TOWARD A REGIONAL APPROACH TO DISASTER RISK FINANCE IN ASIA DISCUSSION PAPER MAY 2016 Contact Olivier Mahul, Global Lead, Disaster Risk Finance Program Manager Disaster Risk Financing and Insurance Program The World Bank Group omahul@worldbank.org Asia is at high risk of catastrophic disaster and climate shocks that cause damage and erode welfare and economic gains. Financial protection strategies have been recognized by countries and their development partners as important tools to protect countries from these e ects and to thereby support them in reducing poverty and increasing shared prosperity. There is currently an active international dialogue on establishing a regional initiative for disaster risk nancing (DRF) for Asia. The topic of an Asian disaster risk nancing facility has been raised in various forums, including the Association of Southeast Asian Nations (ASEAN), Asia-Paci c Economic Cooperation (APEC), the G-7 InsuResilience Initiative, and the Vulnerable 20 Group of Ministers of Finance (V20). This note aims to connect international dialogue with experience on the ground to provide a basis for consultation with countries, donors, and private partners. C H I N A 2008 Sichuan Earthquake 69,195 dead $148 billion MYANMAR PHILIPPINES 2008 2013 Cyclone Nargis Typhoon Yolanda 138,366 dead 6,340 dead $4 billion $12.9 billion NEPAL 2015 Earthquake Over 8,635 dead $10 billion INDIAN OCEAN 2014 Tsunami Over 230,000 dead THAILAND JAPAN 2011 2011 $15 billion Floods Earthquake & Tsunami 815 dead 15,894 dead $45.7 billion $300 billion Asia has seen some of the deadliest and costliest disasters in history. The nature of hazards faced across the region varies significantly, with flood as the most persistent hazard. Analysis indicates that average annual economic losses from Asian flood disasters could surge to US$500 billion or more by 2050. Half of all ASEAN member states have experienced at least one severe flood event costing over US$100 million in the past decade. Countries also face potentially devastating droughts, tropical cyclones, earthquakes, tsunamis, and volcanic eruptions. These di erent hazards generate di erent nancing needs in the aftermath of a disaster. While some governments may deal principally with costs for response and recovery after rapid onset disasters, others may need to fund livelihood assistance and food security responses to ood- or drought-damaged crops, while others still may be dealing with reconstruction of critical infrastructure after earthquake, ooding, or storm damage. Some countries have to be prepared to deal with many di erent types of response costs. Di erences in economic development further amplify the di erent post-disaster nancing needs. The nature of costs arising from disasters varies according to the type of population, asset, or economic activity exposed. Countries with higher levels of economic development will have higher values of assets exposed, and are often concerned with large losses that occur through damage to infrastructure and economic disruption. The 2011 Thailand oods, for example, caused an estimated US$47 billion of damage and losses, of which 70 percent came through the manufacturing subsector. While post-disaster assistance for households was required, the largest nancial challenges faced by the government related to reconstruction of physical assets after the initial emergency period, and to the signi cant drop in revenues due to manufacturing interruptions. Conversely, countries with large populations of vulnerable poor may be focused on post-disaster assistance that supports livelihoods, such as social safety nets. The biggest single contributor to losses from the 2005 earthquake in Pakistan came from damage to private housing, which accounted for almost 50 percent of the total cost of reconstruction. Although there was damage to infrastructure and costs through economic sectors, the bulk of the costs arose from the large humanitarian response to aid the 3.5 million people rendered homeless. Why DRF? To respond immediately and e ectively following a disaster—and to minimize the human, economic, and scal costs that increase rapidly when response is delayed or inadequate—Asian countries need access to su cient money and e ective mechanisms for use during an emergency. A strategy for nancial protection against disasters allows governments to increase their nancial capacity to $ respond to disasters, and can improve nancial inclusion for a ected households and businesses by giving them nancial tools to aid recovery. Such nancial mechanisms can also reduce the impact of disasters on social and economic development. They can smooth nancial shocks and prevent governments and populations from resorting to adverse coping mechanisms that disrupt development progress and livelihoods. Through these positive impacts, strategies for nancial protection against disasters can help to protect welfare and economic gains, thereby contributing to poverty reduction and shared prosperity. A strategy for nancial protection against disasters needs to encompass both sources of nancing and mechanisms for e ective disbursement of nancing in-country. Securing Money for Disaster Response Di erent post-disaster needs require di erent nancing mechanisms; an instrument that provides rapid liquidity to support emergency operations may not be appropriate for funding longer-term, large-scale reconstruction of damaged infrastructure. For example, a reserve fund nanced through the budget could be cost-e ective for dealing with frequently recurring ood events; but pre-funding larger losses from more extreme events through such a fund would incur high opportunity costs of ring-fencing large reserves for infrequent use. Insurance, on the other hand, can be costly to fund more frequent but lower-impact events, since premiums increase with the expected frequency of payouts; but for less frequent, catastrophic events, insurance can be an e ective way to increase nancial capacity beyond the national means. Mechanisms that retain risk (e.g., budgetary mechanisms and credit that keep risk in the budget or on the balance sheet). Mechanisms arranged ex-ante that Countries have various nancial disburse funds quickly (such as options available to them for contingent credit). dealing with post-disaster losses, each suitable for di erent contexts and priorities. Mechanisms that transfer risk (e.g., insurance or international assistance). Funding sources sought after disaster occurrence that may take longer to mobilize, but may provide a large volume of funds (such as reconstruction loans and grants from international partners). A more detailed introduction to DRF is provided in the paper “Financial Protection Against Disasters: An Operational Framework for Disaster Risk Financing and Insurance” (World Bank, 2014). Disbursing Money E ectively It is equally important to ensure that resources for disaster response reach the people who need them the most, when they need them the most. Responsibilities for post-disaster spending may be shared between central and local governments, such that state- or municipal-level entities may lead the nancial response on the ground. Structuring of subnational disaster risk nancing is therefore as important as mobilization of funds. The government of Mexico acknowledges the importance of both in its fund for natural disasters (FONDEN), which is governed by clearly de ned rules on allocation of post-disaster funding for recovery and reconstruction at the federal and state levels. The government of Ethiopia uses a social safety net (the Productive Safety Net Program) to deploy contingent nancing mobilized at the national level down to a ected households. In the event of a catastrophic drought, the safety net program can reach additional a ected households using locally held contingency budgets and a federal-level contingent nancing window. $ $ $ $ $ $ $ $ $ $ $ $ $ The private sector plays an important role in absorbing post-disaster costs and in ensuring that households receive nancial assistance after a disaster. $ $ $ The private sector’s importance was exempli ed after the 2011 Thailand oods. According to World Bank estimates, domestic insurers absorbed losses of around 25 percent of TB 1.4 trillion total economic losses (approx. US$40 billion). One key government response to the disaster was to establish a national Natural Catastrophe Insurance Fund as a back-stop for insurers. The fund was designed to help insurers cover costs should another major natural disaster event strike. The government of Turkey has also focused on using the private sector to provide disaster-a ected households with nancial assistance. The Turkish Catastrophe Insurance Pool is a national program of a ordable earthquake insurance that functions as a public-private partnership. It was established following the devastating 1999 Marmara earthquake, and now covers millions of homeowners in Turkey. Low insurance penetration, however, hampers such a response through the private sector in many Asian countries. Figure 1. Non-life Insurance Penetration as a Percentage of GDP in Asia (selected economies) Figure 2 Source: Based on 2014 data from Swiss Re Sigma and 2013 and 2014 data from AXCO. Disaster Risk Finance in Asia Several Asian countries are already active in strengthening their nancial resilience and any regional initiative should build on existing work. Some countries have already been working on national strategies to better manage the cost of disasters. Annex I provides a summary of progress made by a subset of Asian countries, as identi ed through ongoing World Bank engagements. This summary is not meant to be exhaustive and should be expanded following further consultations with countries and other (donor and private) partners. Knowledge gained through World Bank support to countries, dialogue in the region, and recent disasters suggests that countries’ current priorities for use of nancial instruments can be placed in three broad categories: 1 Access to rapid disaster response 2 Property catastrophe risk 3 Property catastrophe risk nancing (and deployment of insurance for public assets. insurance for private assets. rapid response funds at the subnational level and to households). Rapid response nancing here refers to nancing released within days or weeks of a disaster occurring. This feature tends to be o ered by contingent instruments such as the World Bank’s contingent credit (Cat DDO ) o ering, insurance products such as the Paci c Catastrophe Insurance Pilot, contingent grant windows such as the federal-level contingent nancing window linked to the Ethiopia productive safety net program described above, and donor emergency response facilities. 1. Rapid Response Financing Access to rapid response nancing is a priority issue for economies at risk of severe shocks, such as large earthquakes and tropical cyclones (e.g. Indonesia, the Philippines, Nepal, Myanmar); but it is also important for economies that are exposed to recurrent ooding but have limited resources to respond (e.g. Myanmar, Cambodia, the Lao People’s Democratic Republic, Sri Lanka). The Philippines has already implemented and used disaster-contingent credit instruments with the World Bank (a Catastrophe Deferred Drawdown Option, or Cat DDO; US$500 million) and with the Japan International Cooperation Agency (JICA) (SECURE; US$500 million); it has just nalized a second World Bank Cat DDO (US$500 million) and also is evaluating sovereign catastrophe risk insurance as an option. Indonesia has carried out analysis to develop a national strategy for nancial protection against disasters, including a possible disaster reserve fund. Myanmar, Cambodia, and Lao PDR are in the early phases of exploring sub-regional pooling of risk similar to the Caribbean Catastrophe Risk Insurance Facility (CCRIF) or the Paci c PCRAFI catastrophe insurance program. Sri Lanka has recently put a disaster-contingent credit product in place with the World Bank (Cat DDO). Nepal learned from the devastating April 2015 earthquake of an identi ed gap in rapid response nancing and e ective mechanisms to deploy funding once it is received. Post-disaster rapid response nancing may be mobilized for very di erent purposes across countries. For some, rapid response nancing is valuable as bridge nancing that helps to avoid budget disruption. For others, it allows urgent action (e.g., emergency services, nancial assistance to a ected populations) in the face of severe budget constraints. How such nancing is used will largely be determined by the size of the economy and by budget exibility. Countries may opt to connect rapid response nancing instruments to subnational disaster risk nancing mechanisms, such as local-level disaster funds in Pakistan and the Philippines, or the scalable social safety nets used in Ethiopia. Subnational-level strategies for nancing disaster risk are particularly relevant in Asia, given the region’s increasing urbanization and megacities. For example, strategies at the municipal level can be designed to make allowances for the uneven distribution of exposure within Asian countries. Some of the larger economies (e.g., the Philippines, Pakistan) are focusing on programs for subnational disaster risk nancing that better structure how funds ow to provinces or local government units. The government of the Philippines is in advanced technical preparation of a program of subnational catastrophe risk nancing that will allow local government units to better structure liability and scal transfers, including through a joint catastrophe insurance fund. The government of Pakistan has legislated for the creation of provincial funds for disaster risk management (DRM) administered by provincial governments. Technical work is currently planned to operationalize at least one of these subnational funds. The idea of state-level nancial mechanisms for emergency response to complement state disaster funds has been proposed for India as part of technical discussions, although this is in very early stages of consideration. 2. Public Asset Insurance Catastrophe risk insurance for public assets is a focus for a growing number of countries as they invest in new infrastructure and buildings. In Indonesia and Vietnam, for example, governments are beginning to develop public asset insurance programs. Property catastrophe risk insurance for public assets has already been utilized in the Philippines, and the government is seeking to increase insurance uptake even further. 3. Private Insurance Property catastrophe risk insurance for private assets is becoming a focus for countries where a growing middle class and small businesses are at risk of property damage from disasters. Thailand, for example, set up a National Catastrophe Insurance Fund to shore up the domestic insurance industry following the devastating 2011 ooding. The Philippines is also in the early stages of exploring ways to reach more households and businesses with catastrophe risk insurance through the insurance market. Toward a Regional Disaster Risk Finance Facility for Asia A regional facility for disaster risk nance for Asian countries could contribute substantially to nancial resilience in the region. International experience shows that regional collaborations have opened up access to nancing sources that would not otherwise be available to countries, and have helped countries to smooth the volatility of the cost of disasters and to better structure their nancial response. Initiatives such as the CCRIF and the Paci c PCRAFI insurance program have also demonstrated that regional platforms on disaster risk nancing and insurance have the potential to confer bene ts well beyond increased access to nancing. These regional initiatives have also served as forums for sharing of knowledge and good practice, as vehicles for shared investment in public goods to support understanding of risk, and as engines of political momentum—driving engagement and progress on better managing disaster and climate shocks. Any regional risk nancing solution should build on and strengthen the national priorities of individual Asian countries. Figure 2 shows an initial assessment by World Bank sta of where countries are currently focusing their engagement in disaster risk nancing and what their needs are. Figure 2. Identi ed Country Priorities on Disaster Risk Financing Source: World Bank-GFDRR Disaster Risk Financing and Insurance Program (2015). Any proposal for a regional risk nancing platform should be evaluated against such priority areas and revised jointly with countries in order to provide improved access to the following: • Rapid disaster response nancing for the national government, complementing existing resources. • Rapid disaster response nancing for subnational entities. • Insurance of public assets to transfer risk from the government to dedicated risk carriers. • Private property catastrophe insurance markets that provide access to a ordable and sustainable catastrophe risk insurance for homeowners and enterprises. In Asia, such a platform should be exible enough to accommodate the di erent post-disaster nancing needs arising from the heterogeneity in natural hazards, exposure, and vulnerability across the region. The platform would have to accommodate the needs of both large and small economies (see box 1). Such a platform could prioritize the needs of countries dealing with ood, given that ooding is the most prevalent hazard in the region, but would also have to address the needs of countries struggling with losses from infrequent severe shocks such as earthquakes and tropical cyclones. To add value to countries’ e orts in building nancial resilience, it would need to serve countries focused on assisting vulnerable rural populations as well as countries focused on reconstructing homes and infrastructure. Box 1: Small versus Large Economies: Financial Capacity for Major, Infrequent Events Countries in Latin America and the Caribbean are highly exposed to losses from earthquake, tropical cyclone, and ooding, but their experiences in seeking to access external nancial capacity for post-disaster spending have di ered depending on their size. Mexico—an upper-middle-income country with a relatively large economy—sought to transfer a large amount of risk and was able to do so through a single approach to the international markets. Mexico’s fund for natural disasters (FONDEN) is therefore backed by an excess-of-loss reinsurance program that provides in excess of US$400 million of nancial capacity, while a catastrophe bond provides US$300 million of additional capacity. For smaller countries seeking to place smaller amounts of risk, a single approach to the markets may not be feasible or cost-e ective. The Caribbean Catastrophe Risk Insurance Facility has helped Caribbean countries and more recently Central American countries to access external nancial capacity for post-disaster spending arising from hurricanes and earthquakes. Sixteen Caribbean countries have pooled their risk in CCRIF, which is essentially a joint reserve mechanism that accrues premiums and joining fees from countries and pays out to members when a disaster occurs. As an aggregator of risk, the facility has been able to access risk-bearing capital on the international markets in the form of both reinsurance and catastrophe swap contracts. Although individual country policies with CCRIF are typically of the order of tens of millions of dollars, the reinsurance and catastrophe swap program provides over US$100 million of coverage. Countries have bene ted from lower premiums, from the diversi cation bene ts of risk pooling, and also from shared transaction costs. The facility was originally established with the technical assistance of the World Bank, and with initial funding of more than US$60 million from donors to provide claims-paying capacity in the early years. The success of any regional risk nancing facility will depend in equal measures on political and institutional arrangements as its technical foundation. Any proposal for a regional risk nancing facility will need a strong regional partner organization to support the political and policy coordination between participating governments. This could take various forms. For example, a regional organization could serve as a convening body to ensure coordination, could host a small technical secretariat for policy coordination while the facility maintenance is outsourced to specialist service providers, or could establish a regional facility as an a liate body of the regional organization. An appropriate organization or organizations and their respective roles will need to be identi ed in consultation with interested governments. A regional facility for Asia should build on existing work at the national level to develop strategies for nancial protection against disasters and to ensure e ective deployment as well as mobilization of funds. Financing mobilized within a regional facility must be connected to mechanisms in-country that disburse funds, such as scalable social safety nets/cash transfer schemes or subnational disaster risk nancing mechanisms like those planned in Pakistan and the Philippines. And any regional platform should o er opportunities to develop resilience beyond mobilization of funds—by serving as a platform for capacity building, common investment, and dialogue. Figure 3. Possible Interaction between National DRF Strategy and a Regional Facility INTERNATIONAL INSURANCE INTERNATIONAL MARKETS COMMUNITY ASIA REGIONAL RISK FINANCING FACILITY NATIONAL DISASTER RISK FINANCE STRATEGY For example: National DRM Fund Social Public asset protection insurance system program Subnational Property governments insurance pool insurance pool The proposed structure would enhance national strategies ( gure 3), and funds mobilized through the facility could pass through national reserve funds and potentially into local-level disbursement mechanisms such as scalable social protection networks (as in Ethiopia) or down to subnational government entities (such as cities or provinces). Groups of subnational entities such as cities could also approach the platform directly as part of a national strategy, for collective coverage. Desired outcomes from a regional facility should include: Improved understanding of risk. Standardization of risk data and policy terms across countries is key for the establishment of a regional facility. This initiative can provide an incentive for economies to work together to better understand their risks and to invest in risk data, which can then also serve to inform risk reduction and prevention. This could also lead to the development of global public goods, such as regional risk databases and regional cat risk models, to support the establishment and operations of such a facility. Comprehensive financial protection. A regional facility could be an umbrella to national strategies, o ering additional options for nancing to supplement national capacity, and working in S complement to mechanisms in-country to disburse nancing (such as cash transfer schemes) and thereby reach a ected households and businesses. This regional mechanism could be used to drive the design and implementation of comprehensive national strategies. Improved financial disaster planning. A facility could promote the development/improvement of national contingency planning to allow for a timely and cost-e ective use of funds post disaster. Reduced reliance on disruptive budget reallocations or uncertain humanitarian assistance. A facility could provide rapid funds in the immediate aftermath of a disaster. The facility could include a joint reserve mechanism alongside risk transfer to access international market capacity. Cost savings through diversification of catastrophe risks. Through catastrophe risk pooling $ across countries with di erent risk pro les, such a facility could achieve signi cant diversi cation bene ts, re ected in lower costs for risk nancing. Lower cost for participants will come from diversi cation through pooling, not from cross-subsidization. Risk-based pricing would be applied for participants. $ Access to financial capacity in the international markets. A facility could provide a platform for access to international (re)insurance or capital markets for individual countries looking to place large amounts of risk or for clusters of smaller economies looking to achieve economies of scale by working together. A vehicle for insurance of public assets could also be considered. Proposed Structure for a Regional Disaster Risk Financing Facility for Asia Figure 4 shows a possible structure for such a facility. It combines individual approaches for larger economies with joint approaches for smaller economies. As well as the individual nancial components and technical assistance facility shown, the facility would also comprise a forum for key stakeholders (such as Ministries of Finance) to participate in knowledge-sharing and to foster political momentum in the eld. Figure 4. Proposed Structure for a Regional Disaster Risk Financing Facility for Asia INTERNATIONAL (RE)INSURANCE/CAPITAL MARKETS PLATFORM FOR RISK TRANSFER TO INTERNATIONAL MARKETS TECHNICAL ASSISTANCE JOINT DISASTER FACILITY INSURANCE FUND NATIONAL DISASTER RISK FINANCE STRATEGY ASIAN COUNTRIES BY INCREASING GDP The joint disaster insurance fund would be best suited for smaller economies, with uncorrelated but similar risk, looking to gain from the bene ts of risk pooling. A model similar to that of the CCRIF, where countries enter into an insurance contract with the facility and pay a premium for access to rapid liquidity as bridge nancing post-disaster, could be considered. The risk transfer platform could function as a clearinghouse for transferring sovereign disaster risk in Asia to the international markets. It would allow large economies to approach the market directly and smaller economies to approach the market as a group, as the Paci c countries have. Standardized contracts could be used as well as a standardized process for readying countries for transacting. An approach for collectives of subnational entities (such as cities) to this platform could also be considered. The technical assistance facility would be the home of public goods such as catastrophe risk models that would support the above components. It could also assist countries with their national strategies for nancial protection, and speci cally with mechanisms for disbursing funds in-country to better reach a ected households and businesses. A transparent, rules-based disbursement mechanism could be an attractive option to allow international partners to “pre-commit” post-disaster aid, thereby making the disbursement of funds quicker and more predictable for countries and allowing governments to plan ahead. This could also be considered as a platform component. A regional platform for disaster risk nance in Asia could confer signi cant bene ts to the region. Many discussions—between countries, donors, and development partners and within regional economic platforms—are already taking place on how to improve nancial resilience in Asian countries. The heterogeneity of countries’ needs and di ering levels of existing engagement in this area should form the basis of a transparent and open dialogue on disaster risk nancing for the region. The next step in establishing any regional facility is to engage with individual countries to assess their priorities and needs. N E X T S T E P S The collaboration between the World Bank Disaster Risk Financing and Insurance Program and the Rockefeller Foundation aims to bring together ongoing discussions, identify key questions, and explore options for the institutional, nancial, and operational design of a regional facility. Activities will be carried out over the course of this year toward an outline for a potential regional mechanism, to identify challenges to be addressed in the establishment of such a mechanism, and propose a roadmap for its implementation: a. Country demand assessment and policy dialogue. Review of existing national initiatives; consultations with potential participants to assess demand, readiness, barriers, and political will; and targeted outreach bringing together decision makers and champions to support the necessary political will for implementation. b. Catastrophe risk data assessment. Review to evaluate available catastrophe risk data on which parametric contracts could be settled; evaluate available catastrophe risk models for priority countries and perils, which could be used to support placement of indemnity or parametric contracts; and evaluate the need for investment in building the required risk data tools for such a facility. c. Insurance market consultation. Assessment of private sector view on challenges to –and feasibility of– increased provision of risk transfer products in developing Asian economies, given the ndings with respect to catastrophe risk data availability, especially through a regional vehicle. A nal report will bring together the ndings from these activities, matching country needs with technically feasible solutions to propose a structure for consideration by potential participants and international partners. This will also contain a roadmap for implementation. Annex 1. Progress in Disaster Risk Financing in Select Asian Countries . This note was developed under a partnership between the Rockefeller Foundation and the Disaster Risk Financing and Insurance Program (DFRIP) to explore options for supporting the establishment of a regional disaster risk nance facility in Asia. The Disaster Risk Financing and Insurance Program is a joint program of the World Bank’s Finance & Markets Global Practice and the Global Facility for Disaster Reduction and Recovery (GDFRR), with nancial support from partners including the Rockefeller Foundation, the Swiss State Secretariat for Economic A airs, the UK Department for International Development, the European Union, the Government of Japan, the Government of Germany, and the Swedish International Development Cooperation Agency. DRFIP has provided advisory services on disaster risk nancing and insurance to more than 60 countries worldwide. The program works along four priority areas to support four main bene ciary groups: governments, farmers, homeowners and SMEs, and the poorest and most vulnerable. For more information visit www.worldbank.org/dr .