ECONOM I C S FO R DI SAS TE R P R EVEN TION A N D PR EPA R ED N ES S Financially Prepared: The Case for Pre-positioned Finance in European Union Member States and Countries under EU Civil Protection Mechanism 1 © 2024 The World Bank 1818 H Street NW, Washington, DC 20433 www.worldbank.org Disclaimer This document is the product of work performed by World Bank staff and consultants. The findings, interpretations, and conclusions expressed in this document do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denomination, and other information shown in 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. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. This publication was produced with the financial support of the European Union. Its contents do not necessarily reflect the views of the European Union. Administration Agreement between the European Commission and the International Bank for Reconstruction and Development concerning the Part II Europe 2020 Programmatic Single-Donor Trust Fund (No. TF073835) (EC Contract No. ECHOSUB/IBRD/2022/876156). Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions of this work without permission may be a violation of applicable laws. For permission to photocopy or reprint any part of this work, please send a request with the complete information to either The World Bank or European Commission. Please use the following wording when citing this report: World Bank and European Commission. 2024. Financially Prepared: The Case for Pre-positioned Finance. Translations If you create a translation of this work, please add the following disclaimer along with the attribution: This translation is an adaptation of an original work by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptation If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the authors of the adaptation and are not endorsed by The World Bank. 1 ACKNOWLEDGMENTS This report was developed as part of the technical assistance project ‘Phase 2 - Economic Analysis of Prevention and Preparedness in European Union Member States and Countries under EU Civil Protection Mechanism’, coordinated by the Directorate-General for European Civil Protection and Humanitarian Aid Operations (DG ECHO) and other European Commission (EC) stakeholders. The technical assistance is funded through the UCPM Multi-Annual Work Programme 2021–2024. This report was produced under Component 3: Bringing National and Regional Finance to Scale. This report was prepared under the guidance and supervision of Christoph Pusch (Practice Manager, Urban, Disaster Risk Management, Resilience, and Land, Europe and Central Asia), Marina Wes (Country Director for the European Union Countries), and Sameh Wahba (Regional Director for Sustainable Development, Europe and Central Asia). It was developed under the leadership and coordination of Samantha Cook (Senior Financial Sector Specialist) and Zuzana Stanton-Geddes (Senior Disaster Risk Management Specialist), with expert inputs from Jo Kemp (Senior Public Financial Management Consultant), Daniel McGree (Senior Actuarial Consultant), Yiao Yu (Actuarial Consultant), and Nicolas Pondard (Senior Disaster Risk Management Consultant). The report content drew on data from European Forest Fire Information System (EFFIS), and modelling outputs under Economic Analysis of Prevention and Preparedness in European Union Member States and Countries under EU Civil Protection Mechanism - Phase 1 developed by JBA Risk Management and the Global Earthquake Model. The report also benefitted from comments and guidance from Matija Laco (Senior Economist), Mary Boyer (Disaster Risk Management Specialist, GFDRR), and Tatiana Skalon (Financial Sector Specialist). The report drew on consultations with civil protection agencies, institutional actors, and experts for the country-specific case studies as well as inputs from World Bank staff. The team would like to express its gratitude for the cooperation and guidance of the representatives of the EC and all the stakeholders consulted during the report’s preparation. This includes the Directorates General for Civil Protection and Humanitarian Aid Operations (DG ECHO) for the overall coordination, Climate Action (DG CLIMA), Joint Research Centre (JRC), Economic and Financial Affairs (DG ECFIN), Regional and Urban Policy (DG REGIO), European Insurance and Occupational Pensions Authority (EIOPA), and Environment (DG ENV). The report was edited by Selvaraj Ranganathan and designed by Ariane Kascha and photo imagery was provided by the © European Commission. 2 Contents STATEMENT FROM FROM THE WORLD BANK. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 EXECUTIVE SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 1.1. Context 19 1.2. Methodology 23 2. RISK RETENTION INSTRUMENTS IN EUROPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.1. Regional Instruments 27 2.2. National Level 32 3. RISK-BASED BUDGETING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4. RISK TRANSFER INSTRUMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.1. Sovereign Insurance or Capital Market Instruments 38 4.2. Public Asset Insurance 39 4.3. Household Insurance 40 4.4. Market Overview 41 5. FUNDING GAP ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 5.1. Wildfire Results of the EU-Level Analysis 45 5.2. Drought results of the EU-Level Analysis 50 5.3. Total Drought Cost 51 5.4. Wildfire Results: Country Case Studies 53 5.5. Bulgaria 55 5.6. Croatia 55 5.7. Greece 59 5.8. Italy 59 5.9. Romania 60 6. KEY FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 6.1. Options for Consideration 66 ANNEX 1. CURRENT WILDFIRE MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 ANNEX 2. EXAMPLES OF DISASTER RESERVE FUNDS AND CONTINGENCIES FUNDS IN THE EU . 69 ANNEX 3. FUNDING FROM THE EUSF AND DISASTER DAMAGES OVER 2002–2022. . . . . . . . . . . . . 72 ANNEX 4. INSURANCE PENETRATION IN THE EU MS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 ANNEX 5. ASSUMPTIONS OF THE NATIONAL FUNDING GAP ANALYSIS. . . . . . . . . . . . . . . . . . . . . . . 77 ANNEX 6. ASSUMPTIONS OF THE EU-LEVEL FUNDING GAP ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . 78 ANNEX 7. DERIVING PUBLIC SECTOR COSTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 ANNEX 8. RESERVE FUNDS OF THE EU MS (€, MILLIONS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 ANNEX 9. REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 3 List of Figures Figure 1: Drought funding gap (midpoint scenario) 15 Figure 2: How much damage did 2023’s wildfires do compared to the average historical average? 20 Figure 3: Hectares burned by wildfires in 2023 20 Figure 4: The cost of Europe’s wildfires in 2023 21 Figure 5: Heat stress during the July 2023 heatwave 22 Figure 6: Risk layering 27 Figure 7: How the UCPM works 27 Figure 8: Total MFF budget, 2014–2027 28 Figure 9: UCPM budget, 2014–2027 29 Figure 10: UCPM budget by category, 2014–2022 29 Figure 11. Instruments for DRM prevention and preparedness investments 32 Figure 12: The ladder approach to catastrophe insurance 42 Figure 13: Estimated emergency operation costs 46 Figure 14: Wildfire emergency funding gap - Tier 1 funding 47 Figure 15: Emergency funding gap - Tier 2 funding 48 Figure 16: Total annual cost - funding gap midpoint scenario 49 Figure 17: Total cost funding gap - midpoint scenario (10.6 percent of total costs) 50 Figure 18: Drought funding gap (midpoint scenario) 52 Figure 19: Total Cost - Funding Gap Midpoint Estimate 55 Figure 20: Total Cost - Funding Gap Midpoint Estimate 56 Figure 21: Total Cost - Funding Gap Midpoint Estimate 59 Figure 22: Total Cost - Funding Gap Midpoint Estimate 60 Figure 23: Total Cost - Funding Gap Midpoint Estimate 61 Figure 24: Number and direct damages (€, millions) of natural disasters as reported to the EU for the EUSF 73 Figure 25: Insurance penetration in EU MS by peril 74 4 List of Tables Table 1: Summary of options for consideration 17 Table 2: Risk-based budgeting activities 35 Table 3: Benefits of risk-based budgeting across the budget cycle 35 Table 4: Range of regional-level funding gap from Phase 1 and Phase 2 analysis in € millions 53 Table 5: Amount allocated for item A539025, 2013-2018 (€, millions) 56 Table 6: State expenditure, 2013–2018 (€, millions) 57 Table 7: Romania reserve funds 61 Table 8: Range of funding gap from Phase 1 and 2 analysis (€, millions) 64 Table 9: Funding from the EUSF (2002–2022) versus total damages of the disasters accepted by the EU 72 Table 10: Proportion of households covered by insurance in the EU MS (%) 75 Table 11: Reserve and contingency funds in the EU MS, including the estimated contingency funds 80 List of Boxes Box 1: National fire management fund, Argentina 33 Box 2: Examples of risk-based budgeting practices outside the EU 37 Box 3: Europa Re 39 Box 4: Linking parametric insurance to ecosystems 41 Box 5: Government support to wildfire insurance schemes 42 5 ABBREVIATIONS AAL Average Annual Loss CAT DDO Catastrophe Deferred Drawdown Option CCA Climate Change Adaptation CCR Caisse Centrale de Réassurance CCS Consorcio de Compensación de Seguros CSO Civil Society Organization DRF Disaster Risk Finance/Financing DRM Disaster Risk Management EC European Commission ECMWF European Centre for Medium-Range Weather Forecasts EFFIS European Forest Fire Information System EIOPA European Insurance and Occupational Pensions Authority EU European Union EUSF European Union Solidarity Fund GDP Gross Domestic Product GEM Global Earthquake Model GNI Gross National Income ILS Index-Linked Securities IMF International Monetary Fund JBA JBA Risk Management JRC Joint Research Centre LADWP Los Angeles Department of Water and Power MFF Multiannual Financial Framework MS Member State(s) MTPL Motor Third Party Liability NTMA National Treasury Management Agency OBR Office for Budget Responsibility OECD Organisation for Economic Co-operation and Development PAID Natural Disaster Insurance Pool PFM Public Financial Management PIM Public Investment management RRF Recovery and Resilience Facility SEAR Solidarity and the Emergency Aid Reserve UCPM Union Civil Protection Mechanism WHO World Health Organization WUI Wildland-Urban Interface 6 GLOSSARY Adaptation: Adjustments or changes in economic, preparedness, public awareness, financial resilience social, or environmental approaches in response to (various instruments), and resilient recovery. the effect of present or future climate change. Disaster risk: The combination of the probability of Average annual loss (AAL): The average amount of an event and its negative consequences—that is, the expected (or potential) loss over many years; likelihood of severe disruptions in the normal calculated as the sum of all modelled or simulated functioning of a community or a society over a losses expected over a period divided by the number specified period due to hazardous physical events of years in that period. interacting with vulnerable social conditions, leading to widespread adverse human, material, economic, Average annual loss ratio: AAL relative to the total or environmental effects that require immediate replacement cost of the building stock. emergency response to satisfy critical human needs and may need external support for recovery. Contingent liability: A potential payment obligation (or future expenditure) that may be incurred, Emergency operations: In the context of this report, depending on the outcome of a future event; in the this term refers to the immediate response costs. case of disaster risk for governments, the expenditure may be to pay for emergency response or Ex ante/prearranged risk financing instruments reconstruction in the event of a natural hazard (solutions, mechanisms): In the context of disaster impact. Contingent liabilities can be explicit events, instru­ments (solutions, mechanisms) (underpinned by some form of legal obligation) or arranged before the event. Ex ante decisions are implicit (when there is a social expectation that the decisions made before the event. government will step in as an insurer of last resort). Ex post risk financing instruments (solutions, Cost neutral: Indicates an outcome when the cost is mechanisms): In the context of disaster events, not higher than the revenue it generates. instruments (solutions, mechanisms) arranged after the event. Ex post decisions are decisions made after Disaster risk finance (DRF)/financial resilience the event. (preparedness/protection) to disasters: Financial protection that is planned ahead to better manage Exposure: The situation of people, infrastructure, the cost of disasters, ensure predictable and timely housing, production capacities, and other tangible access to much needed resources, and ultimately human assets located in hazard-prone areas. mitigate long-term fiscal impacts. Exposure includes the number of people or types of assets in an area. These can be combined with the Disaster risk management (DRM): Processes for specific vulnerability and capacity of the exposed designing, implementing, and evaluating strategies, elements to any particular hazard to estimate the policies, and measures to improve the understanding quantitative risks associated with that hazard in the of disaster risk, foster disaster risk reduction and area of interest. transfer, and promote continuous improvement in disaster preparedness, response, and recovery Funding gap: The difference between the available practices, with the explicit purpose of increasing government budget and the probable loss for a given human security, well-being, quality of life, and event size (or return period). sustainable development. DRM investments are understood as investments in risk identification (risk Hazard: The potential occurrence of a natural or assessments and so on), risk reduction (prevention), human-induced physical event that may cause loss early warning, emergency and response of life, injury, or other health impacts as well as 7 damage and loss to property, infrastructure, Resilience: The ability of a system and its component livelihoods, service provision, and environmental parts to anticipate, absorb, accommodate, or recover resources. from the effects of a hazardous event in a timely and efficient manner, including through ensuring the Household insurance penetration (in this report): preservation, restoration, or improvement of its Proportion of households in each country with essential basic structures and functions. catastrophe insurance. Return period: The estimated time between losses of Losses: Quantifiable damages of disasters that can a certain size occurring. For example, a 1-in-10-year be translated into monetary terms. A distinction can return period refers to losses that are expected to be be made between direct disaster losses, which refer exceeded once every 10 years—that is, in any given to directly quantifiable losses (number of people year there is a 10 percent probability of such losses killed; damages to buildings, infrastructure, or natural at least as great as this. The estimates do not mean resources), and indirect losses, which refer to these disasters will occur only once every 10 (or 20 or indirectly quantifiable losses (declines in output or 50) years. revenue, impact on well-being, disruptions to flow of goods and services in an economy). Total cost: The sum of emergency operations costs and damage. Reserve fund (contingency fund): An amount of money set aside to finance—usually—unexpected Vulnerability: The characteristics and circumstances future needs. May be used interchangeably with of the built environment and communities that make contingency fund. However, the latter usually refers them susceptible to damaging impacts (or human to general funds set aside to meet all types of vulnerability). Vulnerability factors include building unexpected spending, while reserve funds might be construction type and socioeconomic context. targeted (for example, dedicated to disasters). 8 Statement from from the World Bank We live in a time when region have set ambitious goals, which require crises have become substantial investment to mitigate and adapt to the normal. In Europe, the projected changes, such as the increased frequency scale of loss and and intensity of extreme weather events. While much destruction from disaster needs to be done, financial resources are scarce, events is staggering. with many urgent and often competing priorities. Recent years recorded multiple concurrent ma­ To respond to these challenges, focused and smart jor dis­asters—inclu­ding investments are needed in climate adaptation and floods, wild­fires, heat­ disaster prevention and preparedness, accompanied waves, and droughts. In 2023 alone, the hottest year by strengthening and adapting infrastructure, on record, economic losses from disasters amounted institutions, societies, and finance at different levels to €77 billion across Europe.1 of government. Europe is warming faster than any other continent in Focused – because while Europe has been taking the world. Recent events indicate a disturbing steps to invest in disaster and climate resilience, trend—- ongoing global warming driving increasingly critical sectors, including those providing civil intense climate extremes. Projections suggest that protection and emergency response, remain highly economic losses from climate-related events in the exposed. If infrastructure fails—because a fire station EU could soar to €175 billion per year in a 3°C is destroyed in an earthquake, critical evacuation warming scenario.2 routes are flooded, or hospitals are evacuated because of wildfires—people, homes and businesses Globally—and in Europe—disasters have far- cannot be saved, magnifying the impacts of an event. reaching effects, with the vulnerable suffering the If public financing is severely affected—or even most.3 Disasters not only have a direct impact on depleted—due to the impact of major catastrophic physical assets and infrastructure, but also increase events, the government cannot provide timely poverty and exacerbate inequality over the long term. emergency, recovery and reconstruction support to When mechanisms to prevent, prepare, respond, and its populations and the economy. recover from disasters are missing or inadequate, these events can erode decades of development and Smart – because while preventive investments make deeply affect society’s welfare. clear economic sense,4 more can be achieved using data and information to scale up prevention, Preparing for this new era of climate challenges is preparedness and adaptation efforts in a cost- critical for safeguarding the well-being of Europe's effective, and targeted manner. In an environment of communities and economies. Many countries in the constrained resources, the region will not be able to 1 Munich Re. 2023. Record thunderstorm losses and deadly earthquakes: The Natural Disasters of 2023. Link. 2 EC (European Commission). 2020. PESETA IV. Link. 3 Kerblat, Yann, et all. 2022. Overlooked: Examining the Impact of Disasters and Climate Shocks on Poverty in the Europe and Central Asia Region. Link. 4 World Bank and European Commission. 2021. Economics for Disaster Prevention and Preparedness: Investment in Disaster Risk Management in Europe Makes Economic Sense. Link. 9 Statement from from the World bank successfully manage current and future risks unless This series of analytical reports, produced as part of a investments to prevent and prepare for disasters are partnership with the European Commission, attests prioritized. At the same time, disaster prevention and to our commitment. climate adaptation efforts are closely interlinked and should be integrated to maximize the benefits of Building on results generated in 2021,5 this set of socioeconomic development and fiscal sustainability.  reports provides new evidence, tools, and examples for countries in Europe to strengthen their disaster At the World Bank Group, we are modernizing our and climate resilience in a focused and smart manner. mission and instruments to ensure better support to By highlighting aspects such as prioritized decision- countries globally and in Europe. In the region, the making, understanding the costs of climate change, World Bank Group has been strengthening and risk-informed budgeting, these reports can be partnerships, providing financing and sharing instrumental in developing and implementing knowledge to help communities manage the risks of nuanced policies and strategic investments that are disasters and climate change. Among these efforts, attuned to the diverse hazards facing Europe. By we support countries to modernize their policy and embracing such new tools and approaches, we can strategic frameworks, and prioritize, design and ensure that communities are more resilient in the finance investments that strengthen disaster and face of ever-evolving climate impacts and help secure climate resilience, including in critical infrastructure a sustainable future for generations to come. and emergency response services. Antonella Bassani Vice President, Europe and Central Asia World Bank 5 World Bank and European Commission. 2021. Economics for Disaster Prevention and Preparedness. Link. 10 Statement from from the World bank Executive Summary This report was developed as part of the technical Disaster Risk in the European Union (EU) assistance program ‘Phase 2 - Economics of Member States (MS) Prevention and Preparedness (EDPP) in European Union (EU) Member States (MS) and Countries Drought and extreme heat have similar impacts on under EU Civil Protection Mechanism’ (UCPM). The economies and often occur simultaneously. For report has been prepared for the Directorate-General example, declines in labor productivity and for European Civil Protection and Humanitarian Aid disruptions to the water-dependent industry such as Operations (DG ECHO) and other European water-intensive manufacturing, agriculture/forestry, Commission (EC) stakeholders. It complements the food production, power generation, and water Phase 1 report that focused on earthquake and flood distribution. Extreme heat is additionally associated risk and revealed that losses from these events at the with added stress on emergency response and health EU level can cost €13–50 billion a year depending on sectors. Due to the relationships between droughts the magnitude of the events. The EDPP Phase 2 and extreme heat, their impacts are often difficult to considers wildfire and drought hazards and current separate in historical event reports (that is, extreme financing mechanisms and recommends options on heat impacts may be considered part of overall how to scale disaster risk financing (DRF) at the drought impacts; GAR 2021). This report focuses on national and regional levels. the financial impacts of drought using the losses reported by the Joint Research Centre (JRC) and This report aims to inform discussions on the discusses potential impacts from extreme heat where development of effective national and regional risk appropriate noting that little to no information is financing mechanisms by identifying funding gaps systematically collected on the impacts of extreme for wildfire and drought response. The identification heat beyond excess hospital admissions and mortality of funding gaps can be used to inform a risk-layering numbers in a few countries (that is, Belgium and Italy). approach, which combines different financial instruments to provide predictable finance when The number of wildfires has been increasing year on needed. This also includes the identification of additional year since 2019 and typically coincides with periods regional funding to complement national finances. of drought and extreme heat. Damage caused by wildfires in Lithuania, Estonia, Austria, and Greece Overall, this report reinforces the findings from was particularly acute in 2023 compared to previous Phase 1 which highlight that too much of the disaster years exceeding the historic average damage costs and climate risk is managed through risk retention by between 83 and 1,758 percent. In terms of actual (for example, budgetary instruments at the EU and costs, the 2023 wildfires estimates of direct damage EU MS levels) and more should be done to incentivize were equivalent to €2 billion in Greece, €1 billion in transfer risk to the private sector at both the EU and Italy, and around €913 million in Spain. the EU MS levels. 11 Executive Summary Heatwaves across Europe have affected the health The indirect losses associated with wildfire and and livelihoods of millions, and this is predicted to drought are expected to be far higher and pose get worse due to climate change. In 2022, the World impacts for the long-term health of society and Health Organization (WHO) European region ultimately businesses as the potential to reduce the estimated that extreme heat claimed more than number of working days increases. However, 60,000 lives and by 2050 this could rise to 120,000 currently, data are not available to substantiate this, heat-related deaths per year.6 The July 2023 and there is a need to start collecting data on these European heatwave, analyzed by the European impacts, for example, the number of hospital Centre for Medium-Range Weather Forecasts admissions due to extreme heat and wildfire events. (ECMWF), resulted in widespread conditions of heat Consequently, the quantification of indirect losses stress across the Mediterranean, reaching extremely from wildfire and drought was beyond the scope of high levels in some areas of Spain, Italy, and Greece. this study which focuses on addressing the financial At present, the EU does not systematically collect needs for response. data on hospital admissions for heat-related illnesses, nor has the UCPM or the European Union Solidarity A key message from this report is should a drought Fund (EUSF) been triggered to respond to these or a wildfire happen in a year where a major events. earthquake or flood has already occurred, there may be no funding available at the EU level to Despite the increasing impacts from wildfires, and respond to a wildfire or drought event. This reinforces drought, there is currently no publicly available the finding from Phase 1 that there is scope for probabilistic risk model for these hazards in Europe. additional financial instruments at the EU level and/ However, some models are currently in development or a need to incentivize national governments to with multiple providers, with delivery timelines invest in DRF. beyond the time frame of this report. It should be noted that wildfire risk, across Europe, has not historically been regarded in the same way as floods Current Disaster Risk Financing Mechanisms and earthquakes, and modeling work has been done in a few countries where corporates have sought to A significant increase in budget allocation to the cover their assets at risk. Annex 1 provides further UCPM reflects the series of crises since 2019. information on the 12 models that are currently Before 2019, the UCPM’s budget from the EU available and/or under development. Multiannual Financial Framework (MFF) was relatively stable ranging from €39 million in 2018 to €51 million It is important to recognize that these risks exist in 2015. A significant increase in the UCPM budget alongside flood, earthquake, landslide, and storm from the MFF from 2019 onward—resulting in an risks and in Europe these risks create additional overall increase of 746 percent over 2014–2022— pressure on already constrained response and has been due to needs associated with the UCPM recovery budgets. Considering all the disaster and revision (including financing of rescEU), COVID-19 climate risks that Europe faces, the financial impacts (repatriations, medical supplies), Afghanistan from earthquakes remain the highest, followed by repatriations, and the war in Ukraine.7 flood. The size of the funding gap from Phase 1 (earthquake and flood) varies between €13 billion Analysis of the UCPM’s budget over time indicates and €50 billion. In comparison, losses from wildfire that most of the expenditure has been for response range from €16 million to €717 million, depending on activities of which wildfire activities are equivalent the scenario and magnitude of the event, while to approximately one-third of response costs, while drought saw a consistent funding gap between €13 no expenditures were found for extreme heat. The million and up to €323 million. response budget increased from €13 million in 2014 6 WHO. Link. 7 Link. 12 Executive Summary to €150 million in 2022. Other notable increases over droughts are often connected to extreme heat and 2014–2022 include a rise in funding for prevention wildfires, compounding overall impacts. It is the and preparedness activities (from €9 million in 2014 damage from this combined risk that leads to the to €19 million in 2022) and for firefighting (from €1 application to the EUSF. million in 2014 to €1.67 million in 2022). RescEU was a new budget line that was added to the overall The EUSF has been triggered eight times since 2002 budget from 2019 onward. No expenditures were for wildfires, in response to damages of €6.1 billion found for extreme heat; therefore, the analysis will for these eight events. Since its establishment, there focus on drought. have been eight successful applications to the EUSF for support costs associated with wildfires, which Established in 2002,8 the EUSF provides financial have received €207.1 million in financing.13 There assistance to emergency and recovery operations in are no reported applications for extreme heat events MS (and accession countries) which complements given that the limited physical damage associated financing available for response.9 The EUSF operates with these events is unlikely to breach the threshold outside of the MFF which allows the EUSF to mobilize of damage in excess of either 0.6 percent of the necessary funds to react to unforeseen events such affected State’s gross national income (GNI) or €3 as crisis and emergency situations. However, as billion in 2011 prices, to submit an application to the found in the Phase 1 report and EC reviews10 of the EUSF. These events are typically ineligible despite mechanism, access to funding takes time, with high emergency response, and information on applications taking 8–10 weeks and disbursement medical costs associated with these events is taking an average of 56 weeks (although advances unavailable. The EUSF is not typically used to support can be provided before the grant is fully disbursed MS following wildfires, in part, due to the low amount and it is possible to include retroactive financing). of eligible direct damage costs for EUSF funding. Since 2002, the most common application to the EUSF is for floods, accounting for 49 percent of total applications, and the most costly events are Wildfire and Drought Emergency Response earthquakes, which have received approximately half Financing Gap of the financing provided under the EUSF. Considering the scale of the risk that EU countries Droughts have had large impacts on economies in face from wildfires and drought, it is critical to MS; since the EUSF was created in 2002 there have assess the financing gap between available budget been four applications submitted and accepted for and the potential costs associated with emergency droughts. Funding was provided to Cyprus in 2008 response from a given event. The funding gap and 2016 and to Romania in 2012 and 2022. In 2022, analysis here considers the national reserves held by Romania which suffered a loss of over €1 billion in the MS, the MFF budget lines associated with emergency agricultural sector11 due to droughts and wildfires in response activities under the UCPM, and the EUSF the south-eastern region received almost €34 million based on record between 2002 and 2020. The from the EUSF12 to cover some of the losses from the resulting funding gap will need to be covered by (i) drought and associated wildfires. Currently, MS are budget reallocation, (ii) debt, or (iii) bilateral donor not easily able to access EUSF funds to support assistance; otherwise, the cost of response will fall on drought losses, due in part to the challenges in the affected populations, for example, farmers who defining the exact start of a drought and the fact that lost crops as a result of drought. 8 Revisions in 2014, 2020, and 2023. 9 Council Regulation (EC) No 2012/2002 of 11 November 2002 establishing the European Union Solidarity Fund (OJ L 311, 14.11.2002, p. 3), Link. 10 Review of the European Union Solidarity Fund, 2021 Link. 11 Link. 12 Almost €34 million in European Solidarity Funds awarded to Romania to repair damages caused by severe drought in 2022 Link. 13 EC. EU Solidarity Fund: Supporting Disaster Recovery 2002–2022. Link. 13 Executive Summary The estimated cost of wildfire emergency response €301 million was in 2017. This funding gap was operations ranges between €41 million and €752 caused by a combination of high losses (€337 million million, depending on the scenario assessed and the midpoint estimate, range of €233–717 million) and a year observed. This estimate is derived from modest level of UCPM funding at the time (€36 calculating the average cost per hectare burned and million). applying it to the total number of hectares documented in the European Forest Fire Information System A funding gap for wildfires of around €190 million (EFFIS) database14 and then using a proportion of the per year could occur based on the assumptions from cost to represent emergency response operations the high scenario. Given the uncertainty surrounding with 13 percent as the low estimate/scenario and the loss estimates for wildfire, this could better reflect 35 percent as the high estimate/scenario.15 future loss. However, given the limited number of events over a short time frame, the three-year average The average cost of emergency response operations between 2019 and 2022 may better represent future is estimated to be between €105 million (low loss, which indicates that enough finance is available scenario) and €294 million (high scenario) per year. to cover all emergency operation costs, although this Between 2014 and 2019, there was an average varies depending on the assumed level of emergency funding gap of €84 million a year based on the operation costs in proportion to the overall cost. midpoint scenario, which suggests that there was insufficient funding available within national reserves The estimated funding gap at the EU level related to and relevant UCPM budget allocations. emergency response to drought is €13 million per year based on current drought losses. That is, if the A repeat of the 2017 wildfire season in the future UCPM and EUSF were used to cover all public and would require a 70 percent increase in the 2023 private sector costs associated with drought hazards UCPM budget16 to completely cover emergency for EU MS, the remaining €13 million would need to response operation costs. The year 2017 saw be covered by the EU MS or the private sector or extensive wildfires in Italy, Spain, and Portugal, households within each country. Figure 1 illustrates resulting in a significant increase in area burned and the potential funding gap using the midpoint estimate associated emergency response costs. Considering for current annual drought losses and potential losses these scenarios and existing financing mechanisms under two future scenarios (1.5°C warmer and 3.0°C between 2014 and 2023, the largest funding gap of warmer). 14 JRC EFFIS. Link. 15 The estimates were developed using data from the EC – cohesion data (Link.) which include total direct damage estimates and emergency operation cost estimates per event in addition to the EUSF. 16 Considering the 2023 UCPM budget of €177 million. 14 Executive Summary Figure 1: Drought funding gap (midpoint scenario) 500 EUSF Annual Allocation for 450 drought events 400 Total UCPM budget (€ m) 350 (Response - 3-year Average) 300 Reserves at EU level €, millions 250 200 150 100 50 0 Current Annual Loss Future scenario Future scenario (1.5° C warmer) (3° C warmer) Source: World Bank. The projected growth of the funding gap for drought start and end date of the drought; and (iii) the asset over time, aligning with anticipated temperature exposure differs greatly. For example, residential, increases in the region, reveals a potential funding commercial, and industrial assets were analyzed in gap between €29 million and €323 million annually, Phase I whereas Phase II looks at the damage and depending upon the climate warming scenario used. loss incurred from firefighting, reforestation, damage, Using the estimates derived for the 1.5°C warmer and cleanup. This is due in large part to the geographic scenario, the midpoint for the annual funding gap locations where wildfire occurs and is expected to estimate increases by 125 percent to €29 million occur. Noting the above, it is expected that the annually (low-range estimate: no gap; high-range analysis conducted in this report underestimates the estimate: €29 million). Under the 3.0°C warmer amount of risk from wildfire and drought. scenario, the midpoint annual funding gap estimate increases 15-fold compared to the current annual loss, reaching €206 million annually (low-range Results of the Country Case Studies estimate: €150 million; high-range estimate: €323 million). While these numbers remain low, it should Five countries were selected for national case be noted that this analysis was based on the midpoint studies in the funding gap analysis: Greece, Italy, scenario and may underestimate the risk. In addition, Romania, Croatia, and Bulgaria. The approach to the the increase in the projected funding gap over time country-level funding gap analysis considers overall suggests that the financial demands for addressing economic losses incurred as a closer representation drought-related costs are likely to escalate of the contingent liabilities member countries will significantly in the future. face in the wake of wildfire events. It should be noted that it is not possible to combine The Bulgaria analysis indicates that there is no the results from EDPP Phase 2 with the Phase 1 immediate funding gap under the midpoint scenario, funding gap analysis, for three reasons: (i) there are although potential gaps may arise in extreme years, limited observations of wildfire and drought events particularly under the high-range scenario. In 2023, compared to the number of observed earthquakes estimated reserves and contingency funds sufficiently and floods; (ii) the limited observations impede the covered the estimated costs, but a shortfall of €22 development of probabilistic risk models for wildfire million would occur if recent three-year average in particular and drought models face challenges in losses from 2021 to 2023 were considered but could identifying the event duration, that is, identifying the be covered by the inclusion of UCPM and EUSF funds. 15 Executive Summary Notably, if high-range assumptions are applied to Options for Consideration historical cost estimates, a funding gap of €130 million would emerge. This report reinforces the findings from Phase 1 confirming that financial instruments to manage Croatia faces no funding gap under the midpoint disaster risk are limited to risk retention instruments scenario, with reserves and contingency funds such as the UCPM and EUSF and more should be covering estimated costs in 2023. However, if high- done to incentivize risk transfer at both the EU and range assumptions were applied to historical cost EU MS levels. At present, there are no risk transfer estimates, a funding gap of €220 million would arise. products at the EU level or in the case study countries. Risk transfer instruments can be expensive and Greece faces a funding gap across all scenarios due require the payment of a premium regardless of to limited reserves and contingency funds relative whether a payment is received from the instrument. to estimated costs. Despite additional financing Risk transfer serves to smooth expenditures over options from the UCPM and EUSF, the shortfall years where no disaster occurs, yet the instrument remains high, ranging from €259 million under the has a cost. When a disaster occurs, a payment that three-year average between 2021 and 2023 to corresponds to the magnitude of the event is received. €1.4 billion under the high-range assumptions. The above findings suggest that more can be done Italy shows no funding gap in any scenario due to at the national and EU levels to promote DRF ample reserves and contingency funds compared to solutions and close funding gaps. Table 1 presents estimated costs. Even in scenarios that consider the some options for consideration; not all need to be year with the highest historical cost estimates, pursued or implemented at the same time. Some of reserves and contingency funds remain sufficient. these options were presented in Phase 1 but remain relevant today based on the analysis and findings of Romania also demonstrates no funding gap across Phase 2. scenarios due to sufficient reserves and contingency funds relative to estimated costs. In 2023, these funds covered estimated costs, with similar results under recent three-year average losses and high- range assumptions. 16 Executive Summary Table 1: Summary of options for consideration EU-LEVEL OPTIONS DETAILS TIME FRAME 1. Develop an EU-level • Clearly establish and articulate EU-level priorities that can be Short to medium overarching DRF awarded finance from the EU-level instruments to delineate what term strategy the EU expects countries to cover from national resources. 2. Increase the allocation • Increase the amount of funding proportionate to the risk appetite Short to medium for the UCPM and EUSF of the EU, that is, identify and articulate how much of the total term cost that the EU intends to cover. 3. Introduce risk transfer • Linked to option 2, the introduction of risk transfer provides a Medium to long instruments mechanism to bring in finance when it is needed most. term • This could be most highly effective as a means to manage cross- border events or where several countries in the region experience concurrent disasters. 4. Improve data on DRF • Start the systematic collection of hospital admissions and/or Short to medium on the number of people requiring treatment following wildfire, term drought, and extreme heat events to help quantify the health impacts on the population. As these events are expected to become more frequent it is highly recommended that this is rectified to explore whether there is a need to provide additional financial support through health insurance and/or directly to hospitals. NATIONAL-LEVEL OPTIONS DETAILS TIME FRAME 5. Develop national DRF • To ensure financial preparedness to disasters. Short to medium strategies term • Determine national priorities in strengthening DRF (such as focusing on households, the poorest members of society, government budget, and so on). • In developing the strategy there is a need to improve data on DRF such as the collection of data on hospital admissions to feed in the EU-level data aggregation. 6. Consider the • Explore whether risk transfer could present a cost-effective way Medium to long introduction of to manage select risks, for example, earthquake and flood in the term sovereign risk transfer first instance. instruments • Once risk models are available, leverage their outputs to determine whether the risk from wildfire is a viable option to finance response. 7. Increase penetration of • For both households and public assets. Medium to long insurance term • This could also be considered at the sector level, for example, agriculture insurance for drought. 8. Strengthen risk-based • Create incentives to invest in DRF by introducing a risk-based Medium to long budgeting budgeting approach to ensure that MS understand and prepare term for the risks they face. 17 Executive Summary 1. Introduction This chapter presents key findings from the Phase 1 report and information on the hazards analyzed in this paper (wildfire and drought). It also sets out the methodology used to determine the funding gaps and analyze national-level disaster risk financing instruments. This report was developed as part of the technical This report is structured as follows: Chapter 1 assistance program ‘Phase 2 - Economics of introduces the report, outlining its context and the Prevention and Preparedness (EDPP) in European methodologies used to elicit findings. Chapter 2 Union (EU) Member States (MS) and Countries provides updated information from the Phase 1 report under EU Civil Protection Mechanism’ (UCPM). The on DRF instruments currently in use at the regional report has been prepared for the Directorate-General and national levels. Chapters 3 and 4 detail risk- for European Civil Protection and Humanitarian Aid based budgeting and risk transfer instruments, Operations (DG ECHO) and other European respectively. Chapter 5 discusses the current funding Commission (EC) stakeholders. EDPP Phase 2 gap, and Chapter 6 concludes with key findings and analyzes the hazards of wildfire and drought and options for consideration. considers how finance for disaster risk can be brought to scale at the national and regional levels. This report aims to inform discussions on the development of effective national and regional risk Disaster risk financing (DRF)17 is defined as the financing mechanisms by identifying funding gaps strategies and instruments used to manage the for wildfire and drought. The identification of funding financial impact of disasters. As noted in the 2021 gaps can be used to inform a risk-layering approach, EU peer review DRF assessment report, a thorough which combines different financial instruments to understanding of risk exposure and risk-bearing provide finance when needed. The identification and capacity, as well as institutional arrangements quantification of funding gaps can demonstrate the creating favorable regulatory and market case for additional regional funding to complement infrastructure are the major components of a com­ national finances. prehensive disaster financing strategy. 17 OECD. 2012a. Disaster Risk Assessment and Risk Financing: A G20/OECD methodological framework. 18 Introduction 1.1. Context EDPP Phase 1 focused on the analysis of fiscal and This report, under Phase 2, complements the economic impacts of earthquakes and floods at the analysis of Phase 1 by analyzing wildfire and regional and national levels and provided high-level drought. It provides insights into how much finance is recommendations to reduce contingent liabilities required to respond to these events at the regional from these risks. The key findings from that report and country levels. However, it is not possible to are as follows: combine the results with the Phase I analysis, for three reasons: (i) there are limited observations of • DRF across the EU MS is limited. Penetration rates wildfire and drought events compared to the number of insurance for public and residential assets are of observed earthquakes and floods; (ii) the limited low, reserve funds are limited, and other types of observations impede the development of probabilistic risk transfer and contingency funding are not risk models for wildfire in particular and drought widely used. Around 40 percent of countries lack models face challenges in identifying the event prearranged funding to manage combined duration, that is, identifying the start and end date of emergency response costs for 1-in-10-year flood the drought; and (iii) the asset exposure differs and earthquake events, that is, for events that greatly. For example, residential, commercial, and occur relatively frequently (EDPP Phase 1 report). industrial assets were analyzed in Phase I whereas Phase II looks at the damage and loss incurred from • Only a few countries in the EU have dedicated firefighting, reforestation, damage, and cleanup. In disaster reserve funds. General contingency funds large part this is due to the geographic locations might be largely unavailable, especially if a disaster where wildfire occurs and is expected to occur. happens toward the end of the fiscal year. There is a 10 percent probability in any given year that the Wildfires tend to damage natural habitats, pollute EU region will experience earthquake or flood watersheds, destroy biological diversity, and events severe enough to produce losses exceeding increase air pollution with serious health effects. countries’ national reserves. On average, damage Agricultural, forestry, transportation, trade, and to residential buildings accounted for over industry sectors tend to record the highest economic 50 percent of the total loss for both flood and losses from forest fires by the resulting breaks in the earthquake risks, which points to an urgent need supply chain that outweigh the value of damaged to increase access to and uptake of catastrophe assets, for example, a productive plantation that is household insurance. For earthquake risk alone, burned will not have high damage in terms of the residential building damage may be even higher; in trees lost but the impacts on the construction sector many countries, residential losses account for over (for example, project delays and potential loss of 50 percent of total loss. jobs) connected to that plantation could be far higher from the resulting delays while the • The sum of the European Union Solidarity Fund reforestation occurs. (EUSF), reserve funds, and contingency funds available to EU MS covers on average less than The number of wildfires has been increasing year 4 percent of total government liabilities each year on year since 2019. Figure 2 shows the percentage when analyzed from an EU perspective (with difference in the hectares burned by wildfires across disasters aggregated for both earthquake and Europe in 2023 compared to the annual average flood and for all EU MS). This suggests that there is burned by wildfires since tracking started in 2006. scope for additional instruments at the EU level As illustrated in this figure, the damage caused in and/or there is a need to incentivize national Lithuania, Estonia, Austria, and Greece was governments to invest in DRF more seriously. particularly acute in comparison to previous years. Figure 3 provides details on the hectares burned by wildfires in 2023. In 2023, wildfires caused damages 19 Introduction costing €2 billion in Greece, €1 billion in Italy, and on the cost of Europe’s wildfires in 2023 by looking around €913 million in Spain. Figure 4 presents data at the average global costs per hectare burned. Figure 2: How much damage did 2023’s wildfires do compared to the average historical average? How much damage did 2023’s Distrelec’s analysis compared how WILDFIRES VS ANNUAL AVERACE (BURNED AREA) many hectares of land were burnt in each European country compared to do compared to the annual the annual average burnt by wildfires 1758% PERCENT DIFFERENCE (9) OF historical average? since 2006. HECTARES BURNT IN 2023 663% 442% 301% -26% -29% -62% -68% -71% -79% -80% -88% -92% 83% 79% 69% 51% 51% 32% 11% 9% 7% 7% Lithuania Estonia Austria Greece Denmark Italy France Bulgaria Germany Ireland Cyprus Spain Finland Belgium Romania Latvia Portugal Slovenia Netherlands Hungary croatia Poland Sweden Source: European Forest Fire Information System (EFFIS) data analyzed by Distrelec. Figure 3: Hectares burned by wildfires in 2023 Denmark Sweden Austria Finland 134 188 398 204 Germany Estonia 975 173 Netherlands Latvia 31 127 Ireland Lithuania 4,302 305 Belgium Poland 240 47 Slovenia Hungary 112 156 France Romania 22,350 16,563 Spain Bulgaria 88,444 15,478 Portugal Cyprus 36,498 1,919 Italy Croatia Greece 97,382 2,768 174,773 Source: EFFIS analysis by Distrelec. 20 Introduction Figure 4: The cost of Europe’s wildfires in 2023 Latvia Lithuania Poland Austria €1.3M €3.2M €468K €4.1M France Portugal €231M Greece €377.2M Estonia €1.8B €1.8M Denmark Croatia €1.4M €28.5M Italy Spain €1B €912M Cyprus €19.8M Romania Belgium Ireland Bulgaria €2.5M €44.5M €1.7B €159.9M Finland Netherlands €2.1M €320K Sweden Germany €1.9M €10.1M Hungary Slovenia €1.6M €1.2M Source: EFFIS analysis by Distrelec. Note: Financial cost calculated using average global cost per hectare. ‘M’ refers to million euros and ‘B’ refers to billion euros. Droughts and extreme heat are among the most Health Organization (WHO) European region complex climate-related hazards, with wide-ranging estimates that extreme heat claimed more than and cascading impacts across hazards, ecosystems, 60,000 lives and by 2050 this could rise to 120,000 and economies. Due to the relationship between heat-related deaths per year.19 Climate change is droughts and extreme heat, their impacts are often increasing the risk of heatwaves, and extreme heat in difficult to separate in historical event reports (that is, the summer months is becoming the norm, not the extreme heat impacts may be considered part of exception. The July 2023 European heatwave analysis overall drought impacts).18 Furthermore, extreme by the European Centre for Medium-Range Weather heat and drought are interconnected because Forecasts (ECMWF) shows widespread conditions of episodes of extreme heat tend to occur during heat stress across the Mediterranean, reaching drought periods, and both hazards can affect extremely high levels in some areas of Spain, Italy, economies through relatively similar channels, such and Greece (see Figure 5). The heatwaves in recent as (i) the direct impact on the health sector, (ii) a years are indicative of a potentially hazardous health/ decline in labor productivity, and (iii) disruption from heat environment where it is imperative for one to the water-dependent industry such as water-intensive cool down immediately and take actions to avoid manufacturing, agriculture/forestry, food production, heatstroke. Going forward, this will likely increase the power generation, and water distribution. number of people who need treatment for these symptoms and health service providers should be Heatwaves across Europe have affected the health prepared to cater for this increase. and livelihoods of millions, and this is predicted to get worse due to climate change. In 2022, the World 18 GAR 2021. 19 WHO, Link. 21 Introduction Extreme heat gives rise to direct costs on the health example, the United States. For instance, the New sector and increased energy expenses due to Mexico Environmental Public Health Tracking heightened cooling demands. It also causes indirect Program develops, monitors, and analyses indicators costs through a decline in labor productivity and of heat stress to document changes in morbidity and impacts on critical infrastructure, such as power mortality across different locations and time. One of plants or roads which may experience service the heat stress morbidity indicators that is tracked— disruptions. At present, the EU does not systematically disaggregated by sex, age, and month—is collect data on the direct and indirect costs of hospitalization for heat stress20. Some EU countries extreme heat. In particular, it does not systematically collect data on hospitalizations for heat stress, for collect data on hospital admissions for heat-related example, Belgium and Italy,21 but not all countries do illnesses nor has the UCPM or the EUSF been this, and it is not currently possible to access EU-wide triggered to respond to these events. Data on heat data. stress are however collected in other countries, for Figure 5: Heat stress during the July 2023 heatwave None Moderate heat Strong heat Very strong heat Extreme heat Source: ECMWF. 2023. Heat stress and the European heatwave of 2023. Link. Note: The map shows locations where the highest conditions of heat stress occurred between July 15 and 27, 2023, in Europe. 20 NM-Tracking - Summary Health Indicator Report - Heat Stress Hospitalizations 21 See Link. for Data from Belgium and Link. for data from Italy. 22 Introduction 1.2. Methodology A N A LY S I S O F W I L D F I R E S A N D D RO U G H T S While there are specific technical challenges to the design and development of wildfire risk models, the EDPP Phase 1 focused on the impacts of earthquakes lack of availability and sophistication of financial and floods at the regional and national levels. The solutions is not a reflection of technical capabilities risk modelling was conducted using two regionally but a market view that wildfire is not a key aggregation consistent probabilistic disaster risk models: the JBA risk unlike windstorms, floods, and seismic events. Risk Management (JBA) model for fluvial and surface Risk modelling capabilities have focused on primary water flood and the Global Earthquake Model (GEM) insurance applications such as risk maps or scores for seismic risk. Both models are fully probabilistic which can be used for pricing the risk and exposure based on a model of individual flood or seismic events management applications that are useful to that account for the often-complex correlation corporates with many assets in a particular area. structures reflecting the physics of the natural peril However, these models do not provide information damages. They are well regarded in the catastrophe on the correlation of damages between locations or risk modelling community and are part of an support estimation of a full risk profile (that is, the ecosystem of commercially available vendor- probability of observing different levels of damages/ developed models as well as proprietary models losses in any given year). (within the insurance sector) to assess, manage, and transfer the risk of these perils. The broad set of At the time of writing, there were no fully probabilistic modelling tools reflects the long-standing recognition disaster risk models available for wildfire in Europe, that these perils are key aggregation risks with but some models are currently in development with potential solvency implications for insurance multiple providers, with delivery timelines beyond companies, that is, so-called ‘catastrophe risks’. the time frame of this report. Annex 1 provides further information on the 12 models that are Commercially available risk models for wildfire have currently available and/or under development. been limited to peak (worldwide) insurance concentrators. For example, the wildfire models for In the absence of a probabilistic model, a straight­ California have only recently been expanded to provide forward methodology has been employed to US-wide solutions. However, wildfire risk in Europe has calculate the funding gap associated with wildfires. not historically been regarded in the same way as This involves calculating the average cost per hectare floods and earthquakes, with modeling done in limited burned and applying it to the total number of hectares countries where corporates have sought to cover their documented in the EFFIS database.22 On the assets at risk. The sophistication of the models financing side, two separate scenarios are analyzed available is also limited, with wildfire models typically based on accessibility to different funding tiers: considered 1st or 2nd generation solutions compared to approximately 5th generation earthquake models. • Tier 1: Estimates of national reserves held by EU The sub-peril damage mechanisms such as smoke MS combined with the UCPM budget allocation damage have only been included in commercial associated with emergency operations costs for solutions in the last couple of years. wildfire events. The existing models on wildfire do not focus on the • Tier 2: Tier 1, combined with an estimation of the generation of ‘full’ event-based catastrophe risk EUSF annual allocation for wildfire events. modelling approaches, and therefore the analysis here will differ from that conducted under EDPP1. 22 Link. 23 Introduction The funding gap analysis is first applied to the region • The high-range scenario assumes that 80 percent (all EU MS). The analysis of the UCPM budget over of wildfire damages are classified as public sector time indicates that most of the expenditure has been contingent liabilities (€3.7 billion per year on on emergency response activities. Therefore, the average based on historical experience between analysis focuses on estimating these costs but 2014-2023), and 3 percent of total wildfire costs recognizes that other costs associated with these are covered by covered the EU MS reserves, UCPM, hazards will emerge to represent the full cost. The and EUSF (€118 million per year on average for the funding gap analysis considers the Multiannual same period).  Financial Framework (MFF) budget lines associated with emergency response activities and the EUSF It should be noted that the results from this funding based on records between 2002 and 2020. The gap analysis differ significantly from those produced analysis assumes that the resulting funding gap will in EDPP1. This is in part because of the different be covered by (i) budget reallocation, (ii) debt, or (iii) asset types held in the affected areas. For example, in bilateral donor assistance. Due to data limitations, EDPP1, the losses were predominantly driven by the the analysis does not review alternative DRF strategies household sector which accounted for over 50 percent that include sovereign risk transfer. of total loss. The areas historically affected by wildfires have less residential assets due to their location so In addition, five selected national case studies are there will be underlying differences in the asset’s analyzed in the funding gap analysis: Greece, Italy, values being analyzed. In addition, the flood and Romania, Croatia, and Bulgaria. The approach to the earthquake models are fully probabilistic. As noted country-level funding gap analysis takes into account above, it was not possible to combine the results with overall economic losses incurred as a closer Phase I due to (i) the differences in the number of representation of the contingent liabilities member observed events, (ii) the lack of probabilistic models, countries will face in the wake of wildfire events. To and (iii) the differences in exposure. However, where facilitate the analysis and assess the potential funding possible the team has tried to draw parallels. gap related to overall estimated contingent liabilities, three scenarios were developed focusing on To calculate the funding gap associated with emergency operation costs and damages: drought, the analysis relies on the estimate provided by the Joint Research Centre (JRC), which suggests • The low-range scenario assumes that 35 percent of that current annual losses from drought in the EU wildfire damages are classified as public sector and United Kingdom amount to approximately €9 contingent liabilities (€1.6 billion per year on billion.23 These losses include various impacts, average based on historical experience between including diminished public water supplies, 2014 and 2023), and 2.2 percent of total wildfire agricultural losses, damage to buildings and costs are covered by the EU MS reserves, UCPM, infrastructure from soil subsidence, decreased inland and EUSF (€35 million per year on average for the water transportation, and reduced energy production. same period).  Although some of these losses may not be considered contingent liabilities for the EU MS, no adjustments • The midpoint scenario assumes that 50 percent have been made to the loss estimates. This is due to of wildfire damages are classified as public sector the lack of sufficient information available to make contingent liabilities (€2.3 billion per year on accurate assessments of these contingent liabilities average based on historical experience between or the proportion of impacts within the United 2014 and 2023), and 2.3 percent of total wildfire Kingdom or other non-EU MS. costs are covered by the EU MS reserves, UCPM, and EUSF (€53 million per year on average for the same period).  23 JRC 2020. Global Warming and Drought Impacts in the EU. Technical Report. PESATA IV Project - Task 7 - Drought. Link. 24 Introduction According to the JRC, if the future climate affects A N A LY S I S O F N AT I O N A L-L E V E L D R F present-day society, the overall drought damage in INSTRUMENTS the EU and United Kingdom will experience a slight rise with a 1.5°C global warming (reaching €9.7 Structured interviews were conducted in two case billion annually). However, this damage escalates study countries—Romania and Croatia—with the significantly with further warming, reaching €17.3 Ministries of Finance, other relevant line ministries billion per year at 3°C. These estimates form the basis which play a part in disaster risk finance, and for providing loss estimates under different scenarios insurance providers and regulators. These interviews in the funding gap analysis. Further information on were designed to deepen the knowledge and the methodology used is provided in Section 5.3. understanding of DRF instruments built under Phase 1. For example, specific attention was paid to existing reserve funds and public asset insurance to provide a more informed view of the level of risk the case study countries can finance. 25 Introduction 2. Risk Retention Instruments in Europe This chapter reviews different financing instruments available in the EU MS to manage disaster costs. It also provides a basis for the funding gap analysis presented in this report. The review focuses on national- level DRF instruments and EU-level DRF mechanisms (EUSF and UCPM) which can provide in-kind assistance and support emergency response in the EU MS (and across the world). A key finding is that financial instruments to manage disaster risk are primarily focused on risk retention and more should be done to incentivize risk transfer at both the EU and EU MS levels. Developing an EU overarching DRF strategy and national DRF strategies could help strengthen financial preparedness to disasters. As noted in the 2021 EU Peer Review Framework needed. From a cost perspective, the use of budgetary report, DRF embraces a variety of instruments instruments is, for instance, more suitable for high- aimed at and capable of achieving different frequency, low-severity events in EU MS. The use of outcomes. Each of these instruments can efficiently risk transfer market-based instruments is more handle only a certain type of risk, depending on its suitable for high-risk events that occur less frequently. frequency, intensity, and impacts. Consequently, a This chapter considers certain instruments at the strategy that builds upon a diversified pool of mutually regional and national levels, noting that high- complementing financial tools and institutions is frequency and low-impact events are more closely better equipped to cope with and respond to a variety aligned with the profile of past events that have of natural and man-made hazards. However, as noted occurred in EU MS. Regional instruments are at in Phase I, there are still significant gaps. present primarily risk retention related. Subject to further analysis, there may be merit in exploring The use of DRF instruments should ideally be guided whether there are some appropriate and cost- by a strategy that selects risk retention and risk effective risk transfer instruments which the EU may transfer instruments for events of different wish to consider. This recommendation is given on magnitudes (see Figure 6). No single DRF instrument the basis that, at present, there is a heavy dependency can cover all losses; a combination of instruments is on risk retention instruments across the EU. 26 Risk Retention Instruments in Europe Figure 6: Risk layering Low Major Disaster risks Disaster risk financing instruments High risk layer Risk transfer: (e.g., major earthquake, major e.g. catastrophe risk insurance, tropical cyclone) capital market instruments Frequency of event Severity of impact Medium risk layer Contingent credit (e.g., floods, small earthquakes) Low risk layer Contingency budget, national reserves, (e.g., localized floods, landslides) annual budget allocation High Minor Source: World Bank. 2.1. Regional Instruments UCPM countries, such as pre-positioning of response goods, financing of transportation costs, deployment of expert The UCPM can provide in-kind rapid assistance to and specialized teams after disasters, and provision of EU MS after disasters (mobilizing support within a grants. Figure 7 provides an overview of how the UCPM few hours/days after a disaster strikes). The UCPM works. includes a number of mechanisms to support affected Figure 7: How the UCPM works Source: EC, “EU Civil Protection Mechanism,” Link. 27 Risk Retention Instruments in Europe Before 2019, the UCPM’s budget from the EU MFF increase of 746 percent from 2014 to 2022. The was relatively stable, ranging from €39 million in increase in budget allocation from 2019 has been 2018 to €51 million in 2015 (see Figure 8). There due to UCPM revision (including rescEU), COVID-19 has, however, been a significant increase in the MFF’s (repatriations, medical supplies), Afghanistan budget from 2019 onward, resulting in an overall repatriations, and the war in Ukraine. Figure 8: Total MFF budget, 2014–2027 700 600 500 400 300 200 100 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 Source: World Bank analysis of UCPM data. As per Council Regulation 2024/765 (February 29, amounts not used for either instrument will be made 2024), it was agreed that the existing Solidarity and available for use in the flexibility instrument in the the Emergency Aid Reserve (SEAR) would be split following year. into two separate instruments with a guaranteed amount per instrument. The European Solidarity Since 2021, the UCPM’s budget has been covered Reserve (the budget line from which the EUSF is through the MFF and NextGenerationEU (NGEU) funded) will have €1,016 million per year in total (in (see Figure 9). NGEU is the EU’s €800 billion 2018 prices), with an increase of €216 million per temporary recovery instrument to support the year (in 2018 prices), for assistance to respond to economic recovery from the coronavirus pandemic emergency situations covered by the EUSF. The and build a greener, more digital, and more resilient Emergency Aid Reserve will have €400 million per future. NGEU’s budget represented 77 percent of the year with an increase of €108 million per year (in total UCPM budget in 2021, 64 percent in 2022, and 2018 prices), that is, €508 million per year (in 2018 was 71 percent of the total UCPM budget in 2023. prices), for rapid responses to specific emergency The UCPM’s total budget for 2021–2027 is €3.7 needs within the EU or in third countries. Annual billion. 28 Risk Retention Instruments in Europe Figure 9: UCPM budget, 2014–2027 600 500 400 300 200 100 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 Year Total MFF budget Total NGEU budget Source: World Bank analysis of UCPM data. Analysis of the UCPM’s budget over time indicates that funding for prevention and preparedness activities (from most of the expenditure has been for response activities €9 million in 2014 to €19 million in 2022) and for firefighting of which wildfire activities are equivalent to approximately (from €1 million in 2014 to €1.67 million in 2022; see one-third of response costs, while no expenditures were Figure 10). RescEU was a new budget line that was added found for extreme heat. The response budget increased to the overall budget from 2019 onward. No expenditures from €13 million in 2014 to €150 million in 2022. Other were found for extreme heat; therefore, the analysis will notable increases over 2014–2022 include a rise in focus on drought. Figure 10: UCPM budget by category, 2014–2022 700,00 600,00 500,00 400,00 300,00 200,00 100,00 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 Year Source: World Bank analysis of UCPM data. 29 Risk Retention Instruments in Europe In 2022, €150 million was allocated for the UCPM’s in principle, is non-insurable. Measures eligible for response activities, firefighting capacities (planes funding are as follows: and helicopters—€145 million), and rescEU transition (firefighting—€40 million). The remaining • The restoration to working order of infrastructure amount of the annual MFF (€47 million) was allocated and facilities providing energy, drinking water, for ‘other’ prevention and preparedness work (€19 wastewater disposal, telecommunications, trans­ million), training and exercises (€15 million), and port, health care, and education. prevention and preparedness grants (€12 million). • The provision of temporary accommodation and the funding of rescue services, to meet the needs EUSF of the population affected. Established in 2002 with revisions in 2014, and • The consolidation of preventive infrastructure and 2020, the EUSF provides financial assistance to protection of cultural heritage sites. emergency and recovery operations in MS (and accession countries), but, due to the time it takes to • The cleaning up of disaster-stricken areas, mobilize funds, it is best suited for recovery.24 including natural zones. The EUSF is one of the mechanisms that operates outside the multiannual framework (MFF). This allows • Rapid assistance, including medical, to the the EUSF to mobilize necessary funds to react to population affected by a major public health unforeseen events, such as crisis and emergency emergency, and the protection of the population situations. However, as found in the Phase 1 report, from the risk of being affected. access to funding takes time, with applications taking 8–10 weeks and disbursement taking an average of Since 2002, the EUSF has mobilized a total of 56 weeks (although advances can be provided before almost €8.6 billion for 110 natural disasters and 20 the grant is fully disbursed). interventions as a response to public health emergencies. It has supported 24 MS (plus the The amount of aid from the EUSF for a given disaster United Kingdom) and four accession countries is determined on the basis of the total direct damage (Albania, Montenegro, Serbia and Türkiye). Italy is caused by that disaster in relation to the relative by far the biggest beneficiary of the fund, having wealth of the affected State as reflected by the received more than €3 billion after a devastating threshold.25 The threshold is the level of total direct earthquake in 2016, followed by Germany, which damage defined by the regulation that must be received over €1.6 billion for flood, and Croatia, exceeded to trigger the intervention of the fund for which received over €1 billion after the two major “major disasters” and is specific to each eligible earthquakes in 2020. In that period, flooding was by State. It is defined as damage in excess of either far the most frequently occurring disaster affecting 0.6 percent of the affected State’s gross national European countries, followed by storm. However, the income (GNI) or €3 billion in 2011 prices, whichever EUSF paid out a similar amount for earthquakes, but is lower. the number of earthquakes was five times less than the number of floods. From 2002 to the mid- 2024, Assistance from the EUSF takes the form of a grant 181 applications were made to the EUSF; 130 were to supplement public spending by the beneficiary successful, 47 were rejected, and 4 were withdrawn.26 State and is intended to finance essential emergency and recovery measures to alleviate damage which, 24 Council Regulation (EC No 2012/2022 of 11 November 2002 establisching the European Union Solidarity Fund (OJ L 311, 14.11.2002, p. 3), Link. 25 EC. EU regional and urban development - Regional Policy, Solidarity Fund. Link. 26 EUSF data on disaster aid provided: Link. 30 Risk Retention Instruments in Europe Analysis of EUSF data indicates that over the affected State’s gross national income (GNI) or 20-year period from 2002, the number of natural €3 billion in 2011 prices, to submit an application to disasters requesting EUSF funding has remained the EUSF. These events are typically ineligible despite roughly the same, but the direct damages reported high emergency response, and information on to the EUSF for these disasters have tripled.27 This medical costs associated with these events is may be partially due to improved assessment of unavailable. The EUSF is not typically used to support damages; however, it could also be that the average MS following wildfires, in part, due to the low amount cost of damages is increasing. If the average severity of eligible direct damage costs for EUSF funding. of losses utilizing EUSF funding is increasing, this indicates that the ability for the EUSF to continue to support MS will reduce. EU INSTRUMENTS FOR DRM PREVENTION A N D P R E PA R E D N E S S I N V E S T M E N T S Droughts have had large impacts on economies in MS; since the EUSF was created in 2002 there have This report primarily examines risk retention tools been four applications submitted and accepted for and advocates for the strategic use of pre-positioned droughts. Funding was provided to Cyprus in 2008 finance. However, it is important to acknowledge the and 2016 and to Romania in 2012 and 2022. In existence of additional funding mechanisms available 2022, Romania which suffered a loss of over €1 to EU MS to strengthen resilience against future billion in the agricultural sector28 due to droughts disasters by investing in mitigation and adaptation and wildfires in the south-eastern region received activities. The EU’s 2021–2027 MFF offers a range of almost €34 million from the EUSF29 to cover some of opportunities for investment in disaster risk the losses from the drought and associated wildfires. management (DRM) and climate change adaptation Currently, MS are not easily able to access EUSF (CCA).31 Furthermore, the EU Green Deal, which funds to support drought losses, due in part to the prioritizes CCA and DRM, also aligns with these challenges in defining the exact start of a drought objectives.32 An overview of the opportunities within and the fact that droughts are often connected to the 2021–2027 MFF is depicted in Figure 11. extreme heat and wildfires, compounding overall impacts. It is the damage from this combined risk The EU created specialized funds after the that leads to the application to the EUSF. recognition of the magnitude of the COVID-19 pandemic including, the Recovery and Resilience The EUSF has been triggered eight times since 2002 Facility (RRF)33 and NGEU initiative. While these for wildfires, in response to damages of €6.1 billion instruments are a direct response to the pandemic’s for these eight events. Since its establishment, there far-reaching economic and social impacts, they focus have been eight successful applications to the EUSF on the provision of finance for mitigation and for support costs associated with wildfires, which adaptation activities. The RRF, in particular, was have received €207.1 million in financing.30 There designed to support MS in their recovery efforts, by are no reported applications for extreme heat events fostering resilience against future shocks. given that the limited physical damage associated Additionally, funding has been earmarked for sectors with these events is unlikely to breach the threshold critical to the EU’s strategic interests, including the of damage in excess of either 0.6 percent of the single market, innovation, and digital; cohesion, 27 EUSF data on disaster aid provided: Link. 28 Link. 29 Almost €34 million in European Solidarity Funds to Romania to repair damages caused by severe drought in 2022 Link. 30 EC. EU Solidarity Fund: Supporting Disaster Recovery 2002-2022. Link. 31 See Forthcoming World Bank and European Commission. 2024. From Data to Decisions: Tools for Making Smart Investments in Prevention and Preparedness. The report serves as a guide for European policymakers, equipping them with the necessary tools and case studies to prioritize initiatives that enhance resilience. 32 EC. 2019. The European Green Deal COM/2019/640. Link. 33 EC. 2023c. The Recovery and Resilience Facility. Link. 31 Risk Retention Instruments in Europe resilience, and values; natural resources and period of crisis underscores the strategic importance environment; and international partnerships. These of pre-positioned finance as having these instruments allocations reflect the EU’s investment in resilience in advance would have saved time and undoubtedly of the Union. The emergence of these funds from a lives during the pandemic. Figure 11. Instruments for DRM prevention and preparedness investments Horizon Europe InvestEU Single market, innovation and digital Other: Euratom Research & Training, Connecting Europe Facility ITER European Space Program European Regional Development Fund Recovery and Rsilience Facility (inc. Technical Support) Cohesion Fund Cohesion, resilience and values REACT EU Union Civil Protection Mechanism European Social Fund+ EU4Health European Agricultural Fund European Agricultural Natural resources for Rural Development Guarantee Fund and environment Programme for Environment and Just Transition Fund Climate Action – LIFE Neighbourhood, Development & Humanitarian Aid Intl. Cooperation Instrument Neighborhood & the world Comon Foreign & Pre-accession Assistance Security Policy Source: World Bank. Note: ITER = International Thermonuclear Experimental Reactor; REACT EU = Recovery assistance for cohesion and the territories of Europe. 2.2. National Level National governments have several financing on economic development. The availability of rapid instruments at their disposal to respond to a liquidity (or the lack of it) may have implications for disaster. National-level instruments are typically timeliness of disaster response and immediate used by EU MS before accessing regional instruments. recovery. National-level instruments are as follows:34 While it is generally assumed that most of the EU MS have some form of a contingency budget or “rainy Reserve funds day fund”, only four EU MS (Austria, France, Hungary, and Italy) have a dedicated disaster reserve fund and Setting aside an adequate amount of budget annually the funds are not earmarked for particular hazards. to meet post-disaster needs can help mitigate Moreover, Phase 1 found that only 10 EU MS have disaster impacts and reduce the need for budget general contingency funds that explicitly cover reallocation in the event of an emergency, in turn disaster relief. lessening the negative impact of budget reallocations 34 Further information on national financing instruments can be found at Link. 32 Risk Retention Instruments in Europe Budget allocations to relevant budget users/ choose to allocate funds in advance to dedicated budget lines budget lines within relevant ministries. These funds may then be spent and reported in line with guidelines To meet needs after a disaster, governments may and relevant legislations. BOX 1: NATIONAL FIRE MANAGEMENT FUND, ARGENTINA In Argentina, a national fire management fund has been actions, including the pilot of new technologies for early established, which after being modified by Law fire detection and warning. The establishment of disaster #27.591/21 has secured a stable source of income of reserve funds, earmarked or not for particular hazards 0.3 percent of all insurance premiums except those of such as in Argentina, may be one avenue to explore in life insurances. With an estimated annual income of strengthening MSs’ ability to finance needs after a around US$70 million, this fund has enabled an increase disaster. in support to local governments’ fire management Budget reallocations ex post capacity to cover a financing gap between a reserve fund and more expensive or longer-disbursing To meet unplanned needs, governments may use sources of funds (such as insurance). budget reallocations between budget lines or users. In times when budget reallocations are the most Contingent credit can be offered by the private sector effective means of meeting unplanned needs, it is or development institutions. For example, the World necessary to ensure that policies to guide reallocation Bank Catastrophe Deferred Drawdown Option (CAT decisions are transparent and adhered to. Having a DDO) provides a line of credit to countries upon plan and process in place can reduce the time successful completion of some policy actions that are required for reallocations and the opportunity costs agreed in advance. The release of the finance is of cancelled or delayed returns from planned contingent upon a disaster event happening and expenditures. Chapter 3 provides more information sequential national declaration of disaster/statement on risk-based budgeting which can be used to of emergency. For example, Serbia used a CAT DDO of strengthen budget allocations to relevant budget €66.1 million to strengthen post-disaster users, budget lines, and reserve funds and make reconstruction and disaster and climate risk informed budget reallocations ex post. management and reduce the fiscal impact of disasters. Contingent lines of credit Insurance (sovereign, asset, and/or sector or stakeholder specific) Contingent credit arrangements offer rapid liquidity that is disbursed following an event of a pre-agreed Risk transfer solutions help in mobilizing private magnitude or based on a pre-agreed trigger (for sector capital (which often exceeds available public example, declaration of a national emergency funds) and can be structured in various ways, for situation). Contingent credits can be fungible or example, as (i) sovereign insurance or capital market conditional by design. As with other sources of credit, instruments (protecting government budgets), (ii) the amount available will depend on the development public asset insurance, or (iii) property insurance for status of the country and the debt-servicing ratio. households. Chapter 4 provides more information on The advantage of contingent credit is its rapidity and insurance. 33 Risk Retention Instruments in Europe 3. Risk-based Budgeting This chapter discusses risk-based budgeting as one way to increase a country’s resilience to shocks. It describes how it can be applied, its potential benefits and how it is currently being applied in some EU MS. Risk-based budgeting practices currently vary across EU MS and more could be done to utilize the benefits of risk-based budgeting. Risk-based budgeting is the consideration of reported in the context of the annual budget. disaster risk throughout the government budget Discussions are currently ongoing to potentially cycle. This enhances a country’s financial resilience amend the EU’s directive (2011/85/EU) on budgetary to shocks. A broad term, it may be interpreted and frameworks of the MS. The amendment calls for the implemented differently by EU MS. Associated terms publishing of disaster and climate-related contingent include risk-informed budgeting, crisis budgeting, liabilities and the assessment and publishing of and disaster-resilient and disaster-responsive public information on the losses incurred and fiscal costs financial management (PFM). due to disasters and climate-related shocks as well as the instruments used to mitigate or cover them. Risk-based budgeting in DRF is a relatively new area of research. There is some guidance on expectations The consideration of disaster risks can occur in all of risk-informed budgeting, but the exploration of stages of the budget cycle. The entry point a country disaster risk across the budget cycle and practices chooses to adopt will depend on capacity and across countries is still evolving. For instance, the priorities as well as the preexisting PFM system. Early International Monetary Fund (IMF) proposes a progress is likely to focus on budget preparation, but three-tier classification for risk disclosure and advanced countries may also consider disaster risks analysis,35 with specific fiscal risks such as disasters during budget authorization, budget execution, and/ encouraged to be regularly monitored and disclosed.36 or budget accountability functions. How a country The Organisation for Economic Co-operation and may consider disaster risk in each stage of the budget Development (OECD) also provides recommendations cycle is detailed in Table 2. Risk-based budgeting and principles for fiscal risk assessments.37 OECD’s may take place at the national, sectoral, or subnational principle 9 on budgetary governance requires fiscal levels. risks to be identified, explained, and classified by type. Fiscal risks should also be quantified and 35 The three layers include basic practice (specific risks to the fiscal forecast are disclosed in a summary report and in qualitative terms), good practice (specific risks to the fiscal forecast are disclosed in a summary report along with estimates of their magnitude), and advanced practice (specific risks to the fiscal forecast are disclosed in a summary report along with estimates of their magnitude and, where practical, their likelihood). 36 IMF, 2018. Fiscal Transparency Handbook. Link. 37 OECD. 2020. OECD Best Practices for Managing Fiscal Risks. Link. 34 Risk-based Budgeting Table 2: Risk-based budgeting activities BUDGET ACCOUNTABILITY BUDGET PREPARATION • Tracking and reporting disaster expenditure • Identifying and quantifying disaster-related contingent liabilities • Evaluating the impact of disaster expenditure • Integrating risk into medium-term forecasts • Auditing disaster risk and expenditure • Mainstreaming risks into annual budgets • Publishing disaster expenditure for civil society organization (CSO) engagement • Budgeting for DRF instruments • Risk-informed public asset management • Risk-informed revenue budgeting BUDGET EXECUTION BUDGET AUTHORIZATION • Emergency procurement procedures and protocols • Ensure Parliament has sufficient information related to disaster risks • Risk-informed budget reallocation processes • Build capacity of relevant parliamentary committees to • Design effective disbursement mechanisms for DRF scrutinize the management of disaster risks instruments The implementation of risk-based budgeting can risk-layering strategy; an approach to cost-effectively strengthen financial resilience. For instance, combine different sources of funding to manage the considering disaster risks in medium-term forecasts financial impacts of disasters. Table 3 shows some of and annual budgets creates greater certainty on how the benefits in implementing the risk-based budgeting unexpected costs may be financed, and risks activities, as means to enhance financial resilience, managed. Information produced through risk-based at various stages of the budget cycle. budgeting can also support development of a disaster Table 3: Benefits of risk-based budgeting across the budget cycle BUDGET ACCOUNTABILITY BUDGET PREPARATION • Supports the transparent tracking and reporting of • Helps address expenditure and revenue risk across disaster expenditure different time scales • Supports the production and use of established • Creates greater certainty around what disaster costs guidelines and procedures for the allocation and use of may be and how they will be financed. funds • Supports government budget planning and the management of financial risks, including the design and use of DRF instruments BUDGET EXECUTION BUDGET AUTHORIZATION • Helps link budgeting to implementation (for example, • Links appropriate sources of funding with predefined emergency procurement) priority activities • Can create a predictable funding source that is • Expediates the funding approval process, resulting in separate from regular budget allocations, which may faster release of funds reduce the opportunity cost of in-year adjustments • Provides greater transparency for parliament and • Supports the effective and efficient use of resources budget users on how funds may be utilized after a disaster 35 Risk-based Budgeting Risk-based budgeting practices vary across EU MS. • Finland: Disaster-related fiscal risks are considered A recent EC report,38 for instance, notes that despite in long-term projections. its considerable relevance, the analysis of climate- related risks has “often been absent from fiscal • Germany: Estimates of disaster damages by 2030 sustainability frameworks of official institutions, for individuals, companies, and critical notably due to inherent difficulties in conceptualizing infrastructure have been provided for the German and quantifying such aspects.” However, some Environment Agency. positive developments are noted in a recent EC report:39 More could be done to utilize the benefits of • Belgium: Establishment of a coordinating body, risk-based budgeting. the Centre of Excellence on Climate, to analyze and evaluate climate-related risks. • Georgia: Inclusion of fiscal risks from natural disasters and climate change in fiscal risk At the EU level, notable initiatives on fiscal matters statement. The statement provides a historical and climate change relate to ongoing work on ‘green perspective on financial losses and numbers of budgeting’ and inclusion in the 2019 Debt people affected by disaster, a forward-looking Sustainability Monitor on how to encompass climate assessment of annual expected damages at change impacts on growth and public finances in the different periods, and an overview of budgetary standard EC’s Debt Sustainability Analysis. Box 2 instruments for DRF. provides further examples of risk-based budgeting side of the activities being undertaken by countries out­ • Spain: The Spanish fiscal council is working to EU. incorporate climate change into macroeconomic projections. 38 EC. 2022a. Fiscal Sustainability Report 2021, Volume 1. Link. 39 EC. 2022b. “Disaster Risk Financing: Limiting the Fiscal Cost of Climate-Related Disasters.” Discussion paper 174. Link. 36 Risk-based Budgeting BOX 2: EXAMPLES OF RISK-BASED BUDGETING PRACTICES OUTSIDE THE EU40 • Budget preparation: Contingent liability management • Budget execution: Advance procurement agreements in Colombia. Colombia’s annual medium-term fiscal in Japan. The 2011 earthquake and tsunami caused framework includes contingent liabilities related to US$43 billion damage to infrastructure and public disasters (La Niña and earthquakes). This information utilities. Rapid reconstruction was key to supporting feeds into projections for revenue, expenditure, and relief activities and saving significant indirect costs. debt dynamics. Mitigation instruments and financing To support rapid reconstruction efforts, the sources are also identified for each contingent government enters into pre-disaster agreements with liability. For disasters, the Ministry of Finance has the private sector for construction, engineering, identified the national fund, agricultural insurance, surveying, telecommunications, and broadcasting.41 CAT DDO and CAT bond, and the financial protection • Budget accountability: Disaster budget tagging in strategy as core mitigation and financing instruments. Ethiopia. Ethiopia has undertaken numerous stand- • Budget authorization: Office for Budget Responsibility alone expenditure reviews on disaster risk reduction/ (OBR) in the United Kingdom. The OBR was climate change spending, but its accounting established in 2010 to provide an independent structure makes it difficult to carry out regular examination on the sustainability of public finances. assessments. A budget tagging system has therefore It has provided a biennial report on fiscal risks since been introduced to flag adaptation/mitigation 2015. For instance, its 2021 Working Paper No. 17 spending as well as spending on the whole disaster report focused on three catastrophic risks— cycle. A dual tagging system has been put in place to COVID-19, climate change, and the cost of dorThis reflect synergies between climate and disaster risk report noted that the projected medium-term legacy reduction. The budget tagging system will be costs of the pandemic are likely to be £10 billion per embedded into the Integrated Financial Management year, which are currently unfunded in the government Information System, with reporting included in the budget. In 2023, the OBR, published Discussion 2023/24 budget. Paper 4, which details the work done to date to understand and analyze the fiscal impacts of climate change on the budget. 40 For further information and case studies, see World Bank. Forthcoming. Disaster Risk Based Budgeting: Introduction and Options for Operationalization. 41 Link. 37 Risk-based Budgeting 4. Risk Transfer Instruments Risk transfer solutions help in mobilizing private sector capital to complement limited public funds and can be structured in various ways, for example, as (i) sovereign insurance or capital market instruments (protecting government budgets), (ii) public asset insurance, or (iii) property insurance for households. These instruments can be developed on a hazard-by-hazard basis or using a multi-peril approach. This chapter presents a general discussion of the instruments to clarify the basis on which these products can be developed. More could be done to incentivize risk transfer at both the EU and the EU MS levels. 4.1. Sovereign Insurance or Capital Market Instruments The risk of disaster losses can be transferred to the products, in particular, can provide fast payouts in a private insurance market (or to capital markets in matter of days or weeks and the proceeds can be the case of CAT bonds) via sovereign insurance— used for emergency relief. that is, an insurance policy where a government (the sovereign) is the policyholder. Together with risk Sovereign insurance can be structured in different retention instruments such as budgetary reserves ways, such as indemnity products or parametric and contingent credit, sovereign risk transfers can be products which are based on an index or modelled a key part of a comprehensive DRF strategy.42 loss. Indemnity-based products require extensive loss assessment processes before a payout can be At present there are no risk transfer products at the confirmed but have the advantage that payout is EU level or in the case study countries, and closely tied to the loss that occurs, so that basis risk43 consideration should be given to their incorporation is minimal. Typical household insurance, for example, in future DRF strategies. There is a trade-off with risk follows an indemnity approach. A modelled loss– transfer products as the initial premium is to be paid based product relies on the assessment of loss using up front regardless of payouts. However, in extreme an agreed independent risk model, with a payout loss years, the payouts from a risk transfer product occurring if a modelled loss threshold is exceeded. A can be many times the premium. Parametric parametric index is a simplified version of this 42 World Bank, “Sovereign Catastrophe Risk Pools: A Brief for Policy Makers,” Link. 43 Basis risk is the level of potential risk that exists when a calculated loss of a model or insurance index differs from the actual incurred loss. Any mismatch between the two will result in a discrepancy in the payout received. This could mean either that a payout is higher than the actual loss incurred (positive basis risk) or that the payout is lower than the actual loss (negative basis risk). 38 Risk Transfer Instruments approach and uses formulas to estimate loss from an that allows the issuer to raise funds in case of a event that has occurred. Parametric products pay out natural disaster and does not count against a when an event occurs that meets a pre-agreed country’s debt ceiling. CAT bonds are high-yield debt definition (in terms of type, location, and hazard instruments that pay out only if a specific event such intensity threshold). Thus, they typically offer faster as an earthquake or a flood occurs. If the insured payouts than other types of sovereign insurance but event occurs and triggers the payment to the bond are subject to the greatest basis risk. Parametric issuer, the principal will be used to cover a part of the products can be a useful tool for budget support or as losses. Investors who are ready to take this kind of a backstop into existing reserve funds for a risk target CAT bonds because they offer attractive government. rates of return that are usually higher than other fixed-income securities. In addition, because losses Governments can also transfer their disaster risk on CAT bonds are not correlated with those of other through capital market instruments. For example, a capital market instruments, they offer portfolio CAT bond is a risk transfer capital market instrument diversification for large investors. BOX 3: EUROPA RE Europa Re was launched in 2009 as a public-private countries has been low. As of 2020, there are three partnership reinsurer for property and casualty risks, participating countries—Albania, North Macedonia, and specializing in catastrophe risks, but the take-up from Serbia—who are shareholders of this entity. 4.2. Public Asset Insurance Some of the biggest risks from climate shocks, like Insurance cover for public assets may be voluntary or drought, extreme heat, and wildfire, come from compulsory, and products may be indemnity-based their impacts on critical systems—food systems, (payout determined by the assessed losses), supply chains, and infrastructure. These so-called parametric (payout based on the occurrence or systemic risks can potentially cause cascading severity and location of a hazard event), or a hybrid; impacts on people and economies. These types of each has its own benefits and challenges. Financial risk can be much more difficult to assess, but their management of public assets can be complemented impacts can be substantial. Many countries are now by risk retention instruments. increasingly aware of the importance of strengthening the resilience of their critical infrastructure and are As found in Phase 1, data on penetration of public looking for ways to go about making the decision on asset insurance in Europe are limited. Bräuninger et which critical infrastructure to protect and how. al. (2011) mention that only one-third of private and public assets in the EU are insured against floods and Public asset insurance, as part of a comprehensive drought (with public assets generally uninsured).44 disaster management strategy, can smooth Fire is usually included as the norm unless explicitly expenditures by helping avoid budget shocks stated overwise. Bulgaria has a legal requirement for through transferring the risk to the private insurance municipalities to buy insurance for their assets sector. It can provide benefits by pooling multiple (excluding against floods in high-flood-risk areas), assets into an insurance scheme, thereby diversifying but the extent to which this law has been implemented the risk and reducing the premium cost per asset. is unclear. 44 Bräuninger et al. 2011. “Application of Economic Instruments for Adaptation to Climate Change.” Link. 39 Risk Transfer Instruments 4.3. Household Insurance Catastrophe insurance for households can be governments should be careful not to disincentivize offered in different ways. Globally, household its uptake. The consensus among insurers on the low insurance against disasters is offered by private penetration is that the government will support sector insurers, but the number of public or private- citizens when there is a large disaster and therefore public schemes is growing. Catastrophe insurance there is a lack of will to purchase insurance. It has (usually referred to as ‘natural catastrophe’ or been suggested from multiple insurers that this ‘NatCat’ cover) is offered as a stand-alone product, provides an argument to make earthquake insurance as part of a general property damage policy or as an coverage mandatory. This would need to be coupled add-on to such a policy. It can be either compulsory with awareness creation on the importance of or voluntary (or voluntary with some elements of financial resilience. The ad hoc payouts from the obligation, for example, through mortgage contracts). government will also affect uptake in other sectors of Depending on the country, flood, storm, or earthquake insurance such as in agriculture insurance which is cover (or some combination of these) is commonly particularly pertinent for drought. provided as standard in a household policy45 in addition to standard cover such as fire, and payout The following are some examples of catastrophe is usually on an indemnity basis, requiring assessment insurance for households in Europe: and adjustment of losses after an event. • Wildfire coverage. It It is generally the case that Property insurance policies often cover damage household fire policies would include damage from caused by wildfires as this is included under the wildfires in most EU MS. However, the team is traditional fire policies. For most EU countries, aware that reinsurers such as Munich Re provide wildfire insurance coverage for homeowners is specific cover for wildfire in Greece and Portugal mandatory either by law or through mortgage where there is sufficient volume of assets at risk. requirements by banks.46 Nevertheless, insurance penetration is lower for agricultural sites or forests. • Agriculture insurance. Another area which can Also, several parametric insurance schemes have affect government contingent liabilities is crops or become available over the past years. However, in livestock which are destroyed from a natural well-established wildfire insurance markets like disaster. It is common in many countries for local California, primary insurers have been challenged by and central governments to offer support to the current inflationary environment and the farmers after a disaster event. Having agriculture resistance of regulators to allow rate increases that insurance, whether for the farmer, agriculture- would cover the estimated cost of this risk in relation lending institutions, or local/central government, to inflation.47 This has seen some insurers raise could support ex ante planning for this sector insurance rates by more than 30 percent or cease instead of making ad hoc payments after the event. offering new home insurance policies altogether, Understanding the government’s exposure to prompting a discourse on the feasibility of private agriculture losses, that is, where ad hoc payments insurability and the need for adequate insurance are being made to farmers after an event, would be rates and regulatory environment.48 the first step in understanding where government- subsidized agriculture schemes might be a more The penetration of private sector catastrophe cost-effective intervention. The Common insurance varies largely between EU MS, and Agricultural Policy paved the way for increased 45 EU JRC 2012. Link. 46 Link. 47 Link. 48 Bloomberg news. Link. 40 Risk Transfer Instruments crop insurance support through premium subsidies with a forecast indicating these subsidies will (Regulation (EU) 2021/2115), and consequently, increase to €523 million by 2027, in recognition of there was a sharp rise in publicly funded insurance the fact that insurance can serve as an important in EU MS from €133 million to €386 million in risk management tool in a sector facing high and 2020.49 It is expected that this trend will continue, increasing risks due to climate change. 4.4. Market Overview W I L D F I R E R I S K T RA N S F E R-L E V E L • Potential risks to public finances posed by wildfires INSTRUMENTS should be evaluated to manage the impacts of disasters on public finances. An approach to Catastrophe bonds are increasingly being used to managing those financial needs should be cover the liability risk associated with the risk of developed, including mechanisms for estimating, wildfires. Many utility companies are now seeking to accounting, and disclosing contingent liabilities use CAT bonds to manage their liability risk from associated with losses to critical sectors. wildfires, recognizing that faults in their equipment could cause a fire and a fire could equally damage • The National NatCat insurance system, if in place their utility plants.50 and effectively enforced, should cover the risk of wildfire. The 2023 EU Wildfire Peer Review assessment further notes that a financing strategy for wildfire • A variety of funding sources should be used at the risk should be based on an integrated, multi-hazard national, subnational, and local levels. The EU approach and cooperation across levels of funding instruments, such as the Resilience and government and with relevant stakeholders should Recovery Facility, Cohesion Policy Funds, be established, with the necessary resources and Agriculture and Rural Development Fund, the LIFE expertise to manage the financial impacts of severe program, the Technical Support Instrument, the and large-scale events. To further strengthen the EU Mission on Adaptation to Climate Change, and financing of wildfire events, the 2023 report made UCPM funding programs, should be fully exploited several recommendations: in the event of wildfires and other disasters.51 BOX 4: LINKING PARAMETRIC INSURANCE TO ECOSYSTEMS Combining ecological action with financial protection Conservancy and Willis Towers Watson (2021) found that can make good economic and financial sense and help residential insurance premiums could decrease by overcome the pricing issues associated with wildfire 41 percent when ecological forestry techniques such as risk. For example, an ecological forestry approach linked forest thinning and prescribed burning were applied to a to parametric wildfire losses could reduce losses for the relevant area. Without such ecological measures, the insurance and reinsurance sector. A study by the Nature risk of wildfire continues to grow. 49 Link. 50 Cook, S., and S. Holliday. 2022. Insuring Nature's Survival: The Role of Insurance in Meeting the Financial Need to Preserve Biodiversity. Link. 51 Opportunities for DRM investments are discussed in World Bank and European Commission. 2024. From Data to Decisions: Tools for Making Smart Investments in Prevention and Preparedness. 41 Risk Transfer Instruments The economic costs of wildfires within the EU are different disaster risk schemes that could be used to carried through different vehicles, with some novel manage climate risk and includes a similar approaches gaining traction over more common recommendation to that made in EDPP Phase 1 to traditional (re)insurance schemes. Figure 12, deve­ introduce a risk transfer scheme at the EU level that loped by the European Insurance and Occupational could be connected to either the UCPM and/or EUSF. Pensions Authority (EIOPA), presents a summary of Figure 12: The ladder approach to catastrophe insurance Low frequency / high impact EU component in excess of national level / (high loss layer) alternative risk transfer PPP (national) / other public (national) meas- ures / alternative risk transfer − supplement- ing coverage by private sector Reinsurance / reinsurance pool / alternative risk transfer (e.g. cat bonds) − private sector High frequency / Insurance / insurance pool − private sector low impact (low loss layer) Source: EIOPA, European Central Bank.52 In the reinsurance market, reinsurers tried to reduce to the United States (especially California), CAT their wildfire exposure over the past few years due bonds as a risk transfer mechanism are less common to the recent spike in (re)insurance claims, that is, in the EU and should be further explored once risk from the megafires in Portugal (2017) or Greece models are available to see if this could provide a (2021) that caused more than €1 billion damages.53 cost-effective option to manage the risk of wildfires. This led to increasing reinsurance prices. Compared BOX 5: GOVERNMENT SUPPORT TO WILDFIRE INSURANCE SCHEMES In some countries, the government provides financial Réassurance (CCR) in France. Furthermore, governments assistance to help cover the costs of wildfires that might support disaster struck regions and communities, exceed the capacity of insurance policies. One example such as Portugal in 2017. However, these financial would be the Consorcio de Compensación de Seguros resources need to be made available through a (CCS) in Spain or the publicly owned Caisse Centrale de reallocation of government funds. Given the prominence of infrastructure assets Department of Water and Power (LADWP) secured located in or around a forest, a specific wildfire US$30 million of wildfire insurance cover from its instrument for these assets may be beneficial for second CAT bond issued via Power Protective Re. Ltd. the EU, and lessons could be learned from The product embeds liability protection in this CAT California.54 For example, the Los Angeles bond deal and recognizes that the assets insured 52 Policy Options to reduce the Climate insurance protection Gap Link. 53 S&P Global. Link. 54 Cook and Holliday 2022. 42 Risk Transfer Instruments may themselves cause wildfires. However, it should Parametric risk transfer solutions are increasingly be noted that while wildfire insurance or reinsurance being used to cover the risk associated with capacity is available from the Index-Linked Securities droughts. Identifying the drought typologies that (ILS) market, agreeing on pricing may be difficult, as drive impact in each EU country (that is, hydrological, ILS funds and investors are demanding higher returns meteorological, and/or agricultural droughts)57 for the California wildfire peril after recent losses. helps determine the associated metrics to consider That said, this is a new peril and pricing may change when designing a risk transfer mechanism, based on as we learn more on the risk itself. information on (i) water security and vulnerability in the context of surface water, ground water, and rainfall reliance as well as on (ii) critical impacts. R I S K T RA N S F E R M E C H A N I S M S F O R E XT R E M E H E AT A N D D RO U G H T Testing and determining parametric triggers in support of a drought risk transfer mechanism Limited risk transfer mechanisms are available for requires engagement with key country stakeholders. extreme heat and drought, particularly compared to These engagement needs include, but are not limited other climate extremes such as floods, storms, or to, addressing questions around the key preferred wildfires. As noted earlier, extreme heat gives rise to (sectoral) impact focus to determine the final drought costs in four ways: (i) direct impacts on the health definition and relevant metrics (for example, sector; (ii) a decline in labor productivity both indoors agriculture/crops, water) as well as further and outdoors; (iii) increased energy expenses due to meteorological and hydrological data and as an heightened cooling demands; and (iv) direct and assessment of their relationship at the individual indirect effects on critical infrastructure, such as country and subcountry levels. power plants or roads, which may experience service disruptions. Similarly, droughts increase costs An assessment of EU countries regarding the cost through (i) a decline in productivity from the water- for immediate response measures helps determine dependent industry such as water-intensive the dominant driver of drought response (for manufacturing, agriculture/forestry, food production, example, distribution of relief supplies) and support power generation, and water distribution and (ii) discussions around coverage needs. Consideration direct impact on the health sector. should be given to (i) testing the feasibility of existing social protection schemes to ensure no opportunities There could be cases where certain economic are missed; (ii) building efficient and rapid payout consequences can be transferred to insurance via delivery structures to increase the value for money of business interruption, for example. Business insurance payouts and thus the return on investment interruption policies can address risks related to of EU countries premium payments; and/or (iii) nuclear power plants when the cooling water combining early action and rapid response temperature rises excessively or to outdoor work interventions to capitalize on the multiplier effects when the temperature reaches hazardous thresholds (for example, by designing a two-step trigger for workers. Additionally, weather derivatives are mechanism), which combines the initial release of employed to hedge against revenue losses within the early action resources and later release of rapid energy sector in the United States.55 response funds into one parametric coverage. Further Lastly, emerging parametric insurance products for research may help decrease overall payout and thus extreme heat are entering the market to safeguard coverage needs, thereby supporting premium the income of (informal) workers against salary losses affordability as well. in India.56 55 Link. 56 Forbes. Link. 57 GAR 2021. 43 Risk Transfer Instruments 5. Funding Gap Analysis This chapter outlines the estimated contingent liabilities related to wildfires and droughts that can be budgeted for, based on existing cover from financial instruments held by the EC and selected EU MS. It provides an indication of what proportion remains to be financed, commonly referred to as the funding gap. Analyzing the funding gap provides a tool to help countries build their DRF strategies. There is an increasing need for informed financial decision-making on how much funding to allocate before disasters, how to evaluate risk transfer instruments, and how much to spend on risk reduction. The funding gap analysis is first applied to the region available for damage incurred and how the amount of (all EU MS). The analysis of the UCPM budget over finance required differs across countries. To facilitate time indicates that most of the expenditure has been the analysis and assess the potential funding gap on response activities. Therefore, the analysis focuses related to overall estimated contingent liabilities. on estimating these costs but recognizes that other costs associated with these hazards will emerge to A fundamental assumption underpinning this represent the true cost. The funding gap analysis analysis lies in the ratio of public sector costs considers the MFF budget lines associated with compared to total costs. Recognizing the inherent emergency response activities and the EUSF based uncertainty and observed fluctuations in the cost on records between 2002 and 2020. The analysis metric, the analysis assesses two additional scenarios assumes that the resulting funding gap will be to highlight the potential variability based on existing covered by (i) budget reallocation, (ii) debt, or (iii) EUSF budget data information: bilateral donor assistance. Due to data limitations, the analysis does not review alternative DRF strategies • The low scenario assumes that 30 percent of total that include sovereign risk transfer. cost is attributed to the public sector. In addition, five selected national case studies are • The high scenario assumes that 80 percent of total analyzed: Greece, Italy, Romania, Croatia, and cost is attributed to the public sector. Bulgaria. The approach to the country-level funding gap analysis considers overall economic losses The scenarios demonstrate the potential range of incurred as a closer representation of the contingent the financial impact associated with different liabilities member countries will face in the wake of proportions of public sector costs, and these are wildfire and drought events. These case studies applied to both emergency operation costs and demonstrate how combining different risk financing damages separately. instruments can affect the amount of finance 44 Funding Gap Analysis Two case study countries—Croatia and Romania— countries. Furthermore, the cost estimates are based are supplemented with information from discussions on economic losses which encompass firefighting, with the respective governments. The case study reforestation, damages, and cleanup. discussions of both countries contextualize the funding gap within the broader picture of what financing Wildfires, extreme heat, and droughts are complex instruments are currently available to each MS. perils to model, due to the broad range of factors and conditions influencing their occurrence and The approach to assessing the loss profile and severity. Therefore, unlike earthquakes and floods, contingent liabilities varies considerably compared there are currently no fires, extreme heat, and drought to other hazards. For example, wildfires result in loss models available in Europe. However, loss data varying damage to different types of assets and occur exist for historical events, with detailed event reports in different locations compared to those affected by (for example, EFFIS). Therefore, wildfire, extreme other perils like earthquakes and floods. In EDPP1, heat, and drought impacts can be derived using the losses were predominantly driven by the damage and loss ratios comparable to historical household sector which accounted for over 50 percent events. In the case of wildfires, models are currently of total loss. The areas historically affected by wildfires under development to give probabilistic estimates have a lower penetration of assets across the sectors (for example, CIMA). As mentioned in Section 1.2, (for example, residential, commercial, industrial, existing models on wildfire do not focus on the health, and education) due to their location so there generation of ‘full’ event-based catastrophe risk will be underlying differences in the asset values modelling approaches and there are no probabilistic being analyzed. Other factors associated with risk models which can be used to estimate future loss, so management, preparedness, and reduction measure the analysis here will differ from that conducted in MS will have a bearing on the overall emergency under EDPP1. In the absence of probabilistic cost associated with these events and the extent to modelling, a straightforward methodology has been which these are effective will likely vary across employed for each hazard. 5.1. Wildfire Results of the EU-Level Analysis M E T H O D O LO G Y • Tier 2: Tier 1, combined with an estimation of the EUSF annual allocation for wildfire events. The methodology applied involves calculating the average cost per hectare burned and applying it to the total number of hectares documented in the E M E RG E N CY O P E RAT I O N S CO S T EFFIS database.58 Details on this methodology are included in Annexes 6 and 7. On the financing side, The total cost for emergency operations associated two separate scenarios are analyzed based on with wildfire events ranges between €41 million and accessibility to different funding tiers: €752 million depending on the scenario assessed and the year observed (Figure 13). The proportion of • Tier 1: Estimates national reserves held by EU MS costs associated with emergency operations varies combined with the UCPM budget associated with across all events but is estimated to be between emergency operations costs for wildfire events 13 percent (low estimate) and 35 percent (high estimate)59 of total costs of the contingent liabilities 58 JRC EFFIS. Link. 59 The estimates were developed using data from EC - cohesion data (Link.) which include total direct damage estimates and emergency operation cost estimates per event in addition to the EUSF. 45 Funding Gap Analysis from each year. The spike observed in 2017 On average, the cost of emergency operations is corresponds to the wildfires in Italy, Spain, and estimated to be between €105 million (low scenario) Portugal, resulting in a significant increase in the area and €294 million (high scenario). burned and associated emergency response costs. Figure 13: Estimated emergency operation costs 800 700 600 500 €, millions 400 300 200 100 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023* Year low scenario Midpoint scenario High scenario Source: World Bank. An initial assessment compared Tier 1 allocations to recent three-year (2021–2023) average costs and estimated emergency operation costs, indicating budget allocations, respectively. The funding gap is that the recent rise in UCPM budget for wildfire visually represented by the red columns in the graph, emergencies has reduced the funding gap since while the green columns indicate years where it is 2019. Figure 14 shows the estimated funding gap "estimated that no funding gap existed (that is, based on historical emergency cost estimates for sufficient funds were available to cover emergency wildfire events across EU MS and Tier 1 funding operation costs). arrangements and an estimate based on the most 46 Funding Gap Analysis Figure 14: Wildfire emergency funding gap - Tier 1 funding Estimated Funding Shortfall Estimated Excess Funding 400 350 300 250 €, millions 200 Total UCPM budget (Response +Firefighting) 150 Reserves at EU level 100 for wildfire 50 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023* 3-yr Average -50 No funding gap -100 Source: World Bank. Note:* Estimates based on area burned by wildfires in Europe as of September 2023. Analysis of the national reserves held by the EU MS Between 2014 and 2019, there was an average combined with the UCPM budget associated with funding gap of €84 million based on the midpoint emergency operations costs for wildfire events shows scenario (range of €59–235 million). This indicates no funding gap was estimated based on the average that there was insufficient funding was available of the last three years (2021–2023). This is due, in within national reserves and relevant UCPM budget part, to the increase in the amount allocated to the allocations (Tier 1). relevant UCPM budget lines between 2019 and 2022.60 A funding gap of around €190 million could occur During the historical period between 2014 and based on the assumptions from the high scenario. 2023, the largest funding gap of €301 million was Given the uncertainty surrounding the loss estimates estimated in 2017, and would require a 70 percent for wildfire, this could better reflect future loss. increase to the 2023 UCPM budget to cover similar However, the three-year average between 2019 and losses in future years. This funding gap was caused 2022 may better represent current practices, which by a combination of high losses (€337 million indicates that enough funds are available to cover all midpoint estimate, range of €233–717 million) and a emergency operation costs, although this varies modest level of funding at the time (€36 million). depending on the assumed level of emergency Comparing the emergency response costs in 2017 to operation costs in proportion to the overall cost. the 2023 UCPM budget of €177 million shows that a 70 percent increase in the 2023 UCPM budget would be necessary to cover losses in similarly severe years. 60 Data from the UCPM indicate an overall increase of €294.9 million between 2019 and 2022. Specifically, in 2022, €149.5 million was allocated for response activities (a 20 percent increase since 2019), and €39.9 million was allocated to the rescEU transition (firefighting) (an 8 percent increase since 2019) within the MFF sub-budget from 2019 to 2022. 47 Funding Gap Analysis The initial analysis of emergency operation costs into the analysis was done to represent the additional was revised to demonstrate the impact of adding support that could be granted to EU MS from the the EUSF annual allocation to fund (or reimburse) regional level. The results of this analysis are shown emergency operation costs. The addition of the EUSF in Figure 15. Figure 15: Emergency funding gap - Tier 2 funding Estimated Funding Shortfall Estimated Excess Funding 400 300 €, millions 200 EUSF Annual Allocation for Wildfire events 100 Total UCPM budget (Response +Firefighting) Reserves at EU level for wildfire 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023* 3-yr Average -100 No funding gap Source: World Bank. The addition of the EUSF funding source indicates TOTA L CO S T (T H E S U M O F E M E RG E N CY A N D that adequate funding was available to cover DA M AG E CO S T S) emergency operation costs for wildfire each year except 2016 and 2017, which required €21 million Additional analysis was undertaken to evaluate the and €227 million, respectively. However, there are adequacy of Tier 2 funding arrangements that limitations on the speed of disbursement from the increase costs to account for emergency costs and EUSF, and funding from the EUSF is not guaranteed61 damage incurred. The extent of public sector for wildfires or any other hazard. In all other years damages fluctuates based on the occurrences of between 2014 and 2023, funding covered wildfire events in any given year. According to EUSF budget events, although there is a chance that resources data, public sector damages typically constitute could have been depleted by earthquakes or floods. around 50 percent of total damages but can range Furthermore, the additional resources provided by from as low as 0.5 percent to as high as 83 percent. the EUSF reduced the 2017 funding gap estimate To accommodate this uncertainty, the low scenario from €301 million to €227 million. Incorporating the assumes public sector damages to be 30 percent of EUSF in the three-year average between 2021 and total damages, while the high scenario increases this 2023 would raise the funding buffer from €12 million estimate to 80 percent. to approximately €86 million, in excess of estimated losses. 61 The EUSF can only be activated at the request of the applicant state within the deadline of 12 weeks from the first damage occurred, demonstrating that the total direct damage exceeds the thresholds specified in Article 2 Regulation (EC) No. 2012/2002. 48 Funding Gap Analysis The Tier 2 arrangements are not intended to cover estimated damages from wildfires (equivalent to the entirety of public damages; hence a funding gap €26 million per year on average) and has consistently is expected. The purpose of this analysis is to hovered in the range of 2–3 percent. However, the emphasize the residual portion to develop the case outcomes of the analysis are highly sensitive to this for pre-positioned finance, that is, where pre- assumed value. If a decision were made to increase positioned DRF instruments could have offered EUSF allocations, even by a few percentage points, to additional financial protection. It also aims to quantify cover a greater proportion of total costs, it could the historical magnitude of protection that may have significantly reduce the funding gap. This underscores been required. Additional assumptions were the importance of allocating resources based on introduced to ascertain the proportion of total costs where the need is greatest, whether for recovery that these Tier 2 instruments will address. (generally provided by the EUSF) or relief efforts (generally provided by the UCPM). Figure 16 shows Historically, the EUSF has covered a relatively small the results of this analysis. fraction of total costs, approximately 2.3 percent of Figure 16: Total annual cost - funding gap midpoint scenario Estimated Funding Shortfall Estimated Excess Funding 500 400 300 EUSF Annual Allocation for Wildfire events 200 €, millions Total UCPM budget (€ m) 100 (Response +Firefighting) - Reserves at EU level 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023* 3-yr Average -100 -200 -300 No funding gap -400 Source: World Bank. The analysis suggests that Tier 2 instruments can wildfires, providing additional protection to EU MS and sufficiently cover 2.3 percent of total costs from wildfire alleviating a significant financial burden. Further analysis including 2017, meaning that 97.7 percent of the cost is necessary to comprehend the implications of such a needs to be met from other financial instruments at the change, considering the variable nature of wildfire events MS level. Based on the average experience from 2021 to and EUSF allocations for costs associated with other 2023, continuing this practice would result in estimated hazards. In addition, the EU should decide upon its risk excess funds ranging between €153 million and 255 appetite to inform what proportion of the damage should million. However, this approach implies that the remaining be covered at the regional level. 97.7 percent would need to be covered by other resources available to the EU MS affected by the event. This simplified A 6.5-fold increase in funds for the UCPM and EUSF analysis indicates potential flexibility for increased would result in a cost-neutral outcome where the allocations from the EUSF to fund total costs related to three-year funding gap (2021–2023) is expected to 49 Funding Gap Analysis be zero based on the midpoint estimate; this could pattern of deficits and surpluses over the observed be targeted to reduce the funding gap. By adjusting years. It is important to note that this calculation aims the percentage of total costs that the EUSF and the to neutralize the funding gap based on the three-year UCPM could cover, it was found that if these average, leading to larger cumulative deficits instruments covered 10.6 percent of total costs the compared to cumulative surpluses over the anticipated funding gap based on the midpoint (€299 observation period (primarily due to the large losses million) would be zero, with a range of €99 million in 2017). Similarly, if the objective were to smooth (surplus) in the low scenario and a €366 million expenditures for a cost-neutral outcome over the funding gap in the high scenario. Figure 17 illustrates observation period (2014–2023), the EUSF and the potential shift in the funding gap profile (midpoint UCPM allocations would be designed to cover a estimate) if this adjustment were implemented. This slightly lower percentage of total costs, specifically adjustment results in a more evenly distributed 9.3 percent. Figure 17: Total cost funding gap - midpoint scenario (10.6 percent of total costs) Estimated Funding Shortfall Estimated Excess Funding 800 700 600 500 €, millions 400 EUSF Annual Allocation for 300 Wildfire events 200 Total UCPM budget (€ m) (Response + Firefighting) 100 Reserves at EU level - 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023* 3-yr Average -100 -200 No funding gap Source: World Bank. 5.2. Drought results of the EU-Level Analysis M E T H O D O LO G Y structure from soil subsidence, decreased inland water transportation, and reduced energy production. The analysis relies on the estimate provided by the Although some losses in these categories may not be JRC, which suggests that current annual losses from considered contingent liabilities for the EU MS, no drought in the EU and United Kingdom amount to adjustments have been made to the loss estimates. approximately €9 billion.62 These losses span various This is due to the lack of sufficient information impacts, including diminished public water supplies, available to make accurate assessments of these agricultural losses, damage to buildings and infra­ contingent liabilities or the proportion of impacts 62 JRC. 2020. Global Warming and Drought Impacts in the EU. Technical Report. PESETA IV Project - Task 7 - Drought. Link. 50 Funding Gap Analysis within the United Kingdom or other non-EU MS. support for drought - only events, partly due to the However, a consistent approach has been upheld by complexities in defining the beginning and end of referencing historical data on the estimated such events.63 To address this, we hypothetically proportion of damages covered by the respective consider the following scenario: given that 15 percent financing instruments for wildfires. This methodology of the annual allocation in the EUSF is designated for helps determine the potential extent to which total wildfire support64 and the average annual total cost of drought-related costs may be covered by these wildfire events amounts to €1.14 billion, while the financing instruments. average annual total cost of drought is approximately €0.650 billion,65 we assume that the EUSF would According to the JRC, if the future climate affects theoretically allocate 8.6 percent to drought events present-day society, the overall drought damage in based on these figures (€69 million). the EU and United Kingdom will experience a slight rise with a 1.5°C global warming (reaching €9.7 Considering the nature of drought hazards, and the billion annually). However, this damage escalates challenges in identifying the exact time frame of significantly with further warming, reaching €17.3 the impact, it is not feasible to determine an billion per year at 3°C. These estimates form the basis emergency cost component and only total cost will for providing loss estimates under different scenarios be analyzed. Therefore, the analysis conducted in the funding gap analysis. here focuses solely on coverage for the total cost component to be covered by the relevant Tier 2 Regarding drought-related financing, the budget instruments. These include estimated national within the UCPM allocated to address drought- reserves held by EU MS combined with the UCPM related expenses is assumed to equal the average budget allocated for general emergency response allocation designated for emergency response activities, along with an estimate of the EUSF annual activities between 2020 and 2022 (€104 million). allocation for drought-related events. This approach This is slightly lower than that for wildfire estimates overlooks the timing of potential fund disbursements, due to the specific allocations earmarked for which could potentially increase as the severity and firefighting purposes. The EUSF has never provided duration of the drought escalates. 5.3. Total Drought Cost M E T H O D O LO G Y timing, considering the chronic and extensive nature of drought hazards. The analysis evaluated the adequacy of the Tier 2 funding arrangements to finance the total costs of Similar to the wildfire analysis, three scenarios were drought events across the EU and United Kingdom. created to address the uncertainty regarding how The analysis provides a comparison of average total much of the total costs the UCPM and EUSF will costs from droughts with financing instruments. cover for drought events. Including these scenarios However, given that these instruments are not in the analysis compensates for the lack of any designed to cover all costs, funding gap arises as assumption regarding what portion of the total costs expected. Further research is needed to isolate the represents contingent liabilities, given the relatively public sector portion of these costs and focus on the small funding provided compared to the magnitude 63 There have been two successful applications for droughts (2012 in Romania and 2016 in Cyprus) however these have been classified as droughts and fires. 64 Link. 65 Link. 51 Funding Gap Analysis of the estimated losses. The scenarios are as follows: Figure 18 illustrates the potential funding gap using the midpoint scenario for current annual drought • The low-range scenario assumes 2 percent of total losses and potential losses under two future scenarios drought costs (equivalent to €180 million per year (1.5°C warmer and 3.0°C warmer). Put simply, if the on average based on current loss estimates). UCPM and EUSF were used to cover all public and private sector costs associated with drought hazards • The midpoint scenario assumes 2.3 percent of for EU MS, the remaining €13 million would need to total drought costs (equivalent to €209 million per be covered by the EU MS or the private sectors within year on average based on current loss estimates). each country. It is important to note that the loss estimate includes some non-EU MS, such as the • The high-range scenario uses a value of 3 percent United Kingdom, for which there are not enough of total drought costs (equivalent to €270 million detailed data to quantify their portion of these losses. per year on average based on current loss However, it is plausible to suggest that their inclusion estimates). means the funding gap is lower in reality. When comparing these values to the alternate scenarios, the low-range estimate indicates no funding gap (€16 Estimated funding gap million of excess funding available on average each year), whereas the high-range estimate suggests that Based on current annual drought losses, the the gap could reach €74 million annually. estimated funding gap is €13 million per year. Figure 18: Drought funding gap (midpoint scenario) 500 EUSF Annual Allocation for 450 drought events 400 Total UCPM budget (€ m) 350 (Response - 3-year Average) 300 Reserves at EU level €, millions 250 200 150 100 50 0 Current Annual Loss Future scenario Future scenario (1.5° C warmer) (3° C warmer) Source: World Bank. The projected growth of the funding gap over time, Under the 3.0°C warmer scenario, the midpoint aligning with anticipated temperature increases in annual funding gap estimate increases 15-fold the region, reveals potential funding gap between compared to the current annual loss, reaching €206 €29 million and €323 million annually depending million annually (low-range estimate: €150 million; upon the climate warming scenario used. Using the high-range estimate: €323 million). This notable estimates derived for the 1.5°C warmer scenario, the increase in the projected funding gap over time midpoint annual funding gap estimate increases by suggests that the financial demands for addressing 125 percent to €29 million annually (low-range drought-related costs are likely to escalate estimate: no gap; high-range estimate: €29 million). significantly in the future. 52 Funding Gap Analysis An additional conclusion drawn from the analysis is The magnitude of losses from wildfire and drought, the limitation of relying solely on averages, which while marginal in comparison to earthquake and may mask variability in annual outcomes and fail to flood, creates additional pressure on already capture the potential severity of extreme drought constrained response and recovery budgets. The events. Without accounting for this variability, the magnitude of losses from earthquake and floods analysis may underestimate the true extent of the aggregated varies between €13 billion and over €50 funding gap, particularly in years characterized by billion for the low liability scenario, where the EU more extreme climate events or natural fluctuations. assumes a smaller proportion of the cost. In Therefore, there is a need for enhanced methodologies comparison, losses from wildfire range between €16 that account for variability and extreme scenarios to million and €717million, depending on the scenario provide a more comprehensive understanding of the and magnitude of the event, while drought saw a potential financial challenges associated with consistent funding gap between €29 million and droughts. This highlights the importance of adopting €323 million (see Table 4). adaptive and flexible funding strategies capable of responding to the unpredictability and variability inherent in climatic conditions. Table 4: Range of regional-level funding gap from Phase 1 and Phase 2 analysis in millions (€) RANGE OF ESTIMATED LOSS EUR € MILLION TYPE OF MODEL Phase 2 – Drought 29–323 10-year historical measures Phase 1 - Earthquake and flood 13,125–49,643 Probabilistic measures In a year where a major earthquake and flood has there is scope for additional financial instruments at already occurred, there would be no funding available the EU level and/or there is a need to incentivize at the EU level to respond to a wildfire or drought national governments to invest in DRF. event. This reinforces the finding from Phase 1 that 5.4. Wildfire Results: Country Case Studies M E T H O D O LO G Y total cost rather than isolating emergency operation costs, as this represents the overall financial burden The methodology applied to calculate potential on each country from wildfire events. Assumptions wildfire losses for each country case study is regarding the proportion of public asset losses to the consistent with the EU-level analysis. This involves total loss are retained to estimate contingent liabilities calculating the average cost per hectare burned and for each country, though this may fluctuate among applying it to the total number of hectares documented countries and wildfire events in any given year. To for each country based on the EFFIS database.66 address this uncertainty, low-range and high-range Details of this overall methodology are included in scenarios are incorporated to provide insight into the Annex 5. The case studies concentrate solely on the sensitivity of the analysis results based on this 66 JRC EFFIS. Link. 53 Funding Gap Analysis assumption. The low scenario assumes public sector Bulgaria: The funding gap analysis for Bulgaria damages to be 30 percent of total damages, while the indicates no immediate funding gap under the high scenario increases this estimate to 80 percent. midpoint scenario, although potential gaps may arise in extreme years, particularly under the high-range Regarding financing, the case studies rely on a scenario. In 2023, estimated reserves and contin­ unified estimate incorporating various available gency funds sufficiently covered the estimated costs, financing instruments for each country. This but a shortfall of €22 million would occur if recent encompasses country-specific estimations of reserve three-year average losses from 2021–2023 were and contingency funds, as detailed in Annex 8. These considered but could be covered by the inclusion of estimations align with those utilized in EDPP Phase 1, UCPM and EUSF funds. Notably, if high-range which examined the fiscal and economic impacts of assumptions were applied to historical cost estimates, earthquakes and floods. These amounts are a funding gap of €130 million would emerge. combined with the UCPM budget associated with emergency operations costs for wildfire events. Croatia. Similar to Bulgaria, Croatia faces no funding gap under the midpoint scenario, with reserves and A historical funding gap analysis similar to the EU- contingency funds covering estimated costs in 2023. level analysis is unfeasible due to limited historical However, if high-range assumptions were applied to data on reserves and contingency funds across each historical cost estimates, a funding gap of €220 country. Conducting this type of analysis in the million would arise. absence of such data could lead to an inaccurate portrayal of the financial capacity to handle losses at Greece. Greece faces a funding gap across all the time. In place of this, each case study evaluates scenarios due to inadequate reserves and potential financial impacts using 2023 total cost contingency funds relative to estimated costs. estimates and a three-year average (2021–2023) Despite additional financing options from the UCPM based on historical loss data. Additionally, a scenario and EUSF, the shortfall remains high, ranging from is included based on the year where the highest €259 million under the three-year average between wildfire losses were recorded in each country, as 2021 and 2023 to €1.4 billion under the high-range identified in the EFFIS database, which serves as a assumptions. proxy for gauging potential funding gaps during extreme years. Italy. Italy shows no funding gap in any scenario due to ample reserves and contingency funds compared to estimated costs. Even in scenarios that consider S U M M A RY O F CA S E S T U DY F I N D I N G S the year with the highest historical cost estimates, reserves and contingency funds remain sufficient. Several high-level differences were observed across each of the scenarios that provide some insight into Romania. Romania, like Italy, demonstrates no each country’s financial preparedness for wildfire funding gap across scenarios due to sufficient events, highlighting areas where additional measures reserves and contingency funds relative to estimated may be necessary to address potential funding costs. In 2023, these funds covered estimated costs, shortfalls in extreme circumstances. with similar results under recent three-year average losses and high-range assumptions. 54 Funding Gap Analysis 5.5. Bulgaria The funding gap analysis conducted for Bulgaria costs. According to EFFIS data, the most severe year indicates no funding gap based on the midpoint was 2000, with 57,406 ha burned, equating to a cost scenario, but a potential gap may emerge in an estimate of €340 million. If a similar year were to extreme year under the high-range scenario. The occur, the combination of financing instruments results of the midpoint estimate analysis are shown in would suffice to cover these costs based on the Figure 19. In 2023, the estimated reserves and midpoint scenario. However, if the high-range contingency funds (€116 million) adequately cover assumptions are applied to the cost estimates from the €55 million estimated costs. When considering 2000, the estimated cost is €546 million and the the recent three-year average (2021–2023) loss combined financial capacity across all available (€138 million), these funds would fall short by €22 instruments would be insufficient to cover the total million, though the additional UCPM and EUSF costs associated with such an event and result in an funding would be sufficient to cover the residual estimated funding gap of €130 million. Figure 19: Total Cost - Funding Gap Midpoint Estimate 800 EUSF Annual Allocation for Wildfire events 600 Total UCPM budget (€ m) 400 (Response + Firefighting) €, millions 200 Reserves at EU level 0 Estimated Excess Funding Historical Max. 2023* 3-yr Average -200 - 400 -600 No funding gap Source: World Bank. 5.6.Croatia The funding gap analysis conducted for Croatia equating to a cost estimate of €403 million. If a similar indicates no funding gap based on the midpoint year were to occur, the reserves and contingent funds scenario, but a potential gap may emerge in an would be insufficient to cover the entirety of these extreme year under the high-range scenario. The costs based on the midpoint scenario, but the results of the midpoint estimate analysis are shown in additional financing available from the UCPM and Figure 20. In 2023, the estimated reserves and EUSF would cover the shortfall. However, if the high- contingency funds (€128 million) adequately cover range assumptions are applied to the cost estimates the €15 million estimated costs. When considering from 2000, the estimated cost is €649 million and the the recent three-year average loss (€107 million), combined financial capacity across all available these funds are still sufficient to cover the total costs instruments would be insufficient to cover the total with no requirement for additional financing from the costs associated with such an event and result in an UCPM and EUSF. According to EFFIS data, the most estimated funding gap of €220 million. severe year was 2000, with 68,171 ha burned, 55 Funding Gap Analysis Figure 20: Total Cost - Funding Gap Midpoint Estimate 800 600 EUSF Annual Allocation for Wildfire events 400 Total UCPM budget (€ m) (Response + Firefighting) €, millions 200 National Reserves 0 Historical Max. Estimated Excess Funding 2023* 3-yr Average -200 - 400 No funding gap -600 Source: World Bank. To finance expenditure after a disaster, Croatia year. Dedicated budget lines that may also be used utilizes funds from the budget reserve and dedicated include item A539025 “compensation for damages budget lines.67 A budget reserve is set aside each caused by natural disasters” and A539020 year for unforeseen purposes; it is not earmarked for “assessment of damages from natural disasters.” disasters and may be used for a range of different Table 5 indicates the amount allocated to item purposes. The budget reserve may amount to A539025 over 2013–2018. 0.5 percent of planned budget revenue in any given Table 5: Amount allocated for item A539025, 2013-2018 (€, millions) CONFIRMED THE AMOUNT OF ASSISTANCE ALLOCATED FOR THE REPAIR OF DAMAGE (€, MILLIONS) YEAR DAMAGE (€, MILLIONS) 2013 2014 2015 2016 2017 2018 Total 2013 27.14 2.63 0.05 — — — — 2.68 2014 236.99 — 4.67 2.57 — — — 7.25 2015 246.5 — — — — — — 0 2016 163.51 — — — 2.65 — — 2.65 2017 330.9 — — — — 13.21 — 13.21 2018 24.02 — — — — — 2.65 2.65 Total   2.63 4.72 2.57 2.65 13.21 2.65   Source: Primorac, 2019 with figures converted from kuna to euro. Table 6 presents analysis of the amount spent from necessary equipment damaged during the both budget lines over 2013–2018. 1.35 million was rehabilitation of damage from floods and winter paid from the budget reserve in 2015 to the State storms in 2014. In 2017, €1.01 million was paid from Office for Reconstruction and Housing Care for the the budget stock to the Croatian Fire-Fighting rehabilitation of flood damage and to the Croatian Community for work carried out to address wildfires. Mountain Rescue Service for the restoration of 67 Assistance after a disaster and related risks are regulated by the Act on Mitigation and Elimination of Consequences of Natural Disasters (OG 16/19) and the Ordinance on the Registrar of Damages of Natural Disasters (OG65/19). 56 Funding Gap Analysis Table 6: State expenditure, 2013–2018 (€, millions) DAMAGE ELIMINATION OF TOTAL COMPENSATION EXPENDITURE ASSESSMENT FROM CONSEQUENCES GOVERNMENT FROM NATURAL FROM THE YEAR NATURAL NATURAL EXPENDITURE DISASTERS (1) BUDGETARY STOCK DISASTERS (2) DISASTERS (3) = (1) (3) + (4) (€, MILLIONS) (4) (€, MILLIONS) (€, THOUSANDS) + (2) (€, MILLIONS) (€, MILLIONS) 2013 2.56 2.52 2.57 — 5.20 2014 5.19 1.99 5.20 — 5.20 2015 2.57 1.99 2.57 1.35 3.92 2016 2.65 0.80 2.66 — 2.66 2017 13.21 3.19 13.21 1.01 14.22 2018 2.65 9.56 2.66 — 2.66 Source: Primorac, 2019 with figures converted from kuna to euro. The Ministry of Finance also uses budget National Assessment identifies 28 risks spread across reallocations to finance unplanned needs. 11 perils, with scenarios drawn up for each of the Amendment of the PFM Act, Articles 58 and 59, risks, and the consequences of the event evaluated removes restrictions on the use of budget reallocations by impacts on people’s lives and health, the economy, after a disaster. The use of budget reallocations— social stability, and politics. The listed 11 selected from which budget user to which budget user—is risks are plant diseases, animal diseases, extreme reported twice a year. temperatures, epidemics and pandemics, industrial accidents, floods caused by spills of terrestrial water At the subnational level, funding for disasters comes bodies, earthquake, open-type fires, snow and ice, from existing budget lines and/or budget drought, and water salinization. The Ministry of reallocations. Local and regional governments can Finance holds a Damage Register which includes also ask for additional assistance from the state self-reported data from budget users on the type of budget if the “value of total direct damage is at least damage, the time of the disaster, and the area 20 percent of the value of the original income of the affected by the disaster. Data from the Damage local self-government unit for the previous year or if Register are published on the Ministry of Finance’s the disaster has reduced the value of property in the website. There is not, however, at present, a central area of local self-government units by at least coordinating body for all data. Moreover, weaknesses 30 percent.”68 in the data that are being collected and analyzed have also been noted. For instance, it was noted by Data on disaster risks are currently being collected the Ministry of Interior that more data are needed on by different entities, but the quality of data differs expected damage and losses, vulnerabilities, and greatly. For instance, the Ministry of Agriculture is exposure. currently collecting data on the potential impact of disasters on agricuIture. The city of Zagreb is Disaster risks are not currently included in the fiscal calculating the risk of earthquakes. The Ministry of strategy, medium-term forecasts, or annual budgets. Interior carries out risk assessments for different Moreover, line ministries and subnational perils on a regular basis as well as in charge of governments are not currently being asked to provide coordinating national risk assessments (2015, 2019, this information in their plans and budget submissions. and an update ongoing). The Ministry of Interior’s They do, however, provide information on damages/ 68 Law on Mitigation and Elimination of Consequences of Natural Disasters, OG 16/19. 57 Funding Gap Analysis losses as needed through the registrar of damages sector. For Croatia the purchase of insurance for held in the Ministry of Finance and are required to public assets will be at the behest of local governments conduct a regular risk assessment, which includes and thus the amount of coverage may differ. However, the consideration of disaster risks. it was not believed that many local governments would not have insurance for their assets. The Ministry of Finance is not tracking and reporting on disaster expenditure or the impact of disaster Wildfire risk is included in household fire policies, expenditure. The Auditor General has however as there is no specific exclusion for wildfires. This carried out financial, compliance, and performance has been confirmed by speaking to private insurance audits that relate to disaster risks. For instance, the companies in Croatia. It is therefore expected that budget reserve is audited, reports are produced twice household fire policies would include damage from a year on the use of in-year budget reallocations, and wildfires in most EU MS. Approximately 25 percent of special reviews have taken place, for example, on the homeowners have household insurance that includes use of EUSF funds and the state inventory. Moreover, wildfire with 16 percent of this proportion with cover the Auditor General has given recommendations on for earthquakes. The consensus among insurers on how to improve the performance of certain financing the low penetration is that the government will instruments, for example, to improve clarity on the support citizens when there is a large disaster and purpose and guidelines in the use of the budget therefore there is a lack of will to purchase insurance. reserve. The ad hoc payouts from the government will also affect take-up in other sectors of insurance such as in agriculture insurance which often provides cover for Insurance drought. A new law is being introduced in Croatia which The motor third party liability (MTPL) market in mandates landlords to have property catastrophe Croatia is the largest non-life insurance class of insurance for the buildings. This is a good step to business. Croatia has compulsory MTPL schemes, ensure protection against disasters, but it is not clear and the non-life insurance industry would like to build how payouts would be split between landlords and off this coverage to increase awareness of the value tenants, if at all. Currently for accessing home loans, of insurance in the population, especially for banks will recommend a purchase of a household catastrophes, particularly earthquakes in Croatia. insurance policy however the banks do not require catastrophe insurance to be purchased, only fire Croatia has a government-subsidized agriculture risks. insurance scheme which does not have high levels of penetration. The low levels of penetration could be State-owned enterprises, especially larger com­ due to ad hoc payments made to farmers by the local panies that own roads, ports, and airports, will governments, after an event. The publicly available purchase insurance for these assets which will statistics from insurance regulators in Croatia do not include catastrophe cover. In Croatia, the airports provide the breakdown of this information, but it is are insured with Croatia Insurance. Buying insurance understood that the business is relatively unprofitable is up to the commercial entity because of which the and written only by a handful of insurers. level of protection can vary by type of asset and 58 Funding Gap Analysis 5.7. Greece The funding gap analysis conducted for Greece total costs and result in a shortfall of €259 million. indicates a funding gap based on the midpoint According to EFFIS data, the most severe year was estimate across each of the scenarios. The results of 2007, with 225,734 ha burned, equating to a cost the midpoint estimate analysis are shown in estimate of €1.335 billion. If a similar year were to Figure 21. In 2023, the estimated reserves and occur, an estimated funding gap of €587 million contingency funds (€447 million) are inadequate to would emerge despite the availability of additional cover the €952 million of estimated costs. When financing from the UCPM and EUSF. Based on the considering the recent three-year average loss (€976 high-range assumptions, the cost estimates from million), the combined financial resources available 2000 equate to €2.148 billion, resulting in an across reserves, contingency funds, and the UCPM, estimated funding gap of €1.4 billion. and EUSF funding would be insufficient to cover the Figure 21: Total Cost - Funding Gap Midpoint Estimate 2.500 EUSF Annual Allocation for 2.000 Wildfire events €, millions 1.500 Total UCPM budget (€ m) (Response + Firefighting) 1.000 National Reserves 500 0 Historical Max. 2023* 3-yr Average Source: World Bank. 5.8. Italy The funding gap analysis conducted for Italy average loss (€883 million). According to EFFIS data, indicates no funding gap based any of the scenarios the most severe year was 1981, with 229,850 ha due to the high level of reserves and contingency burned, equating to a cost estimate of €1.359 billion. funds relative to the total estimated costs from If a similar year were to occur, the reserves and wildfires. The results of the midpoint estimate contingency funds would suffice to cover these costs analysis are shown in Figure 22. In 2023, the based on the midpoint scenario. If the high-range estimated reserves and contingency funds (€2.791 assumptions are applied to the cost estimates from billion) adequately cover the €391 million estimated 2000, the estimated cost is €2.187 billion, and the costs. These funds would also be sufficient to cover estimated reserves and contingency funds would the estimated costs based on the recent three-year suffice to cover these costs. 59 Funding Gap Analysis Figure 22: Total Cost - Funding Gap Midpoint Estimate 7.000 EUSF Annual Allocation for 6.000 Wildfire events 5.000 Total UCPM budget (€ m) (Response + Firefighting) €, millions 4.000 National Reserves 3.000 2.000 Estimated Excess Funding 1.000 0 Historical Max. 2023* 3-yr Average -1.000 -2.000 -3.000 -4.000 No funding gap Source: World Bank. Analysis of expenditure by the National Fire Brigade in 2020 to €104 million in 2023), in general budget over 2020–2023 indicates that the government has lines for strengthening forest firefighting (€7 million increased expenditure on forest firefighting. In in 2020 to €19 million in 2023), and in purchasing 2020, it spent an estimated €151 million, which rose firefighting vehicles (€2 million in 2020 to €28 million to €178 million in 2023.69 Over this period, the in 2023).70 government invested more in its air fleet (€86 million 5.9. Romania The funding gap analysis conducted for Romania year for wildfires according to EFFIS data, with 15,308 indicates no funding gap based on any of the ha burned. These reserves and contingency funds scenarios due to the high level of reserves and would also be sufficient to cover the estimated costs contingency funds relative to the total estimated based on the recent three-year average loss (€64 costs from wildfires. The results of the midpoint million). If the high-range assumptions are applied to estimate analysis are shown in Figure 23. In 2023, the cost estimates from 2023, the estimated cost is the estimated reserves and contingency funds (€354 €146 million, and the estimated reserves and million) adequately cover the €91 million estimated contingency funds would suffice to cover these costs. costs. In Romania, 2023 was also the most severe 69 The figure is an estimate as it does not include related expenditure which may be captured under additional budget lines. 70 Government of Italy. 2024. 60 Funding Gap Analysis Figure 23: Total Cost - Funding Gap Midpoint Estimate 1.400 EUSF Annual Allocation for 1.200 Wildfire events 1.000 Total UCPM budget (€ m) (Response + Firefighting) €, millions 800 National Reserves 600 Estimated Excess Funding 400 200 - Historical Max. 3-yr Average 2023* -200 -400 -600 No funding gap -800 Source: World Bank. To finance expenditure after a disaster Romania fluctuated over recent years. The Intervention Fund is utilizes funds the from the Government Reserve earmarked for post-disaster expenditure and can be Fund and the Government Intervention Fund. These topped up from the Reserve Fund (which is not funds are specified in Article 30 of the 2002 Public earmarked). The amount allocated to the Intervention Finances Act (500/2002); the size of the Intervention Fund is determined on analysis of historical Fund is not set out in legal provisions and has expenditure. Table 7: Romania reserve funds DIMENSION GOVERNMENT RESERVE FUND GOVERNMENT INTERVENTION FUND Scope Urgent or unexpected spending Urgent action to mitigate the effects occurring during the budget cycle of natural disasters and supporting affected persons 2019 initial allocation (RON, 99.0 1.0 millions) 2021 initial allocation (RON, 597.7 1.0 millions) Source: UNICEF. 2022. Policy Brief on Financing Health Emergencies in Romania: The Response to COVID-19. Different line ministries are responsible for different and Forests is responsible for floods. Should additional disasters and will utilize their own budget lines as funding be required after a disaster, the Ministry of needed. For example, the Ministry for Development, Finance uses budget reallocations, as needed, to Public Works, and Administration is responsible for meet unplanned needs. Moreover, legal provisions the aftermath of an earthquake, the Ministry of stipulate that 10 percent of all public institutions’ Agriculture and Rural Development is responsible for yearly budget is withheld and only released in the droughts, and the Ministry of Environment, Water, second semester, subject to the government’s review 61 Funding Gap Analysis of budget execution.71 This cautious spending The information on disaster risks, contained in the enables the government to relocate funds, as needed, Fiscal and Budgeting Strategy, is not currently and/or withhold funds for unplanned events. updated to inform annual appropriations of budget users. Moreover, as noted in a recent review,74 there Local governments largely depend on is a need to strengthen Romania’s Medium-Term intergovernmental transfers for funding their Expenditure Framework to deliver credible capital expenditures.72 Approximately 70 percent of local expenditure ceilings and incorporate disaster risks. government revenue is made up of allocations from Weaknesses in Romania’s public investment the state budget; the rest consists of own revenue.73 management (PIM) system have also been noted, Total revenue is equally divided between non- notably in ensuring access to budget funding for new earmarked revenue, which local governments are projects and securing adequate funding for existing free to allocate according to local priorities, and projects. Line ministries and subnational governments earmarked revenue which must be spent on the are not requested to consider and incorporate services and expenditure items specified in the disaster risks into their plans and budget submissions. underlying conditionalities and regulations. In the In making improvements to budgeting practices over context of disasters, local governments will use their the medium term, particularly for annual and capital existing resources and budget lines, make budget expenditure, more could be done to explicitly consider reallocations where needed, and request additional disaster related risks. funding from the Reserve Fund should the need arise. The Ministry of Finance is implementing In addition to budgetary instruments, the World performance-based budgeting in select line Bank provided Romania with CAT DDO in 2018 as a ministries with the objective of enhancing the contingency credit, which served as an additional efficiency of public spending.75 Three ministries funding source alongside the national disaster should execute their budgets based on budget reserves and EUSF mechanism. However, the programs in 2024. This process will improve COVID-19 pandemic led to the complete exhaustion coordination of public policies and spending and lay of the credit line. A new CAT DDO is under development the ground for a systemic approach to priority setting, but is not expected to be available until July 2024. planning, and budgeting. This again, as part of reforms to strengthen current planning and budgeting Disaster risks are currently included in Romania’s practices could support the greater integration of Fiscal and Budgeting Strategy. The strategy disasters into the budget cycle—shifting the focus estimates the budget impact of four different disaster from the management of inputs to a process based scenarios. The estimates were calculated through on results and achievement of policy objectives. utilization of fiscal risk model, based on disaster scenarios and utilizing data from the RO-RISK At present, the Ministry of Finance is not currently platform. Romania’s RO-RISK project facilitates the tracking disaster-related expenditure or the impact exchange of data and information on risk exposure, of disasters on government revenue. There is no vulnerabilities, and existing risks for authorities and systematic reporting of disaster risks apart from what the population. The fiscal and budgeting strategy it included in the fiscal and budgeting strategy. notes that the most important risks Romania may Moreover, data on disasters and climate change are face are earthquakes and floods. currently held within different institutions and 71 The Reserve Fund and Intervention Fund are exempt from this rule. 72 The share of local government expenditure in general government expenditure was 22 percent in 2021. 73 World Bank. 2022b. Technical Support to the Ministry of Romania. Modernizing the Intergovernmental Transfers Systems - Assessment of the Current Situation. 74 Nadoll, J. 2017. Unblocking Public Investment Bottlenecks in Romania: A Report on Systemic Causes of Delays and Inefficiencies in the Preparation and Implementation of World Bank Financing Investment Projects. 75 World Bank. 2022a. Advisory Services Agreement on Strengthening Planning and Budgeting Capacity II: Output 5. Final Report Presenting the Recommendations of the RAS. 62 Funding Gap Analysis hampered by not having a standard policy framework Romania has private insurers who write agriculture on digitalization, data storage and protection.76 insurance schemes. The publicly available statistics from insurance regulators in Romania do not provide The Court of Accounts, Romania’s Auditor General, the breakdown this information, but it is understood is primarily responsible for financial and compliance that the business is relatively unprofitable and written audits. To strengthen the application of risk-based only by a handful of insurers. budgeting, the Court of Account’s work focus would need to shift toward performance audits and checking In Romania, there have been four insurer collapses whether internal control frameworks are functioning in eight years, which has decreased public trust in properly. Twinning arrangements with other Supreme the insurance industry, which presents a case for Audit Institutions could be considered to changes public intervention. In countries where having current practices. insurance is not part of the culture, having instances where there are constant insurer collapses will regress any positive awareness that has been built by Insurance other private insurers and the insurance associations. While MTPL schemes account for 73 percent of the Household property policies in Romania generally non-life insurance market,77 nonmandatory schemes cover ‘fire’ risks which would include cover for do not have high levels of penetration. A lack of wildfire losses. There is no explicit exclusion for confidence in the insurance industry due to insurance wildfires and thus, it is expected that these policies collapses could be one of the reasons for this. will cover damage from wildfires under the same terms and limits as other fires losses. As a result of the reinsurance market hardening with large rate increases on the same terms, insurers In Romania, catastrophe insurance for flood and have struggled to maintain the same level of earthquake for households is mandatory. It is profitability when allowing for this additional cost. implemented by private insurers who formed an PAID in Romania have mentioned that their association, the Natural Disaster Insurance Pool reinsurance rates have increased dramatically and (PAID). This is expected to be a first loss catastrophe as they want to maintain their strong level of solvency cover. The insurance is offered with two fixed sets of their combined ratios have deteriorated. This has premium rates/coverage that differ for urban and resulted in them submitting to the regulators an rural areas. Beyond this policy, households can increase in their premium rates. The reinsurance purchase multi-peril top-up property insurance cover market has had rate hardenings due to large from private insurers with market-based premiums. catastrophe losses and utilizing reinsurers in the However, despite the mandatory nature of the policy, region such as Europa Re could provide cheaper penetration is about 20 percent due to lack of reinsurance to MS. Seeking alternative ways to perceived insurance value, limited enforcement of reinsure their risk considering the current state of the the ‘mandatory’ aspect (which requires municipalities market could be a good option. to fine those not in compliance), and lack of public awareness. This insurance is mandatory for accessing home loans, but many people in Romania own their Summary of Case Study Findings houses and the number of loans to buy houses is limited. PAID is looking to increase penetration using Noting the differences in the quality of inputs and as awareness programs with a target of 40 percent mentioned earlier in the discussion, combining the penetration over the next five years. analysis to develop a multi-hazard view of the funding gap is not possible. However, to help 76 World Bank. 2023. Romania Country Climate and Development Report. Link. 77 AXCO 2022. 63 Funding Gap Analysis understand the differences in the order of magnitude occurs at the same time as a wildfire or flood, there created by the hazards, Table 8 provides a summary are insufficient resources to respond. However, if only which reinforces the key findings of the regional a single wildfire or drought occurs, sufficient finance analysis by indicating that if a flood or earthquake should be available in most years. Table 8: Range of funding gap from Phase 1 and 2 analysis (€, millions) BULGARIA CROATIA GREECE ITALY ROMANIA METHOD Phase 2 0–130 0–220 259–1,400 0 0 10-year historical measures Phase 1 140–1,478 9–815 42–773 741–3,933 0–4,160 Probabilistic measures 64 Funding Gap Analysis 6. Key Findings This report reinforces the findings from Phase 1 for support costs associated with wildfires, which confirming that financial instruments to manage have received €207.1 million in financing. disaster risk are limited to risk retention and more should be done to incentivize risk transfer at both Droughts have had large impacts on economies in the EU and the EU MS levels. At present there are no MS; since the EUSF was created in 2002 there have risk transfer products at the EU level or in the two been four applications submitted and accepted for case study countries. There is a trade-off with risk droughts. Funding was provided to Cyprus in 2008 transfer products as the initial premium is to be paid and 2016 and to Romania in 2012 and 2022. In 2022, upfront regardless of payouts. However, in extreme Romania which suffered a loss of over €1 billion in the loss years, the payouts from a risk transfer product agricultural sector due to droughts and wildfires in can be many times the premium. the south-eastern region received almost €34 million from the EUSF to cover some of the losses from the Direct damage and loss associated with wildfire and drought and associated wildfires. Currently, MS are extreme heat are significantly lower than for other not easily able to access EUSF funds to support hazards, for example, earthquake and flood. This is drought losses, due in part to the challenges in due in part to the assets exposed to the hazards that defining the exact start of a drought and the fact that consist of natural assets with some limited droughts are often connected to extreme heat and infrastructure for wildfire and the fact that extreme wildfires, compounding overall impacts. It is the heat by itself has little physical impact on assets. damage from this combined risk that leads to the application to the EUSF. Indirect losses from wildfire and drought (including extreme heat as a sub-hazard of these events) are Wildfire accounts for over one-third of the UCPM expected to be far higher and pose impacts for the response costs, while no expenditures were found long-term health of society and ultimately for extreme heat. The response budget of the UCPM businesses as the potential to reduce the number of increased from €13 million in 2014 to €150 million in working days increases. However, the data are not 2022. Other notable increases over 2014–2022 available to substantiate this and there is a need to include a rise in funding for prevention and start collecting data on the number of hospital preparedness activities (from €9million in 2014 to admissions due to extreme heat and wildfire events. €19 million in 2022) and for firefighting (from €1 million in 2014 to €1.67 million in 2022). The EUSF has been triggered eight times since 2002 for wildfires, in response to damages of €6.1 billion The budgets of the UCPM and the EUSF have both for these eight events. Since its establishment, there been exhausted in recent years, and there is a need have been eight successful applications to the EUSF to revisit the budget of the UCPM to strengthen 65 Key Findings response capacity and complement preparedness. It where possible the team has tried to draw parallels. is recommended that these are complemented by the introduction of risk transfer instruments as a cost- The magnitude of losses from wildfire and drought, effective tool that enables prearranged finance to be while marginal in comparison to earthquake and released when it is needed most. A phased approach flood, creates additional pressure on already to the introduction of risk transfer instruments is constrained response and recovery budgets. The recommended, noting that the funding gap from size of the funding from Phase 1 (earthquake and earthquake and flood risk is far higher than that for flood) varies between €13 billion and over €50 billion wildfire and drought. This will also provide time for for the low liability scenario, where the EU assumes a the wildfire models under development to be released smaller proportion of the cost. In comparison, losses and further analysis to ascertain whether risk transfer from wildfire range between €16 million and €717 for this risk is cost-effective. Such instruments could million, depending on the scenario and magnitude of be linked to the UCPM and/or EUSF.  event, while drought saw a consistent funding gap between €29 million and €323 million. The analysis in this report was hindered because, at present, no probabilistic models for wildfire in the Should a drought or a wildfire happen in a year EU exist. However, models are under development, where a major earthquake and flood has already but for the time being their absence limits our occurred, there would be no funding available at the understanding on the future risk that wildfire events EU level to respond to a wildfire or drought event. may pose. As noted throughout, it was not possible to This reinforces the finding from Phase 1 that there is combine the results with Phase I due to (i) the scope for additional financial instruments at the EU differences in the number of observed events for level and/or there is a need to incentivize national wildfire and drought, (ii) the lack of probabilistic governments to invest in DRF. models, and (iii) the differences in exposure. However, 6.1. Options for Consideration The above findings suggest that more can be done at to reinforce the application of the 2021 EU Climate the national and EU levels to promote DRF solutions Adaptation Strategy. and close funding gaps. Below are some options for consideration; not all need to be pursued or 2. Increase the allocation for the UCPM and EUSF. implemented at the same time. Some of these options Both instruments have proven an important source were presented in Phase 1 but remain relevant today of finance for disaster damages in EU MS and based on the analysis and findings of Phase 2. accession countries. Though the funding they provide covers only a small fraction of needs, 1. Develop an EU level overarching DRF strategy. designed to complement national DRF instruments, This was recommended in Phase 1 and as noted the budget allocation to both remains insufficient the introduction of a coherent and comprehensive as shown in the analysis. EU-wide policy on DRF would benefit the region by defining common priorities and practices and 3. Introduce risk transfer instruments. These would identifying the level of loss the EU can and is willing complement the limited funding from the EUSF to cover via its EU-level instruments. Having simple and the UCPM and could be structured to bring in and clear messaging at the EU level could additional finance when needed most such as incentivize investments in DRM, including DRF, at cross-border events where multiple countries are the national level. A DRF strategy can be developed affected. 66 Key Findings 4. Improve data on DRF. To inform decision-making 7. Increase penetration of insurance. National on DRF at the EU level, it is important to have governments should consider options for reliable data and analytics. As noted in this report, increasing catastrophe household insurance and it is suspected that there are significant health public asset insurance. This could be done through costs associated with wildfire, extreme heat, and public-private partnerships. However, each drought. However, no data are collected on hospital country has its unique set of circumstances and admissions associated with these events which will face different complexities when introducing limited the analysis. Systematically collecting data household insurance (that is, making insurance on health impacts could help identify whether mandatory may not be feasible for all countries). At additional financial solutions such as health the same time, in some countries, a political insurance could help cover some of these costs. If decision may be made on providing different ways found effective, this could be subsidized at the of supporting households after disasters, such as national and/or regional level. through public compensation. Therefore, a decision on how to increase penetration of At the national level: household insurance will be context specific. 5. Develop national DRF strategies. To complement 8. Strengthen risk-based budgeting. It was also the EU-level DRF strategy, consider the recommended in Phase 1 and is further justified in introduction of comprehensive national DRF this report. The implementation of risk-based strategies to ensure financial preparedness to budgeting can strengthen financial resilience and disasters. The first step would include determining could be explored more fully across the EU, with national priorities in strengthening DRF (such as guidance and good practice shared across MS. focusing on households, the poorest members of While this report was unable to identify EU society, and government budget). Improving data contingent liabilities based on an all-hazards for DRF, including the collection of heat stress– approach, a funding gap was still found for wildfire related hospital admissions should also be and extreme heat. The identified funding gap for considered during development of DRF strategies. these hazards was significantly lower than for earthquake and flood, but, as mentioned earlier, 6. Consider the introduction of sovereign risk this was based on a small number of observed transfer instruments. As mentioned at the EU events and should be revisited once modelling level, while this is not a new recommendation, this capacity improves. This would help develop analysis has served to reinforce the need for quantified estimates of risk that could be applied finance when it is needed most and the introduction to a risk-based budgeting framework; this would in of risk transfer instruments at the sovereign level turn create incentives to invest in DRF by could present a cost-effective way to manage this supporting MS to know, understand, and plan for risk. While this may not be a viable option for the risks they face. wildfire or drought at this time, it is overdue for earthquake and flood risk. 67 Key Findings Annex 1. Current wildfire models COMPANY MODEL DESCRIPTION PROBABILISTIC Munich Re Higher resolution to better resolve wildland-urban interfaces (WUIs), key No zones for high-risk and high-value locations. The WUI is 1.5 miles (2.4 km) wide and divided into 5 detailed zones. Easy-to-visualize maps for risk acceptance rather than complex stochastic models. API available for flexible integration into your existing tools and processes. Underpinned by the claims analysis of one of the world’s leading reinsurers. Taking into account the latest knowledge on the impacts of climate change. RMS No model for Europe Somewhat Descartes Our parametric wildfire product is structured using a combination of Somewhat satellite imagery, long-term climate, and weather data. We develop an index calibrated to historical wildfire impact and the distribution of value across the property or areas insured. Verisk (AIR) No model for Europe No Celsius Pro Probabilistic model under development Swiss Re Apply WUI layer, similar to Munich Re No Guy Not available for Europe. No Carpenter/ The GC Wildfire Risk Score assesses risk on a site-by-site basis, capturing Marsh climate, meteorological and landscape data across multiple spatial scales summarized in a simple, single metric. Covering the lower 48 states at a 30-meter resolution. CORELogic Not available for Europe. They also just made a score. CoreLogic developed Somewhat highly granular 30 m resolution deterministic and probabilistic models to create a score that differentiates hazard within neighborhoods down to the parcel level. The new model uses artificial intelligence and machine learning to measure risk reduction from a dozen mitigation factors that influence structure vulnerability to wildfire. The advanced model offers more precise risk assessment, differentiation, and reduction to reflect the unique characteristics of every structure. AON Only for the United States Yes Considers expected fire propagation, various fire spread azimuths, and a wide range of fuel loads. Loss reporting allows a choice of views of multiple aggregated expected losses. Examines the effects of fuel and slope on fire behavior as well as the possibility of multiple large fires within an extensive fire outbreak during extreme fire weather conditions. AXA XL Appears to be under development. Some probabilistic components. Focus on Somewhat the United States 68 Annex 1. Current wildfire models COMPANY MODEL DESCRIPTION PROBABILISTIC Mitiga Mitiga, use a hybrid approach for wildfire forecasting. That includes using Somewhat physics-based models to determine how real-time factors like wind, topography, or vegetation may affect the behavior of wildfires. Liberty Trained on past burning experience No Mutual Annex 2. Examples of Disaster Reserve Funds and Contingencies Funds in the EU NAME PURPOSE COUNTRY Dedicated disaster funds National disaster Established with the Disaster Fund Act of 1996, the fund can cover Austria reserve fund both ex ante risk management and post-disaster needs. The fund (Katastrophenfonds) can be used to finance large-scale protection infrastructure (ex ante risk management) but also serves to compensate the affected population. The fund can cover damages from flood, avalanche, earthquake, landslide, hurricane, and hail.78 In addition to this fund, ministries and agencies in Austria seem to be able to use their reserves (savings or additional revenues) at their own discretion for different purposes, including post-disaster financing.79 National Recovery Managed by the Ministry of Finance, the fund can support Finland Fund municipalities if disaster damages to public infrastructure exceed their own budgets. However, the amount of funding available annually is challenging to estimate. 80 Fonds de prévention The fund can cover emergency housing or temporary rehousing France des risques naturels and relocation, prevention measures, information measures, and majeurs (‘Fonds research activities (the local governments are the beneficiaries). 81 Barnier’) 78 OECD 2015, 2016, and 2018. 79 Steger, Gerhard. 2010. “Austria’s Budget Reform: How to Create Consensus for a Decisive Change of Fiscal Rules.” OECD Journal on Budgeting 2010/1. 80 DRIVER Finland 2015. 81 French Ministry of Ecology Report 2019. Other special purpose reserve funds covering disaster relief expenditures are available in France but were not considered for the analysis: (i) the Fonds national de gestion des risques en agriculture covers agricultural producers for uninsurable crop lost due to natural hazards or disease outbreak and (ii) the Fonds de secours outre-mer covers the reconstruction of uninsured private assets, uninsurable subnational assets, and for immediate disaster relief in overseas territories (purchase of basic necessities). OECD. 2019. Fiscal Resilience to Natural Disasters. 69 Annex 2. Examples of Disaster Reserve Funds and Contingencies Funds in the EU NAME PURPOSE COUNTRY General contingency funds Interagency The commission is allocated with regular budget that is aimed at Bulgaria Commission for covering exceptional and unanticipated costs that may occur from Relief and Recovery disasters (natural or man-made) or other events such as the mass to the Council of migration of refugees.82 Ministers Budgetary reserve Article 56 of the Budget Act provides for the establishment of a Croatia budgetary reserve covering expenditures which emerged during the alleviation of the consequences of natural disasters, epidemics, environmental mishaps, or extraordinary events and other unforeseen purposes for which no funding has been secured in the budget, or for which it is ascertained during the course of the year that insufficient funds were established for them because they were impossible to foresee during budget planning. Although the government can use the reserve discretionally, it has to report use of the reserve to the Sabor (the Croatian Parliament). Its amount cannot exceed 0.5 percent of budget revenues (including taxes but excluding receipts such as user charges and fees).83 General contingency Two general contingency funds are available: (i) a budgetary Czech Republic funds allocation for emergency and immediate measures, based on Regulatory Acts No. 239/2000 Sb and No.240/2000 Sb, that can cover rescue and health protection of affected population and (ii) a budgetary allocation for property reconstruction and revitalization (State Aid for Territorial Restoration), based on Acts No. 12/200 Sb and No.186/2002 Sb, that can cover reconstruction of destroyed property in the form of interest-free loans to municipalities, firms, and households.84 Stabilization Reserve The fund can be used in case of “resolution or prevention of an Estonia Fund emergency situation, a state of emergency, a state of war or other extraordinary situation or a crisis with material effect.” Unexpected costs of small-scale emergencies are covered by the government liquidity reserve.85 Force Majeure Fund The fund can cover reconstruction of government owned assets Hungary destroyed by natural disasters.86 Rainy Day Fund The fund set up by National Surplus Bill 2018 and managed by the Ireland National Treasury Management Agency (NTMA) on behalf of the Ministry of Finance aims to mitigate severe economic shocks, in excess of the normal fluctuations of the economic cycle. Fondo per le The fund was set up by art. 5.5-quinquies Legge 225/1992 to cover Italy emergenze nazionali “situations of national emergencies.” The fund can, for instance, finance relief and assistance to the population, restoration of public services and network infrastructure, needs and damages assessments, fiscal exonerations for victims and temporary suspensions of repayments of real estate loans on destroyed or damaged buildings, and so on.87 82 Bulgaria Assessment of DRM sector 2018. 83 Budgeting in Croatia 2006. 84 World Bank. 2011. Financial and Fiscal Instruments for Catastrophe Risk Management Addressing Losses from Flood Hazards in Central Europe (Poland, Czech Republic, Hungary, and Slovakia). 85 Estonian Ministry of Finance website. 86 World Bank 2011. 87 PROMETEIA 2019. 70 Annex 2. Examples of Disaster Reserve Funds and Contingencies Funds in the EU NAME PURPOSE COUNTRY Fiscal Safety Reserve The fund was set up by the Civil Protection and Disaster Latvia Management Law and can cover fiscal risks with funding of maximum 0.1 percent of GDP.88 Stabilization Reserve The fund can cover state budget expenditures in case of “exceptional Lithuania Fund circumstances.”89 Annual reserve for It can cover disaster risk prevention activities such as flood Poland prevention and protection works, as well as liquidation of property damages recovery caused by natural disasters through financial assistance to local governments for housing or infrastructure reconstruction, and post- disaster assistance to individuals.90 Fundo de Apoio The Fundo de Apoio Municipal financial can cover recovery of Portugal Municipal and the the municipalities in financial state of imbalance through the Emergency Bank implementation of municipal adjustment programs. This includes Account loans to municipalities to finance reconstruction of infrastructure and equipment damaged as a result of disasters. The Emergency Bank Account can support individual citizens affected by certain disasters.91 Fondul de rezervă The fund was set up by Lege 500/2002 to cover disaster costs as well Romania budgetară as other contingencies.92 Special budget The reserve can cover general contingencies and disaster relief Slovenia reserve programs.93 Ad hoc reserve funds Funds established Following disasters, several ad hoc reserve funds were established, For example, in after disasters for instance in Germany, (i) Sonderfonds Aufbauhilfe, amount Germany, France €7.1 billion, was set up to cover damages from the 2002 flood and (ii) Aufbauhilfegesetz, amount €8 billion, was set up to cover damages from the June 2013 flood.94 In France, ad hoc emergency relief funds are usually set up by the Ministry of Interior to provide immediate disaster relief (purchase of basic necessities, for example, food, clothing, accommodation) to the affected individuals with compensations, capped at €300 per affected adults and €100 per affected child. These funds are excluded from the analysis. Subnational disaster reserve funds Funds to cover In some countries, like in Belgium, the national disaster reserve fund For example, in disasters costs at the was abolished and instead replaced with funds at the subnational Belgium, Germany, local level level. In Germany, disaster risk reduction and DRM are under the Slovakia responsibility of the Landers. In Slovakia, self-governing regions can also establish crisis funds to finance a potential damage, even if the obligation to do so does not exist.95 These funds are excluded from the analysis. 88 Civil Protection and Disaster Management Law. 89 OECD. 2012b. Budgeting in Lithuania. 90 Act on Crisis Management of April 26, 2007. 91 EC, EUROPA 2019 and Fundo de Apoio Municipal website, Link. 92 România Consiliul Fiscal 2019. 93 Final account of the budget of the Republic of Slovenia for 2019. 94 MERZ, ELMER, KUNZ et al. 2014. 95 DRIVER Slovakia 2015. 71 Annex 2. Examples of Disaster Reserve Funds and Contingencies Funds in the EU Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Table 9: Funding from the EUSF (2002–2022) versus total damages of the disasters accepted by the EU TOTAL DIRECT DAMAGE OF DISASTERS SUM OF EUSF AID PAID (2002–2022) COUNTRIES WHICH WAS ACCEPTED (2002–2022) (€, MILLIONS) (€, MILLIONS) Albania 1 Austria 214 4,872 Belgium 125 5,560 Bulgaria 42 1,184 Cyprus 15 515 Czech Republic  178 3,578 Germany 1,645 53,351 Estonia 5 48 Spain 154 7,167 France 403 13,289 Greece 160 5,026 Croatia 1,033 17,882 Hungary 77 1,441 Ireland 57 521 Italy 3,081 69,762 Lithuania 20 423 Luxembourg 5 189 Latvia 28 573 Montenegro 0 North Macedonia - Malta 1 30 Netherlands 5 500 Poland 125 3,773 Portugal 198 4,277 Romania 141 4,389 Serbia 72 1,106 Sweden 82 2,297 72 Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Slovenia 48 1,285 Slovakia 26 785 United Kingdom 223 7,724 Grand Total 8,164 211,547 Figure 24: Number and direct damages (€, millions) of natural disasters as reported to the EU for the EUSF 18 1.400 16 1.200 14 Direct damages (EUR m) 1.000 Number of Natural Disasters 12 10 800 8 600 6 400 4 200 2 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Number of natural disasters Total direct damages, accepted by COM: EUR m Source: Authors based on EUSF data. Link. Note:* This includes applications which have been rejected by the EUSF. 73 Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Annex 4. Insurance penetration in the EU MS Figure 25: Insurance penetration in EU MS by peril Earthquake Flood* Windstorm Wildfire Coastal Flood Source: EIOPA dashboard. Note:* Flood excluding coastal flood. 74 Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Table 10: Proportion of households covered by insurance in the EU MS (%)   INSURANCE PENETRATION - EIOPA DASHBOARD INSURANCE PENETRATION - FROM NATIONAL DISCUSSIONS FLOOD EX COASTAL FLOOD EX COASTAL   WINDSTORM EARTHQUAKE WILDFIRE* WINDSTORM EARTHQUAKE WILDFIRE COASTAL FLOOD COASTAL FLOOD Austria 33 15 0 0 0 Belgium 57 20 0 0 0 Bulgaria 13 0 0 0 0 Croatia 2 6 0 0 0 25 25 25 4 25 Cyprus 19 0 0 0 0 Czech 15 38 0 0 0 Republic Denmark 62 72 75 0 0 Estonia 18 0 0 0 0 Finland 54 5 50 0 0 France 41 37 86 58 0 Germany 51 23 0 24 0 Greece 1 3 0 2 0 Hungary 34 1 0 0 0 Iceland 0 98 0 98 0 Ireland 24 28 0 0 0 Italy 4 3 0 3 0 75 Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Latvia 12 0 0 0 0 Liechtenstein 100 0 0 0 0 Lithuania 15 0 0 0 0 Luxembourg 64 49 0 0 0 Malta 13 0 0 0 0 Netherlands 55 2 0 10 0 Norway 98 59 100 100 0   INSURANCE PENETRATION - EIOPA DASHBOARD INSURANCE PENETRATION - FROM NATIONAL DISCUSSIONS FLOOD EX COASTAL FLOOD EX COASTAL   WINDSTORM EARTHQUAKE WILDFIRE* WINDSTORM EARTHQUAKE WILDFIRE COASTAL FLOOD COASTAL FLOOD Poland 11 8 0 0 0 Portugal 35 3 0 0 5 Romania 0 2 0 40 0 13 20 20 20 13 Slovakia 2 6 0 0 0 Slovenia 16 55 0 8 0 Spain 27 33 32 66 2 Sweden 28 100 0 0 35 EEA Countries 41 20 52 3 3 Source: EIOPA dashboard, Croatian and Romania based on discussions with the insurance industry in the country. Note:*Generally speaking, wildfires are covered under household fire policies, that is, not explicitly excluded. 76 Annex 3. Funding from the EUSF and Disaster Damages over 2002–2022 Annex 5. Assumptions of the National Funding Gap Analysis Government Liabilities applied to ascertain the government’s pro­ portionate share of these losses. An average of The funding gap analysis focuses on the national- 50 percent has been used in the analysis based on level contingent liabilities as follows: historical events where the EUSF has provided support, but the proportion to damages associated • Emergency operations. These include essential with the public sector fluctuates considerably infrastructure restoration, temporary across these events—ranging from the minimum accommodation, emergency service costs, of 0.5 percent of total damage to a maximum of prevention infrastructure, measures for protecting 83.3 percent.96 Given the observed variability, the cultural heritage, and cleanup operations. While overall funding gap assessment incorporates two not explicitly stated in EUSF documentation, separate scenarios, each applying alternative suppression and prevention activities are assumed values to this assumption. The low scenario to fall within this category given the nature of assumes that public damages constitute wildfire events. The government is assumed to be 35 percent of total damages, while the high 100 percent liable for all emergency operation scenario posits an increased percentage of costs. 80 percent. • Government liability for public sector damage. It is assumed that the government is 100 percent Financial Instruments liable for the costs associated with damage to public assets. Additional costs associated with To understand the funding gap in EU MS, the reconstruction of these assets are not considered analysis considers the following financial but increase the potential costs associated with instruments: these liabilities if included. • National reserve funds. Information on relevant • Government liability for non-public sector damage. reserve funds is included in each case study where Encompassing residential, commercial, and possible, but these funds are multipurpose in industrial losses. In the absence of data and nature, and it would be unrealistic to assume they information on the distribution of these losses can be solely used to fund wildfire costs. The among categories, it is assumed that the remaining national funds are estimated based on losses are not contingent liabilities of the government 0.075 percent of national revenues based on data and will either be covered through insurance or between 2019 and 2022. This estimate was absorbed directly by the private sector. calculated by assuming a reserve level of 0.5 percent held in the respective EU MS, which • Proportion of public sector damage relative to was the basis for the analysis conducted in Phase total damage. The data used to estimate damages 1. Subsequently, it was reduced by 85 percent to resulting from wildfires do not differentiate accommodate the likelihood of the majority of between losses incurred by the public and private these reserved funds being utilized for other sectors. Consequently, an assumption must be hazard events beyond wildfires. This serves as a 96 EC - cohesion data. Link. 77 Annex 5. Assumptions of the National Funding Gap Analysis proxy for overall general contingencies funds between the EUSF aid and the Emergency Aid of (including any dedicated national disaster the EU. For all scenarios, the analysis assumes a reserves) held in each country. total available balance of €800 million based on the information contained in the special meeting • EUSF. Given the complexity of the EUSF, the level of the European Council (February 1, 2024). Based of loss required to activate the EUSF (0.6 percent on historical data, the EUSF has predominantly of GNI or a max level of disaster loss) is analyzed to supported other hazards such as floods and understand (i) where the activation point is relative earthquakes, accounting for approximately to the national reserves and (ii) if lowering the 85 percent of total assistance. Therefore, an activation level allows for an improved risk layering assumption of €123.4 million from the EUSF is of instruments for less severe but frequent events. made available for wildfire events. Second, EUSF financing is capped due to the split Annex 6. Assumptions of the EU-Level Funding Gap Analysis Besides the assumptions already presented in instruments. The national funds are estimated Annex 4, some more simplifying assumptions were based on 0.075 percent of national revenues needed to be able to run an ‘all MS’ analysis. based on data between 2019 and 2022. This estimate was calculated by assuming a reserve level of 0.5 percent held in the respective EU MS. Government Liabilities The allocation for wildfire and drought was subsequently reduced to 15 percent and • Eligible emergency operations costs. The baseline 8.6 percent, respectively, to account for the scenario assumes that such costs represent likelihood of the majority of these reserved funds 16 percent of total economic costs, which is the would be used for other types of hazards beyond historical average for the eight wildfire events wildfires and drought. This serves as a proxy for where the EU has provided support where eligible overall general contingencies funds (including any emergency operations and the costs are estimated. dedicated national disaster reserves) held in each The low scenario and high scenario assume that country. These amounts are then aggregated up to eligible emergency operation costs are 12 percent the EU level. (median) and 35 percent (maximum) across the respective scenarios. • Budget allocations under the MFF. Within the MFF, the proportion of budget allocated to • Proportion of public sector damage relative to response, the ECP pool, and the rescEU transition total damage. The same method has been applied are the main DRF instruments available for eligible as stated in Annex 4. emergency operations costs. These budget allocation instruments are first assessed individually to determine their adequacy and then Financial Instruments collectively in a separate emergency funding gap analysis which combines the MFF and the EUSF • National reserves. Each country is assumed to resources. have a national contingency fund and/or a dedicated disaster reserve to respond to disasters, • EUSF. The assumption is that the EUSF serves as and for the EU-level analysis it is assumed that the the primary DRF instrument to address costs EU will benefit from a country’s national DRF beyond the capacity of national reserves and the 78 Annex 6. Assumptions of the EU-Level Funding Gap Analysis MFF. In recognition that the EU faces multiple p The total public damage estimate from these hazards in any given year it is unlikely that countries events is €2.3 billion. will receive the full amount of €500 million in 2011 ‚ Combining these values and based on the prices97 which has resulted in the need for some associated assumptions previously stated, the plausible assumptions. As a result, the EUSF is total government contingent liabilities are capped at an annual limit of €123 million for estimated at €3.6 billion. wildfire, which represents 15 percent of the EUSF annual budget and aligns with the approach used p From these six events, the EUSF support paid for natural reserves. Similarly, the fund is capped was €157.3 million. at €69 million for drought, equivalent to 8.6 percent ‚ On average, this equated to 36.4 percent of of the EUSF annual budget and consistent with the emergency operation costs per event. treatment of the national reserves. ‚ On average, this equated to 2.3 percent of the estimated contingent liabilities. Other Key Assumptions The style of analysis is based on a low number of wildfire events; however, no clear alternative was • Percent of losses covered by the EUSF. Historically available for use. To account for this uncertainty, the the EUSF has only paid a fraction of the estimated high scenario assumes that the EUSF will aim to cover total costs associated with wildfire events. Six 3.2 percent (the maximum proportion observed in wildfire events were identified between 2002 and the six historical events), while the low scenario 2020 where the EUSF was activated and where assumes that the EUSF will aim to cover 2.2 percent public damages were estimated.98 (the second lowest proportion99 observed in the six historical events) of the estimated contingent p The total emergency operation cost estimate liabilities in future. from these events is €1.3 billion. Annex 7. Deriving public sector costs The EFFIS data indicate that 394,962 (approxi­ sector damages of €5,914 per ha or 56.7 percent of mately 400,000) ha of land across the EU was the total estimated economic cost. Additional spot destroyed by fire in the first nine months of 2023. A checks were conducted by examining historical corresponding cost of €4.1 billion was estimated for a wildfire events both in the region and globally, subset of EU countries by Distrelec100—giving the comparing the area of hectares burned to overall cost ratio of cost per hectare burned of €10,332.50. The estimates. The analysis revealed that, although some ratio estimates for economic impact include costs of ‘cost per area burned’ ratios were similar, the limited firefighting, reforestation, damages, and cleanup. number of data points for this comparison and Hence, the portion of private sector losses was substantial variations in other instances raised calculated and the ratio was adjusted to account for concerns about the validity of the value used in this costs associated with emergency costs and public analysis. 97 Plus any unspent allocation from the proceeding year which is raised over and above the normal EU budget. Link. 98 Link. 99 The lowest proportion recorded was 0.3 percent from the 2012 wildfire in Romania which is considered an outlier. 100 Link. 79 Annex 7. Deriving public sector costs Annex 8. Reserve Funds of the EU MS (€, millions) Table 11: Reserve and contingency funds in the EU MS, including the estimated contingency funds Source: World Bank (2021) 80 Annex 8. Reserve Funds of the EU MS (€, millions) Annex 9. References Bräuninger, Michael, Butzengeiger-Geyer, Sonja, Dlugolecki, Andrew, Hochrainer, Stefan, Köhler, Michel, Linnerooth-Bayer, Joanne, Mechler, Reinhard, Michaelowa, Axel, Schulze, Sven. 2011. “Application of Economic Instruments for Adaptation to Climate Change.” Link. Casartelli, V., and J. Mysiak. 2023. Union Civil Protection Mechanism - Peer Review Programme for disaster risk management: Wildfire Peer Review Assessment Framework (Wildfire PRAF). Cook, S., and S. Holliday. 2022. Insuring Nature’s Survival: The Role of Insurance in Meeting the Financial Need to Preserve Biodiversity (English). Washington, DC: World Bank Group. Link. Distrelec. 2024. The Cost of European Wildfires 2023 Report. Link. EC (European Commission). 2021. “Disaster Risk Financing: Main Concepts and Evidence from EU MS.” Discussion Paper 150. Link. EC (European Commission). 2022a. Fiscal Sustainability Report 2021, Volume 1. Link. EC (European Commission). 2022b. “Disaster Risk Financing: Limiting the Fiscal Cost of Climate-Related Disasters.” Discussion paper 174. Link. EC (European Commission). 2023. Proposal for Council Directive amending Directive 2011/85/EU on Requirements for Budgetary Frameworks of the MS. Link. EU JRC (European Union Joint Research Centre). 2012. Natural Catastrophes: Risk Relevance and Insurance Coverage in the EU. Luxembourg: Publications Office of the European Union. Link. EUSF. 2022. EU Solidarity Fund: supporting disaster recovery 2002-2022. Link. GAR (Global Assessment Report). 2021. GAR Special Report on Drought 2021. United Nations. Link. Government of Italy. 2024. Expenditure data for National Fire Brigade, 2020-2023. Government of Romania. 2021. Fiscal and Budget Strategy for the Period 2022–2024. IMF, 2018. Fiscal Transparency Handbook. Link. JRC (Joint Research Center). 2020. Global Warming and Drought Impacts in the EU. Technical Report. PESETA IV Project - Task 7 - Drought. Link. Mysiak, J., V. Casartelli, and S. Torresan. 2021. Union Civil Protection Mechanism - Peer Review Programme for Disaster Risk Management: Assessment Framework. Nadoll, J. 2017. Unblocking Public Investment Bottlenecks in Romania: A Report on Systemic Causes of Delays and Inefficiencies in the Preparation and Implementation of World Bank Financing Investment Projects. The Nature Conservancy and Willis Towers Watson. 2021. Wildfire Resilience Insurance: Quantifying the Risk Reduction of Ecological Forestry with Insurance—Summary of Insights. Link. 81 Annex 9. References OECD (Organisation for Economic Co-operation and Development). 2012a. Disaster Risk Assessment and Risk Financing: A G20/OECD methodological framework. OECD (Organisation for Economic Co-operation and Development). 2012b. Budgeting in Lithuania. OECD (Organisation for Economic Co-operation and Development). 2019. Fiscal Resilience to Natural Disasters. OECD (Organisation for Economic Co-operation and Development). 2020. Best Practices for Managing Fiscal Risks. Link. Office for Budget Responsibility. 2021. “Working Paper No. 17. Evaluating Forecast Uncertainty with Stochastic Simulations.” Link. Office for Budget Responsibility. 2023. “Discussion Paper 4. Next Steps for Climate Change Analysis.” Link. Primorac, M. 2019. Fiscal Risks of Natural Disasters in Croatia. Steger, Gerhard. 2010. “Austria’s Budget Reform: How to Create Consensus for a Decisive Change of Fiscal Rules.” OECD Journal on Budgeting 2010/1. UNICEF (United Nations Children’s Fund). 2022. Policy Brief on Financing Health Emergencies in Romania: The Response to COVID-19. World Bank. 2011. Learning from Megadisasters: A Decade of Lessons from the Great East Japan Earthquake. Link. World Bank. 2018. Romania - Systematic Country Diagnostic: Background Note - Climate and Disaster Management. World Bank. 2021. Financial risk and opportunities to build resilience in Europe. Link. World Bank. 2022a. Advisory Services Agreement on Strengthening Planning and Budgeting Capacity II: Output 5. Final Report Presenting the Recommendations of the RAS. World Bank. 2022b. Technical Support to the Ministry of Romania. Modernizing the Intergovernmental Transfers Systems - Assessment of the Current Situation. World Bank. 2023. Romania Country Climate and Development Report. Link. World Bank. Forthcoming. Disaster Risk Based Budgeting: Introduction and Options for Operationalization. World Bank and European Commission. 2024. From Data to Decisions: Tools for Making Smart Investments in Prevention and Preparedness. 82 Annex 9. References 83