The Government Spending Multiplier: A Survey of Empirical Literature

What do empirical studies tell us about the effect of government spending on economic output?

By Jack Salmon of Mercatus. Excerpts:

"This literature review examines the wide-ranging estimates of the fiscal multiplier—the effect of government spending on economic output. Despite extensive empirical exploration, estimates of the fiscal multiplier span from –3.00 to 3.00 and are influenced by model assumptions, theoretical innovations, state-dependent factors, and dataset choices. This review discusses methodological advancements, including state-dependent estimates, and explores how the multiplier varies under economic slack, at the zero lower bound (ZLB), and with large public debt burdens. The review finds that multipliers are generally within the range of 0.50 to 0.90, with higher estimates during economic slack and at the ZLB and lower estimates for regimes with high public-debt ratios. The degree of state dependence is more modest than suggested by earlier research. Robust evidence that ZLB multipliers are larger than 1.00 is scarce. The paper concludes by calling for more dynamic models and cross-country comparisons to lead to a better understanding of the nuanced effects of fiscal stimulus." 

"In the aggregate, government spending multipliers broadly fall within the range 0.50 to 0.90, in line with the range of estimates offered by Ramey (2019) and consistent with recent meta-analyses of large datasets.4 Some studies find higher multipliers (close to or above 1.00) during periods of economic slack, but the evidence is mixed, with many results suggesting only modest differences compared with expansions. At the ZLB, where monetary policy is constrained, multipliers are often found to be slightly higher, but robust findings indicate that they remain below or near unity. High public debt generally reduces fiscal multipliers, with some studies showing negative long-term effects as debt levels rise significantly. Although some theoretical models predict higher fiscal multipliers for government investment spending compared with consumption, robust empirical data often show minimal differences, with investment multipliers frequently constrained by significant crowding out of private investment. Overall, the degree of state dependence is more modest than suggested by earlier research. The literature also underscores the importance of recognizing distinct phases of multiplier effects—impact, peak, and long term. Although short-term impacts may boost output, long-term effects often diminish or turn negative due to reduced private investment and consumption, emphasizing the role of anticipatory effects and private-sector responses."