The Consumption Function Modern Stotisticd Evidence

While the above generalizations give us some guides to consumption behavior, they are too rough to be satisfactory for national forecasting purposes—for example, when we want to predict the probable effects of a tax cut or tax increase. Thus, modern econometricians (economists who are interested in both economic theory and the statistical testing of their theories) have devoted a great deal of attention to analyzing precisely the empirical relationship between the various causes listed above and consumption spending in the past.

A number of modern studies agree that we can generally get a good approximation of the likely change in consumption for the quarter or year ahead if we can predict changes in disposable income and changes in wealth (or perhaps money) held by consumers. As is suggested above, this prediction seems to improve if we substitute for current disposable income a somewhat more sophisticated measure of “permanent” or “lifecycle” income. Rough approximations of these concepts have been made by assuming that people will take an average of their income over the past several years as a predictor of their permanent or life-cycle income, once adjustments are made for their age, family status, and the like.

Another approach is to break down consumer spending into a number of major components—spending on services, on nondurable goods, on automobiles, on other durable goods, and the like. One well-known econometric model, for example, relates changes in spending on nondurable goods to disposable income, the level of spending on nondurables in the preceding period, and holdings of liquid assets in the preceding period, plus a “catch-all” factor for other variables. Note that the inclusion of spending on nondurables in the preceding period reflects the importance of previous consumption levels; this reflects the fact that consumers’ current consumption spending is considerably influenced by what they spent last quarter or in other recent quarters. They are reluctant to cut back their spending drastically, even if their income level falls.

PRIVATE INVESTMENT

EXPENDITURE

So far we have assumed that investment spending is autonomous. Thus we have been able to consider private investment spending (together with government spending) as an independent driving force, to which (through the multiplier effect) total g.n.p. and consumer spending adjust. Now we must look at the determinants of private investment expenditure in more detail. We shall see that while private investment expenditure is partly autonomous, it is also related closely to the level of consumer spending. That is, rising consumption may induce more investment, as well as the other way around. Thus, a more thorough look at the system will require us to complete the circle: Investment spending —~ more consumption —~ more private investment —* more consumption . . . etc.

Investment spending is largely business spending—on plant, equipment, machinery, and inventories. Only one component of private investment is made directly by households—new single-family housing—and this seldom accounts for over a quarter of the total. Private investment spending reflects primarily the decisions made by businessmen—presidents, finance committees, boards of directors, and others, all the way down to the corner grocer.

Private investment spending has been the most dynamic and unstable major component of the gross national product. Most economists think it plays a central role in explaining both economic growth and fluctuations. Figure 7-3 shows the fluctuation in private investment spending since 1929. The top line is gross, or total, investment. The lower line is net investment, which subtracts the allowance to replace capital goods worn out (depreciated) during that year. Only the lower line represents the net addition to our capital stock each year.