Editors’ note: This column is based on CEPR Discussion Paper 21723 “From Stocks to Flows: Debt Service and Fiscal Sustainability”.

Standard debt sustainability analysis typically focuses on debt dynamics and the interest rate-growth rate differential. Yet historically, governments have often sustained large debt stocks for prolonged periods under favourable financing conditions, whereas episodes of fiscal stress have sometimes emerged at much lower debt levels (e.g. Reinhart and Sbrancia 2015, Blanchard 2019, Eichengreen et al. 2021, Mauro and Zhou 2021). In a new paper (Eichengreen et al. 2026), we argue that these seemingly contradictory observations reflect a common underlying mechanism.

Figure 1 illustrates this tension. After WWII, federal debt exceeded 100% of GDP, yet debt ratios declined despite the absence of large and persistent primary surpluses, largely because strong growth and favourable financing conditions limited debt-service costs. During the Volcker disinflation of the early 1980s, in contrast, debt levels remained comparatively modest, while soaring real interest rates sharply increased debt-service costs, coinciding with a major fiscal consolidation episode. More recently, debt ratios surged following the Global Crisis and the COVID-19 pandemic, yet historically low interest rates muted fiscal pressures despite persistent deficits. Across these episodes, fiscal adjustment coincides less with high debt itself than with periods when debt becomes expensive to finance.

Figure 1 Public debt and debt-service burdens in the US, 1800-2023

Notes: The figure reports the evolution of the U.S. federal debt-to-GDP ratio (solid blue line) and debt-service costs as a share of GDP (dashed red line) over the period 1800–2023. Debt-service costs are scaled by a factor of 25 for visualization purposes. Gray shaded areas correspond to major wartime periods. The vertical dashed line marks the year 1913, often associated with the emergence of the modern U.S. fiscal and monetary regime. Blue shaded areas highlight the post-World War II adjustment, the Volcker disinflation, and the post-2008 period.

Our hypothesis is that primary-surplus adjustment occurs not when debt is high, but when servicing debt becomes fiscally costly. This observation sits uneasily with the dominant empirical framework used to evaluate fiscal sustainability. Following Bohn (1998), the literature typically assesses sustainability through a fiscal reaction function in which governments increase primary surpluses in response to rising debt ratios. This approach has become the benchmark empirical framework for studying fiscal sustainability and fiscal space (e.g. Ghosh et al. 2013, Mendoza and Ostry 2008, Checherita-Westphal and Žďárek 2017). It also underlies much of the operational debt sustainability analysis conducted by international policy institutions such as the International Monetary Fund and the European Commission. 

Yet governments do not repay debt stocks directly; they service them by paying interest. Fiscal pressure therefore stems not from the stock of outstanding liabilities per se, but from the debt-service burden it generates. Unlike debt ratios, debt-service burdens directly absorb fiscal resources and crowd out other public expenditures. They depend not only on debt levels but also on financing conditions, maturity structures, inflation, and sovereign risk premia, so that identical debt ratios can generate markedly different fiscal pressures. 

Our empirical analysis exploits this variation to distinguish the fiscal burden associated with servicing public debt from debt accumulation itself. It revisits fiscal reaction functions through this lens. The key result is that primary surpluses are more strongly associated with debt-service costs than with debt ratios. 

We formalise this idea in a simple framework in which governments react to realised financing pressures. Debt sustainability depends on whether fiscal authorities generate sufficiently strong and persistent primary surpluses when debt-service costs rise. Debt stabilisation may therefore emerge either through sharp short-run consolidations or through smaller but sustained fiscal corrections. 

Our empirical analysis combines long-run US historical evidence with cross-country panel estimation. For the US, we construct a continuous annual dataset spanning 1800–2023, allowing us to trace fiscal adjustment across major monetary, financial, and institutional regimes. We complement baseline fiscal reaction functions with dynamic time-series evidence based on vector autoregressions and state-dependent local projections to examine how fiscal responses evolve following debt and debt-service shocks. We then extend the analysis to a long-run panel of 12 advanced economies using the IMF’s Public Finances in Modern History database. 

Across specifications, periods, and estimation strategies, the evidence points in the same direction. First, primary surpluses are more closely associated with debt-service costs than with debt ratios. Primary surpluses continue to be strongly associated with debt-service costs, but the effect of debt becomes small and statistically insignificant once service is taken into account. Over 1800–2023, a 10% increase in debt-service costs is associated with an increase in the primary surplus of approximately 2.8% in the US and 1.35% in the cross-country panel. Dynamic estimates further show that debt-service shocks are followed by persistent increases in primary surpluses, whereas debt shocks generate substantially weaker responses. 

Second, fiscal responses intensify when financing conditions deteriorate. Governments respond more forcefully when the interest rate-growth differential turns unfavourable and sovereign risk premia rise. State-dependent local projections show that fiscal responses to debt-service shocks become significantly stronger under adverse financing conditions. This pattern is consistent with the intertemporal budget constraint: as financing costs rise relative to growth, stabilising debt dynamics requires stronger fiscal adjustment. 

These results suggest that debt ratios alone are not sufficient statistics for the fiscal pressures associated with public debt. 

Third, the historical evidence points to changes in fiscal adjustment over time. Before 1913, debt-service pressures remained closely tied to wartime financing. After 1913, deeper financial markets and modern monetary institutions coincided with a looser link between debt accumulation and fiscal pressure, allowing governments to sustain larger debt ratios, given favourable financing conditions. 

More broadly, our findings suggest that fiscal sustainability should be assessed not only through the evolution of debt stocks but also through the cost of servicing them. Debt-service costs represent the immediate claim of public debt on government resources and therefore provide a more direct measure of fiscal pressure than debt ratios alone, which may remain manageable under favourable financing conditions.

References

Blanchard, O (2019), “Public debt and low interest rates”, American Economic Review 109: 1197–1229.

Bohn, H (1998), “The behavior of U.S. public debt and deficits”, The Quarterly Journal of Economics 113: 949–963.

Checherita-Westphal, C and V Žďárek (2017), “Fiscal reaction function and fiscal fatigue: evidence for the euro area”, ECB Working Paper No. 2036.

Eichengreen, B J, A El-Ganainy, R Esteves and K J Mitchener (2021), In Defense of Public Debt, Oxford University Press.

Eichengreen, B, M Menuet and G Donnat (2026), “From Stocks to Flows: Debt Service and Fiscal Sustainability”, CEPR Discussion Paper 21723.

Ghosh, A R, J I Kim, E G Mendoza, J D Ostry and M S Qureshi (2013), “Fiscal fatigue, fiscal space and debt sustainability in advanced economies”, Economic Journal 123(566): F4–F30.

Mauro, P and J Zhou (2021), “r − g < 0: Can we sleep more soundly?”, IMF Economic Review 69: 197–229.

Mendoza, E G and J D Ostry (2008), “International evidence on fiscal solvency: Is fiscal policy ‘responsible’?”, Journal of Monetary Economics 55: 1081–1093.

Reinhart, C M and M B Sbrancia (2015), “The liquidation of government debt”, Economic Policy 30(82): 291–333.