Don’t Confuse Me With Facts: Denying the Lack of Growth in U.S. Economic Concentration

Robert D. Atkinson April 1, 2022
April 1, 2022

(Ed. Note: The “Innovation Fact of the Week” appears as a regular feature in each edition of ITIF’s weekly email newsletter. Sign up today.)

Perhaps the most important question swirling around the heated debates about antitrust is whether economic concentration has increased. Anticorporate Neo-Brandeisians have a big stake in painting a dystopian picture of rampant monopolies—killing small businesses, jacking up prices, and crushing wages—all in their attempts to achieve a wholesale restructuring of U.S. antitrust law and practice.

But these claims to date have largely been hortatory. Concentration is up… Because I and 100 other people said so. To attempt to remedy this erroneous groupthink, ITIF conducted a study using the most recent Census data on over 850 industries at the six-digit level of NAICS codes. We found that from 2002 to 2017 there was almost no increase in concentration. If fact, there was a decline among more concentrated industries. The U.S. Chamber of Commerce commissioned a study by NERA Economic Consulting group that used a similar methodology and it actually found a decline in concentration.

But for various reasons some have tried to refute ITIF’s analysis. Case in point: A recent article by antitrust scholar Herbert Kovenkamp, who criticizes the study because “it does not examine the period prior to that, thus missing the period of greatest increase.” The reason is simple: The farther one goes back in time, the harder it is to get apples-to-apples NAICS-code comparisons. This is why the base period for the ITIF study was 2002.

But even more striking is his claim that the period of greatest increase in concentration was pre-2002. If this were the case, then why all the panic now out-of-control growth in monopoly? If the problem was in the Clinton years, then why didn’t advocates complain then? The reality, of course, is those raising Cain about monopoly are referring to the last decade or so. Eduardo Porter of the New York Times writes, “There is plenty of evidence that corporate concentration is on the rise.” The Center for American Progress writes, “America faces a problem of rising market concentration across the economy.” These and others aren’t folks who were asleep for much of the last two decades and just realized what happened when they were watching Friends in the original. They are writing about the present, not the past.

So, for Kovenkamp to say that the ITIF study is irrelevant because it doesn’t focus on the right period is to ignore current claims, or to acknowledge that there has been no increase in 20 years. If the latter is true, then clearly there is no major problem in recent antitrust policy.

He goes on to argue that NAICS code industries, even at the most detailed level, are not accurate representations of actual markets. This may well be, but in the absence of any empirical evidence to the contrary, the fact that these two studies referenced above show no increase in concentration should mean at least something. And if he wants to dismiss NAICS code studies to assess economy-wide concentration, then how should we assess concentration changes?

He doesn’t say, other than referring to a study purporting to assess changes in price/cost margins. He writes margins have risen, corporate profits have risen as a share of GDP, and labor participation has declined, as if these were not only all true, but evidence of concentration. But he is guilty of sloppy analysis at best. The study purporting to show higher markups has been widely and effectively critiqued, including by ITIF. Its principal error is to fail to consider the significant increase in the growth of corporate intangible capital (e.g., research and development), which, when controlled for, shows no increase in markups. When it comes to profit increases, U.S. government data paints a different picture. As a share of GDP, overall corporate profits are now lower than they were the 1960s, when antitrust enforcement was at a peak. Moreover, domestic profits in nonfinancial firms actually have fallen as a share of net value-added. Finally, while labor’s share of income has declined slightly over the past two decades, it is not principally because capital’s share of income has increased. Most of the decline is attributable to an increase in the share of rental income—what renters pay and what the imputed rent homeowners pay for their house—and an increase in self-employment income (a form of labor income).

But data and evidence be damned. I feel like there must have been an increase in market power over the last decade, so there must have been one. Case closed.