(Bloomberg Opinion) -- When both experts and the general public think about monopoly power, they tend to think of huge companies dominating a nationwide market — like Standard Oil or Bell Telephone. Those companies were the subject of two of the most spectacular and high-profile antitrust cases in U. S. history. Right now, much of the talk of high-profile antitrust action centers around huge technology companies like Amazon.com and Facebook that claim large national market shares. And controversial mega-mergers, like the proposed AT&T takeover of Time Warner, tend to involve companies that are nationally prominent.
Much of economists’ research, too, has been focused on the nationwide level. A number of recent papers sounding the alarm about increased industrial concentration, rising price markups and profits, and falling corporate investment have all looked at on aggregate trends. For example, a 2017 paper by David Autor, David Dorn, Lawrence Katz, Christina Patterson and John Van Reenen received a lot of attention for finding a correlation between the degree to which a U.S. industry has become more concentrated in recent years and the degree to which labor’s share of income in that industry has fallen. They show increasing concentration in most industries:
But what if these national-level trends just don’t matter as much as one might think? After all, most competition doesn’t happen at the national level. For example, Wal-Mart can — and often does — charge different prices in different locations. If one town also has a Target, a Best Buy and other competitors, the local Wal-Mart might slash its prices to stay competitive. But in a neighboring town where Wal-Mart is the only big-box retailer, it can raise prices and squeeze local consumers. Because driving to the cheaper Wal-Mart would cost time and money, these markets are to some degree cut off from each other. This is especially true in industries like health care.
A similar dynamic is in play with wages. A trio of recent papers has found that in towns where there are only a few employers in an industry, workers tend — all else equal — to get paid less. That makes sense — it’s very costly and difficult to switch industries, retrain for a different kind of work, or move from town to town. But it also emphasizes that the key arena where competition for workers happens is often local, not national. Top software engineers, doctors or academics may move across the country for a new job, but they’re the exception.
Realizing this, some economists are now looking more closely at local labor market concentration. A new paper by Esteban Rossi-Hansberg, Pierre-Daniel Sarte and Nicholas Trachter looks at how many companies are in an industry nationwide versus how many are in the average ZIP code or county. Like Autor et al., they confirmed that national-level concentration going up in almost every industry. But they found that at the local level, the exact opposite was true — the average number of business in a given industry in each area went up between 1990 and 2014 in about 77 percent of industries.
Rossi-Hansberg et al. have a simple story for why this is the case. As top companies like Wal-Mart expand into more and more areas, they increase local competition, because suddenly all the existing smaller companies in the area have to compete with them. But by expanding, Wal-Mart also increases its national market share. Looking at cases where a top company moves into a town, the authors confirm that local competition tends to be higher seven years later. This story will ring true for the many Americans who have seen Wal-Mart move in and squeeze all the other stores’ margins.
Kevin Rinz, an economist at the U.S. Census Bureau, confirms Rossi-Hansberg et al.’s result of decreasing local concentration using extremely detailed and comprehensive Census data. He finds, like others, that — all else equal — local concentration acts to keep wages down. But because he finds that local concentration is falling, he asserts that employer market power — at least, in local labor markets — is unlikely to have been a big factor behind the recent wage stagnation. In fact, more competitive local economies may have been a factor fighting against whatever else is acting to push wages down.
So the work of Rinz and Rossi-Hansberg et al. gives us less reason to sound the alarm about monopoly power. But it doesn’t mean rising national-level concentration is nothing to worry about. First of all, many big businesses have supply chains that stretch across the country — increasing national concentration may allow them to squeeze their suppliers more. Second, big companies may hold down wages for higher-earning employees who move from city to city for jobs. Third, it may be that in the long run — say, 20 years instead of seven — big businesses end up reducing local competition as they finally muscle all the smaller fish out of the local pond for good. Additionally, many internet companies like Amazon and Facebook sell products nationally, and therefore compete at the national level. And finally, large companies may amass political power that allows them to rewrite local rules in their favor, to the detriment of competitors, consumers, and workers.
So it’s not time to stop worrying about national-level concentration. But it does look like local labor-market concentration might not be the chief villain in the wage stagnation story.
To contact the editor responsible for this story: James Greiff at jgreiff@bloomberg.net
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Noah Smith is a Bloomberg Opinion columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
©2018 Bloomberg L.P.