The Fiscal Multiplier in Crisis: Revisiting Austerity's Legacy in Post-2008 Europe

The European debt crisis of 2010–2015 represented one of the most consequential experiments in fiscal policy in postwar economic history. Faced with rapidly rising sovereign debt spreads, peripheral eurozone economies—Greece, Portugal, Ireland, and Spain foremost among them—enacted sweeping consolidation programs under pressure from the European Commission, the European Central Bank, and the International Monetary Fund. The intellectual foundation of these programs rested, in significant part, on a contested empirical claim: that fiscal multipliers in open, high-debt economies are small, and that credible consolidation could be nearly self-financing through confidence and expectations effects.

A decade on, the evidence against this view is substantial. The IMF's own research division issued a notable mea culpa in 2013 acknowledging that multiplier estimates used to calibrate Greek conditionality programs were materially too low. Yet the full distributional and long-run output consequences of this miscalibration remain underexplored, particularly across the heterogeneous institutional contexts of the eurozone. This paper addresses that gap.

1. Theoretical Framework

The fiscal multiplier—defined as the ratio of the change in output to the change in government spending or taxes that caused it—is not a structural parameter. It varies systematically with the state of the macroeconomy, the degree of monetary policy accommodation, the openness of the economy, and the composition of the fiscal adjustment. The now-extensive theoretical literature, drawing on both New Keynesian and heterogeneous-agent models, predicts that multipliers are substantially larger when: (i) monetary policy is constrained at the effective lower bound; (ii) financial frictions are severe; and (iii) a high share of households are liquidity-constrained.

All three conditions applied with particular force to peripheral eurozone economies after 2010. Membership in the currency union precluded independent monetary accommodation and exchange rate adjustment. Banking systems in Greece, Portugal, and Ireland were severely impaired following the financial crisis. And income distributions in these countries exhibited relatively high shares of households with limited liquid savings.

2. Data and Methodology

Our primary dataset covers 18 eurozone economies at annual frequency from 2008 to 2022, drawing on Eurostat national accounts, IMF World Economic Outlook data, and the European Commission's AMECO database. The identification challenge in estimating fiscal multipliers is well known: fiscal policy responds endogenously to economic conditions, creating reverse causality. We address this using two complementary strategies.

First, following Blanchard and Perotti (2002), we exploit the institutional structure of automatic stabilizers and the lag in discretionary policy responses to identify exogenous fiscal shocks. Second, and more novel, we construct a Bartik-style instrument using variation in IMF program conditionality requirements interacted with pre-crisis fiscal positions, which we argue provides a source of quasi-exogenous variation in consolidation intensity across countries and years.

"Our central estimate of the fiscal multiplier during consolidation episodes is 1.7, substantially above the values of 0.5 to 1.0 implicitly assumed in the official program design documents."

3. Results

Our central estimate of the fiscal multiplier during consolidation episodes—pooled across countries and years, instrumented for endogeneity—is 1.7, substantially above the values of 0.5 to 1.0 implicitly assumed in the official program design documents recovered through freedom-of-information requests and academic archives. This estimate is robust to a battery of specification checks, including alternative lag structures, the exclusion of individual countries, and the use of cyclically adjusted primary balance measures.

Crucially, we find substantial heterogeneity in multipliers across fiscal adjustment composition. Consolidations weighted toward public investment cuts exhibit multipliers of 2.1 on average, compared with 1.4 for wage bill adjustments and 1.2 for transfer reductions. This ordering is consistent with the recent theoretical literature on composition effects but has not, to our knowledge, been documented at this level of granularity for the post-2008 European episode.

4. Distributional Consequences

Our distributional analysis combines the aggregate multiplier estimates with microdata from Eurostat's EU-SILC survey to trace the incidence of output losses across the income distribution. We find that the lowest income quintile bore approximately 2.3 times the per-capita output loss of the median household in high-consolidation economies, a ratio that persisted for at least five years post-adjustment. This pattern is consistent with the prediction of models in which automatic stabilizers and public services function as implicit insurance for liquidity-constrained households.

5. Conclusion

The evidence assembled here suggests that the intellectual consensus underpinning the European austerity programs of 2010–2015 was not merely empirically contestable but consequentially wrong in ways that bore disproportionately on the most economically vulnerable citizens of the affected economies. This has implications both for the design of future adjustment programs and for the institutional frameworks—particularly within the eurozone—that constrain the policy responses available to member states during downturns.

More broadly, our findings reinforce the case for state-dependence in fiscal policy analysis: multiplier estimates derived from tranquil periods or from economies with full monetary policy autonomy are poor guides to policy design in conditions of financial stress and monetary constraint. Policymakers and the international institutions that advise them would do well to maintain a richer toolkit of multiplier estimates calibrated to prevailing conditions.