Recent fiscal projections for the US forecast a rising debt-GDP ratio which exceeds the peak at the end of World War II. This column shows that over the last 20 years, there is no longer evidence of a stabilising fiscal feedback, where higher projected deficits lead to more deficit reduction. It argues that the current fiscal trajectory is unstable, and adverse fiscal shocks would only worsen the prospects. However, if a gradual fiscal feedback were to be re-established, then the trajectory would be stable. This feedback would imply deficit reductions in the range of 0.5% to 1.1% of GDP over the next decade.
America’s current fiscal path is unsustainable. Under current law, the most recent projections by the Congressional Budget Office (CBO 2025) forecast primary deficits persistently above 2% of GDP and a debt-GDP ratio (net of US government holdings) approaching 120% by the end of 2035, which would be far above the ratio’s peak at the end of WWII. Moreover, these forecasts assume that the massive Trump tax cuts adopted in 2017 will soon expire, as current law dictates, even though a process to extend most or all of the tax cuts is currently underway. Even more problematic, the projections ignore uncertainty, leaving out the budget consequences of a recession or, even worse, a crisis of the scale experienced twice in the last two decades, due to the Great Recession and the Covid-19 pandemic.
But projections based on current law also assume that the government will take no actions to stabilise the national debt trajectory. Any fiscal path is sustainable if future fiscal policy responds sufficiently to high deficits. Thus, a judgement about whether US fiscal policy is sustainable requires both accounting for the risks along the fiscal path and the strength of government responses. In a classic paper on the subject, Bohn (1998) found for the US that higher levels of the debt-GDP ratio led to deficit reductions, indicating a stable fiscal path. More recently, Auerbach (2003), estimating the effects of projected deficits and the GDP gap on legislative changes to the deficit, found that legislative changes responded strongly to projected deficits over the period 1984-2003, behaviour consistent with Bohn’s results. Thus, one might conclude that the current Congressional Budget Office projections, though concerning, do not presage a fiscal crisis.
However, US fiscal policy has changed since the periods studied by Bohn and Auerbach. Here, we rely on changes in actual policy responses, rather than in formal budget rules, an approach that aligns with recent work finding that budget rules that seek to contain fiscal policy lack credibility and enforcement (Potrafke 2023). Examples include the EU’s Stability and Growth Pact (Blanchard et al. 2021, Buti et al. 2022) and the US Gramm-Rudman-Hollings deficit targets of the mid-1980s and targets for discretionary spending and so-called ‘PAYGO’ rules for taxes and entitlement spending beginning in the 1990s (Auerbach 2008).
As shown in our recent paper, Auerbach and Yagan (2025), there is no longer any evidence of stabilising fiscal feedback. Figure 1 shows residuals from a regression of the legislated primary deficit reduction on the lagged GDP gap versus residuals from a regression of the projected deficit on the lagged GDP gap, having added back the respective mean to each. The green circles in the figure plot the observations used in Auerbach’s original analysis, with the green regression line showing the strong positive relationship found there – higher projected deficits lead to more deficit reduction. By contrast, the orange squares and regression line plot the observations and relationship for the period 2004-2024, showing not only that projected deficits were higher during this period, but also that the positive relationship between projected deficits and deficit reduction has completely disappeared. In short, policy has moved toward higher deficits and away from reacting to them. Whatever the cause of this change in behaviour, it has important implications for fiscal stabilisation.
Figure 1 Legislative responses to projected deficits
Modelling the US fiscal trajectory
Considering the disappearance of fiscal feedback, how likely is it that the US fiscal path is unsustainable? One cannot answer this question without accounting for the shocks that may hit the US economy. We therefore model the fiscal trajectory as being consistent with current CBO projections, but subject in addition to two stochastic shocks. One involves the difference between the interest rate and the economic growth rate. As emphasised in recent contributions (e.g. Blanchard 2019), a low interest rate, in particular an interest rate that is smaller than the rate of economic growth, is a force for fiscal stabilisation, as lower debt service is more easily met by a growing economy. But even if the interest rate is projected to fall below the economic growth rate, i.e., if ‘r – g’ is negative, history shows us that there may be prolonged periods when this is not so. A more positive gap between interest rates and growth rates than is currently forecast would worsen the chances of fiscal stability.
A second shock which we account for is best illustrated by a graph showing the recent path of the US debt-GDP ratio, in Figure 2. The series exhibits relative stability except during two crises: first during the Great Recession, when the debt-GDP ratio doubled, from 35% to 70% between 2007 and 2012, and during the COVID-19 pandemic and its aftermath, with the debt-GDP ratio rising by 20 percentage points between 2019 and 2020. These jumps reflect the combination of automatic stabilisers and discretionary fiscal actions, but the observed pattern suggests that one can think of the shocks to the budget as taking the form of large, infrequent jumps that are asymmetric in nature. That is, during this period, there were no offsetting declines in the debt-GDP ratio outside of the episodes when the debt-GDP ratio jumped.
Figure 2 US debt-GDP ratio since 2000
Incorporating both types of shocks in our projections of the fiscal path for the US, we ask how likely it is that the fiscal path will be unsustainable and consider how our conclusions would change if fiscal feedback of the strength observed in earlier eras were practised. As it is difficult to know exactly when a fiscal trajectory becomes unsustainable, we use as a benchmark whether the debt-GDP ratio exceeds 250% at the end of the next 100 years and classify the fiscal path as unsustainable if it exceeds this limit in more than 5% of the trajectories, given our estimates of the distributions of the two types of fiscal shocks.
How strong does fiscal feedback have to be?
Even without shocks, the fiscal trajectory based on CBO projections is unstable based on our criterion – the debt-GDP ratio would exceed 250% after one hundred years. With the possibility of adverse fiscal shocks, prospects are worse. Surprisingly, though, we also find that, were fiscal feedback of the strength estimated by Bohn and Auerbach to be re-established, then the current fiscal trajectory is stable, at least for the relatively lax debt-GDP ratio ceiling of 250%. If the US government were to resume a policy of gradual fiscal feedback sufficient to put the budget on a stable trajectory, what would this imply for legislative changes in the short term? Based on our estimates of the minimum feedback required, deficit reductions in the range of 0.5% to 1.1% of GDP would be needed over the next decade, relative to the baseline that also excludes extending the 2017 tax cuts.
Is there an alternative to such steady fiscal feedback? Some might argue that, rather than forsaking fiscal feedback in recent decades, US policymakers have simply adopted a different approach of enacting occasional large fiscal consolidations, when necessary, the implication being that no action has been necessary in recent years. How feasible would such a policy approach be? Based on the experience of the past several decades, a fiscal consolidation of 1.5% of GDP – a permanent primary deficit reduction of this size – is about as large a change as one can expect at once. Suppose that such consolidations could occur periodically whenever the US fiscal situation reached a troubling point, which, based on the recommendation of Furman and Summers (2020), we take to be when real debt service exceeds 2% of GDP. How frequently would the government have to undertake such ‘sudden’ consolidations to achieve fiscal stability? We find that a policy of sudden fiscal feedback would produce a stable trajectory only if consolidations could occur as frequently as every 12 years, which seems inconsistent with the idea of a policy based on infrequent action. In short, there seems to be little alternative to the resumption of a more responsive fiscal policy for a stable fiscal path to be achieved.
source : cepr.org
