VoxEU Column Macroeconomic policy

Fiscal stimulus in times of high public debt: Reconsidering multipliers and twin deficits

Fiscal stimulus can either be a boon for an economy or its bane. This column discusses new empirical research on how the effects of fiscal stimuli interact with public debt-to-GDP ratios. Using data on real GDP, private investment and the trade balance, evidence suggests that the cumulative effect on real GDP is positive at moderate debt-to-GDP ratios but turns negative as the ratio increases. The evidence also suggests that the cumulative trade-balance effect is negative at first but switches sign at higher degrees of indebtedness.

The stimulus impact of a government-spending shocks varies with the economic environment (Ilzetzki, Mendoza, and Végh 2013). One important determinant is the level of government indebtedness – particularly when the nation’s fiscal sustainability may be in doubt. The economic logic of this is straightforward.

Fiscal multipliers and twin deficits: How it all fits together

The national income accounting identities tell us that current account equals the gap between total domestic saving and investment. The public-sector deficit measures public dis-saving. The reaction of the ‘twin deficits’ (government and current account) to a fiscal shock is thus linked to how private saving and investment choice react to the shock.

In particular, private agents’ behaviour depend, inter alia, on how fiscal shocks are perceived. These perceptions will normally depend upon the economic environment – especially debt sustainability.

  • If expansion of the budget deficit strongly stimulates private consumption but does little to discourage private investment, the current account will swing towards negative – resulting in a twin deficit situation.

This is the traditional Keynesian paradigm (see Corsetti and Müller 2006).

However, the current-account deficit can move in a divergent direction.

  • If the fiscal situation is viewed as unsustainable, a bigger fiscal deficit may dampen private consumption by boosting precautionary savings. If the shock also dampens private investment sufficiently, the current account deficit will swing towards positive just as the public deficit is swinging towards negative (i.e. is getting bigger).

In such situations, the external balance and the public balance may diverge (see Kim and Roubini 2008, Nickel and Vansteenkiste 2008).

This is one clear illustration of how the impact of fiscal shocks can be non-monotonic and how the sign of the reaction can depend upon measures of a nation’s debt sustainability. A fiscal shock’s impact on external balance and GDP (i.e. the fiscal multiplier) can change in non-linear ways. It can shift from being ‘Keynesian’ or expansionary at low debt-to-GDP ratios to contractionary as government indebtedness rises.

This shift may happen continuously without any sort of threshold. For instance, in a model with credit-constrained consumers, Perotti (1999) shows that the effect of government expenditure shocks on aggregate consumption may depend on initial conditions such as public indebtedness. The non-linearity arises since credit-constrained and unconstrained individuals perceive wealth effects differently. Those that are credit-constrained only experience the positive income effect; those that are unconstrained experience also experience a wealth effect from future anticipated changes in fiscal policy. As debt levels rise, the latter will dominate the overall effect, so additional government spending tends to be contractionary.

Measuring the effects of fiscal stimuli with varying debt-to-GDP ratios

In recent research, we empirically investigate the effects of fiscal stimuli on macroeconomic variables such as real GDP, private investment and the trade balance (Nickel and Tudyka 2013). In particular, we explicitly allow the behaviour of all variables included in our specification to vary with the degree of indebtedness in response to a fiscal impulse. This is important since it allows government debt to have direct effects as well as indirect effects through the other variables in our specification.

To this end, we use an interacted panel vector autoregression as in Towbin and Weber (2011) which is estimated in Bayesian fashion for 17 European countries. The baseline specification includes government consumption, real GDP, real private investment, the debt-to-GDP ratio and the trade balance as endogenous variables (see Nickel and Tudyka 2013 for more details).

Figure 1 reports cumulative impulse responses to a fiscal shock, focusing on the flow variables for government consumption, real GDP, private investment and the trade balance. The shock is a rise in government consumption equal to one percentage point of GDP.

  • The columns represent our assumption as to the initial public debt-to-GDP ratio (we show results for the ratio equal to 0.4, 0.67 and 1.09).
  • The rows show the variables of interest, namely government consumption (‘G’) in the first row, real GDP (‘Y’) in the second, private investment (‘I’) in the third and the trade balance (‘TB’) in the final row.
  • The solid line corresponds to the median (50th percentile) impulse response and the dotted lines are the 16th and 84th percentiles of the respective posterior distribution as (one standard deviation) probability bands.
  • In all plots, horizontal axes indicate periods after the shock and vertical axes are in percentage shares of GDP.

Figure 1. Cumulative impulse responses to a one percentage point of GDP shock to government consumption at various interaction levels

To summarise our findings, we focus on three points:

  • First, the impact of government consumption is highly dependent on the initial debt ratio.
  • Second, it becomes far more self-reversing pattern at higher debt ratios (beyond approximately 60%).

This can be seen in the increasingly hump-shaped cumulative impulse responses (Chung and Leeper 2007, Corsetti et al. 2012). The hump indicates that the shock initially provides stimulus but this can turn into a contractionary force as time passes.

  • Third, while the overall effect on real GDP is expansive even at very long horizons, the higher the debt ratio, the less this is the case.

Eventually, at debt ratios beyond 90%, the overall effect on real GDP becomes significantly negative (see Ilzetzki et al. 2010).

The cumulative response of private investment turns increasingly negative as government indebtedness increases. In particular, the cumulative effect on private investment turns negative for the first time at debt ratios of approximately 57%. Moreover, at low debt-to-GDP (less than 40%) ratios the overall effect on the trade balance is negative (see Corsetti and Müller 2006). However, at very high debt-to-GDP ratios (more than 85%) the cumulative effect on the trade balance becomes significantly positive at longer horizons after the shock (see Kim and Roubini 2008).

Concluding remarks: Mind the debt!

Our results indicate that the private sector increasingly internalises the government budget constraint as the degree of indebtedness rises. This qualifies debt as an important variable to watch when thinking about the impact of fiscal stimulus.

  • In particular, while fiscal stimuli exert expansionary effects on macroeconomic activity at low debt-to-GDP ratios, the overall effect on real GDP becomes less positive or even negative points at higher debt-to-GDP ratios.
  • Depending on which debt scenario we examine, our results accommodate the inconclusive results of previous studies, which either document positively correlated or divergent behaviour of government activity and the trade balance or the current account.

In sum, our findings lead us to conclude that the contradictory findings of previous studies may be the result of estimation within a static debt regime when indeed the debt regime is dynamic.

A consequence of this may be, for instance, that when estimating responses over the entire range of debt ratios, the effect on the current account may well show up as insignificant because positive and negative effects cancel out. Consequently, the effectiveness of fiscal stimuli to boost economic activity or resolve external imbalances may fade with increasing debt-ratios.1 In fact they may even be counterproductive. From a policy perspective, these results therefore lend additional support to increased prudence at high public-debt ratios.

Editor’s note: The views expressed are the authors’ and do not necessarily reflect those of the ECB.


Chung, H and E M Leeper (2007), “What has financed government debt?”, Working Paper 13425, National Bureau of Economic Research.

Corsetti, G, A Meier, and G Müller (2012), “Fiscal stimulus with spending reversals”, The Review of Economics and Statistics 94(4), 878–895.

Corsetti, G and G J Müller (2006), “Twin deficits: Squaring theory, evidence and common sense”, Economic Policy 21(48), 597–638.

Giavazzi, F and M Pagano (1990), “Can severe fiscal contractions be expansionary? Tales of two small European countries”, NBER Macroeconomics Annual 5, 75–111.

Ilzetzki, E, E G Mendoza, and C A Végh (2013), “How big (small?) are fiscal multipliers?”, Journal of Monetary Economics 60(2), March, 239-254.

Kim, S and N Roubini (2008), “Twin deficit or twin divergence? Fiscal policy, current account, and real exchange rate in the US”, Journal of International Economics 74(2), 362–383.

Nickel, C and A Tudyka (2013), “Fiscal stimulus in times of high debt: Reconsidering multipliers and twin deficits”, Working Paper Series 1513, European Central Bank.

Nickel, C and I Vansteenkiste (2008), “Fiscal policies, the current account and Ricardian equivalence”, Working Paper Series 935, European Central Bank.

Perotti, R (1999), “Fiscal policy in good times and bad”, The Quarterly Journal of Economics 114(4), 1399–1436.

Towbin, P and S Weber (2011), “Limits of floating exchange rates: The role of foreign currency debt and import structure”, IMF Working Papers 11/42, International Monetary Fund.

1 Due to symmetry of the results, in line with the literature on expansionary fiscal consolidations, contractionary policy measures may exert expansionary effects on medium-run economic activity when debt-to-GDP ratios are high (Giavazzi and Pagano 1990).

3,150 Reads