by Andreas Hoffmann*
There has been a long-lasting debate about the effects of fiscal stimulus in times of crisis. Proponents of government stimulus argue that the fiscal multiplier is quantitatively more significant than the possible negative effects. On the other hand, free market economists are in doubt of these effects and fear possible crowding-out, high future costs or malinvestment.
While Keynesians and free market economists discuss effects on welfare, allocation and sustainability of government investment or consumption, Chicago’s Harald Uhlig recently presented an interesting paper at NYU (“Fiscal Stimulus and Distortionary Taxation,” coauthored with Thorsten Drautzburg – an earlier version is here), in which he tries to quantify the fiscal multiplier and its growth effects in the short, medium and long-run.
Therefore he calibrates a neoclassical as well as a new-Keynesian general equilibrium model to US data since the post-WWII period and compares the effects, under the assumption that the money spent has to be recollected via taxes in later periods. Most interestingly, he also takes into account the effects of a stimulus when interest rates reach the zero-bound.
Two findings are interesting: First, both models show a positive effect in the short-run. However, in the long-run the accumulated effect of the fiscal stimulus on growth is negative in both models. The assumption of sticky prices in a New-Keynesian model seems to lower the multiplier, which is an interesting result. In general, the fiscal multiplier is below 1. This is also true for fiscal stimulus when interest rates are zero.
Second, the longer rates are low, the lower the negative effect of government spending on medium and long-term growth and consumption. The pro-market economist concludes that we should have an eye on the medium and long-term effects of government spending (as opposed to only short-run effects).
Uhlig explicitly does not go into a welfare analysis, but solely focuses on the fiscal multiplier and its growth effects in this forthcoming paper. Thus he does not have to distinguish between wanted and unwanted consumption or good and mal-investment. Further he also does not have to address the institutional implications that arise from the model. But what is the intuition one gets from his results for future government and central bank policies (that seem to trust this kind of model)?
Even though it may seem obvious, the models show that there is a great incentive for government to put pressure on the central bank to keep rates low, in order to improve the fiscal multiplier.
Further the effects largely depend on whether the debt is recollected via taxes. Thus, as following previous recessions, the incentive for anti-cyclical policies is small (obviously this is not in his model). Instead, it is likely that macroeconomic policy making will remain expansionary in the bust period and less restrictive in the following boom. This asymmetry is likely to increase the debt level and constitute a low interest rate policy to monetize some of the debt (as proposed by Blanchard). This is especially worrisome, as long periods of easy money may induce new risk-taking, mal-investment and new bubbles.
*Andreas Hoffmann is visiting the Department of Economics, New York University from the Institute for Economic Policy, University of Leipzig during the current academic year. He is a Bradley Fellow and a Fellow of the Friedrich Naumann Foundation. Some of his work can be found here.