Fiskalpolitik versus Geldpolitik

Abstract

Fiscal Policy versus Monetary Policy On the question of their relative importance

The hypotheses of Andersen and Jordan that the effects of monetary policy make themselves felt, not only more strongly and more rapidly, but also more reliably than those of fiscal policy could not be confirmed for the FRG in 1960 - 1970. In contrast to the assertions of Andersen and Jordan, in the FRG the effects of fiscal policy on economic activity have been roughly equally strong and at the same time both more rapid (shorter lag) and more reliable (more predictable) than the effects of monetary policy. Furthermore, government expenditures do not merely displace and supplant private expenditures, but result on balance in expansive effects. The study proceeds from the argument that the effects of fiscal policy have an impact in less than two years, while the effects of monetary policy need no more than two and a half years. From the (greater) dependence of the FRG on foreign trade, the author derives hitherto neglected theoretical points of departure for explaining the deviation of results from those of Andersen and Jordan. The method applied is regression analysıs of an explanatory equation for the national product, in which not only immediate, but also lagged effects of both monetary and fiscal policy are possible a priori. For econometric estimation of the effects distributed over time, the so-called Almon lag technique is used without, and for the purpose of comparison also with, end-point restrictions: for fourth-degree polynomials. Compared with the procedure of Andersen and Jordan, there is a series of methodological differences, which are listed in a systematic compilation. The use of the initial stimulus instead of the fullemployment budget surplus as an indicator of fiscal policy impetus, and of the extended money base instead of the quantity of money as an indicator of monetary policy impetus deserves emphasis. To allow for the FRG’s marked dependence on foreign trade, exports were also used as a measure of impulses from abroad. For several reasons, growth rates of series of observations should be given preference over simple changes (first-order differences) in such series. The difference in the results with and without end-point restrictiofnosr polynomials is negligible. These conclusions must be considered in conjunction with the results obtained by Waud for the U.S.A. and by Artis and Nobay. for Britain, which similarly do not confirm the monetaristic hypotheses of the type investigated ‚in this study. Hence, economic-policy recommendations for the FRG must be based on empirical findings which contradict the monetaristic hypotheses.