Asymmetries in Fiscal Policy

Print pagePDF pageEmail page

Perry Gogas 15xx (2)

Dr. Periklis Gogas is a frequent contributor to The Business Thinker magazine. He is an Assistant Professor of Economic Analysis and international Economics, Department of International Economics and Development, Democritus University of Thrace, Greece


Ioannis PragidisDr. Ioannis Pragidis is a co-author for the Business Thinker magazine. He is a Lecturer of Economic Analysis at the  Department of Business Administration, Democritus University of Thrace, Greec

 

 

Introduction

Fiscal and monetary policies are the cornerstone of policy-making.  However, until 2000 the main bulk of empirical research was dedicated solely to the effects of monetary policy to real economic activity. In the aftermath of the global financial crisis of 2008 there is a renewed interest and a growing debate of whether governments should run fiscal stimulus packages (based on Keynesian macroeconomic theory) in order to restore previous growth rates or run an austerity program to reduce deficits and in the long-run the debt-to-GDP ratio. Recently for example, highly indebted Eurozone countries (Greece, Ireland, Portugal and Spain) are required to implement strict fiscal austerity measures in order to balance their balance sheets. In this context it is interesting to see whether and how Keynesian principles may apply.

According to economic theory, the impact of an expansionary fiscal policy on GDP can range between negative and positive values and be large or small depending on the broader macroeconomic setting. Thus, we can identify five potential sources of nonlinearities/asymmetries from the implementation of fiscal policy:

a) the phase of the business cycle: whether the economy is expanding or contracting,

b) the GDP to debt ratio: the effect of fiscal policy may differ when this ratio is small or large

c) the sign of the fiscal policy shock: positive versus negative shocks of the same instrument). For example an equal in absolute magnitude expansionary versus a contractionary tax policy.

d) the nature of the fiscal policy shock: whether an expansionary or contractionary fiscal policy is implemented through spending or revenues (taxes).

e) the magnitude of the shock: “small” and “marge” fiscal policy shocks may have asymmetric effects with respect to real GDP.

Despite the many possible sources of non-linearities and divergence of economists’ opinions, the empirical research is still too narrow and only recently has started to take into consideration the possible asymmetries reported above. Trying to reconcile economic theory with the empirical results many practitioners analyse the effects of fiscal policy on economic activity over the business cycle -case a) above- for many countries. Most of these studies report a spending multiplier around unity in boom times and 0.36 in recessions. However, the rest of the possible non-linearities have not be tested adequately yet. In order to fill this gap in empirical macroeconomics we employ several econometric methodologies in order to test for the existence of asymmetries that may be associated with the conduct of fiscal policy.  In doing so, we try to detect two types of fiscal policy asymmetries. First, whether equal in magnitude contractionary or expansionary fiscal shocks –case c) above- have the same multiplier impact on real output and second whether theoretically equal –in terms of their impact on the government budget- fiscal policy tools, such as a tax cut or an increase in government spending – case d) above-  have the same impact on output. Using quarterly data that span the period 1967Q1 to 2011Q4 for the U.S we uncover some very interesting results.

Empirical Methodology

We reveal that an increase in government spending is by far more effective in stimulating the real economy than a decrease in government revenue, i.e. a tax cut. Moreover, an interesting result is that we uncover an asymmetry in the size of the government revenue shocks. It appears that a large increase in government revenue (increase in taxes) during periods of increasing government expenditures may have a positive and lasting effect on the growth rate of real output. This may be the case since such a shock may be perceived as a strong signal of the government’s commitment to fiscal consolidation. This improves the credibility of the implemented policy and alters the public’s expectations on future economic conditions. On the other hand, during periods of decreasing government spending the decrease of taxes cannot boost the economy into a growth path. Finally, we find evidence on asymmetries in terms of the persistence of fiscal policy shocks. Increments in spending and revenue shocks, i.e. increased government spending or an increase in taxes, are more persistent as they take seven to eight quarters to fade, whereas their negative counterparts, a decrease in government spending and a tax cut, take only three to four quarters.

Empirical Results

We reveal that an increase in government spending is by far more effective in stimulating the real economy than a decrease in government revenue, i.e. a tax cut. Moreover, an interesting result is that we uncover an asymmetry in the size of the government revenue shocks. It appears that a large increase in government revenue (increase in taxes) during periods of increasing government expenditures may have a positive and lasting effect on the growth rate of real output. This may be the case since such a shock may be perceived as a strong signal of the government’s commitment to fiscal consolidation. This improves the credibility of the implemented policy and alters the public’s expectations on future economic conditions. On the other hand, during periods of decreasing government spending the decrease of taxes cannot boost the economy into a growth path. Finally, we find evidence on asymmetries in terms of the persistence of fiscal policy shocks. Increments in spending and revenue shocks, i.e. increased government spending or an increase in taxes, are more persistent as they take seven to eight quarters to fade, whereas their negative counterparts, a decrease in government spending and a tax cut, take only three to four quarters.

Policy Implications

The policy implications of the empirical evidence we found with the two alternative empirical methodologies are very important. To summarize them here:

a)  It appears that the most effective fiscal policy instrument in terms of its impact on real private GNP is government spending. The empirical evidence provided in our analysis shows that government spending has the sign and impact predicted by standard Keynesian macroeconomics.

b) On the other hand, government revenue shocks, i.e. through taxation, are found to have a much smaller (approximately four times) multiplier than equal in size spending shocks.

c) Finally, our results highlight the importance of the credibility of the fiscal policy. It appears that an increase in government revenue sends a strong signal to the public that the government is committed to fiscal consolidation and is moving towards prudency. This alters the public’s expectations about future economic activity increasing confidence that may a positive impact on real private GNP.

According to these results a policy maker would be far more successful in implementing fiscal policy through government spending than revenue. Moreover, to stimulate the economy, an expansionary fiscal policy through an increase in government spending appears that it would be the most efficient instrument of choice for the policy maker. An equal in size stimulus package through lower taxes would be marginally successful as the corresponding multipliers reveal.

References

  • James P. Cover, (1992). Asymmetric Effects of Positive and Negative Money Supply Shocks.  Quarterly Journal of Economics, 107(4), 1261- 1282.
  • Robert Lucas, (1972). “Expectations and the Neutrality of Money”. Journal of Economic Theory 4 (2): 103–124. doi:10.1016/0022-0531(72)90142-1.
  • John F. Muth. (1961). “Rational Expectations and the Theory of Price Movements”, Econometrica 29, pp. 315–335.

 

Leave a Reply

Your email address will not be published. Required fields are marked *