The rate of capital formation by businesses has long been among the most closely watched elements of the national accounts. During the last decade, this component of investment attracted considerable interest as capital spending helped support our uncommonly high rate of economic growth. Not only did this spending lift the growth of aggregate demand, it also increased our capacity for supplying goods and services, which in turn could allow output to continue growing rapidly in the future.
This article analyzes the performance of conventional models of investment spending by comparing their abilities to describe this spending from 1960 to 1990 as well as their abilities to forecast spending during the 1990s. The authors find that recent shifts in the composition of the stock of capital goods and in the relative prices of capital goods have undermined the performance of these models of aggregate spending. In many ways, aggregate capital spending seems to depend more heavily than it has in the past on industries unique circumstances and changing technologies. The authors suggest that errors of the models, the changing composition of capital, and new methods of measuring the stocks of capital warrant considering more disaggregated descriptions of investment spending.