “Honduras and Nicaragua Research Institute, Washington, DC. Catastrophe Risk Insurance Project. De Janvry, del Valle, and Sadoulet. Project Appraisal Document.” Report No. Ghesquiere and Mahul. 2010. PAD994. http://projects.worldbank.org/ Hallegatte et al. 2013. P149895/?lang=en&tab=documents& Hallegatte, Stephane, Mook Bangalore, Laura subTab=projectDocuments. Bonzanigo, Marianne Fay, Tamaro Kane, Ulf ———. 2014b. Turn Down the Heat: Confronting the Narloch, Julie Rozenberg, David Treguer, and Adrien New Climate Normal. Washington, DC: World Bank. Vogt-Schilb. 2016. Shock Waves: Managing the ———. 2016. Annual Report 2016. World Bank. http:// Impacts of Climate Change on Poverty. Climate www.worldbank.org/en/about/annual-report. Change and Development. Washington, DC: World World Bank Group. 2014. “Financial Protection Against Bank. Natural Disasters: An Operational Framework for Hallegatte, Stephane, Adrien Vogt-Schilb, Disaster Risk Financing and Insurance.” World Mook Bangalore, and Julie Rozenberg. Bank, Washington, DC. 2017. Unbreakable: Building the Resilience of World Bank and Government of Thailand. 2012. the Poor in the Face of Natural Disasters. Climate World Bank and Rockefeller Foundation. 2016. Change and Development. Washington, DC: World “Toward a Regional Approach to Disaster Bank. Risk Finance in Asia.” Discussion Paper. IPCC (Intergovernmental Panel on Climate Change). http://documents.worldbank.org/curated/ 2013. en/584961480930535198/pdf/110702-WP- DRFRockefellerFINAL-PUBLIC.pdf. 76 SOVEREIGN CATASTROPHE RISK POOLS