³Ô¹ÏºÚÁÏÍø

Debt sustainability analysis (DSA) is a tool used by institutions such as the International Monetary Fund and the European Commission to assess whether countries can continue with their fiscal policies without running into debt that is too large to service without major corrections. Under European Union fiscal rules updated in April 2024 ¹ Zsolt Darvas, ‘What are the implications of the EU new fiscal rules?’ The Why Axis, 24 June 2024, /²Ô±ð·É²õ±ô±ð³Ù³Ù±ð°ù/·É³ó²¹³Ù-²¹°ù±ð-¾±³¾±è±ô¾±³¦²¹³Ù¾±´Ç²Ô²õ-±ð³Ü²õ-²Ô±ð·É-´Ú¾±²õ³¦²¹±ô…. , the European Commission uses DSA to assess whether countries are likely to breach debt and deficit limits of 60 percent and 3 percent of GDP, and if countries that already exceed those levels are likely to be able to rein in their debts within a reasonable timeframe. 

Such DSA-based analysis is expected to be a reliable watchdog of public finances. It is based on economic data, integrating macro, budgetary and financial variables into a transparent risk-based assessment of debt. However, a significant element is missing from DSA as applied by the Commission and other institutions: political risk. This remains absent even though institutions are aware of this risk, which has been receiving increasing attention in their reports.

Political events such as snap elections, collapsing government coalitions and regional conflicts put pressure on public finance, as do issues that must be managed at political level, such as inequality (Alesina and Perotti, 1996). Climate change could also lead to increased political instability and therefore fiscal unpredictability (Dell et al, 2014). Political risk is already a factor in sovereign ratings, but is it also relevant for DSA?

In this analysis, we discuss how political risk can influence sovereign debt. We look first at how political risk can be measured and second at how it might alter debt sustainability projections ² Our analysis is based on Ajovalasit et al (2025). .

Measuring political risk

Political risk is the probability of disruptions caused by political decisions or events that can impact economic and business processes (Sottilotta, 2016). Disruption can stem from government instability or changes in government policies, and these two are not necessarily correlated (Gala et al, 2023). Political risk has often been equated with government instability. The World Bank proxies this risk for emerging markets by the numbers of political assassinations, coups or revolutions ³ See . . For developed markets, political risk is considered to be heightened around elections and is proxied by political cycles (Alesina et al, 1997). But coups and revolutions are extreme events, and elections come about only every few years. More granular and regularly updated indicators would therefore be more useful. 

A comprehensive proxy of political risk is provided by International Country Risk Guide (ICRG) ratings ⁴ Produced by a private company, PRS Group. See . Bekaert et al (2014) validated the ICRG as measuring political rather than economic risks, and Gala et al (2024) documented that a global political risk factor (P-factor) using ICRG carries higher risk premia in stocks, bonds and currency markets than other ratings. ICRG is the most widely used rating. . These weigh several political factors that can disrupt the economy and business, including government instability, internal and external conflict, corruption, the presence of the military in politics, religious and ethnic tensions, undermining of law and order, lack of democratic accountability and weak bureaucracy. The factors are aggregated into a score of 0 to 100, with a higher value indicating lower political risk. Political risk is a well-documented determinant of economic growth, investments and asset prices ⁵ Barro, 1991; Alesina and Perotti, 1996; Bekaert et al, 2014; Pástor and Veronesi, 2020; Gala et al, 2023. .

Significantly, political risk is not only an emerging markets problem. It also matters for developed markets, as seen in the volatile political risk ratings for Italy and France (Figure 1). The volatility is quite significant for both countries. Italy’s rating improved notably from 2014 to 2018 reflecting reforms enacted by prime ministers Matteo Renzi and Paolo Gentiloni. France’s rating dropped steeply in summer 2024 as a consequence of the snap parliamentary elections.

Figure 1: Aggregate ICRG political ratings for Italy (top) and France (bottom)

^{Source: Bruegel based on ICRG (see Ajovalasit et al, 2025).}

The impact of political risk on debt sustainability

Political risk can modify debt sustainability through three channels:

1. Through its impact on the expected values of debt drivers, the most important of which is economic growth. For example, political risk raises uncertainty, which is bad for investment and innovation and, hence, growth.

2. By impacting the volatility of debt drivers, political risk increases risk premia.

3. Political uncertainty undercuts the willingness to pay, as opposed to the ability to pay determined by the economic fundamentals. This is consistent with the increase in risk premia. Even when economic variables are all that matters for debt sustainability, political risk could influence those variables. But it could also influence sovereign risk beyond the influence of economic variables through willingness to pay. 

We assessed the impact of high levels of political risk (ie low ICRG ratings) on the growth rates and sovereign spreads of 46 countries, including high and low political-risk countries, high- and low-debt countries, and emerging and developed economies (see Annex 1 and Ajovalasit et al, 2025). The impact of high levels of political risk is economically large and statistically significant. A ten-point deterioration in a country’s ICRG rating (ie an increase in political risk) is associated with an average annual increase in sovereign spreads of 106 basis points, with more marked impacts for high-debt and high-risk countries. The average reduction in GDP growth with such an increase in political risk is two percentage points.

Developed countries have, on average, higher ICRG ratings and therefore exhibit less political risk, but they also experience rating crashes, and their spreads are only slightly less sensitive to political risk than those of emerging economies.

When political risk is incorporated into long-term debt assessment (by incorporating the ICRG political rating as a variable into the DSA model), the effects on debt trajectories are significant (Figure 2; Ajovalasit et al, 2025).

Figure 2: Debt trajectories accounting for or disregarding political risk

^{Source: Ajovalasit et al (2025). Note: ‘No political risk' refers to a debt trajectory in which political risk is not taken into account.}

1. As a result of political risk, the interquartile range of debt trajectories widens. This is relevant for a sustainability assessment that examines the long-term distribution of debt and asks if the extreme (typically 75th) percentile is excessively large or will keep increasing.

2. When political risk is factored in, the long-term debt mean value also worsens even if the ICRG rating fluctuates around its current value without any shock to the long-term political outlook. It is not only the level of the political rating that impacts debt, but also its uncertainty, because of the non-linear effect of debt stocks on spreads. Even if the ICRG changes have zero mean and zero average political spread, the negative shocks (ie greater political risk) are more impactful than the positive changes (ie lower political risk), since the spreads increase nonlinearly after adverse shocks. 

3. Because of political risk, long-term debt may be put an on increasing trajectory and hence become unsustainable. Figure 2 shows that the 75th percentile is stable when the political risk is not factored in, but increases when the political risk is accounted for (blue curve).

4. Political risk effects are more pronounced for high-debt countries in a high-yield environment, but can also be significant for low-debt countries under low yields. 

Reforms and policy implications

It follows that lower political risk should have a positive impact on debt sustainability. But what measures can be taken to improve political risk ratings and soften their impact on debt sustainability? One way to answer this question is to see whether structural reforms incorporated in the International Monetary Fund’s Structural Reforms Database aggregate index (SRD) lead to an improvement in ICRG ratings. The SRD covers domestic and external finance, trade, product market and labour market reforms (Alesina et al, 2020). Our analysis (see Annex 2 and Ajovalasit et al, 2025) suggests that structural reforms lead to improved political risk ratings one year ahead, alleviating debt-sustainability concerns.

However, structural reforms can be difficult. Reforms that bring benefits for an economy may also increase inequality and cause political unrest, thus adversely affecting debt sustainability through the political-risk channel. For example, the French government’s attempt in 2018 to increase fuel taxes as part of its agenda of transitioning to cleaner energy ushered in a period of political unrest and ultimately a government climbdown ⁶ Kim Willsher, ‘Macron scraps fuel tax rise in face of gilets jaunes protests’, The Guardian, 5 December 2018, . . Reforms cannot be the sole response to debt problems, but our analysis shows they can play a significant role.

To test the impact of reforms, we developed debt trajectories for Italy and France. For Italy, we looked at the 2014-2019 period when political risk ratings improved (Figure 1) and projected debt trajectories with and without these ratings improvements (filtering out fiscal spending effects by using the actual primary balance). For France, we projected debt trajectories that, first, take into account the heightened political risk relating to the 2024 snap parliamentary election, and second, that disregard that risk.

For Italy (Figure 3, left panel), our projections show that accounting for improved political risk ratings (red line) generates a result much closer to ex-post observed actual debt values (red triangles). The projection that disregards political risk (or in this case, reduced political risk) produces a less accurate result (blue line). Furthermore, the projected debt dynamics with political risk are sustainable, whereas ignoring the improved political risk ratings would predict steadily rising debt. Thus, accounting for lower political risk through reforms gives better predictions.

For France, political risk ratings, which had been stable for five years, crashed as a consequence of the 2024 snap election (Figure 1). Our projection assumes the worse political risk rating will persist until the 2027 elections, and suggests heightened political risk will add to the debt pressure (red line in Figure 3, right panel). This is in contrast to a scenario in which political risk is disregarded, and also in contrast to IMF projections for France (red triangles in Figure 3).

Figure 3: Impacts on debt of reforms in Italy and snap elections in France (debt/y, %)

^{Source: Ajovalasit et al (2024). Note: ‘No political risk' refers to a debt trajectory in which political risk is not taken into account.}

In summary, political risk can threaten debt sustainability. The effects are more pronounced for high-debt countries in a high-yield environment but can also be significant for low-debt and low-yield countries. Reforms can lead to political rating improvement with a first-order effect on debt, providing one possible path to alleviate debt concerns.

These findings suggest the need for a rethink of debt sustainability analysis by international institutions and the European Commission, to incorporate political risk. However, the treacherous political economy of reforms should be navigated with care.

^{Comments from Hans Geeroms, Ivo Maes, Jonathan Ostry, Lucio Pench, Lennard Welslau and Jeromin Zettlemeyer are gratefully acknowledged.}

References

Ajovalasit, S., A. Consiglio, G. Pagliardi and S. Zenios (2025) ‘Are bad governments a threat to sovereign defaults? The effects of political risk on debt sustainability’, Working Paper, Bruegel, forthcoming

Alesina, A.F., D. Furceri, J.D. Ostry, C. Papageorgiou and D.P. Quinn (2020) ‘Structural Reforms and Elections: Evidence from a World-Wide New Dataset’, NBER Working Paper 26720, National Bureau of Economic Research

Alesina, A., N. Roubini and G.D. Cohen (1997) Political Cycles and the Macroeconomy, MIT Press

Alesina, A. and R. Perotti (1996) ‘Income distribution, political instability, and investment’, European Economic Review 40: 1203–1228

Barro, R.J. (1991) ‘Economic growth in a cross-section of countries’, Quarterly Journal of Economics 106: 407–443

Bekaert, G., C.R. Harvey, C.T. Lundblad and S. Siegel (2014) ‘Political risk spreads’, Journal of International Business Studies 45: 471–493

Dell, M., B. Jones and B. Olken (2014) ‘What do we learn from the weather? The new climate-economy literature’, Journal of Economic Literature 52: 740–798

Gala, V.D., G. Pagliardi, I. Shaliastovich and S.A. Zenios (2024) ‘Political Risk Everywhere’, mimeo, available at 

Gala, V.D., G. Pagliardi and S.A. Zenios (2023) ‘Global political risk and international stock returns’, Journal of Empirical Finance 72: 78–102

Pástor, L. and P. Veronesi (2020) ‘Political cycles and stock returns’, Journal of Political Economy 128: 4011-4045

Sottilotta, C. (2016) Rethinking Political Risk: Concepts, Theories, Challenges, Taylor & Francis

Annex 1: Political risk effects on yields and growth

These tables are from Ajovalasit et al (2025). Countries with ICRG ratings above the post-2008 median value of 77 are classified as high risk (HR), and those below that level as low risk (LR). High-debt (HD) countries are those with average temporal debt ratios above the median of 48 percent of GDP, with low-debt (LD) countries below that level, with respective means of 85 percent and 35 percent. EM and DM denote emerging and developed markets.

Table A1: Political risk and GDP growth

Table A2: Political risk and bond yield spreads

Annex 2: Structural reforms and political ratings

The table reports results from panel regressions of the ICRG political ratings on the SRD index. In column (1), we run pooled OLS regressions; in column (2), we control for country fixed effects; in column (3), we add year fixed effects. Standard errors are robust and clustered at the country level. The asterisks (***), (**), and (*) denote statistical significance at the 1 percent, 5 percent and 10 percent levels, respectively. The sample spans the overlapping ICRG and SRD data on 46 countries over the period 1984–2014 using yearly observations.

^{Source: Ajovalasit et al (2025).}