The measurement and future of U.S. productivity growth
Productivity is a hard statistic to pin down. Never mind conceptual distinctions such as the difference between labor productivity and total factor productivity. The actual measurement of productivity itself—defined broadly as the effectiveness of producing a good or service—and its growth in the economy is debatable. The recent slow growth in productivity raises concerns that it is being mismeasured, but that problem may mask other problems in the economy that also may be crimping productivity growth.
But first, let’s look at the mismeasurement argument. The latest angst involving productivity measurement is captured in a Wall Street Journal article by Timothy Appel. Appel talks to several entrepreneurs and innovators in Silicon Valley, including Google Inc. chief economist Hal Varian, most of whom think that the official measurements by the U.S. Bureau of Labor Statistics undervalue many firms’ recent innovations. Varian’s employer is a perfect example: search engines can help workers easily access information that before would have taken much longer to find in a full research library. But these services are provided at no monetary cost to the consumer, thus not reflected in productivity statistics.
How much this “free problem” is biasing down productivity growth is an open question. It’s entirely possible, as Appel notes in his piece, that the seeming mismatch between the innovation and disruption we read about, and the official data is a matter of timing. When Appel quotes Nobel Laureate and Massachusetts Institute of Technology economist Robert Solow back in 1987 noting that “you can see the computer age everywhere but in the productivity statistics” the problem was that the benefits of the computer age hadn’t kicked in yet. Productivity eventually did increase later in the 1990s as the benefits of personal computer spread out to the broader economy.
This “diffusion” explanation just requires some patience on the part of economists and policymakers to wait for the fruits of recent innovation to filter into the broader economy. Eventually the innovations made at vanguard firms will flow outward, and the gains will end up staring back at us in the Bureaus of Labor Statistics data. At least that’s the rosy scenario.
The not so rosy one is this—the ability of productivity to diffuse outward might now be impaired. A recently released book by the Organization for Economic Co-operation and Development shows that productivity growth has been quite quick at many firms operating among its developed- and leading developing-nation members. But these innovations aren’t moving out to the rest of the firms in those economies. In other words, diffusion isn’t working as promised, moving from the innovative firms to the broader economy in the way computers did in the 1990s.
What exactly is impairing this diffusion? First, it’s important to note that economists haven’t exactly locked down a good understanding of how ideas diffuse, as University of Houston economist Dietz Vollrath points out. But one potential culprit, suggested by Timothy Taylor, the managing editor of the Journal of Economic Perspectives, is the slowdown in the rate of start-up firms. He suggests that bevy of new, quickly growing, and highly competitive firms could be one way to get new innovative ideas out into the broader economy. It is certainly the case that despite all the new innovation news out of northern California, the rate at which new firms are started is on the decline. There is no consensus on whether this trend is a cause of slower growth, or is because of slower growth.
Economists aren’t sure what’s causing the slowdown in startups, but their role in the diffusion of new technologies that spur productivity growth is a good reason to figure it out sooner rather than later. There may be some flaws with our current measurement system, but we’d be better served by thinking about the root causes of slower productivity growth rather than tinkering with the productivity calculations.