The ‘crowding out’ effect of federal government outlay decisions: An empirical note

Summary and conclusions

This note has addressed the empirical issue of crowding out by examining the proportion of GNP devoted to private investment in new physical capital as a function of the proportion of GNP devoted to federal government outlays. Three alternative models were estimated, all of which found evidence of (a) a definite pattern in which private investment is crowded out by government spending and (b) only partial, i.e., incomplete, crowding out. These findings are, in principle, compatible with the studies by Arestis (1979), Abrams and Schmitz (1978), and Zahn (1978).

We may infer at least two important policy implications from the above findings. First, federal government decisions which act to raise federal outlays tend to diminish private-sector investment in new physical capital. To the degree that this form of crowding out occurs, private sector unemployment is generated. This clearly acts to weaken the stimulatory direct effects of the increased federal spending. Second, to the extent that federal government spending decisions lead to diminished investment in new physical capital, the rate of capital formation is diminished. This tends to worsen long-term inflation by cutting down on the ability of aggregate productive capacity to keep pace with aggregate demand.

The two implications stated above cast potentially grave doubts upon the wisdom of federal government decisions that lead to increased federal outlays. Ideally, at the very least, each such spending decision should be scrutinized for its particular impact on investment in new physical capital. Clearly, although federal government expenditures in the aggregate lead to diminished private investment, certain specific forms of federal spending may not change private investment at all, whereas other forms of federal spending may even lead to increased investment. The latter could well be characteristic of federal outlays for new highway construction. Thus, there appears to be a pressing need to disaggregate according to federal spending type.