Impact of climate risk on fiscal space: Do political stability and financial development matter?

Beirne, John; Park, Donghyun; Saadaoui, Jamel; Uddin, Mohammed Gazi Salah Working Paper Impact of climate risk on fiscal space: Do political stability and financial development matter? ADB Economics Working Paper Series, No. 748 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: Beirne, John; Park, Donghyun; Saadaoui, Jamel; Uddin, Mohammed Gazi Salah (2024) : Impact of climate risk on fiscal space: Do political stability and financial development matter?, ADB Economics Working Paper Series, No. 748, Asian Development Bank (ADB), Manila, https://doi.org/10.22617/WPS240467-2 This Version is available at: https://hdl.handle.net/10419/310382 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ ASIAN DEVELOPMENT BANK ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org ADB ECONOMICS WORKING PAPER SERIES NO. 748 October 2024 Impact of Climate Risk on Fiscal Space Do Political Stability and Financial Development Matter? This paper explores how climate risks adversely affect fiscal space using panel local projections covering 199 economies spanning from 1990 to 2022. The findings highlight the impact on economies most vulnerable to climate change. The results suggest that factors such as political stability and financial development have the potential to alleviate these effects. It reveals that the influence of climate risk on fiscal capacity is more significant in situations of limited fiscal space. Implementing fiscal consolidation emerges as a crucial factor in mitigating the negative impact of climate risks on fiscal capacity, with political stability and financial development also playing pivotal roles. About the Asian Development Bank ADB is committed to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific, while sustaining its efforts to eradicate extreme poverty. Established in 1966, it is owned by 69 members —49 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. IMPACT OF CLIMATE RISK ON FISCAL SPACE DO POLITICAL STABILITY AND FINANCIAL DEVELOPMENT MATTER? John Beirne, Donghyun Park, Jamel Saadaoui, and Gazi Salah Uddin ASIAN DEVELOPMENT BANK The ADB Economics Working Paper Series presents research in progress to elicit comments and encourage debate on development issues in Asia and the Pacific. The views expressed are those of the authors and do not necessarily reflect the views and policies of ADB or its Board of Governors or the governments they represent. ADB Economics Working Paper Series John Beirne, Donghyun Park, Jamel Saadaoui, and Gazi Salah Uddin No. 748 | October 2024 John Beirne ([email protected]) is a principal economist and Donghyun Park ([email protected]) is an economic advisor at the Economic Research and Development Impact Department, Asian Development Bank. Jamel Saadaoui ([email protected]) is a full professor of economics at the Université Paris 8. Gazi Salah Uddin ([email protected]) is an associate professor at Linköping University. Impact of Climate Risk on Fiscal Space: Do Political Stability and Financial Development Matter? Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) © 2024 Asian Development Bank 6 ADB Avenue, Mandaluyong City, 1550 Metro Manila, Philippines Tel +63 2 8632 4444; Fax +63 2 8636 2444 www.adb.org Some rights reserved. Published in 2024. ISSN 2313-6537 (print), 2313-6545 (PDF) Publication Stock No. WPS240467-2 DOI: http://dx.doi.org/10.22617/WPS240467-2 The views expressed in this publication are those of the authors and do not necessarily reflect the views and policies ofthe Asian Development Bank (ADB) or its Board of Governors or the governments they represent. ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The mention of specific companies or products of manufacturers does not imply that they are endorsed or recommended by ADB in preference to others of a similar nature that are not mentioned. By making any designation of or reference to a particular territory or geographic area inthis document, ADB does not intend to make any judgments as to the legal or other status of any territory or area. This publication is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) https://creativecommons.org/licenses/by/3.0/igo/. By using the content of this publication, you agree to be bound bytheterms of this license. For attribution, translations, adaptations, and permissions, please read the provisions andterms of use at https://www.adb.org/terms-use#openaccess. This CC license does not apply to non-ADB copyright materials in this publication. If the material is attributed toanother source, please contact the copyright owner or publisher of that source for permission to reproduce it. ADB cannot be held liable for any claims that arise as a result of your use of the material. Please contact [email protected] if you have questions or comments with respect to content, or if you wish toobtain copyright permission for your intended use that does not fall within these terms, or for permission to use theADB logo. Corrigenda to ADB publications may be found at http://www.adb.org/publications/corrigenda. ABSTRACT We analyze the relationship between climate risk and fiscal space in a systematic and rigorous way. To do so, we use panel local projections to examine the role of political stability and financial development in the relationship. For a sample of 199 economies in 1990–2022, we first empirically confirm that climate risks adversely affect fiscal space. We find that such effects are most pronounced for the economies that are most vulnerable to climate change. However, our evidence indicates that political stability and financial development can mitigate such effects. We also identify nonlinearities in the climate risk– fiscal space nexus. More specifically, the impact of climate risk on fiscal space is greater when fiscal space is most constrained—i.e., in the upper quantile of the distribution. While fiscal consolidation is the key to mitigating the adverse effect of climate risks on fiscal space, our results suggest both political stability and financial development can contribute as well. Keywords: climate risk, institutional quality, fiscal space, bond yields, sovereign ratings JEL codes: F32, F41, F62 1. Introduction Climate risks, which refer to the potential adverse socioeconomic impacts of climate change, entail substantial fiscal risks, especially through their adverse effects on fiscal space. For instance, a big disaster caused by climate change is likely to necessitate large fiscal outlays for relief and recovery efforts. Or extreme heat resulting from global warming could cause extensive agricultural damage, forcing governments to provide subsidies to hard-hit farmers. At a broader level, public spending on climate change adaptation and mitigation is coming to represent one of the biggest sources of fiscal demand around the world. In combination with other large looming fiscal demands, such as those related to population aging, climate change-related fiscal expenditures pose a major threat to fiscal space and fiscal sustainability in the future. The primary original contribution of our paper to the literature is that we empirically examine the link between climate risk and fiscal space in a systematic and rigorous way. To do so, we first investigate whether climate risk has an adverse effect on fiscal space and whether this effect depends on the relative amount of fiscal space. We then examine the role of political stability and financial development in mitigating such climate-related fiscal risks. Political stability is likely to mitigate these risks since it increases the likelihood of more sustainable fiscal policy, for example in the form of a more robust medium-term fiscal framework. As a result, a more stable political environment is likely to reduce the impact of climate shocks and other shocks on fiscal sustainability. In addition, political stability is conducive to more careful, rational, and cost-effective government planning in response to potential climate shocks, which will help preserve fiscal space. Financial development is also expected to mitigate climate-related fiscal risks. In particular, in a financially well-developed economy, firms and households will have access to insurance and other financial instruments that protect them from the negative effects of climate shocks. This, in turn, reduces the need for large fiscal outlays and thus mitigates the negative effect on fiscal space. Furthermore, financial development increases the amount of credit available to firms and households to help them cushion the impact of potential climate shocks. An additional, indirect, channel through which climate risk may adversely affect fiscal space is via its impact on economic uncertainty. There is growing consensus that climate change poses a serious risk to humanity, a consensus that is buttressed by real-world climate events such as record-high temperatures hitting the world in 2024. Nevertheless, there remains significant uncertainty about the future trajectory of climate change as well as its impact on economic activity. Intuitively, political stability and financial development can mitigate the adverse impact of uncertainty on the economy. Political stability prevents political uncertainty from exacerbating economic uncertainty and thus enables economic agents to respond better to the latter. Financial development provides economic agents with various financial instruments, for example hedging instruments such as futures contracts, to protect themselves against economic uncertainty. Therefore, both political 2 stability and financial development can help mitigate the negative economic effects of uncertainty associated with climate change. To empirically examine the role of political stability and financial development on the climate–fiscal nexus, we perform a cross-country analysis. That is, we investigate whether politically more stable and financially more developed economies are less vulnerable to climate-related fiscal risks. To do so, we use a large global sample of 199 economies for the period from 1990 to 2022. Our empirical analysis is based on panel local projections, and two measures of fiscal space—namely, sovereign bond yields and ratings on foreign currency long-term sovereign debt. The two measures are widely used in the literature, and they reflect the financial markets’ assessment of an economy’s fiscal space. Our empirical analysis involves a two-stage estimation. In the first stage, we investigate the relationship between climate vulnerability and fiscal space. We find that climate vulnerability adversely affects fiscal space and that the effects are most pronounced for those economies that are most vulnerable to climate change. The analysis of the first stage confirms the findings of the existing literature, although our country sample is substantially larger than that of other studies. In the second stage, we empirically investigate the role of political stability and financial development in mitigating the negative spillovers of climate exposure to fiscal space. This is the main original contribution of our paper to the literature. We also assess whether the adverse effect of climate risk on fiscal space is nonlinear in the sense that the effect depends on fiscal space. That is, we analyze whether the effect depends on the extent to which an economy is fiscally constrained. Our review of the literature yields only one major study that is somewhat related to our paper—namely, You et al. (2014). In this, the authors empirically investigate the link between democracy, financial openness, and global carbon dioxide (CO2) emissions. They also examine whether the impact of independent variables on CO2 emissions varies throughout the CO2 emission distribution. That is, they analyze whether the impact of these variables on CO2 emissions depends on a country’s CO2 emissions level. Their cross-country empirical analysis covers a global sample of 98 advanced and developing economies and spans the period from 1995 to 2005. The authors find that, among the economies that emit the most, more democracy reduces CO2 emissions but greater financial openness does not. Our paper is fundamentally different, notwithstanding some superficial similarities. First, our political variable is not democracy but political stability, defined as absence of domestic and external conflict. Second, we look at climate risks, which refer to the potential negative impact of climate change, instead of CO2 emissions. Climate risk is a much broader concept than CO2 emissions, which represent a specific component of the global environmental crisis. A deeper difference is that, in our paper, climate risk is an independent rather than a dependent variable. Finally, our key variable of interest is fiscal space rather than climate risk. More precisely, we examine the extent to which political stability moderates the adverse effects of climate risk on fiscal space. In light of the huge 3 challenge that climate change poses for fiscal sustainability, our paper helps us identify the factors that render the challenge more manageable. We also look at nonlinearities but in terms of fiscal space rather than climate risk. Specifically, we test whether the impact of fiscal space on climate risk depends on the extent to which an economy is fiscally constrained. Overall, we find political stability reduces the adverse spillover effects of climate risks on fiscal space. More precisely, our evidence indicates that climate risks are associated with lower-bond-risk-premium lower and higher sovereign ratings in economies that suffer less from both external and internal conflict. In addition, we find that financial development weakens the link between climate risks and fiscal space. Financially more developed economies do not experience a climate-related bond risk premium or a persistent deterioration of sovereign ratings owing to climate vulnerability. Finally, we identify asymmetric effects in the sense that the most fiscally constrained economies are subject to the largest climate-related risk premia. To sum up, we find that financial development and political stability can serve as important buffers against the adverse fiscal impacts of climate change. The rest of the paper is structured as follows: Section 2 reviews the literature; Section 3 outlines the data and methodology; Section 4 reports and discusses the empirical findings; and Section 5 concludes. 2. Literature Review In this paper, we investigate the global challenge of climate risks and climate change, focusing on its impact on fiscal space. Previous studies show that climate change is not just an environmental issue. It also has a significant and adverse impact on economic growth (Oppenheimer et al. 2004; Tol et al. 2004; Mendelson, Dinar, and William 2006; Diffenbaugh and Burke 2019; Dasgupta, Emmerling, and Shayegh 2023) and exacerbates inequality in developing economies (Cappelli, Costantini, and Consoli 2021; Dasgupta, Emmerling, and Shayegh 2023). Given that developing economies are the most vulnerable to climate change, strategic resource allocation is imperative to enhance their resilience (Paglialunga, Coveri, and Zanfei 2022). This includes implementing adaptation policies and risk reduction measures and expanding access to precautionary tools and health services (Paglialunga, Coveri, and Zanfei 2022; Cevik and Tovar Jalles 2023). Furthermore, redistribution and the introduction of social safety nets must be ensured in the affected economies (Cevik and Tovar Jalles 2023). Climate change insurance funds, investments in economic development, and crossnational compensation for low-latitude economies more prone to climate change shocks promote economic resilience in developing economies against climate vulnerabilities (Mendelson, Dinar, and William 2006). To better adapt to and mitigate the socioeconomic impact of climate change and rising temperatures, economies must possess a high adaptive capacity (Tol et al. 2004), a diversified economy (Dissart 2003), political stability (Dell, Jones, and Olken 2012), and strong institutional leadership (Pike, Dawley, and Tomaney 2010). Unfortunately, these attributes are often lacking in developing 4 economies, owing to financial constraints and adverse geographical conditions. Additionally, these economies tend to have more limited fiscal space, meaning their governments are less capable of assisting those affected by climate change shocks (Cevik and Tovar Jalles 2023). One strand of literature examines the asymmetric effects of climate vulnerability and resilience on sovereign risk and public finance. For example, Beirne, Renzhi, and Volz (2021) gauge the impact of climate vulnerability and resilience on sovereign borrowing costs using a panel dataset from 40 advanced and emerging economies. Their findings suggest that climate vulnerability significantly influences sovereign borrowing costs more so than do resilience factors, with bond yields increasing progressively for highly climatevulnerable economies. Cevik and Tovar Jalles (2022) expand this discourse by analyzing data from 98 advanced and developing economies between 1995 and 2017. They find that both climate vulnerability and resilience affect government borrowing costs, with the impact more pronounced in developing economies owing to their weaker adaptive capacities and higher sovereign risk costs. In a similar vein, Boitan and Marchewka-Bartkowiak (2022) analyze the impact of various climate change metrics—performance, exposure to extreme events, vulnerability, readiness, and climate debt— on government borrowing costs in European Union economies from 2000 to 2020. They find that economies with higher climate vulnerability and lower capacity to manage climate disasters face higher sovereign risk premia, underscoring the importance of effective climate disaster management in maintaining favorable borrowing conditions. Zenios (2022) addresses the broader implications of climate risks for fiscal stability, particularly in advanced economies. He argues that climate risks to fiscal stability remain unanswered, providing evidence of divergent climate risks across advanced economies. Moreover, his findings delineate the transmission channels through which climate change impacts public finance, emphasizing the need for comprehensive risk management strategies. Carattini, Heutel, and Melkadze (2023) delve into the impact of macroeconomic stability on climate risk. In particular, they examine whether ambitious climate policies can induce macroeconomic instability and propose efficient climate and macroprudential policies to manage these risks over the long run and across business cycles. Their findings suggest well-designed climate policies can mitigate long-term fiscal risks without destabilizing the economy. Other studies have highlighted additional factors in the ramifications of climate risk. For instance, Cevika and Tovar Jalles (2023) explore the social dimension of climate vulnerability by examining the link between climate risk and economic inequality. Their study reveals that a 1% increase in climate change vulnerability results in a 1.5% increase in inequality, as measured by the Gini coefficient. Yang, Caporin, and Jiménez-Martin (2024) contribute to the discourse by investigating climate transition risk spillovers among six major financial markets from 2013 to 2021. They identify the United States as the primary net contributor to climate transition risk, and the People’s Republic of China (PRC) and Japan as net risk recipients. Their study highlights that climate risk spillovers 11 In Figures 3 and 4, we plot the correlation between our two fiscal space variables and the vulnerability score. We observe that this graphical evidence points toward a positive correlation between climate vulnerability risks and bonds yields and a negative correlation between climate vulnerability risks and sovereign ratings. These observations are in alignment with empirical findings in the related literature (e.g., Beirne, Renzhi, and Volz 2021; Cevik and Tovar Jalles 2022). Figure 3: Scatter Plot for Vulnerability Score and Bond Yields IMF = International Monetary Fund, JPM EMBI = J.P. Morgan Emerging Market Bond Index, ND-GAIN = Notre Dame Global Adaptation Initiative. Source: Authors’ calculations. Figure 4: Scatter Plot for Vulnerability Score and Sovereign Ratings ND-GAIN = Notre Dame Global Adaptation Initiative. Source: Authors’ calculations. 12 3.2. Methodology We use the local projections (LP) approach (Jordà 2005; Jordà and Taylor 2024) to empirically examine the effects of vulnerability shocks on two measures of fiscal space.2 The LP approach presents several advantages, including enabling (i) the estimation of impulse responses directly at each horizon, with no cross-period restrictions, (ii) a simple inference for impulse response coefficients, (iii) ease of application to nonlinear frameworks, and (iv) ease of scaling to panel data. Regarding our research question, all features of the LP approach will help us provide dynamic evidence on the impact of vulnerability shocks on fiscal space variables. We can formulate the LP approach as follows: 𝑦𝑦𝑖𝑖,𝑡𝑡+ℎ =𝑏𝑏ℎ𝑆𝑆𝑖𝑖,𝑡𝑡+𝛾𝛾ℎ𝑦𝑦𝑖𝑖,𝑡𝑡−1 +𝛼𝛼′𝒛𝒛𝑖𝑖,𝑡𝑡−1 +𝑣𝑣𝑖𝑖,𝑡𝑡+ℎ IRF(ℎ) = 𝑏𝑏 ˆℎ (1) where 𝑦𝑦 is the dependent variable of interest, ℎ is the time horizon, 𝑆𝑆 is the impulse variable (a unit shock on the vulnerability score), 𝐳𝐳 is a vector of control variables, IRF denotes the impulse response function, and 𝑣𝑣 is the error term. In our case, the dependent variable will be either government bond yields or sovereign ratings (Kose et al. 2022). The control variables will be based on Beirne, Renzhi, and Volz (2021) and include domestic factors: current account balance, consumer price inflation, government debt, and deficit; and global factors: US bond yields, the MSCI world index, and the VIX. In order to improve identification, we also control for three types of crises (banking, current, and debt), drawn from Laeven and Valencia (2020). 4. Empirical Results 4.1. Panel Local Projection Regressions In Figure 5, our baseline case across all economies indicates a statistically significant premium on sovereign bond yields as a result of climate risk vulnerability, reflecting the surplus return demanded by investors for holding that debt. Further, we split the sample between low and high climate risk vulnerability depending on the value of the vulnerability score. For the less climate-vulnerable economies, a statistically significant effect is not found. This is in line with economic intuition—i.e., low levels of climate exposure will not lead to climate-related premia on sovereign bonds. For economies that are highly exposed to climate change, the impact on bond yields is significant, as expected. Interestingly, the effect is broadly in line with that for the panel as a whole in terms of magnitude, suggesting the economies that are highly vulnerable to climate change may be driving the overall results. 2 We implement the Stata package of Ugarte Ruiz (2023) to estimate the panel LP results. 13 In Figure 6, we perform the same baseline analysis for sovereign ratings, our second measure of fiscal space. We find a consistent result to that carried out on bond yields, whereby a climate vulnerability shock will lead to a persistent decrease in the sovereign ratings for the full sample and the highly climate-vulnerable economies. For the less vulnerable economies, we do not observe such a persistent deterioration in sovereign ratings, as expected. In the following subsections, we investigate the influence of institutional variables and the impact of financial development in light of the preliminary evidence in Table 2. In fact, the level of financial and institutional is significantly lower in the group of highly vulnerable economies. Thus, we go beyond the recent literature (e.g., Beirne, Renzhi, and Volz 2021; Cevik and Jalles Tovar 2022) on the fiscal space–climate nexus in the following by investigating various form of nonlinearities in the climate-related premia on sovereign bond yields and ratings thanks to interaction terms and quantile local projections. Figure 5: Panel Local Projections for Impact of Vulnerability on Bond Yields CI = confidence interval, IRF = impulse response function. Note: Low/high Vulnerability is defined as below/above Q1 for vul100. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 14 Figure 6: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings CI = confidence interval, IRF = impulse response function. Note: Low/high Vulnerability is defined as below/above Q1 for vul100. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. 4.2. Interactions with Institutional Variables In this section, we explore the role of institutional factors as a buffer for climate-related premia on sovereign bond yields and ratings—i.e., the extent to which institutional factors can alleviate the extent of climate risk impacts on the fiscal space. Our institutional variable of interest is a measure of external conflict (from the International Country Risk Guide [ICRG] database, see www.prsgroup.com). Our panel LP impulse responses in Figure 7 reveal striking results. We find that climate-related risk premia on sovereign bonds are significantly greater for economies that score less favorably in terms of external conflict. We thus infer that high political risk (a lower score in the ICRG database) amplifies the compression on fiscal space owing to climate vulnerability by a factor of around 2 at the 1-year horizon compared with the full sample results or those with external conflict. A consistent narrative is found in the case of sovereign ratings (Figure 8).3 3 In Figures C1 and C2 in Appendix C, we present the results using internal as opposed to external conflicts. The results are broadly consistent, although in the case of sovereign ratings there is a lack of statistical significance. This may be related to the measure used, which is foreign currency sovereign ratings. Vulnerability – (shock on vul100) 15 4.3. Interactions with Financial Institution and Financial Conditions In Figure 9, we use Svirydzenka (2016)’s financial institution development index to investigate the influence of financial institution development on the impact of vulnerability shocks on fiscal space. For economies with mature financial institutions, climate vulnerability shocks do not trigger any increase in bond yields. In Figure 10, we can see that climate vulnerability shocks do not have any significant impact on sovereign ratings for economies with elevated levels of financial institution development. On the other hand, climate vulnerability shocks provoke a persistent deterioration in sovereign ratings for economies with low financial institutions and for the full sample, underscoring the importance of sound financial institutions. The mitigating impact of enhanced financial development on the climate–fiscal nexus follows intuition, whereby there is greater depth and liquidity in local financial markets and insurance markets are better developed. We can make some observations on this new result in the literature. Relative to the baseline results bond premia that we find in Figure 5, the upper limit of the impulse response functions for a low level of financial institution development is similar, slightly above 1% at a horizon of 2 years. So an increase of 1 in the vulnerability score implies an increase in bond yields 2 years later. Whereas for economies with a higher level of financial institution development, vulnerability shocks do not imply any increase in bond yields. This is an important result. The development of sound financial institutions makes it possible to preserve fiscal space in the wake of vulnerability to climate risk shocks. In a different context, Aizenman et al. (2024) find that sound financial institutions buffer exchange rate instability. These related results may be viewed as complementary since acquiring hard currency (i.e., US dollar, euro, yen, etc.) for highly vulnerable economies is a question of utmost importance. 16 Figure 7: Panel Local Projections for Impact of Vulnerability on Bond Yields (External Conflicts) CI = confidence interval, IRF = impulse response function. Note: Low/high External Conflicts is defined as below/above Q2 for extconf. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 17 Figure 8: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (External Conflicts) CI = confidence interval, IRF = impulse response function. Note: Low/high External Conflicts is defined as below/above Q2 for extconf. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 18 Figure 9: Panel Local Projections for Impact of Vulnerability on Bond Yields (Financial Institutions) CI = confidence interval, IRF = impulse response function. Note: Low/high Financial Institutions is defined as below/above Q3 for FI. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. 19 Figure 10: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (Financial Institutions) CI = confidence interval, IRF = impulse response function. Note: Low/high Financial Institutions is defined as below/above Q3 for FI. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. 4.4. Quantile Local Projections In this section, we explore the effect of climate risk across the distribution of fiscal space. To do so, we use the approach in Jordà et al. (2022) to estimate quantile panel local projections. We first adjust the dataset to obtain a balanced panel dataset. For the bond yields, we have 32 economies observed from 2000 to 2019 with complete observations. For the sovereign ratings, we have 71 economies observed from 2000 to 2019 with complete observations. We consider the following traditional panel LP estimation function: Δ𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡+ℎ = 𝛼𝛼𝑖𝑖,ℎ+𝛽𝛽ℎ𝑉𝑉𝑉𝑉𝑉𝑉𝑖𝑖,𝑡𝑡+𝛿𝛿ℎ𝑥𝑥𝑖𝑖,𝑡𝑡𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 + 𝜖𝜖𝑖𝑖,𝑡𝑡, = 𝜔𝜔𝑖𝑖,𝑡𝑡𝜃𝜃ℎ,𝜏𝜏, (2) where 𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡 is the fiscal space variable, 𝛼𝛼𝑖𝑖,ℎ is the country-fixed effects, 𝑉𝑉𝑉𝑉𝑉𝑉𝑖𝑖,𝑡𝑡 the vulnerability variable, and 𝑥𝑥𝑖𝑖,𝑡𝑡 the vector of control variables. Let 𝜔𝜔𝑖𝑖,𝑡𝑡 collect the shock, control variables, and fixed effects. Quantile local projections are calculated based on: Vulnerability – (shock on vul100) 20 𝜃𝜃 ˆ𝜏𝜏=arg min 𝜃𝜃𝜏𝜏 ∑ 𝑇𝑇−ℎ 𝑡𝑡=1 �𝜏𝜏𝜏𝜏�Δ𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡+ℎ ≥ 𝜔𝜔𝑖𝑖,𝑡𝑡𝜃𝜃𝜏𝜏��Δ𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡+ℎ − 𝜔𝜔𝑖𝑖,𝑡𝑡𝜃𝜃𝜏𝜏� +(1 − 𝜏𝜏)𝜏𝜏�Δ𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡+ℎ <𝜔𝜔𝑖𝑖,𝑡𝑡𝜃𝜃𝜏𝜏��Δ𝑓𝑓𝑓𝑓𝑖𝑖,𝑡𝑡+ℎ − 𝜔𝜔𝑖𝑖,𝑡𝑡𝜃𝜃𝜏𝜏��. (3) where 𝜏𝜏 (.) denotes the indicator function and 𝜏𝜏 ∈ (0,1) the 𝜏𝜏th quantile. The quantile of Δℎ𝑓𝑓𝑓𝑓𝑖𝑖𝑡𝑡(𝑝𝑝)+ℎ conditional on 𝑋𝑋𝑖𝑖𝑡𝑡(𝑝𝑝) is then obtained as follows: 𝑄𝑄�Δℎ𝑓𝑓𝑓𝑓 ∣ 𝑋𝑋𝑖𝑖𝑡𝑡(𝑝𝑝)�=𝑋𝑋𝑖𝑖𝑡𝑡(𝑝𝑝)𝜃𝜃ℎ,𝜏𝜏≡ 𝑞𝑞𝜏𝜏,𝑡𝑡 ℎ. (4) The results are presented in Figures 11 and 12, which confirm that the impact of vulnerability risk shocks is most evident for higher quantiles of bond yields. A similar pattern emerges for the case of sovereign ratings, thereby implying nonlinear effects and larger relative effects when fiscal space is most constrained—i.e., at the upper quantiles in the distribution for bond yields and the lower quantiles for sovereign ratings. These results confirm the preliminary evidence that we provide in Table 2. In fact, ability to maintain low yields on government bonds and, thus, fiscal space is related to the level of financial institution development. Ability to access hard currency during episodes of financial stress is also key in preserving fiscal space and remaining attractive and credible on international financial markets. Our results, in Figure 11, are in line with the impulse response functions presented in Figure 9 for low level of financial institution development. In addition, for the lower quantile of the distribution for bond yields, the bond premium is not observed after vulnerability shocks at the 2-year horizon. However, a strong change in bond yields is observed after 3 years, indicating that policymakers should remain vigilant in the wake of vulnerability shocks, even when starting from a sound fiscal position. 27 Appendix D. Baseline Impulse Response Functions with an Extensive Set of Controls We add to four variables the original set of controls—namely, capital account openness index, exchange rate stability index, and shares in total trade of fuel imports and exports, as described in Section 2. Figure D1: Panel Local Projections for Impact of Vulnerability on Bond Yields (Extensive Set of Controls) Note: The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. 90% and 95% confidence intervals in dark blue and light blue, respectively. Source: Authors’ calculations. 28 Figure D2: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (Extensive Set of Controls) Note: The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. 90% and 95% confidence intervals in dark blue and light blue, respectively. Source: Authors’ calculations. 29 Appendix E. Impulse Response Functions with Financial Market Development Indicator Figure E1: Panel Local Projections for Impact of Vulnerability on Bond Yields (Financial Markets) CI = confidence interval, IRF = impulse response function. Note: Low/high Financial Markets is defined as below/above Q3 for FM. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 30 Figure E2: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (Financial Markets) CI = confidence interval, IRF = impulse response function. Note: Low/high Financial Markets is defined as below/above Q3 for FM. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 31 Appendix F. Nonlinearities in Impulse Responses According to Government Stability Figure F1: Panel Local Projections for Impact of Vulnerability on Bond Yields (Government Stability) CI = confidence interval, IRF = impulse response function. Note: Low/high Government Stability is defined as below/above Q3 for govstab. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 32 Figure F2: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (Government Stability) CI = confidence interval, IRF = impulse response function. Note: Low/high Government Stability is defined as below/above Q3 for govstab. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 33 Appendix G. Nonlinearities in Impulse Responses According to Involvement of Military in Politics Figure G1: Panel Local Projections for Impact of Vulnerability on Bond Yields (Military in Politics) CI = confidence interval, IRF = impulse response function. Note: Low/high Military in Politics is defined as below/above Q3 for milpol. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 34 Figure G2: Panel Local Projections for Impact of Vulnerability on Sovereign Ratings (Military in Politics) CI = confidence interval, IRF = impulse response function. Note: Low/high Military in Politics is defined as below/above Q3 for milpol. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 35 Appendix H. Nonlinearities in Impulse Responses According to Risk of Ethnic Tensions Figure H1: Panel Local Projections for Impact of Vulnerability on Bond Yields (Ethnic Tensions) CI = confidence interval, IRF = impulse response function. Note: Low/high Ethnic Tensions is defined as below/above Q3 for ethnictens. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100) 36 Figure H2: Panel Local Projections for Impact of Vulnerability on Sovereign Rates (Ethnic Tensions) CI = confidence interval, IRF = impulse response function. Note: Low/high Ethnic Tensions is defined as below/above Q3 for ethnictens. The shock is a unit-shock on the vulnerability variable. Fixed effects are included, and standard errors are obtained through bootstrapping. Source: Authors’ calculations. Vulnerability – (shock on vul100)