Survey of the Factors Contributing to the Decline in U.S. Productivity Growth

Since the first half of the nineteenth century, the secular rate of growth in real gross product per labor hour in the U.S. domestic business economy gradually accelerated from about 0.5 percent a year to a maximum average annual rate of 3.5 percent in the subperiod 1948-66 (see Table 1). Since then, it declined to about 1 percent during the period 1973-78, and then fell absolutely in 1979 and will probably drop again in 1980 due largely to cyclical influences and continuing oil price increases. The declining trend-rate of productivity growth after 1966, and the absolute declines since 1978 have become an increasing matter of concern to policy-makers and informed citizens in the United States. Since productivity gains are the chief source of increases in real income per capita, the slowing has meant lesser gains in living standards. Since increases in factor productivity are an offset to increasing factor prices, the slowing has been a significant element in the acceleration of inflation in unit factor costs and product prices. Although overall productivity.changes are only indirectly involved in balance-of-payments problems, the industries whose relative productivity growth has slowed the most have had the greatest difficulties in meeting foreign competition. Clearly, policies to promote productivity would be of considerable assistance in helping this country meet some of its more pressing economic problems. As background for policy formulation, it is essential that we understand the chief sources of productivity growth, and thus the causes of the slowdown after 1966. A convenient and useful classification of sources of economic growth in general, and of productivity growth in particular, has been provided by Edward F. Denison, together with estimates of the percentage point contributions of the various sources from 1929 through 1976.~ I have made use of his schema, with some modifications described below, as well as many of his estimates, supplemented by my own estimates for selected variables, and for all of them in the subperiod 1973-78 since most of his series end in 1976.