Comments and Discussion

What, another paper on the new economy? When financial markets are raking through the debris of $8 trillion in lost equity value, and ".com" is a reviled four-symbol word, a paper on the impact of the new economy on productivity would seem as welcome as an analysis of the role of whales in the lighting revolution.

In fact, the new economy (or, more precisely, information technologies) continues to raise important puzzles about productivity growth. Variations in productivity growth have proved to be one of the most durable puzzles in macroeconomics. After a period of rapid growth following World War II, productivity stagnated in the early 1970s. There was no shortage of explanations offered, including rising energy prices, high and unpredictable inflation, rising tax rates, growing government, burdensome environmental and health regulation, declining research and development, deteriorating labor skills, depleted possibilities for invention, and societal laziness.¹ Yet these explanations seemed increasingly inadequate as inflation fell, tax rates were cut, regulatory burdens stabilized, government's share of output fell, research and development and patents granted grew sharply, real energy prices fell back to pre-1973 levels, and a burst of invention in the new economy and other sectors fueled an investment boom in the 1990s.

The productivity slowdown puzzle of the 1980s evolved into the Solow paradox of the early 1990s: computers were everywhere except in the productivity statistics. The penetration of the American workplace by increasingly sophisticated and powerful computers and software apparently failed to give an upward boost to productivity growth, for through thin and thick, labor productivity growth seemed to be on a stable track of slightly over 1 percent a year.

Then, in the mid-1990s, productivity growth rebounded sharply. Beginning in 1995, productivity in the business sector grew at a rate close to that in the pre-1973 period. The causes of the rebound were widely debated, but at least part was clearly due to astonishing productivity growth in the new economy sectors of information technology and communications. This period led to yet another paradox, identified by Robert Gordon, who argued that, after correcting for computers, the business cycle, and changes in measurement techniques, there was no productivity rebound outside the computer industry.

This paper attempts to sort out the productivity disputes by using a new technique for decomposing sectoral productivity growth rates and using a new data set that relies primarily on value added by industry. In addition to examining the recent behavior of productivity, the paper adds a few new features to the analysis.

First, it lays out a different way of decomposing productivity growth, one that divides aggregate productivity trends into factors that increase average productivity growth through changes in the shares of different sectors. Second, it develops an alternative way of measuring aggregate and industrial productivity based on industrial data built up from the income side rather than the product side of the national accounts. By relying on the industrial data, I can focus on different definitions of output and get sharper estimates of the sources of productivity growth. Third, by working with the new industrial data, I can make more accurate adjustments for the contribution of the new economy than has been possible in earlier studies. Finally, this new data set allows creation of a new economic aggregate, which I call "well-measured output," that excludes those sectors where output is poorly measured or measured by inputs.

Productivity Accounting

Measuring productivity would appear to be a straightforward issue of dividing output by inputs. In fact, particularly with the introduction of chain-weighted output measures, disentangling the different components of productivity growth has become quite complex. In this section I explore how to decompose productivity growth into three components: a fixed-weight aggregate productivity index, a "Baumol effect" that reflects the effect of changing shares of output, and a "Denison effect" that reflects the effect of differences between output and input weights.²

Consider indexes for the major aggregates. Define aggregate output as X_t, composite inputs (here, hours of work) as S_t, and aggregate productivity as A_t = X_t/S_t. The share...