In spite of the fact that the consumption relationship plays the central role in the Keynesian multiplier analysis of the effects of fiscal policy, economists are not in agreement about the effects government deficits have on consumer expenditures. In contrast to the popular economic policy descriptions, recent theoretical work indicates that, through adjustments in private intergenerational transfers, current government surpluses, which imply lower future tax liabilities, could have an expansionary effect on current private consumption demand. (See, for example, Barro, 1974.) In the 1968 experiment relating to this problem, the tax surcharge, which had little effect on consumption and aggregate demand, was interpreted as giving evidence in favor of the monetarists' position and against the traditional Keynesian view of policy prescriptions.-; The monetarists claimed victory based upon this experiment because national income rose with the rising money supply, and the fiscalists pleaded nolo contendere ex post since the tax surcharge was temporary in nature and, therefore, had little effect on permanent income and thus, little effect on aggregate consumption demand (Eisner, 1969). However, some macro theorists now would explain the rise in the aggregate marginal propensity to consume in this period as resulting from the lower future taxes caused by the higher current tax surcharge. Thus, rather than being evidence on the Keynes versus classical theories of the income determination controversy, the 1968 tax experiment may be evidence on the private sector's ability to perceive the effects on future tax liabilities of current fiscal actions. For, if consumers do foresee these effects, current tax increases or decreases (whether viewed by policy makers as being temporary or permanent) will have little effect on aggregate demand. As a consequence, and with the large volume of continuing discussion of the role of fiscal policy as a stabilizing tool, a closer examination of consumption expenditures in the context of changing government fiscal policies is required. II. Theoretical Issues Reviewed

Economists are often puzzled as to why profit-maximizing firms buy professional forecasts when statistics such as the root-mean-squared error or the mean absolute error often indicate that a naive model will forecast about as well. This paper argues that the reason is that these traditional summary statistics may not be closely related to a forecast's profits. Using profit measures, the authors find only very weak relationships between such summary error statistics and forecast value. If these results are robust, then least-squares regression analysis may not be appropriate for many studies of economic behavior.

Economists are often puzzled as to why profit-maximizing firms buy professional forecasts when statistics such as the root-mean-squared error or the mean absolute error often indicate that a naive model will forecast about as well. This paper argues that the reason is that these traditional summary statistics may not be closely related to a forecast's profits. Using profit measures, we find only very weak relationships between such summary error statistics and forecast value. If these results are robust, then least-squares regression analysis may not be appropriate for many studies of economic behavior.