Two papers I saw presented this morning took very different approaches to reach a similar conclusion—inside a currency union there can be big fiscal policy multipliers. In other words, large positive effects of running budget deficits and negative effects of fiscal consolidation.
First Olivier Blanchard and Daniel Leigh presented "Growth Forecast Errors and Fiscal Multipliers" which takes advantage of the fact that economic forecasters typically make mistakes. Specifically, they looked at the IMF's forecasts for growth for different European countries and asked how far off the mark the IMF got. It turned out that the scale of the error was both variable, and systematically correlated with the amount of deficit reduction the countries did. The conclusion is simple—previously the IMF had been underestimating the "multiplier" effect whereby deficits help bolster a depressed economy.
Daniel Shoag's "Using State Level Pension Shocks to Estimate Fiscal Multipliers" is very different but reaches a very similar conclusion. He looks at the fact that different US state pension funds lost different amounts during the big stock market crash that coincided with the recession. That's a big shock to state fiscal capacity that's independent of the local economic situation. But it turns out that a bad hit to your pension fund is correlated with bad economic performance, seemingly because there's a state spending multiplier greater than one.
In both cases, the policy implications are suggestive but not conclusive. The worst-hit European countries generally didn't have the option of doing fiscal stimulus. And states obviously can't choose to get well-timed stock market windfalls. But in both cases the implication of the research is that timely bailouts could be very helpful to the recipients.