Back in 1987, economist Robert Solow observed that "you can see the computer age everywhere but in the productivity statistics." At the time, output per hour was still creeping higher at an annual rate of 1.1% or so.
Since 1995, however, it has been galloping forward at a 2% clip--reinforcing the view that a technology-driven productivity revival is for real. In fact, when Federal Reserve Chairman Alan Greenspan recently told Congress that the economy's potential growth rate was now 3%, he attributed two percentage points of that to a pickup in productivity.
None of this means that the productivity debate is settled. Indeed, two recent economic studies suggest that the evidence of widespread gains remains murky at best. Intriguingly, however, one concurs with Greenspan's estimate of a 3% potential growth rate while rejecting the New Economy paradigm.
Many believers in a productivity rebound point to the proliferation of new products in recent years. What they forget, however, writes Brookings Institution economist Jack E. Triplett in a recent issue of Business Economics, is that "to have an impact on productivity, the rate of new-product introductions must be greater than in the past."
In other words, the more products already in existence, the greater the number of new ones required just to keep productivity growth constant. As it is, he says, "there's no strong evidence that the rate of product innovation is rising."
Triplett also questions the common claim that underestimation of the output of such major computer-using service industries as banking, insurance, wholesale trade, and business services is limiting measured productivity gains. Because much of the output of these industries is in intermediate products sold to other businesses, he notes, any true productivity gains should still flow through to final output measures and overall economic growth even if intermediate output is mismeasured.
As for the doubling of annual productivity growth to 2% since 1995, economist Robert J. Gordon of Northwestern University attributes it equally to three factors. The first is improvements in the measurement of inflation, via revisions in the consumer price index and a decision to use the slower growing producer price index rather than the CPI to measure roughly half of price changes for medical care.
The second factor is the normal pickup in labor productivity that occurs in periods such as 1997 and 1998 when output grows faster than its long-term trend. Because employers do not anticipate such surges, they are often reluctant to accelerate hiring when they occur--and in tight labor markets, such hiring is impeded anyhow.
Finally, Gordon's analysis indicates that an acceleration in long-term productivity growth of about 0.3% has indeed taken hold in recent years--but it has occurred entirely within the computer industry. On balance, he says, "there has been no productivity acceleration at all in the 98.8% of the economy located outside of computer and related technology manufacturing."
So what does all this say about the economy's potential growth rate? According to Gordon, the productivity explosion in computers, along with the effects of improved output measurement, suggest that the rate has indeed risen to 3%--a number that would be even larger if a true widespread productivity revival were eventually to develop.