Money Demand and the Effects of Fiscal Policies: Comment

Abstract

Mankiw and Summers (1986) argue on theoretical and empirical grounds for consumption as the appropriate scale variable in the money demand function. This serves as the primary basis for their provocative result that personal income tax cuts may be contractionary on equilibrium real income in a short-run, IS-LM model. Moreover, they cite evidence on the interest elasticity of investment demand and on the interest and income elasticities of money demand for the U.S. economy that leads them to conclude that tax cuts are, indeed, contractionary in the United States. In this paper, theoretical and empirical arguments are combined in order to show that the tax cut is likely to be expansionary on real income even though consumption is taken as the scale variable in the money demand function. In section 2, a theoretical model of the interaction of taxes and the opportunity cost of holding money is sketched, with the implication that a cut in personal income taxes, ceteris paribus, reduces the excess demand for money, whereas a cut in corporate income tax rates likely has no effect (although it possibly reduces money demand in one case examined). This is shown in section 3 to generalize the Mankiw-Summers elasticity condition determining the sign of the (personal) tax cut multiplier. In section 4, empirical evidence is brought to bear on this elasticity condition. It is seen that a crucial magnitude is the short-run interest elasticity of investment demand (and, hence, the slope of the IS curve); empirical estimates of this elasticity depend on the nature of the frictions that prevent an immediate adjustment to the new desired capital stock following a change in interest rates. Such frictions include delivery lags and the putty-clay nature of certain types of capital. Upon evaluation of the empirical evidence, it is concluded that a cut in income tax rates is, indeed, expansionary on aggregate demand. Finally, in section 5, some previously omitted factors that may affect outcomes in the short run are briefly discussed; these include supply-side and expectational effects.