© 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. 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. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO), http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: 2021. The Impacts of Disaster Risk on Sovereign Asset and Liability Management. EFI Insight - Finance, Competitiveness and Innovation with Macroeconomics, Trade and Investment. Washington, DC: World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third- party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Graphic Designer: Xtempo Communications 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. Statement on COVID-19 This report does not cover the impacts of COVID-19 on sovereign asset management and liabilities. However, some of the analysis of the challenges involved in identifying natural disaster–related impacts on sovereign asset management and liabilities will apply to the current situation—and indeed to any situation where governments seek to prepare for and respond to crises while facing declines in revenue and higher-than-expected expenditures. >>> Contents Acknowledgments 5 Abbreviations 6 Executive Summary 7 1. Introduction 9 2. Public sector balance sheets and SALM 11 3. How natural disasters impact sovereign balance sheets 14 4. Current state of knowledge on the impact of natural 17 disasters on public finances 5. Preparedness: What countries can do to understand the potential impact of natural disasters on the sovereign balance 19 sheet Modeling issues 19 Risk analysis and stress testing 21 Data issues 21 Coping with inadequate models and data 22 6. Preparedness: What countries can do to understand the potential impact of natural disasters on the sovereign balance 23 sheet New Zealand earthquakes 23 Serbia: 2014 floods 29 Peru: 2017 coastal El Niño flooding 34 7. Conclusions 38 References 40 >>> Acknowledgments This paper was prepared by a team led by Samantha Cook (Senior Financial Sector Specialist) and Cigdem Aslan (Lead Debt Specialist), team comprised of Phillip Anderson (Senior Consultant), David Bevan (Senior Consultant), Mellany Pintado (Consultant), and Jelena Kostic (Consultant). The report was edited by Anne Himmelfarb. The team is sincerely thankful for the comments received from the peer reviewers: Erik Feyen (Lead Financial Sector Specialist), Sebastien Boitreaud (Lead Financial Sector Specialist), Rodrigo Cabral (Senior Debt Specialist) Luis Alton (Senior Financial Sector Specialist), and Tatiana Skalon (Disaster Risk Finance Consultant). The team is grateful for financial support received from the Swiss Economic Agency for Cooperation trust fund for Sovereign Disaster Risk Finance in Middle Income Countries which enabled preparation of this paper. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 5 >>> Abbreviations ALM Asset Liability Management ARCC Authority for the Reconstruction with Changes (Peru) CERA Canterbury Earthquake Recovery Authority CERF Canterbury Earthquake Recovery Fund DRFI Disaster Risk Financing and Insurance DSGE Dynamic Stochastic General Equilibrium EQC Earthquake Commission (New Zealand) FEF Stabilization Fund (Peru) GDP Gross Domestic Product GIMF Global Integrated Monetary and Fiscal Model GST Goods and Services Tax IMF International Monetary Fund LTGM Long-Term Growth Model MEF Ministry Of Economy and Finance (Peru) MFMod Macro-Fiscal Model NDF Natural Disaster Fund (New Zealand) OCR Official Cash Rate SALM Sovereign Asset and Liability Management SOE State-Owned Enterprise EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 6 >>> Executive Summary Implicit contingent liabilities, such as those generated by natural disasters, are often not quantified in the government balance sheet. However, when they materialize, they place pressure on government finances that may raise interest expenditures and financial risks. Understanding the impacts of disaster risk on sovereign assets and liabilities plays a key part in understanding the potential impact of sovereign disaster risk finance strategies which allow governments to reduce the costs and risks of disasters using prearranged financing and insurance methods. Applying the Sovereign Asset and Liability Management (SALM) framework is a new and comprehensive way of looking at the potential impact of a disaster on the public sector balance sheet through assets and liabilities. Its implementation can help build key practical recommendations for understanding risk in its multiple dimensions (economic, fiscal, and financial). This paper introduces a framework that identifies three channels through which natural disaster will impact SALM; (i) the impact on the value of public sector assets and liabilities e.g. depreciation in asset value due to direct damage incurred or changes in the market variables that increase the cost of debt; (ii) direct fiscal costs that are the actual costs incurred as a result of the disaster such as the cost to rebuild infrastructure, and triggering of contingent liabilities, such as loan guarantees to state-owned enterprises (SOEs); and (iii) indirect fiscal costs, such as the impact on government revenue and (non-disaster) expenditure, which arise from the disaster’s impact on the national economy. This framework is applied in three case studies, Peru, Serbia and New Zealand to derive lessons about the potential impact of natural disasters on the sovereign balance sheet and highlight the importance of accounting for disaster impacts across public sector balance sheets. Estimating the impact of a disaster on public assets and liabilities is skewed by the fact that the reconstruction is viewed as an investment, as such loss to net worth will occur only if the asset is completely destroyed. However, accrual accounting with a public sector balance sheet provides higher-quality information about the value of government assets that are susceptible to disaster risk and as such can be used to develop and implement disaster risk finance policies. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 7 In practice, it can be difficult to identify the total direct cost of a disaster with any precision. The Peru, Serbia, and New Zealand cases demonstrated that reconstruction can last for many years; there might be reallocation within existing budget baselines that is difficult to track, and replacement assets might be of a higher standard. Still, the establishment of specific institutions to oversee the response in Peru and New Zealand made it easier to track expenditure than otherwise would have been the case. Assessing the indirect fiscal costs resulting from disasters is specific to each case depending on various factors, such as the relative size of the local economy that is impacted, damage to critical infrastructure, nature of supply chains, sector interdependencies, substitution, and other resilience measures. In the New Zealand case, the lack of national economic impact surprised most observers, but was at least partly attributed to minimal damage to agriculture, manufacturing, and transport. The case studies demonstrate that estimating the potential impact of disasters on the national economy and the sovereign balance sheet is complex requiring significant data and modeling. However, they demonstrate that viable mechanisms to assist timely post disaster response and reconstruction can have very high payoffs, especially when assisted by an appropriate SALM framework, moreover, that the lack of these may be very costly. Reserve funds were found to mitigate the need to borrow after the event, as demonstrated by both the Peru and New Zealand cases. In the case of New Zealand, the Natural Disaster Fund, in essence the capital of a government-run insurance scheme for households, covers the first tier of losses for most natural disasters. In Peru, the Stabilization Fund may be accessed to finance severe economic and disaster shocks. Reinsurance can play a major role in reducing the economic impact of disaster. The New Zealand case study included the use of the global reinsurance market, both by the government scheme and private sector insurers; it showed that claims on foreign reinsurers improved the net international investment position of New Zealand by around eight percentage points of GDP. Without this, the government would have been required to borrow more, as the NDF was exhausted (for the first time in 70 years). The application of SALM can increase countries’ resilience to financial shocks posed by disaster risk through improved understanding of the impacts of disaster risk on both sides of the sovereign balance sheet. Going forward it could even be used to define a country’s risk tolerance to disaster risk, monitor changes in this position and help to inform policy design on disaster risk and where needed support the introduction of financial instruments to manage disaster risk. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 8 1. >>> Introduction When sovereign disaster risk financing and insurance (DRFI) is limited or absent, governments act as insurer of last resort, and as such carry much of the financial burden of natural disasters. Consequently, the impacts posed by disaster risk can result in fiscal pressure which suggests that disasters could, and perhaps should, be considered contingent liabilities. In the event of a disaster, governments tend to rely on increased borrowing, increased taxation, or—most likely—budget reallocation, in which budgeted lines of public spending are reduced to release resources for the unbudgeted post-disaster categories that need to be increased. There is a growing body of literature on the need for pre-arranged finance to help manage these unforeseen expenditures, but there is a sizeable gap on the public financial management issues associated with this. By applying the Sovereign Asset and Liabilities Management (SALM) approach a new and comprehensive way of looking at the impacts of disasters on public assets and liabilities is presented. This can serve as a useful tool to design disaster risk finance policies to help create additional fiscal space when needed. It is useful at the outset of this paper to distinguish between the impacts of natural disasters and climate change on government balance sheets. Natural disasters are probabilistic events, the risk of which can be transferred to insurance markets. For governments, natural disasters represent a fiscal risk or shock. By contrast, the impacts of climate change occur gradually over longer time periods. From a fiscal standpoint, climate change leads to fiscal pressure, as opposed to shocks, though it may increase the frequency or severity of shocks. This paper focuses on the problems posed by disaster shocks, setting aside the trend problems posed by climate change. Implicit contingent liabilities, such as those generated by natural disasters, are often not quantified in the government balance sheet. However, when they materialize, they place pressure on government finances that may raise interest expenditures and financial risks. Understanding the impacts of disaster risk on sovereign assets and liabilities plays a key part in understanding EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 9 Peruvian seniors knitting Image by kolibri5 from Pixabay the potential impact of sovereign DRFI strategies which In practice, even when governments set out to implement can allow governments to reduce the costs of disasters SALM, it has proved to be challenging, and the implementation using prearranged financing and insurance methods. When has usually been partial in nature. It is often restricted to governments understand the impacts from natural disasters financial assets and liabilities and does not include physical including the increased risks posed by climate change, for assets (such as infrastructure) or the government’s future example, they can link their debt and cash management revenue-generating capabilities. strategies with their DRFI strategy by taking into account the country’s risk profile. In many applications, this has not been too problematic, since the risks under review - interest rate risk or exchange rate When the impacts of disasters on balance sheets is unknown risk, for example - have no impact on physical assets and it is difficult to approach disaster risk comprehensively and only muted ones on revenue capability. For natural disasters, to make adequate financial decisions on how to protect and however, the case is quite different. Hence extending SALM restore public assets with finite public funds. Traditionally, many to incorporate natural disasters is likely to be even more governments have treated the management of their assets challenging conceptually, in terms of both data requirements and liabilities separately, usually with separate institutional and modeling approaches.2 responsibilities for different classes of assets and for different classes of liabilities. Taking no account of net positions, this This paper will serve to build the body of evidence on the approach can lead to inefficient management of risk and impacts of disaster risk on SALM and, in turn, support inefficient implementation of policy more generally. countries’ efforts to mitigate the impact and occurrence of fiscal shocks. It seeks to increase countries’ resilience to financial The main objective of the SALM approach is to develop a shocks from disaster risk through improved understanding comprehensive public sector balance sheet, and to use this of the impacts of disaster risk on both sides of the sovereign to develop a coordinated management strategy that reflects balance sheet. Going forward it could even be used to define government’s various objectives.1 An example of how this a country’s risk tolerance to disaster risk, monitor changes in issue might be tackled is offered by Amante et al. (2019); this position and help to inform policy design on disaster risk. they provide an overview of the strategic, operational, and However, it should be noted that the application of SALM takes institutional challenges involved, using Uruguay as an time and, dependent upon the existing accounting procedures illustration. in a country, it can take significant resources. 1. For an extended discussion, see for example Das et al.(2012) and IMF (2018a). 2. While this paper includes consideration of some very elaborate formal models, the terms “model” and “modeling” are taken to include any systematic framework for thinking through the problem at hand, including informal ones. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 10 2. >>> Public sector balance sheets and SALM There has been growing awareness that the analysis of public sector (sovereign) balance sheets provides a valuable tool to improve the implementation of fiscal policy and management of public sector assets. For example, the International Monetary Fund (IMF 2018a) notes that standard fiscal analysis misses much government activity by focusing on flows-revenue, expenditure, and deficits-and debt. Furthermore, it can encourage illusory fiscal practices, such as establishing pay-as-you-go pension schemes for public employees, which may improve fiscal balance outcomes in the short run, but lead to an expanding liability that is not recognized in traditional fiscal measures. An approach that takes account of the full public sector balance sheet can improve outcomes in a number of ways. First, by revealing the full extent of public sector assets, it shines a light on how effectively these are being managed. And they are large: an IMF (2018a) analysis of 31 countries covering 61 percent of the global economy estimates they are worth US$101 trillion, or 219 percent of gross domestic product (GDP) in the sample. Even modest improvements in the return on these assets could yield significant fiscal benefits. A second improvement to outcomes involves improved fiscal policy making. The balance sheet approach supports a more thorough and systematic evaluation of the impact of policies on public finances, recognizing the effects on both assets and liabilities in the long run. For example, cutting back on expenditure for maintenance and investment in building and infrastructure may reduce debt, but would also cause the value of assets to decline due to depreciation. This decline signals that higher expenditure lies in the future, if services are to be maintained. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 11 The overall strength of a balance sheet, i.e., the level of net and military equipment. This observation has led some worth, can also support fiscal decision making. Research authors and practitioners to include the present value of has shown that countries with stronger public sector balance government expenditure and revenues in SALM. Das et al. sheets experience shallower recessions and recover faster (2012), in a review of the conceptual issues, note analysis from economic downturns (Yousefi 2019). This can be that uses this broader definition of a balance sheet to shed explained by there being more room for countercyclical fiscal light on economic policy, such as the smoothing of fiscal policy when net worth is high than when it is low or negative. balances or tax rates over time.3 This blending of accounting The third improvement is in the identification and management and economic concepts has been dubbed the “intertemporal” of financial and other risks. Examining both sides of the public balance sheet (IMF 2018a), or the “comprehensive” balance sector balance sheet-which is a consolidation of central sheet (New Zealand Treasury 2018).4 government and other entities-may reveal mismatches. Or it may show natural hedges across assets and liabilities in In practice, the application of SALM has been to subsets of separate entities, reducing the need for risk management public sector assets and liabilities that are financial in nature at the level of an individual entity. An example of a natural and of material size, such as public debt, foreign currency hedge would be where the public sector has foreign currency reserves, and other financial asset portfolios. Amante et al. debt and foreign currency financial assets; the exposure to (2019) describe four situations, each with a number of country changes in exchange rates would be the net value of these examples: (i) coordinated management of foreign currency positions. The currency mix of these assets and liabilities can reserves and foreign currency debt; (ii) management of asset be adjusted to minimize the exposure to individual foreign levels to provide a buffer against adverse market conditions; currencies. (iii) transactions between the central bank and government that strengthen policy outcomes, reduce cost, and reduce Activity in this third area entails a SALM approach, which is risk; and (iv) analysis of the variables that drive government based on the asset liability management (ALM) approach revenues and the fiscal balance to inform decisions about the undertaken in the private sector, in particular by financial composition of public debt. In addition, IMF (2018a) provides institutions, with the goal of maximizing return subject to an examples of using the balance sheet framework to conduct acceptable level of financial risk (such as currency, interest stress tests of fiscal sustainability. rate, and liquidity risks).The application of ALM to the public sector balance sheets has been a fairly recent development, One practical constraint in implementing SALM may be a perhaps reflecting the lack of information in most countries lack of information. Unlike advanced economies (e.g., New about the assets and liabilities that make up their balance Zealand, Australia, the United States, Canada), developing sheet. The New Zealand government was an early adopter countries in general do not produce comprehensive balance of SALM, as it published its first balance sheet in 1991 sheets, which would require them to consolidate individual and applied ALM principles during the 1990s to guide the balance sheets of various public institutions. Many countries composition of public debt (Anderson 1999). lack a complete inventory of nonfinancial assets. Furthermore, consistent pricing of financial and nonfinancial assets is In applying the ALM approach to public sector balance sheets, complicated, since different accounting principles, accrual SALM needs to reflect the unique nature of governments. based or cash based budgeting, may be used (See Box 1 The strength of a government’s balance sheet (and indeed for more information). Producing a balance sheet based on the government’s creditworthiness) arises from the sovereign accrual accounting is important to ensure that policy makers power to tax residents and citizens. At the same time, a can assess and monitor effectively the mismatches between significant share of public sector assets does not directly stocks of assets and liabilities. produce revenue, for example national parks, cultural assets, 3. For example, Barro (1995) notes that the structure of public debt matters for tax rate smoothing. 4. The New Zealand Treasury notes that the government’s financial statements are compiled according to the rules of generally accepted accounting principles (GAAP). In order to analyze prospects for long-term fiscal policy and the impact of shocks, this is combined with the “fiscal balance sheet” to produce a “comprehensive balance sheet.” EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 12 Box 1: Key Differences in Cash Based and liabilities. The government’s objectives of moving to accrual accounting can include strengthening monitoring and and Accrual Accounting Systems control of expenditure arrears, getting a clearer picture of the fiscal position of public entities outside the central government budget, or gaining a better understanding of the long-term While accrual accounting has been the norm among private sustainability of the public finance. corporations, most governments (around 60% (Cavanagh et al 2017)) still prepare their budgets and report their accounts The implementation of accrual accounting is about much more on a pure cash basis. The development of accrual based than adopting new standards and principles. The biggest international standards for government fiscal and financial challenge is implementing them, which requires the collection of reporting including Government Finance Statistics Manual additional data, reforms to business processes, modernization (GFSM) and International Public Sector Accounting Standards of IT systems, and capacity building, both within and outside (IPSAS) has helped the recent spread of accrual accounting of the government. This takes significant time and at times, in the public sector. money. Many countries have taken more than ten years to implement an accrual-based system. The cost depends on Cash-based budgeting system records receipts and outlays different factors such as starting point, scope, and speed of at the same time cash is received or paid, without regard the transition. Developing and low-income countries are likely to when the activity generating the revenue, consuming the to face higher implementation costs due to the greater need resources, or increasing the liability occurs. for capacity and IT systems. Accrual-based budgeting system records transactions in the Accrual accounting and a public sector balance sheet provide period when the activity generating the revenue, increasing higher-quality information about the value of government the liability, or consuming the resources occurs – regardless of assets that are at risk of damage from climate and disaster when the associated cash is actually paid or received. versus a cash accounting. Accrual based budgeting system enables financial steering using the SALM approach. Countries Governments that follow cash accounting tend not to maintain with cash accounting and no complete data set on government comprehensive and up-to-date records of the value of their assets, face challenges in adopting and implementing the assets and liabilities. Producing a balance sheet based on SALM approach. Countries that are only starting to consider accrual accounting enables policy makers to effectively SALM should start with simple analysis like monetary policy assess and monitor the mismatches between stocks of assets and debt analysis. There are also institutional and policy complexities in (World Bank, 2018). In the cases where a SALM framework implementing SALM, as the public sector balance sheet is is implemented, there are significant differences across managed by separate entities to deliver a range of policy countries. A survey of 28 countries found that the objective outcomes. Further, some of these entities have constitutional, of countries who have developed a SALM framework is often statutory, or policy independence. For example, the assets and limited to monitoring sovereign assets and liabilities rather liabilities of a central bank accumulate in order to implement than determining mismatches between them. monetary policy and other objectives; often central banks are granted independence to pursue these. The governance of In this context, Amante et al. (2019) state that SALM can publicly managed financial asset portfolios (such as sovereign be carried out only through negotiation between central wealth funds, pension funds, and insurance companies) tend to government and other entities, and may require a “meeting emphasize granting boards and fund managers independence of minds”to pursue outcomes that are beneficial at the level to pursue agreed objectives. This arrangement is designed of the entire public sector balance sheet. Accordingly, they to address historic under performance, for example from a define the objective of SALM as “to improve the efficiency lack of contestability, imposition of noncommercial objectives, of policy implementation in terms of reducing risk and/or and political interference in asset allocation. However, most cost, consistent with the objectives and policy frameworks countries include state-owned enterprises in the sovereign of monetary and fiscal policies, conventional public debt balance sheet, but only a minority also consider central banks, management, state-owned enterprises (SOEs), and publicly in some cases only international reserves and sovereign funds managed financial-asset portfolios”. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 13 3. >>> How natural disasters impact sovereign balance sheets A natural disaster will impact the public sector balance sheet through three channels: 1. Impact on the value of public sector assets and liabilities. For example: • Loss of or damage to publicly owned infrastructure and buildings. For this loss or damage to be recognized, the value of these assets prior to the disaster needs to have been recorded. (The valuations would reflect the age and condition of the assets—i.e., they would be depreciated accordingly.) • Changes in market variables (such as exchange rates and interest rates)—for example,an increase in the value of foreign currency public debt if the disaster triggers exchange rate depreciation. 2. Direct fiscal costs. These are the actual costs incurred as a result of the disaster. Examples include disaster relief and other financial support to citizens, cost to rebuild infrastructure, and triggering of contingent liabilities, such as loan guarantees to state- owned enterprises (SOEs) or subnational governments that are badly impacted by the disaster. The impact on the balance sheet will depend on the nature of the expenditure. Capital expenditure results in the creation of an asset, which increases net worth; if such expenditure is funded by debt, then the impact on net worth is neutral (the value of the debt and the new asset are equal initially). Operating expenses related to the disaster, such as grants and other assistance, would result in more borrowing than otherwise would have been the case, thereby decreasing net worth. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 14 3. Indirect fiscal costs. These are the costs incurred by to meet the fiscal costs that cannot be fully financed in the the disaster’s impact on government revenue and (non- local market, meaning that the government is forced to disaster) expenditure, which arise from the disaster’s borrow externally. This may lead to increased currency risk impact on the national economy. in the debt portfolio, and interest rate risks in case debt is contracted at variable rates. In addition, a government that Figure 1 provides a summary of these impacts on the is limited to short-term borrowing from local markets would public sector balance sheet, as well as the analysis that see increased refinancing risks. is required to estimate the size of them. The availability of data on disaster risks and the assets exposed to perils The magnitude of these impacts can be mitigated by will be a challenge in many countries. In particular, there measures taken by the government before disaster will be a need for data on public sector assets, including strikes, and as a result many Ministries of Finance now infrastructure. Estimating the economic impact of disasters, view financial resilience as a core component of macro- and therefore the indirect fiscal costs, is also challenging.5 fiscal policy. This formed a key discussion during the G20 Finance Ministers’ and Central Bank Governors’ The impact on SALM could Meeting in 2019. The discussion highlighted that a growing arise through rapid depletion of number of countries are developing financial protection strategies leveraging different financial instruments (such Government contingency funds as contingency funds and risk transfer mechanisms such and cash balances, increasing as catastrophe risk insurance, catastrophe bonds etc.)to secure timely and efficient access to funds for governments the liquidity risk faced by central to respond to shocks (World Bank, 2019). More generally, government. the availability of “fiscal space” through prudent debt levels, as well as careful management of risks in the debt portfolio, can provide a buffer for natural disasters as well as other Additional borrowing by the government may be required shocks. Aerial view of Lima, Peru Image by David Ress on Unsplash 5. Wouter Botzen, Deschenes,and Sanders (2019) provide a useful stock take of models and empirical studies and note shortfalls in existing approaches, particularly in relation to geographical and spatial detail. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 15 > > > FIGURE 1 - Framework for impacts of disasters on the public sector balance sheet Disaster Probability of natural risk disasters model (of various magnitudes) Economic impact Exposure Probability of damage Macro • Local model (at various magnitudes) model Government • National decisions ex ante and ex post Valuation of public Fiscal sector assets and Direct fiscal costs Indirect fiscal impact model liabilities Impact on PS balance sheet • Net worth • Debt levels Source: World Bank Note: PS = Public Sector EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 16 4. >>> Current state of knowledge on the impact of natural disasters on public finances There is a vast literature on the financial impacts of natural disasters and on SALM, although there is limited information on the interrelationships of the two. This section presents a sample of the empirical literature that addresses considerations of balance sheet management following a natural disaster and the impacts on the broader macro-fiscal context. In the past, the idea of any relationship between macro-fiscal risk and disaster risk was often dismissed. However, with the emergence of sustainable finance, the literature on these links is increasing. Feyen et al. (2020) discuss the implications of disaster risk posed by climate change for balance sheet and macro-financial management. They find that the two forms of risk may be correlated, so that many countries face a form of “double jeopardy.”Further, the current COVID-19 pandemic serves to demonstrate that countries may face a triple shock—to health, economic, and financial risk; disasters will further compound this risk. Two studies published by the World Bank go further, examining how resilience might be increased by appropriate fiscal policy (Forni, Catalano, and Pezzolla 2019) and how the fiscal risks associated with natural disasters might be managed (Schuler et al. 2019). These studies show that early, preventive action to address disaster risk —for example, ensuring that public spending in risk reduction (or adaptation)is complemented with public debt reduction, or the accumulation of savings in a reserve fund—is always superior to late, remedial action. Investing in adaptation increases the resilience of the capital stock, while containing or reducing the debt burden improves financial sustainability and eases future borrowing constraints. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 17 In trying to build an understanding of natural disasters’ impact then goes on to outline a framework for coordinated action. on public expenditures, and how this impact might be valued, Clarke et al. (2016) take this one step further and propose Bevan and Cook (2015) make some suggestions toward a framework to help governments choose between different developing an operational framework to address these issues, financial instruments, or combinations of instruments, to fund while stressing that the available evidence is extremely disaster losses in an efficient way. They favor a tiered approach, incomplete. They also provide a demonstration of the problems with different instruments being utilized in sequence as shocks with the Cobb-Douglas assumption and show that outcomes increase in severity. When combining instruments,prearranged may be very different if the assumption of complementarity financing is key, whether this finance is domestic or from the between public and private capital is made instead. To identify international donor community. Both Cantelmo, Melina, and the budgetary savings required to create fiscal buffers for Papageorgiou (2019) and Marto, Papageorgiou, and Kluyev self-insurance, Nishizawa, Roger, and Zhang (2019) use (2018) find that it is more cost-effective for donors to contribute estimates of revenue loss and increased spending pressures to the financing of resilience before a disaster than to disburse from disasters combined with information on frequency; they aid afterward. They also find, however, that welfare gains find an average annual fiscal cost of 1–1.5 percent of GDP for to countries that self-finance investments in resilient public an event that occurs approximately every 14 years. This cost infrastructure are negligible, and international aid must be would then need to be financed by the budget and additional sizable to alter this. financial instruments, including external borrowing, to meet the associated costs of the disaster. The concept of public balance sheet strength was introduced by Yousefi (2019), whose empirical work suggests that Many governments have moved toward proactive risk financial markets take into account government assets and management and seek to prearrange financial instruments so net worth in addition to their liabilities when pricing sovereign that finance can be released immediately after a disaster. IMF bonds. Moreover, given that countries with a strong balance (2019 a), which examines ways of building resilience to large sheet recover faster in the aftermath of shocks, they have natural disasters, proposes a three-pillar strategy emphasizing incentives to improve their financial resilience in addition to in turn structural, financial, and post-disaster resilience, and their handling of SALM. Landslide after heavy raining on whanganui river road, New Zealand. Photo from Freepik EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 18 5. >>> Preparedness: What countries can do to understand the potential impact of natural disasters on the sovereign balance sheet The previous section provided a brief summary of the literature on the impact of natural disasters on public finances, based mostly on empirical analysis. But how can a country obtain an understanding of the potential impact of natural disasters on its economy and public finances, given that each has a unique risk profile? This section canvases the issues relevant to this question, in particular the alternatives for modeling and stress testing and the challenges that arise from lack of data. Modeling issues It is clear that natural disasters have impacts on the real economy and have fiscal and financing implications. To examine these thoroughly would require a pretty comprehensive modeling approach, or more plausibly several complementary approaches. The reason for the latter is that different aspects of the analysis require quite varied features, including a disaggregated model of the real economy, capable of addressing structural issues as well as fiscal interactions with these; a model capable of tracking financial links with the macro economy;and a model capable of addressing public and private responses, ex ante and ex post, to a set of surprises and other matters requiring revision of expectations in a stochastic world. The economics profession simply has not developed an integrated model capable of handling all this; it will not do so any time soon, and perhaps should not attempt to do so. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 19 The current study does not aim to develop purpose-designed using insurance, building a domestic contingency fund, and models, so it will have to use what is currently available, from drawing down pre-shock indebtedness. The dynamics of GDP, two sources. The first is whatever set of models is currently potential GDP, consumption, and debt are examined, and being utilized in a case study country. This may include welfare conclusions are drawn. The results of this exercise are computable general equilibrium (CGE) models utilizing input- still preliminary and are currently being discussed within the output information and social accounting matrixes, econometric World Bank (World Bank, forthcoming). models, and dynamic stochastic general equilibrium (DSGE) models, among others. These models are usually located in Addressing disaster risk involves a long horizon, and recovery different institutions (for the above trio, typically in academic from a major natural disaster may also be protracted, which institutions, the ministry of finance, and the central bank makes it sensible to utilize some form of growth model. One respectively); accessing the full set may therefore be option is the LTGM, which is an Excel-based tool based on the complicated. The other source involves adaptation of one or Solow-Swan model that seeks to provide an accessible way of more of the “generic” models that have been developed by studying the relation between investment, savings, and growth. international agencies, suitably adapted and calibrated to the The extended version, LTGM-PC, differentiates private and country in question.Both the World Bank and the IMF have public capital rather than treating them as perfect substitutes. developed generic models that can be adapted in this way. The treatment of public capital is quite sophisticated, allowing They have both strengths and weaknesses and are also for a degree of congestibility as well as inefficiency. While the complex; only the briefest descriptions are provided here. model was designed to examine the consequences of different investment paths, it could equally well be used to study capital World Bank models losses from natural disasters and subsequent recovery and There are two World Bank models of interest in the current rehabilitation; however, while there are two capital stocks in context, the Macro-Fiscal Model (MFMod) and the Long-Term addition to effective labor, the specification of production is Growth Model (LTGM).6 again Cobb-Douglas. MFMod consists of individual country models for (currently) In addition to these economic models, the World Bank has also 181 countries, which are used by the World Bank’s country been involved in developing an extended nonlinear certainty economists to generate forecasts and simulate various equivalent approximation method (ENLCEQ), which can policies. The models share a similar structure and functional handle large-scale dynamic stochastic problems by repeated form, but with parameters estimated at the country level. iteration.7 One limitation is that the approach cannot handle The approach is to estimate a structural econometric model Epstein-Zin preferences where risk aversion and the inter of the type that was popular up to a quarter of a century temporal elasticity of substitution are separated. ago, then fell out of favor, and is now making something of a comeback. The estimation follows an error correction IMF models approach. There are at least two potential problems in the The IMF has a wide range of models. These are typically present context. The first concerns how well suited this history- DSGE models, with the dynamics largely driven by consumers based approach is to studying the impact of rare shocks and with perfect foresight or, in the stochastic case, rational future stochastic change. The second concerns the underlying expectations. In the present context, three may be particularly Cobb-Douglas production function, which may underestimate relevant. the consequences of damage to capital, and the difficulty of structural change. The Global Integrated Monetary and Fiscal Model (GIMF) has an overlapping generation (OLG) modeling of the household MFMod is being extended to compare the economic impact sector, with some households being credit-constrained or of different policy options for building fiscal resilience to otherwise hand-to-mouth consumers. These features mean hurricanes in Caribbean countries. The model assumes a that the model has non-Ricardian properties and so can be shock to productive capital and abstracts from other forms used to study fiscal policies, from the perspective of either of economic disruption. It compares four types of prior short-run stimulus or long-run sustainability; and this has been management options: adapting toward more resilient capital, done in a number of country studies (Kumhof et al. 2010). 6. For the former, see Burns at al. (2019); and Burns and Jooste (2019). For the latter, see Pennings (2018), and for a version extended to cover public capital (LTGM-PC), see Devadas and Pennings (2018). 7. Cai et al. (2020) illustrate this with an application to natural resource allocation, but the model could be applied to macro-fiscal management of natural disasters and climate change. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 20 Tractable aggregate consumption optimality conditions exist framework for low-income countries, implemented in July 2018, only for perfect foresight or, in the stochastic case, first-order includes stress scenarios for natural disasters (IMF 2018b). approximations. The GIMF uses a variety of constant elasticity of substitution (CES) production relations, so there is no reliance on In the present context, developing appropriate scenarios is the Cobb-Douglas form. One oddity is that, while private agents particularly difficult. Much of the existing work on natural disasters (firms and some households) are capable of highly sophisticated postulates that the economy starts in an unshocked steady state inter temporal optimization, the government itself follows relatively (or more often, some balanced growth path) and is then hit by simple fiscal and monetary rules. a shock of some severity early in the simulation, with no further shocks occurring within the simulation horizon. The exercise Because of the complexity of GIMF, a simpler version, the can then be repeated for single shocks of different magnitudes, Flexible System of Global Models (FSGM), has been introduced and the implications of different combinations of financing (Andrle et al. 2015).This also derives private consumption and instruments (ex post and ex ante) can be examined over the investment from micro-foundations, but includes other relations recovery period. In forward-looking models, it is assumed that the (for example, those involving trade, labor supply, and inflation) shock was unanticipated but that perfect foresight then prevails. that have reduced form representations. The properties of the While convenient and tractable, these assumptions are hardly model are displayed under a wide range of experiments, including plausible. The economy’s initial state is likely to be characterized monetary, financial, demand, supply, fiscal, and international by an awareness that there is some probability distribution of shocks. future shocks, and it may already be in recovery from an earlier shock. Also, suffering one shock does not preclude further shocks The IMF has also been exploring a sequence of models attempting within the simulation horizon. to address debt sustainability in a more comprehensive way; see Zanna et al. (2019) for a recent example. These models allow Data issues financing schemes that mix concessional, external commercial, and domestic debt, while taking into account the impact of public investment on growth and constraints on the speed and Problems of inadequate data loom large in the present context. magnitude of fiscal adjustment. While they were developed Two aspects of this problem are the paucity of information about primarily to examine the feasibility of “big pushes” in investment, very rare events, and the lack of information about an uncertain they are also well-suited to examine recovery programs following future evolution, exacerbated by climate change. There are also asset damage due to natural disasters. the familiar issues of incomplete information about production relationships, and about the relationship between economic Risk analysis and stress testing observables and population well-being. The discussion here, however, focuses on data problems specific to the public sector. There is now a substantial literature on the analysis of risk and On the side of public expenditure, the main problem is one of the development of appropriate scenarios within which to conduct tracking relevant changes, both in composition and in levels. In stress tests. Much of the focus has been on financial risks (e.g., many countries, there is a mismatch between what is budgeted Adrian, Morsink, and Schumacher 2020), but attention has more and what gets spent, even in the absence of major shocks. recently been devoted to fiscal risks, as in the approach to fiscal This may reflect technical implementation problems, a lack of risk analysis and management developed by IMF (2016).This proper budgetary controls, or a political imbalance where some methodology looks at the impact of shocks on fiscal flows and spending departments are raided by other more powerful ones. balance sheet aggregates, including fiscal solvency, government This means that, post-shock, it may not be possible to infer liquidity, and the government financing burden. reallocations by comparing outcomes with budget. Further, budget categories may not be well aligned with the economic As regards debt, the World Bank and IMF framework for debt activity of interest. For example, it may be difficult to infer the sustainability analysis has been upgraded recently; it now takes true level of maintenance spending from the budgetary codes. a much more systematic approach to incorporating risk into Even when doing so is possible, it may be impossible to infer the analysis and pays greater attention to country specifics. In how much of this spending involves post-shock rehabilitation as particular, it recommends deeper and more extensive analysis for opposed to routine maintenance. In principle, there are ways of countries that appear to be in most danger of debt sustainability solving these problems, but they are likely to take a very long problems (IMF 2013). The new debt sustainability analysis time to implement. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 21 On the side of revenue, there is the problem of gauging to model design or a genuine insight. The best rule of thumb in what extent revenues were affected at given tax rates, and these circumstances is not to rely on a model’s output unless it another of assessing the revenue consequences of any is possible to provide a plausible (verbal) analytic explanation changes in these rates. Disentangling these from the record is of what is happening. If we are to accept results that are not challenging. Using this information on a forward-looking basis intuitive, we need to gain some understanding of why our also involves some judgment as to what changes in rates intuition is wrong. might be feasible and desirable. When this sort of difficulty arises, it may be appropriate to To the extent that the deficit was altered, partly as a supplement the analysis using a very simple model, possibly consequence of the shock itself, and partly as a result of policy a part of the larger model, focusing on the direct impacts only. responses, there will also have been changes in financing. Regarding in adequate data, there are broadly two ways to These may have been ex ante, such as a decision to carry proceed. The first procedure is to have some mechanism extra precautionary foreign exchange reserves or take out for generating a substitute for the missing information. This sovereign insurance; or they may have been ex post, such as is central to the “calibration” exercises that accompany much increased domestic or international borrowing. For the ex post modeling, frequently with reference to “the literature.” Either changes, there will also be the question of the extent to which empirical data are borrowed from other countries, or recourse interest premiums rose. is had to theoretical priors. Though far from ideal, this may not be too problematic provided the missing information is not too Coping with inadequate models and extensive; but it becomes more problematic when the missing information covers much of the model’s required input. The data second way to proceed is to reconfigure the model so that it does not require the missing data. Once again, this might involve using only part of a larger model. If available models are seen as complicated but still inadequate, it may be hard to interpret the results they Given the challenges presented by the existing theoretical generate. They may suggest some general equilibrium frameworks, several case studies were conducted to feedback that is unexpectedly powerful or unexpectedly weak, understand what can now be inferred from the application of and the underlying mechanisms may be quite opaque. It is existing data in country. These case studies are discussed in then difficult to decide whether this result is an artifact of the the next section. Skyline of Zemun, Belgrade, Serbia Photo by Nikola Cirkovic on Unsplash EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 22 6. >>> Preparedness: What countries can do to understand the potential impact of natural disasters on the sovereign balance sheet In light of the channels through which natural disasters impact sovereign balance sheets (outlined in section 3) and the insights from the literature (sections 4 and 5), three cases are examined. The objective is to derive lessons from these cases that may help other countries understand and prepare for the potential impact of natural disasters on the sovereign balance sheet. New Zealand earthquakes Situated on the Pacific Ring of Fire, New Zealand is particularly prone to disasters that are caused by forces at a tectonic plate boundary, namely earthquakes, tsunamis, and volcanoes. The New Zealand case study is in two parts. The first part describes the impact of the 2010–11 Canterbury earthquake series on the government’s finances and balance sheet. The second part focuses on the potential impact of a larger event, an earthquake centered in the capital city, Wellington. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 23 A. Canterbury earthquake series NZ$11.4 billion (Insurance Council of New Zealand 2020; New Zealand EQC 2019). The level of insurance relative to losses 2010–11 at around 75 percent was high compared to earthquakes in other high-income countries – for example one study surveyed events in Japan, Chile, and the United States, and observed The Canterbury region of New Zealand was impacted by a coverage ranging from 3 percent to 35 percent of damage series of destructive earthquakes between September 2010 (Wood, Noy, and Parker 2016). Both the EQC and private and December 2011. Although technically an aftershock, the sector insurers had substantial reinsurance in the international most damaging and deadly tremor was in February 2011, markets. Claims on foreign reinsurers improved the net which resulted in 185 deaths and considerable destruction in international investment position of New Zealand by around the city of Christ church, near where it was centered. eight percentage points of GDP, although this unwound as payments were made (Parker and Steenkamp 2012). The cost of the damage caused by the earthquakes has been estimated at around NZD 40 billion, equivalent to around 20 Despite damage estimated at 20 percent of GDP, the percent of GDP and 7 percent of the nation’s building stock Canterbury earthquake had very little negative impact on the at the time.8 There are uncertainties associated with this national macro-economy in the short run.9 A number of factors estimate and it does not include items such as interruption to have been identified as to why the impact was muted: (i) the business, which is an insurable risk, and central government relatively high level of insurance cover relative to comparable expenditure to provide support for citizens through a range cases in high-income countries; (ii) the nature of supply of measures. On the other hand, uncertainty also arises from networks in the region and the central role of agriculture, which differences between the value of the assets destroyed and were largely undisturbed; (iii) the region’s manufacturing hub value of replacements—for example, the additional value escaped significant damage; (iv) transportation was largely of rebuilding to a higher standard or other discretionary unaffected – the local port recommenced activity within four improvements (Parker and Steenkamp 2012). days and volumes reached their previous peak within a few months, supporting exports; (v) the monetary easing by The insurance liability for the Canterbury earthquakes totaled the central bank shortly after largest earthquake may also just over NZ$32 billion. Private sector insurers bore NZ$22 have buffered the impact (Doyle and Noy 2015, Parker and billion and the government-backed scheme for households— Steenkamp 2012). the Earthquake Commission (EQC) (see box 1)—bore around Box 2: New Zealand’s Earthquake citizens, as commercial insurance against earthquakes was expensive at that time and not taken up by many households. Commission Since its establishment, the EQC has undergone a number of changes, including the entities and perils that are covered. The Earthquake Commission (EQC) is a New Zealand The EQC is funded by a levy added to fire insurance, provided government entity that provides disaster risk insurance to by private insurers. At present the levy is 0.2 percent for a residential property owners; it also invests in natural disaster maximum coverage of NZ$150,000. For disaster cover beyond research and education. It provides cover for damage caused this amount, households rely on their private sector insurers. by earthquakes, natural land-slips, volcanic eruptions, hydrothermal activity, and tsunamis. For residential land, there The levies collected by EQC are used to (i) fund its operations; is also cover (within limits) against damage caused by storms (ii) contribute to the National Disaster Fund managed by EQC; and floods. and (iii) purchase reinsurance in the international market. If the EQC is unable to meet all claims as a result of a very large The EQC was founded in 1945 as the Earthquake and War event, it can fall back on the unlimited guarantee provided by Damages Commission, following destructive earthquakes government to make up the shortfall. The EQC pays a fee for in 1929, 1931, and 1942. Recovery and reconstruction after this guarantee. these events had been funded by government support to 8. The New Zealand Treasury (2013), the Insurance Council of New Zealand (2020), and Wood, Noy, and Parker(2016) use this figure. 9. To provide perspective on the scale of the Canterbury earthquakes, the world’s most expensive natural disaster, the Great East Japan (Tōhoku) Earthquake of 2011, had estimated damage of US$ 210 billion, equivalent to around 3.5 percent of GDP (Ranghieri and Ishiwatari 2014). EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 24 As would be expected after a disaster of this nature, there higher relative to the rest of the country.10 Overall, therefore, was a rebound in activity once reconstruction started. Based the impacts of the earthquakes on the economy were relatively on a total cost of NZD 40 billion, one study estimated that localized and then further offset through reconstruction gains. rate of rebuilding activity would average around 1.5 percent of potential GDP from 2012 to beyond 2020, peaking at just The disaster had a sizable impact on the government’s finances. below 2 percent of potential GDP in 2014 (Wood, Noy, and By June 2017, the amount recognized in the government’s Parker 2016). However, official forecasts cautioned that it financial statements, including both operating and capital was difficult to isolate the effect of the earthquakes from other expenditure, was NZD 15.1 billion, equivalent to 7.5 percent of factors during both the short-run and long-run (New Zealand GDP at the time of the disaster (New Zealand Treasury 2017). Treasury 2011c, 2012a, 2013, 2014, 2015). At the time of the This was spread out over many years, and indeed beyond earthquakes, the economy was still recovering slowly from the 2017. The government’s financial statements are prepared global financial crisis of 2008-2009 (GFC) and the Eurozone on an accrual basis, which results in many expenses being crisis was intensifying; on the other hand New Zealand’s terms recognized before cash is paid out. For example, in the 2011 of trade were improving. During the period of reconstruction, financial statements, earthquake expenses of NZD 9.1 billion the economy was also benefiting from a surge in tourism, were recognized, but net cash payments were only NZD 1.7 strong inward migration, near-record terms of trade, and billion (New Zealand Treasury 2011a). Of the NZD 15.1 billion strong labor income growth. in expenditure, NZD 0.7 billion was absorbed within exising budget baselines. In the Canterbury region, building activity rose by 150 percent from pre-earthquake levels by 2016, compared to 20 percent Just under half the direct fiscal costs were insurance payouts in the rest of New Zealand, and nominal GDP growth surged to households by the EQC. This would have been much higher, to 10.5 percent in 2014 (Wood, Noy, and Parker 2016). The around NZD11.2 billion, were it not for the NZD 4.5 billion recovery of small and medium-size enterprises can be tracked recouped from reinsurance. Other significant costs included from goods and services tax(a value-added tax), which rose support for local government to restore infrastructure, capital from 11.6 percent of the national total in 2011 to 13.4 percent expenditure for government-owned assets, compensation for by 2015 (New Zealand Inland Revenue Department 2015), land deemed unsuitable for rebuilding, and welfare support. this indicates that Canterbury’s share of the national total is Christchurch Earthquake 2011 - New Zealand Photo by Adwo on Shutterstock 10. The percentage had been below 12 in the eight years prior to the earthquakes. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 25 Christchurch Earthquake 2011 - New Zealand Photo by Nigel Spiers on Shutterstock The government established the Canterbury Earthquake to assume that the short-term impact of the earthquakes Recovery Fund (CERF) in the 2011 Budget, which helped was negligible. Over the longer term, the indirect impact on provide transparency around the level of expenses. the government’s finances could be expected to have been Nevertheless, the 2017 financial statements (New Zealand positive, given the boost to activity from the reconstruction. Treasury 2017), which was the last year that estimates of The government’s budget statements in the 2012 to 2014 expenditure relating to the earthquakes were provided, note refer to this, and estimates of the boost to the value-added that as time elapses, the ability to directly attribute costs to tax ranged up to NZD 1.3 billion, but it was noted that this the original events in 2010 and 2011 becomes more diffcult.11 was partly offset by refunds to insurers, as a large part of the rebuild was funded by insurance claims (New Zealand In addition to direct fiscal expenditure, the government impaired Treasury 2012b, 2013, 2014). a total of NZD 375 million against the asset valuation reserve in respect of damage to assets owned by the government. To summarize, based on information provided in financial The impact of the earthquakes on the value of Crown assets statements, the public sector’s net worth is NZD 12.1 and liabilities caused by changes to variables such as interest billion lower and public debt NZD 7.7 billion higher than if rates and exchange rates was likely to have been negligible. the Canterbury earthquakes had not occurred, assuming While the currency dipped initially after the February 2011 no second-round effects and other things being equal. To earthquake, it had fully recovered within a month. The New provide a sense of scale, in June 2017, when the majority of Zealand equity and bond markets were not impacted. the expenses had been recognized and cash paid out, the Crown’s net worth was NZD 116.5 billion (around 9.6% lower The New Zealand Treasury has not estimated the full indirect than otherwise) and gross central government debt was NZD fiscal costs, such as the impact on tax or other revenues as 87.1 billion (around 10% higher than otherwise). At the time, a result of the earthquakes, and this is noted in the financial gross central government debt was 32.5 percent of GDP; statements. As the government’s tax revenues are driven to without the earthquakes it would have been 29.6 percent of a large extent by changes in nominal GDP, it is reasonable GDP. 11. For the purposes of this case study, the 2017 amounts are assumed to reflect the cost of the disaster to the government. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 26 > > > TA B L E 1 Net worth Expenditure Assets Liabilities Net worth Item Accounts percentage (NZ$ billion) (NZ$ billion) (NZ$ billion) (NZ$ billion) of 2017 GDP) Valuation of Impairment of Crown assets - -0.375 - -0.375 -0.1 assets and liabilities Valuation impact from - Negligible Negligible Negligible - market variables Total expenditure 15.1A - - - - Absorbed in existing budget 0.7 - - - - Direct fiscal cost Depletion of NDF - -6.9 - -6.9 -2.6 Funded by debtB - - 7.5 -7.5 -2.8 Capital expenditure - 2.6 - 2.6 1.0 Indirect Negligible Negligible Negligible Negligible Negligible fiscal cost Total -4.7 7.5 -12.1 -4.5 Source: World Bank A. The expenditure figure is net of reinsurance payouts totaling NZ$ 4.5 billion. B. In some fiscal years there may be cash surpluses—the NZ$ 7.5 billion represents the increase in debt compared to a base case without earthquake expenditure and all other expenditure remaining the same. The increase in public debt levels was not on a scale that 1945 to provide insurance for households against a range of would have an impact on the composition of public debt, natural hazards. During the next 65 years the Natural Disaster particularly as the cash impact of the expenses was spread Fund (NDF) grew to over NZD 6 billion, funded by levies on over a number of years. households and investment returns; in addition some of its revenue was used to purchase reinsurance in the international The ability of the New Zealand Government to comfortably market. The NDF was completely depleted, for the first withstand the impact of Canterbury earthquakes on the public time in its history, by the Canterbury earthquakes, resulting sector balance sheet was shaped by two policy actions. The in a loss of net worth on the public sector balance sheet. first was building fiscal space during economic expansion Nevertheless, the NDF shielded the government from some prior to the GFC – on the eve of the crisis, central government additional borrowing after the event. Reinsurance payouts of debt was 17 percent of GDP. The combined shocks of the around NZD4.5 billion provided some protection to net worth. crisis and the earthquakes raised this to 38 percent of GDP Without the NDF and reinsurance, a further NZD11.4 billion by 2012, with the GFC accounting for a much greater share would have been borrowed between 2012 and 2018 to settle of the increase. the claims by households. This would have increased central government debt by 7.0 percent of GDP, compared to the 2.8 The second policy action was the establishment of the EQC in percent that actually occurred. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 27 B. Potential impact of a large is not strong enough to lift output to the level in the baseline forecast: as a result, after five years, nominal GDP is projected earthquake on the balance sheet to be a cumulative NZD 44 billion lower than in the baseline projection. The total impact on the government’s finances is projected at NZD68 billion, with indirect fiscal costs comprising The New Zealand Government undertakes stress tests of the largest share at NZD 38 billion. The government’s operating fiscal resilience. Analysis undertaken in 2018 examined three (or budget) balance is projected to be around three percent of plausible shocks, including a severe earthquake in Wellington, GDP worse than the baseline, in each of the five years after the capital city (New Zealand Treasury 2018). The scenario the event. The level of central government debt is projected to assumes significant casualties and injuries, damage to most be 13 percent of GDP higher than the baseline, at the end of houses, a closure or relocation of a large number of businesses, the fifth year – see Figure 2. and that essential services take months to be restored. Unlike the experience with the Canterbury earthquakes, the scenario The stress testing based on scenarios provides an input to assumes that business activity across the country is weak in fiscal policy, in particular judgments about the level of debt the near term, with both consumers and business being very that is sustainable for the New Zealand Government. The cautious, and with a drop in tourist and migrant arrivals. Major scale of the potential impact on debt levels provides a sense reconstruction does not commence until 18 months later and of the fiscal space that could be required, in order to respond takes over ten years to complete. to shocks without pushing debt levels towards unsustainable levels - or forcing fiscal adjustments that blunt the effectiveness Over the medium term, the economic boost from rebuilding of the response. > > > F I G U R E 2 - Impact of Wellington earthquake on net central government debt compared to baseline (2016–25) 35 30 Percentage of GDP 25 20 15 10 5 0 2020 2026 2023 2025 2024 2022 2018 2016 2019 2021 2017 Year ended June 30 Baseline 2017 Wellington Earth quake Source: New Zealand Treasury 2018. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 28 Serbia: 2014 floods The Human Development Index also fell to 0.77, pushing Serbia back to 2012 levels (World Bank 2016). Following the floods, the government of Serbia launched In May 2014, Serbia suffered flash floods and landslides a significant response and reconstruction operation with as a result of heavy rains across the region. During the extraordinary support from the international community. third week of May, record-breaking levels of rainfall were Various sources were used to finance the emergency response, recorded: more than 200 mm of rain fell in western Serbia reconstruction, and recovery: a combination of government within a week, equivalent to three months of rain under normal funds, private sector resources (including personal and conditions. Already-high levels of soil saturation before the enterprise contributions, family remittances from abroad, and rains increased the presence of unstable soils in hilly areas limited insurance proceeds), cash grants and donations from and led to landslides in both inhabited and uninhabited areas. the international community, and new and rescheduled loans The landslides destroyed houses, roads, bridges, and other from international financial institutions. The total funding raised infrastructure works. The 2014 floods are considered the most to implement recovery and reconstruction activities over the severe in 120 years, impacting more than 38 municipalities period May 2014–October 2015 was €514 million (US$423 and affecting more than 1.6 million people, or 23 percent of million), of which €227 million (US$187 million) was from the total population (Government of Serbia 2014). international borrowing, €193 million (US$159 million) was European Union funds, €42 million (US$35 million) was from Post-disaster needs assessment and individual donations, and €40 million (US$33 million) was from response bilateral international donations. According to the National Bank With the support of the European Union, the United Nations, of Serbia, only €16.9 million (US$14 million) had been paid out and the World Bank, the Government of Serbia conducted a by private insurance companies by the end of 2014, and total post-disaster needs assessment that estimated damages and post-flood insurance claims amounted to only €38.8 million losses in the affected municipalities at €1.7 billion (US$1.4 (US$32 million)—less than 2.5 percent of the total damages and billion), equivalent to 4.8 percent of GDP. Of this amount, losses and less than 2.9 percent of the recovery needs (World €0.9 billion (US$0.7 billion) represents the value of destroyed Bank 2016). The size of this contribution highlights the fact that physical assets, and €0.8 billion (US$0.6 billion) refers to the overall insurance market in Serbia is underdeveloped and losses in production. The hardest hit economic sectors were dominated by a state-owned company which may deter other energy, mining, and agriculture, but significant damages market entrants. This suggests that there is an opportunity for were also inflicted on transport infrastructure (roads, bridges, the government to incentivize the insurance sector to improve and railways). The public and private sectors were affected product offering on flood insurance to reduce future government differently by the disaster, although the damages and losses liabilities. However, the affordability and attractiveness of any they incurred were similar in size (Government of Serbia new products would need to be carefully assessed. 2014). Sovereign asset and liability and disaster Financial requirements were estimated for all sectors of shock social and economic activities, under both public and private Like many developing countries, Serbia does not prepare domains. Post-disaster needs were valued at €1.3 billion comprehensive balance sheets, does not have a complete (US$1.1 billion), of which €403 million (US$332 million) data set on nonfinancial assets, and does not consolidate was for recovery activities (e.g., ensuring the recovery of SOEs. Financial statements are prepared by the Treasury’s personal income) and €943 million (US$777 million) was Budget Accounting and Reporting Department. Quarterly and for reconstruction needs. Financing needs for recovery year-end aggregated financial statements are based on the and reconstruction were estimated to last into at least 2016 balance sheet and on budget execution information submitted (Government of Serbia 2014). both electronically and manually by direct and indirect budget beneficiaries. Accounting and financial reporting in Serbia The disaster led Serbia into an economic recession and are currently maintained on a modified cash basis. The main deteriorated its fiscal position. As a result of the ensuing difference between cash - and accrual - based accounting recession (mostly caused by the floods), the Serbian economy lies in the timing for recording revenues and expenditures. contracted by 1.8 percent in 2014, rather than growing by 0.5 Cash-based budgeting systems record receipts and outlays percent as previously projected. After the floods, an estimated when cash is received or paid, without regard to when the 125,000 people fell below the poverty line, an increase of activity generating the revenue, consuming the resources, or almost 7 percent compared with the level of the previous year. increasing the liability occurs. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 29 This form of accounting has traditionally been recognized for its are described below for the case of the Serbian floods. emphasis on compliance with the annual budget law. Several measures have been introduced into Serbian public sector Valuation of assets and liabilities accounting to supplement cash-based data with noncash Since no accurate asset valuation is available, as a proxy, the information. Some issues with the accuracy of nonfinancial analysis assumes that €450 million (US$371 million), or half of asset valuation have been identified. According to the Republic the total damages, falls within the public sector. Property Directorate, it is solely the responsibility of the budget user to enter accurate data on nonfinancial assets, such as their Over 2014, the Serbian dinar depreciated around 2.8 percent value, changes in value, and information related to the disposal (see figure 3), influenced by developments in the international of assets. The directorate does not assume responsibility for financial markets, reduced foreign exchange inflow from data quality; does not validate or verify information in the asset investments, and deterioration in the foreign trade deficit in the registry; and does not demand it when missing (World Bank second half of the year. In November 2014, the National Bank 2017). of Serbia lowered its key policy rate by 0.5 percentage points to the level of 8 percent (see figure 4). This decision was due The three channels affecting the balance sheet—valuation of mainly to low inflationary pressure. assets and liabilities, direct fiscal cost, and indirect fiscal cost— > > > F I G U R E 3 - Exchange rate: Euro/Serbian dinar(2014–January 2018 126.0 124.0 122.0 Serbian dinar 120.0 118.0 116.0 114.0 112.0 110.0 May May May May Sep Nov Sep Nov Sep Nov Sep Nov Mar Mar Mar Mar Jan Jan Jan Jan Jan Jul Jul Jul Jul 2014 2015 2016 2017 2018 Source: National Bank of Serbia. House and street in Obrenovac under water Photo by Nesa Cera on Shutterstock EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 30 > > > F I G U R E 4 - Serbian monetary policy rate (December 2013–March 2018) 10 9 8 7 6 Percent 5 4 3 2 1 0 1 Dec 2013 1 Mar 2014 1 Jun 2014 1 Dec 2014 1 Mar 2015 1 Jun 2015 1 Sep 2015 1 Dec 2015 1 Mar 2016 1 Jun 2016 1 Sep 2016 1 Dec 2016 1 Jun 2017 1 Sep 2017 1 Dec 2017 1 Mar 2018 1 Sep 2014 1 Mar 2017 Source: National Bank of Serbia. Direct fiscal cost Compared to 2013, the total debt stock in 2014 increased Direct costs are those incurred as a result of the damage, from 61.1 percent of GDP to 71.9 percent of GDP (around including emergency support and recovery. The availability US$4.3 billion) (see figure 5). Public debt levels increased due and quality of data are key determinants in adopting a SALM to the 6.2 percent increase in the budget deficit, lower real framework, and in Serbia, data on post-disaster expenditures GDP growth rate, and depreciation of local currency (dinar) are limited and fragmented. Much of the spending on disasters against foreign currencies. In October 2014, the World Bank remains embedded in other budget lines like operations and approved the Floods Emergency Recovery Loan to Serbia maintenance budgets. The primary financial sources in 2014 in the amount of US$300 million, and Serbia issued 10-year were government revenues, debt (US$300 million), and dinar-denominated bonds for the first time. The issue amount grants (US$182 million) (World Bank 2016). In addition, public was RSD 10 billion (US$0.1billion) with an effective yield rate utility companies financed reconstruction from their own funds of 12.99 percent and 10 percent coupon (see figure 6). The and loans backed with state guarantee. Based on a review issuance of the bond was planned before the floods with a of balance sheets, other current expenditures were RSD 14 strategic goal of a maturity extension, but it likely also covered billion (around US$95 million) higher than initially planned as the financial requirements from the flood. a result of floods and early elections. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 31 > > > F I G U R E 5 - General government public debt in Serbia as a share of GDP (2013–17) 80 1.3 1.4 1.1 70 7.2 6.2 0.9 1.5 7.9 60 4.9 8.4 50 Percent 40 67.4 65.7 30 62.6 59.3 51.2 20 10 0 2013 2014 2015 2016 2017 Non-guarenteed debt of local self-government units, Development Fund, and PE Roads of Serbia Indirect liabilities Direct liabilities Source: Ministry of Finance of Serbia. Note: PE = Public Enterprise. Indirect liabilities include guarantees issued by Serbia for other legal entities. > > > F I G U R E 6 - Dinar weighted average accepted rate on primary auctions (2014–17) 14 12.99 11.99 12 10.93 11.99 10.16 9.21 Percentage of GDP 10 8.49 8.62 7.35 7.54 8 7.06 6.5 5.94 5.83 5.38 5.68 6 4.96 4.42 4.56 5.83 3.48 5.12 4 2.78 3.13 4.78 4.83 3.93 2 2.64 0 3M 6M 53W 2Y 3Y 5Y 7Y 10Y 2014 2015 2016 2017 Source: Serbian Public Debt Department. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 32 Indirect fiscal cost that no reallocation between specific budget positions was Disaster shocks are expected to decrease the revenue allowed. In the first year of the program’s implementation, a base. The Serbia damage and loss assessment estimated a strong turnaround occurred in fiscal policy, with results higher reduction in revenues of €130 million (US$107 million). Based than expected. The improved fiscal position of the country on the 2014 balance sheet, the collection of total revenues has reduced the need for borrowing and the costs of servicing was RSD 35 billion (US$247 million) lower than expected, liabilities. while tax revenues were RSD 53 billion (US$358 million) lower than expected. The exact impact of floods here cannot Since there is no comprehensive balance sheet, list, or be determined, as lower collection of revenues was greatly valuation of nonfinancial assets, it is very difficult to determine influenced by other factors, such as slower nominal growth of in what way the 2014 floods affected the government’s private consumption (partially caused by lower inflation) and balance sheet. According to the available data, it can be growth of activity in the gray market, especially in the market assumed that €450 million (US$371 million), half of the total for tobacco products. It is impossible to determine the disaster damage for physical assets, falls within the public sector. As impact over the following years, given that Serbia concluded a direct result of floods, public debt increased by US$300 a Precautionary Arrangement with the IMF and started million. Considering that the deficit in 2014 was 6.2 percent implementing fiscal consolidation measures and structural (see figure 7), mainly caused by the floods, it is fair to assume reforms at the end of 2014. All the budget positions—like that the direct effect on public debt was significantly higher. current expenditures, capital expenditures, and deficit— Total reduction of revenues in 2014 was around €300 million were agreed with the IMF in the budget preparation process. (US$247 million), and it was estimated that €130 million The approved numbers were treated as limits, which means (US$107 million) was caused by the floods. > > > F I G U R E 7 - Consolidated fiscal balance in Serbia as percentage of GDP (2014–17) 2 1.1 1 0 -1 -2 -1.2 Percent -3 -4 -3.5 -5 -6 -7 -6.2 2014 2015 2016 2017 Source: Serbian Ministry of Finance. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 33 Peru: 2017 coastal El Niño flooding The coastal El Niño had severe impacts on production, mainly in non-primary sectors; effects were widespread in manufacturing, construction, and transport services. Disaster event Impacts in primary industry were narrower, affecting some In the first half of 2017, an El Niño costero (coastal) event, one agricultural products and the production chain of some mines. of the strongest El Niño events documented in Peru, caused Economic activity was affected by the closing of roads, the major impacts in the country. A sudden and unexpected damage to physical capital, and lower demand. According to increase in the temperature of the Pacific Ocean created heavy a preliminary analysis performed by the Ministry of Economy storms and rainfall, which triggered floods and landslides that and Finance (MEF), the shock was expected to shrink GDP continued for nearly four months. The impacts were unevenly growth by 1.2 percentage points (Government of Peru 2017d). distributed in the country, affecting mainly the coast, with half of The impact was difficult to isolate, however, given the Lava the geographic regions declaring a state of emergency. These Jato corruption case in 2017, which compounded the shock events ultimately triggered the overpopulation of mosquitos by decreasing investors’ confidence and paralyzing public- that spread dengue and chikungunya virus. An El Niño event private partnership investment projects. with such a localized impact had not been documented since 1925 and is comparable to the strongest ones in the 20th The recovery process started with the establishment of an century (Government of Peru, 2017a). agency, the Authority for the Reconstruction with Changes (ARCC),to lead implementation of the 2017–21 reconstruction Governmental assessments documented the damage plan. The entire plan’s allocation is S/. 26.7 billion (US$7.8 experienced by the population in terms of loss of lives and billion, 3 percent of 2017 GDP), which is programmed into four physical assets across the country. The event affected 1.7 components: (i) public infrastructure (73 percent of the total million people (around 5 percent of the population), caused amount), (ii) mitigation projects (21 percent), (iii) houses (3 132 deaths, and damaged 413,000 houses and 132,000 ha percent), and (iv) capacities of recipient entities (3 percent). of crops. Buildings and infrastructure were severely damaged, The reconstruction of public infrastructure focuses mainly including 2,600 km of national roads, 192 bridges, 7,000 km of on transport (US$2.5 billion), education (US$1 billion), and regional roads, about 1,500 school buildings, and 726 health drainage and sewerage systems (US$1 billion). In the case facilities (Government of Peru 2017a). There are no official of houses, ARCC is financing 100 percent of the cost for records about the cost of these impacts. However, Macro- rebuilding or replacing around 41,000 dwellings among the consult, a local consulting firm, estimated such physical most vulnerable families affected by the event (Government damages at US$3 billion (equivalent to 1.6 percent of the 2017 of Peru 2017c). GDP); roads and bridges alone accounted for 48 percent of the total (Macro-consult 2017). This amount is to be interpreted as The Government of Peru has a combination of instruments a lower bound, as it does not include total physical damages in to finance disaster recovery, aligned with its fiscal policy and infrastructure (e.g., the collapse of sewage systems). financial strategies. Fishing village residents work to clean their town of mud after flooding brought on my El Niño rains in Cabo Blanco, Peru Photo by CLJ Giordano on Shutterstock EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 34 The Multi-annual Macroeconomic Framework 2020–2023 necessary to understand how Peru defines the public sector emphasizes that the current fiscal policy, which encompasses balance sheet and what the basis is for its financial reporting. two decades of effort, has been vital to mitigate the effects of Financial statements in Peru are a consolidation at the public adverse shocks. As part of this effort, the Stabilization Fund sector level. The Accounting Department of the MEF (DGCP) is (FEF) was established in 1999 with the aim of creating fiscal responsible for annually compiling and publishing the financial savings to respond in adverse scenarios (Government of statements, which covered the following units by 2017: Peru 2019). Peru officially launched its national Disaster Risk Financing and Insurance (DRFI) strategy in 2016, after several • 2,505 public sector entities, of which 2,345 were general years of work. The strategy builds on extensive analytical government units and 160 were public corporations, support, and includes the following financial instruments for including the reserve bank recovery: (i) the FEF, (ii) contingent credit lines, and (iii) debt • 272 central government units (including the health (Government of Peru 2016). In 2016, a pass-through disaster insurance system, the military pension scheme, and three fund was established (FONDES). In addition, in 2017 the housing funds), 27 regional government units, and 2,046 country joined a three-year catastrophe bond that provides local government units US$200 million in seismic coverage (World Bank, 2018b).The management of these financial instruments is governed by the In Peru, public financial accounts are based on a combination Global Asset and Debt Strategy (Government of Peru 2017b). of accrual and cash-based methods, with accounts still pending to move into full accrual. In 2015, an IMF mission identified The financial needs raised by the coastal El Niño are valuation methods for pension, infrastructure, building, and expected to be covered mainly by assets. According to the land as areas for strengthening (IMF 2015). According to the MEF, 80 percent of total financial needs due to the event IMF report, infrastructure and equipment are recorded at the will be covered by assets—including FEF, Treasury, and historical value and often fully depreciated. This approach non-Treasury resources12 as well as donations—with a mild mainly affects the valuation of buildings and structures, many impact on the debt stock. Indeed, a total of US$2.8 billion of which have residual value in the balance sheet of only S/. 1. was mobilized from the FEF to FONDES over the 2017–19 The government has been working to strengthen the valuation period.13 These resources were authorized by law on an of public financial accounts. For example, an IT module was annual need basis, with US$1.8 billion in 2017 followed by created for registering the updated value of buildings and lower amounts thereafter. It is important to highlight that FEF structures at the general government level. had accumulated US$5.4 billion by the end of 2019. Table 2 summarizes the impacts of the El Niño event, Sovereign asset and liability and disaster including the valuation of assets and liabilities and direct and indirect fiscal costs. Net worth is estimated to decline by S/. shock 13.2 billion, with an increase in debt by S/. 2.5 billion over Financial statements are published annually by the MEF, the 2017–19 period. The estimations are limited, given the covering the entire public sector. Before discussing the impacts data and assumptions, and should be interpreted to illustrate of the El Niño event on sovereign assets and liabilities, it is trends rather than as offering precision. Flood water coming from Andes causing serious damage and fatalities in Lima, Peru Photo by Tom Carpenter on Shutterstock 12. Non-Treasury resources include non-tax revenues collected and accumulated from national institutions and subnational governments, such as fees, property levies, and supply of goods and services, among others. 13. Stabilization Fund Report. Reports from 2017, 2018 and 2019. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 35 > > > T A B L E 2 - Estimated impacts of 2017 El Niño event, 2017–19 (S/. billion) Net Item Accounts Asset Liability Worth Valuation of assets and liabilities Building and properties -10.3 - -10.3 Fixed assets +3.5 - +3.5 Direct fiscal cost Financial assets, debt and operative results -3.9 2.5 -6.4 Indirect fiscal cost - - - - Total - -10.7 2.5 -13.2 Valuation of assets and liabilities to the disaster shock. See figure 8 for more details on the The 2017 financial statement did not specify a decrease in the evolution of variables. Over the first half of 2017, the Peruvian valuation of properties due to the shock (Government of Peru sol appreciated about 3.4 percent, influenced by the recovery 2018a). This is probably related to difficulties in the valuation of commodity prices and global depreciation of the dollar.14 of fixed assets mentioned above. As a proxy, damage of US$3 In May 2017, the monetary policy rate was cut by 25 basis billion (S/. 10.3 billion) can be considered a lower bound. In points to 4 percent, and then further reduced to 3 percent over the balance sheet analysis, the damages due to the disaster the same year. On the debt side, the 10-year bond interest (destruction) lead to a reduction in the fixed assets, thus rates declined 84 basis points in the first half of 2017, implying reducing the assets and net worth by S/.10.3 billion. a reduction in the cost of new debt. At the same time, the overnight interest rate was cut in line with the monetary rate, Changes in the exchange rate and interest rate may affect decreasing the rentability of Treasury’s deposits in the reserve the valuation of financial assets and debt. However, Peru’s bank and overall rentability of the public sector’s deposits in financial statements did not report any impact attributed commercial banks (Government of Peru 2018a). > > > F I G U R E 8 - Peru’s exchange rate, interest rate, and 10-year bond yield (January 2016–October 2019) a. Exchange rate (sol/U.S. dollar) b. Interest rate (percent) 3.9 8 Monetary policy rate 3.7 10Y bond Yield 6 3.5 3.3 4 3.1 2 2.9 2.7 0 Jan-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Jul-18 Oct-18 Jul-19 Oct-19 Jan-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Jul-18 Oct-18 Jul-19 Oct-19 Apr-18 Apr-18 Apr-16 Apr-19 Apr-16 Apr-19 Jan-19 Jan-19 Source: Central Reserve Bank of Peru. 14. Debt variation was registered at S/. 1.4 billion due to the sol appreciation, but there is no evidence on a relation with the disaster shock. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 36 Direct fiscal cost Smaller shares were contributed by non-treasury commodity This section covers the actual cost incurred as a result of the resources and donations.15 By type of expenditure, capital damage, including emergency support and recovery. Table investment amounted to S/. 3.5 billion. Thus, the government 3 summarizes the executed expenditures through FONDES fixed assets and net worth increased S/. 3.5 billion. At the over the 2017–19 period. The amount recognized for operating same time, financial assets decreased by S/. 3.9 billion, debt and capital expenditures is S/. 6.3 billion. The primary financial increased by S/. 2.5 billion, and net worth decreased by S/. 6.4 sources were central government revenues and debt, in the billion (total expenditures in the operating results). The result amounts of S/. 3.3 billion and S/. 2.5 billion, respectively. is a total decrease of S/. 2.9 billion in net worth. > > > T A B L E 3 - Peru’s emergency and reconstruction expenditure following 2017El Niño event (S/. million) Item 2017 2018 2019 Total Total expenditure 896 1,906 3,551 6,353 By financing source Revenues and buffers* 830 1,541 1,479 3,850 Debt 66 365 2,072 2,503 By financing source Operating 846 1,103 939 2,888 Capital 50 804 2,612 3,465 Expenditure in table 3 is less than a quarter of that programmed 2018, a more decentralized approach was implemented, and in the reconstruction plan due to significant delays. According subnational governments are now in charge of executing 50 to the Audit Office, delays arose in the first years of the plan’s percent of the total budget. This new approach has raised implementation because the central government was in the levels of budget execution from 15 percent in 2017 to 35 charge of executing around 84 percent of the total budget, percent in 2018 and 2019 (Government of Peru 2018b). with minimum participation of local governments. Then, in 15. Government of Peru, Ministry of Economy and Finance, data tracking expenditure from FONDES, http://apps5.mineco.gob.pe/seguimiento_fondes/Navegador/default.aspx. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 37 7. >>> 7. Conclusions The preceding discussion emphasizes the importance of accounting for disaster impacts across public sector balance sheets, and the case for the implementation of SALM to help build understanding of the impacts of disaster risk. This can then be used to provide key practical recommendations for understanding risk in its multiple dimensions (economic, fiscal, and financial). The case studies demonstrate nicely that viable mechanisms to assist timely post disaster response and reconstruction can have very high payoffs, especially when assisted by an appropriate SALM framework, moreover, that the lack of these may be very costly. These conclusions provide a summary of the lessons learned and the implications for applying the SALM framework to help manage disaster risk. Public sector balance sheet Accrual accounting and cash based accounting both recognize reconstruction as an investment, however, one identified advantage of accrual accounting over cash based accounting is that it allows better identification of when and how reconstruction occurs, and hence of the associated costs and benefits, which can be used to inform measures to build financial resilience against disasters. Recognizing reconstruction as an investment; the value of public assets increases as old assets are replaced with new—even if funded by debt, this is neutral in terms of net worth. The loss of net worth arises from the impairment or write-off of the old assets. However, accrual accounting with a public sector balance sheet provides higher- quality information about the value of government assets that are susceptible to disaster risk and as such can be used to develop and implement disaster risk finance policies. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 38 Countries that are only starting to consider SALM should Managing Disaster Risk: lessons start with simple analysis (e.g., debt analysis). Countries like Serbia, with cash accounting and no complete data set learned from the case studies. on government assets, face challenges in adopting and implementing the SALM approach. However, as highlighted in Reserve funds can mitigate the need to borrow after the the discussion not all aspects need to be included at once, and event, as demonstrated by both the Peru and New Zealand having some basic level of understanding on how disaster risk cases. In the case of New Zealand, the Natural Disaster Fund, can impact the structure of your debt portfolio would benefit in essence the capital of a government-run insurance scheme many countries, perhaps now more than ever in the wake of for households, covers the first tier of losses for most natural the COVID-19 pandemic. disasters. In Peru, the Stabilization Fund may be accessed to finance severe economic and disaster shocks. The long In practice, it can be difficult to identify the total direct cost reconstruction period also resulted in a phased requirement of a disaster with any precision. The Peru, Serbia, and New for cash, allowing additional borrowing to be spread over Zealand cases demonstrated that reconstruction can last for time. Establishing an off-budget fund, specially designated for many years; there might be reallocation within existing budget natural disasters, allows governments to keep and accumulate baselines that is difficult to track, and replacement assets resources over the years, thus mitigating the need to borrow might be of a higher standard. Still, the establishment of after the disaster event. specific institutions to oversee the response in Peru and New Zealand made it easier to track expenditure than otherwise Disasters create an opportunity to embark upon a more would have been the case. systematic approach to disaster risk management. Following the floods in Serbia the government developed a The impact of disasters on the national economy, and more systematic approach to disaster risk management, with therefore the indirect fiscal costs, are specific to each case technical assistance from the World Bank. The government depending on factors such as the relative size of the local developed a disaster risk financing strategy and established economy that is impacted, damage to critical infrastructure, a dedicated fiscal risk management unit. Embedded within nature of supply chains, sector interdependencies, the strategy was the inclusion of a contingent credit, a World substitution, and other resilience measures. In the New Bank Catastrophe Deferred Draw Down Option (CAT-DDO) to Zealand case, the lack of national economic impact surprised strengthen the financial response capacity to natural disasters most observers, but was at least partly attributed to minimal by having a pre-agreed line of credit that can be used following damage to agriculture, manufacturing, and transport. an event. Given these complexities, it is a significant challenge to Reinsurance can play a major role in reducing the economic estimate the potential impact of disasters on the national impact of disaster. The New Zealand case study included the economy and the sovereign balance sheet, as it requires use of the global reinsurance market, both by the government modeling many variables and relationships. Nevertheless, scheme and private sector insurers; it showed that claims on broad-brush scenario analysis, such as that undertaken in the foreign reinsurers improved the net international investment New Zealand case, provides useful input to the development position of New Zealand by around eight percentage points of of long-term fiscal policy, in particular the degree of fiscal GDP. Without this, the government would have been required space that may be required to accommodate the realization of to borrow more, as the NDF was exhausted (for the first time large, credible fiscal risks. in 70 years). The application of SALM can increase countries’ resilience The creation of fiscal space in addition to specific to financial shocks posed by disaster risk through improved financing measures for natural disasters reduces the risk understanding of the impacts of disaster risk on both sides of a parsimonious response to a disaster, as shown in New of the sovereign balance sheet. Going forward it could even Zealand’s response to the Christ church earthquakes. The be used to define a country’s risk tolerance to disaster risk, creation of fiscal space in the form of lower central government monitor changes in this position and help to inform policy debt (as a percentage of GDP) during periods of reasonable design on disaster risk and where needed support the to strong growth can provide additional financial resources to introduction of financial instruments to manage disaster risk. finance response to disasters. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 39 >>> References Adrian, T., J. Morsink, and L. Schumacher. 2020. “Stress Testing at the IMF.” Monetary and Capital Markets Department, Departmental Paper 20/04, International Monetary Fund, Washington, DC. Amante, A., P. Anderson, T. Jonasson, H. Kamil, and M. Papaioannou. 2019. “Sovereign Asset and Liability Management in Emerging Market Countries: The Case of Uruguay.” IMF Working Paper WP/19/290, International Monetary Fund, Washington, DC. Anderson, Phillip R. D. 1999. “Debt Management in New Zealand Using an Asset-Liability Management Framework.” Presentation at the Second Sovereign Debt Management Forum, Washington, DC. Andrle, M., P. Blagrave, P. Espaillat, K. Honjo, B. Hunt, M. Kortelainen, R. Lalonde, D. Laxton, E. Mavroeidi, D. Muir, S. Mursula, and S. Snudden. 2015. “The Flexible System of Global Models—FSGM.” IMF Working Paper WP/15/64, International Monetary Fund, Washington, DC. Barro, Robert J. 1995. “Optimal Debt Management.” NBER Working Paper 5327, National Bureau of Economic Research, Cambridge, MA. Bevan, D. and S. Cook. 2015. “Public Expenditure Following Disasters.” World Bank Policy Research Working Paper 7355, World Bank, Washington, DC. Burns, A., B. Campagne, C. Jooste, D. Stephan, and T. T. Bui. 2019. “The World Bank Macro-Fiscal Model: Technical Description.” Policy Research Working Paper 8965, World Bank, Washington, DC. Burns, A., and C. Jooste. 2019. “Estimating and Calibrating MFMod: A Panel Data Approach to Identifying the Parameters of Data Poor Countries in the World Bank’s Structural Macro Model.” Policy Research Working Paper 8939, World Bank, Washington, DC. Cai, Y., J. Steinbuks, K. L. Judd, J. Jaegermeyer, and T. W. Hertel. 2020. “Modeling Uncertainty in Large Natural Resource Allocation Problems.” Policy Research Working Paper 9159, World Bank, Washington, DC. Cantelmo, A., G. Melina, and C. Papageorgiou. 2019. “Macroeconomic Outcomes in Disaster-Prone Countries.” IMF Working Paper WP/19/217, International Monetary Fund, Washington, DC. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 40 Clarke, D., O. Mahul, R. Poulter, and T. Teh .2016. “Evaluating Sovereign Disaster Risk Finance Strategies: A Framework.” Policy Research Working Paper 7721, World Bank, Washington, DC. Cavanagh, J., S. Flynn, D. Moretti, 2016. “Implementing Accrual Accounting in the Public Sector”, Fiscal Affairs Department, International Monetary Fund, Washington, DC Das, U.S., Y. Lu, M. G. Papaioannou, and I. Petrova. 2012. “Sovereign Risk and Asset and Liability Management: Conceptual Issues.” IMF Working Paper WP/12/241, International Monetary Fund, Washington, DC. Devadas, S., and S. Pennings. 2018. “Assessing the Effect of Public Capital on Growth: An Extension of the World Bank Long-Term Growth Model.” Policy Research Working Paper 8604, World Bank, Washington, DC. Doyle, Lisa, and Ilan Noy. 2015. “The Short-Run Macroeconomic Effects of the Canterbury Earthquakes.” New Zealand Economic Papers 49 (2): 134–56. Eaqub, Shamubeel. 2012. “Canterbury after the Earthquakes: April 2012 Update.” New Zealand Institute of Economic Research, Insight 34/2012. Accessed March 30, 2020. https:// nzier.org.nz/publication/canterbury-after-the-earthquakes-april-2012-update. Feyen, E., R. Utz, I. Z. Huertas, O. Bogdan, and J. Moon. 2020. “Macro-Financial Aspects of Climate Change.” Policy Research Working Paper 9109, World Bank, Washington, DC. Forni, Lorenzo, Michele Catalano, and Emilia Pezzolla. 2019. “Increasing Resilience: Fiscal Policy for Climate Adaptation” with “Appendix C. Increasing Resilience: Fiscal Policy for Climate Adaptation.” In Fiscal Policies for Development and Climate Action, edited by M. A. Pigato, 115–32; 197–205. Government of Peru. 2016. “Estrategia Integral de Protección Financiera ante el Riesgo de Desastres Asociados a Fenómenos Naturales.” Ministerio de Economía y Finanzas. Government of Peru. 2017a. “Boletín Estadístico Virtual de la Gestión Reactiva-Julio 2017.” Instituto Nacional de Defensa Civil. Government of Peru. 2017b. “Estrategia Global de Activos y Pasivos 2018–2021.” Ministerio de Economía y Finanzas. Government of Peru. 2017c. “Plan para la Reconstrucción 2017–2020.” Autoridad para la Reconstrucción con Cambios. Government of Peru. 2017d. “Proyecto de Ley que Propone la Aplicación de la Cláusula de Excepción a las Reglas Macrofiscales del Sector Publico No Financiero.” Government of Peru. 2018a. “Cuenta General de la Republica 2017.” Ministerio de Economía y Finanzas Government of Peru. 2018b. “Informe de Avance de Ejecución Física y Financiera al I Trimestre 2018” Autoridad para la Reconstruccion con Cambios https://www.rcc.gob.pe/2020/ informe-de-avance-de-ejecucion-fisica-y-financiera-al-i-trimestre-2018/ Government of Peru. 2019. “Multi-annual Macroeconomic Framework 2020–2023.” Ministry of Economy and Finance. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 41 Government of Serbia. 2014. “Serbia Floods 2014.” Government of Serbia, Belgrade. https://www.gfdrr.org/sites/default/files/Serbia%20rna%20report.pdf. IMF (International Monetary Fund). 2013. “Staff Guidance Note for Public Debt Sustainability Analysis in Market-Access Countries.” https://www.imf.org/external/np/pp/eng/2013/050913.pdf. IMF (International Monetary Fund).2015. “Peru: Fiscal Transparency Evaluation.” IMF Country Report No. 15/294, International Monetary Fund, Washington, DC. IMF (International Monetary Fund). 2016. “Analyzing and Managing Fiscal Risks: Best Practices.” International Monetary Fund, Washington, DC. IMF (International Monetary Fund). 2018a. Fiscal Monitor: Managing Public Wealth. Washington, DC: International Monetary Fund. October. IMF (International Monetary Fund). 2018b. “Guidance Note on the Bank-Fund Debt Sustainability Framework for Low Income Countries.” International Monetary Fund, Washington, DC. IMF (International Monetary Fund). 2019a. “Building Resilience in Developing Countries Vulnerable to Large Natural Disasters.” IMF Policy Paper, International Monetary Fund, Washington, DC. Insurance Council of New Zealand. 2020. “Canterbury Earthquakes.” Accessed March 19, 2020. https://www.icnz.org.nz/natural-disasters/canterbury-earthquakes/. Kumhof, M., D. Laxton, D. Muir, and S. Mursula S. 2010. “The Global Integrated Monetary and Fiscal Model (GIMF)—Theoretical Structure.” IMF Working Paper WP/10/34, International Monetary Fund, Washington, DC. Macroconsult. 2017. “Daños del FEN 2017.” https://sim.macroconsult.pe/danos-de-el-nino- us3-124-millones-hasta-ahora-macroconsult/. Marto, R., C. Papageorgiou, and V. Kluyev. 2018. “Building Resilience to Natural Disasters: An Application to Small Developing States.” Journal of Development Economics 135: 574– 86. New Zealand EQC (Earthquake Commission). 2019. “Insurance Liability Valuation as at 30 June 2019: Final Report.” Accessed March 19, 2020. https://www.eqc.govt.nz/sites/public_ files/documents/ILVR/Insurance_Liability_Valuation-June2019.PDF. New Zealand EQC (Earthquake Commission). 2020. “The Natural Disaster Fund.” Accessed April 1, 2020. https://www.eqc.govt.nz/about-eqc/our-role/ndf. New Zealand Government. 2011. “Canterbury Earthquake Recovery: Proposed Governance Arrangements.” Memorandum to Cabinet, Office of the Minister for Canterbury Earthquake Recovery and Office of the Minister of State Services. Accessed April 8, 2020. https://treasury.govt.nz/sites/default/files/2017-11/b11-2081711.pdf. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 42 New Zealand Government. 2017. “Whole of Government Report: Lessons from the Canterbury Earthquake Sequence.” Department of Prime Minister and Cabinet. Accessed April 8, 2020. https://dpmc.govt.nz/sites/default/files/2017-07/whole-of-government-report- lessons-from-the-canterbury-earthquake-sequence.pdf. New Zealand Inland Revenue Department. 2015. “The Impact of the Canterbury Earthquakes on Canterbury’s Small & Medium Enterprises (SMEs).” Accessed March 21, 2020. https:// www.classic.ird.govt.nz/resources/e/6/e6eeadd1-cb2a-47b8-b9a3-c4618436d472/2015+Adver se+Events+Synthesis.pdf. New Zealand Treasury. 2008. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2011.” Accessed April12, 2020. https://treasury.govt.nz/sites/default/ files/2008-10/fsgnz-year-jun08-pt2of6.pdf. New Zealand Treasury. 2011a. “Budget Economic and Fiscal Update.” Accessed March 17, 2020. https://treasury.govt.nz/sites/default/files/2011-05/befu11-pt3of8.pdf. New Zealand Treasury. 2011b. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2011.” Accessed March 17, 2020. https://treasury.govt.nz/sites/ default/files/2011-10/fsgnz-year-jun11.pdf. New Zealand Treasury. 2011c. “Pre-election Economic and Fiscal Update 2011.” Accessed March 17, 2020. https://treasury.govt.nz/sites/default/files/2018-02/prefu11-pt2of10.pdf. New Zealand Treasury. 2012a. “Budget Economic and Fiscal Update.” March 30, 2020. https://treasury.govt.nz/sites/default/files/2012-05/befu12-pt4of11.pdf. New Zealand Treasury. 2012b. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2012.” Accessed March 17, 2020. https://treasury.govt.nz/sites/ default/files/2017-11/fsgnz-year-jun12.pdf. New Zealand Treasury. 2012c. “Half Year Economic and Fiscal Update, 18 December 2012.” Accessed March 25, 2020. https://treasury.govt.nz/sites/default/files/2012-12/hyefu12- pt4of11.pdf. New Zealand Treasury. 2013. “Budget Economic and Fiscal Update.” Accessed March 18, 2020.https://treasury.govt.nz/sites/default/files/2013-05/befu13-pt3of11.pdf and https://treasury. govt.nz/sites/default/files/2013-05/befu13-pt4of11.pdf. New Zealand Treasury. 2014. “Budget Economic and Fiscal Update.” https://www.treasury. govt.nz/budget/forecasts/befu2014. New Zealand Treasury. 2015. “Half Year Economic and Fiscal Update, 15 December 2015.” Accessed March 31, 2020. https://treasury.govt.nz/sites/default/files/2015-12/hyefu15.pdf. New Zealand Treasury. 2016. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2016.” Accessed March 17, 2020. https://treasury.govt.nz/sites/ default/files/2016-10/fsgnz-year-jun16.pdf. New Zealand Treasury. 2017. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2017.” Accessed March 17, 2020. https://treasury.govt.nz/sites/ default/files/2018-02/fsgnz-year-jun17.pdf. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 43 New Zealand Treasury. 2018.He Puna Hao Pātiki, 2018 Investment Statement: Investment for Wellbeing. Wellington: The Treasury. New Zealand Treasury. 2019. “Financial Statements of the Government of New Zealand for the Year Ended 30 June 2019.” Accessed April 12, 2020. https://treasury.govt.nz/sites/ default/files/2019-10/fsgnz-2019.pdf. New Zealand Treasury. 2020. “Investment Statements.” Accessed April 2, 2020. https:// treasury.govt.nz/publications/strategies-and-plans/investment-statements. Nishizawa, H., S. Roger, and H. Zhang. 2019. “Fiscal Buffers for Natural Disasters in Pacific Island Countries.” IMF Working Paper WP/19/152, International Monetary Fund, Washington, DC. Parker, Miles, and Daan Steenkamp. 2012. “The Economic Impact of the Canterbury Earthquakes.” Reserve Bank of New Zealand Bulletin 75(3): 13–25. Pennings, S. 2018. “Long Term Growth Model (LTGM v4.3): Model Description.” World Bank. https://www.worldbank.org/en/research/brief/LTGM. Ranghieri, Federica, and Mikio Ishiwatari, eds. 2014. Learning from Mega disasters: Lessons from the Great East Japan Earthquake. Washington, DC: World Bank Schuler, Philip, Luiz Edgard Oliveira, Gianluca Mele, and Matias Antonio. “Managing the Fiscal Risks Associated with Natural Disasters.” In Fiscal Policies for Development and Climate Action, edited by M. A. Pigato, 133–54. Washington, DC: World Bank. Wood, Amy, Ilan Noy, and Miles Parker. 2016. “The Canterbury Rebuild Five Years on from the Christchurch Earthquake.” Reserve Bank of New Zealand Bulletin 79(3): 3–16. World Bank.2016. “Disaster Risk Finance Country Note: Serbia.” World Bank, Washington, DC. http://documents.worldbank.org/curated/en/830671468184737730/Disaster-risk-finance- country-note-Serbia. World Bank. 2017. “Serbia—Public Sector Accounting Review: Report on the Enhancement of Public Sector Financial Reporting.” World Bank, Washington, DC. http://documents. worldbank.org/curated/en/637021520634430574/Serbia-Public-sector-accounting-review- report-on-the-enhancement-of-public-sector-financial-reporting. World Bank. 2018. “How Do Countries Use an Asset and Liability Management Approach? A Survey on Sovereign Balance Sheet Management.” World Bank, Washington DC. http://documents1.worldbank.org/curated/en/818281540481513145/pdf/WPS8624.pdf World Bank 2018b. Sovereign disaster risk finance in middle-income countries: annual report 2017–2018.World Bank, Washington, DC. http://documents.worldbank.org/curated/ en/590781533884356379/Sovereign-disaster-risk-finance-in-middle-income-countries-annual- report-2017-2018 World Bank. Forthcoming. “Policy Options for Natural Disaster Risk Management: Modeling the Economics of Adaptation, Insurance, Contingency Funds, and Debt Reduction in the Caribbean.” Working paper (in preparation), World Bank, Washington, DC. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 44 Wouter Botzen, W. J., Olivier Deschenes, and Mark Sanders. 2019. “The Economic Impacts of Natural Disasters: A Review of Models and Empirical Studies.” Review of Environmental Economics and Policy 13 (2): 167–88. https://academic.oup.com/reep/article- abstract/13/2/167/5522921. Yousefi, Seyed Reza. 2019. “Public Sector Balance Sheet Strength and the Macro Economy.” IMF Working Paper WP/19/170, International Monetary Fund, Washington, DC. Zanna, Luis-Felipe, Edward F. Buffie, Raphael Portillo, Andrew Berg, and Catherine Portillo. 2019. “Borrowing for Growth: Big Pushes and Debt Sustainability in Low-Income Countries.” World Bank Economic Review 30(3): 661 –89. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT 45