The fragile roots of productivity growth

J. Bradford DeLong Professor of Economics University of California Berkeley Project Syndicate

How fast is the U.S. economy now growing? Smart investors are now betting that the most recent economic data will show an annual GDP growth rate of 3 percent for the second quarter of 2003. They are also betting that the current third quarter may well show a growth rate as high as 5 percent per year. July saw American industrial production jump by half a percent, while Intel, the big microprocessor maker, is reporting very strong growth in demand for its key products, suggesting that business investment spending is finally accelerating.

Although Japan continues to stagnate, and Western Europe hovers on the edge of recession, a solid "output-growth" recovery in the U.S. should prove to be a big help in boosting demand in the rest of the world. But this good news about the U.S. business cycle has been accompanied by bad news about employment. Hours worked in American business fell at a 2.7 percent annual rate in the second quarter. Employment-particularly manufacturing employment-fell in July. It is very likely that hours worked will continue to fall in the third quarter.

What accounts for this wide divergence? How can America see reasonably rapid output growth and yet rising unemployment? The answer is that the underlying trend of productivity growth in the American economy continues to be exceptionally positive. The 5.7 percent annual productivity gain recorded in the second quarter seems likely to be matched in July-September. With America's workers producing more in less time, it is no wonder that a wedge is being driven between the (good) output news and the (bad) employment news.

In the short run, rapid productivity growth poses dilemmas for macroeconomic management, because what would otherwise be seen as reasonably strong demand growth is proving to be insufficient to keep unemployment low. But the short-run view is not the important one.

Rapid productivity growth is-in the long run-good news for America: A higher-productivity economy is better at enhancing human welfare. In the long run, this is also good news for America's trading partners: There is more value to be gained by trading with a richer economy than with a poorer one.

The continuation of rapid U.S. productivity growth through the recent recession and into the subsequent low-wattage recovery is a very strong piece of evidence that America's long-run rate of GDP and productivity growth has shifted upward permanently, or, if not permanently, at least for a period of time likely to be measured in decades.

America, of course, is not alone among industrial nations that have seen productivity growth accelerate since the second half of the 1990's. Australia, Ireland, and the Scandinavian countries all had their own sustained bursts of unexpectedly rapid growth. But the rest of Western Europe and Japan have not.

Western Europe and Japan have vibrant high-tech sectors, able executives, and cultures easily as capable as the U.S. of taking advantage of the boom promised by the rapidly falling prices of information technology. So you would think that the same productivity revolution would happen in those countries. Yet there has been no sign of it. Why not?

The current conventional wisdom, most strongly advocated by the American economist Robert Gordon, is that the burst of productivity growth that the U.S. is currently experiencing (and that will turn into a full-fledged economic boom whenever demand growth becomes rapid enough) is due to synergy. According to this view, the U.S. is especially well positioned to benefit from rapidly falling prices of information technology, owing to its openness to competition and new ways of doing business, particularly in distribution.

The comparison between the late 1990's and early 2000's in America and in Western Europe compels us to reflect on how fragile the underlying institutional mechanisms needed to support rapid economic growth truly are. Western Europe and the U.S. are both post-industrial democracies. Both possess market economies. Both have large and powerful social insurance systems.

Of course, institutional differences between America and Western Europe do exist: In the regulation of labor, in restrictions on land-use and redevelopment, in the tolerance of competition authorities for resale price maintenance and related practices. But are these differences really so large as to affect macroeconomic performance so dramatically? In Robert Gordon's estimation-and in mine-the roots of an economy-wide high-tech productivity boom like the one seen in the U.S. are so delicate that they flourish in American but not in continental European soil.