Fiscal Multipliers in Advanced and Developing Countries: Evidence from Military Spending 
To study the effects of fiscal policy in different economic environments, the authors compile a novel dataset containing output, government spending, military spending, unemployment rates, trade shares, and many other variables for 129 advanced and developing countries during the period 1988–2013. They estimate fiscal multipliers by using military-spending shocks as an instrument for government spending. The data enable them to differentiate between the multipliers in different states of the economy and the business cycle, in different exchange-rate regimes, and at different degrees of openness to trade, as well as in advanced and developing countries.
Key Findings
- When the data for advanced and developing countries are combined, the one-year government-spending multiplier is estimated to be in the 0.75–0.85 range. For three years after the shock, the cumulative multipliers are significantly different from zero but not from one.
- When advanced and developing countries are considered separately, the baseline multiplier in advanced economies is more than twice as large as it is in developing countries.
- Government spending is more effective in recessions than in expansions. The one-year response of output to a unit shock in government spending is 1.7 in recessions and 0.3 in expansions, with statistically significant differences at horizons of as long as four years. Multipliers are also larger during episodes of slack (when the unemployment rate is relatively high) than when unemployment is low.
- Countries that peg the value of their currency are associated with larger multipliers than those for countries that let the exchange rate float. The multiplier is also larger for closed economies than for open economies.
Implications
These findings have important implications for policymakers because the decisions on whether and how to use fiscal policy should account for particular economic circumstances and environments. Specifically, the authors indicate the various economic conditions, such as recessions and passive monetary policy, for which expansionary fiscal policy can be a particularly effective stabilization tool.
This paper’s results also have implications for the identification of government-spending shocks. The authors find that the military-shocks approach produces multipliers that are comparable to those from other methods and are robust to tests involving various controls and samples that could potentially violate the identifying restrictions.
Abstract
Using novel data on military spending for 129 countries in the period 1988–2013, this paper provides new evidence on the effects of government spending on output in advanced and developing countries. Identifying government-spending shocks with an exogenous variation in military spending, we estimate one-year fiscal multipliers in the range 0.75–0.85. The cumulative multipliers remain significantly different from zero within three years after the shock. We find substantial heterogeneity in the multipliers across groups of countries. We then explore three potential sources leading to heterogeneous effects of fiscal policy: the state of the economy, openness to trade, and the exchange-rate regime. We find that the multipliers are especially large in recessions, in closed economies, and under a fixed exchange rate. We also discuss other potential reasons for heterogeneous effects of fiscal policy, such as its implementation and coordination with the monetary authority.