Everybody hates Wall Street, and with good reason: Excessive risk-taking on Wall Street led to the financial crisis and the Great Recession, but Wall Street has so captured both parties that no one went to jail and the banks were bailed out on generous terms while ordinary people lost their homes and life savings. Not even ten years later, the already-not-as-tough-as-they-could-have-been rules imposed in the wake of the crash are being rolled back by a sham-populist President who smeared his opponents as stooges of Goldman Sachs but now employs half of Goldman’s executive suite in his White House. The financial industry is a corrosive force in American politics and society, and it needs to be brought to heel.

But it is not the only such force. The accumulation of wealth and power in the financial industry since the 1980s has been accompanied by another unsettling economic trend: A smaller and smaller number of bigger and bigger firms are taking over more and more of the American economy and the global economy as a whole. Economists and pundits have noted this trend with some alarm, but for the most part, public debate over economic reform has focused more on the power of Wall Street than on the increasing power of large corporations in general. That’s unfortunate, because market concentration is a serious threat to the welfare of American workers and consumers, and one requiring public action if it is to be remedied.

Many different numbers point to the trend towards market concentration. For example:

  • Between the late 1980s and 2014, the percentage of American workers employed by companies with at least 1,000 employees rose from 40% to 46%
  • Between 1997 and 2007, the market share of the 50 largest companies increased in three-quarters of broad industrial sectors in the American economy
  • The rate of new business formation has declined each decade since the 1970s
  • Over that same period, the number of businesses employing fewer than 50 workers has risen by roughly 50%, while the number of businesses with 1000 or more workers has risen by 100% (see chart).

These general trends have produced plenty of illustrative anecdotes: Wal-Mart owns a quarter of the grocery market; four companies collectively take in more than $7 of every $10 spent at pharmacies. Four companies control 80% of the seats on domestic flights; beer giants Anheuser-Busch and SABMiller are currently moving towards a merger that, if allowed, would create a company responsible for $3 out of every $10 in beer sales in the world.

Screen Shot 2017-03-24 at 4.59.02 PM

But my aim isn’t to catalogue the increasing concentration of the American economy; I take that as a given. Rather, I want to suggest that this market concentration might be connected — through its effect on wages, inequality, productivity, and small-town life — to the social and economic malaise that have upended politics around the globe in recent years. What’s more, market concentration and monopoly power might be a natural feature of a free market economy — one requiring aggressive government intervention in order to restore the balance of power between workers and consumers on the one hand, and employers on the other.

Other people have discussed in greater depth the effects of market concentration on wages, productivity and inequality, so I’ll just outline the problems broadly. First, wages: When firms face decreased competition for labor, they can keep wages below their competitive level — that is, they can pay workers less than those workers are worth, without fear of another company hiring them away. A firm that keeps wages low will hire fewer workers and produce less output than a firm paying an efficient wage, but makes up for this through savings on labor and the higher prices it can charge for the products it does sell. Firms in non-competitive markets can exercise power over workers in other ways as well: For instance, by requiring employees to sign arbitration clauses that prevent employees from suing their employers in court, or by requiring employees to sign non-compete agreements that forbid them from working for a competitor for some time after leaving the firm. Firms in concentrated markets will also find it easier to collude to keep wages down. As this graph shows, while workers at large companies used to receive better wages than workers at small companies, that trend is reversing:

wages and firm size

An implication of non-competitive wage-setting is that wages will become uncoupled from productivity: If competitor firms don’t bid wages up to their marginal product, employers will pay workers less than those workers produce. Indeed, that’s what’s happened since middle-class wages began to stagnate in the 1970s:

screen-shot-2017-03-25-at-12-47-16-pm.png

The methodology behind this last chart spurred some controversy, but even critics had to concede that wages and productivity have drifted apart since around 1973:

Screen Shot 2017-03-25 at 1.00.13 PM

This second chart serves as a nice segue into a brief discussion of inequality and market concentration. One obvious possibility, which Joseph Stiglitz has suggested, is that the owners (including shareholders) and executives of market-dominating companies will become incredibly rich and leave everyone else in their dust. But some of the benefits of monopoly might make it to a firm’s workers as well, contributing to inequality not just between workers and CEOs but between workers in concentrated versus competitive markets. The difference between average and median wages in this chart suggests that some high-paid workers have seen their wages increase along with productivity, as happens in competitive labor markets, while workers in the middle of the income scale have not. Jason Furman and Peter Orszag, former economic policy officials in the Obama administration, have suggested that monopolistic firms charging monopoly prices then pass some of those increased profits on to their employees. People who work for monopolistic firms see their wages rise; those who don’t, don’t.

But don’t firms in non-competitive markets pay low, non-competitive wages? Yes — that is, firms in non-competitive labor markets. Firms are buyers in the labor market and sellers in the product market, and the same firm might face little competition in one market but plenty in the other (although as fewer, bigger firms come to dominate the economy, both markets will become less competitive). A firm that faces little competition when hiring workers or selling products — like Wal-Mart — can reap immense profits without passing them on to its workers. By contrast, a firm that faces more competition when hiring workers than it does when selling its products — like Apple, which dominates the U.S. smartphone market but has to hire skilled engineers to design its phones — will have to pass some of its monopoly profits on to its workers. Depending on the relative levels of competition a firm faces in the product and labor markets, its employees might benefit from market concentration, or they might suffer. That’s how you end up with a graph like the one above (although it should be said that the workers most hurt by market concentration likely also bear the brunt of outsourcing and automation). This might help explain why the returns to skill increased between the 1970s and 1990s: Skilled workers, whose labor firms continued to fight over, could better resist downward pressure on their wages than unskilled workers could.

A small subset of workers have seen their wages rise along with productivity even as labor markets overall have become less competitive. Yet even these workers may be feeling some ill effects of market concentration, which could be partially responsible for the mysterious productivity slowdown that has afflicted the U.S. economy since the 1970s (with a break during the tech boom of the ‘90s) and the global economy since the turn of the century. Monopolistic firms don’t feel the same pressure to innovate that firms in competitive markets do; there are no rivals to out-compete. Accordingly, firms in non-competitive markets don’t focus as much on developing efficient ways of combining labor and capital into output. But firms will only pay workers as much as the workers produce; thus, a worker’s productivity sets the ceiling on her pay. And if firms don’t find ways to increase worker productivity, wages won’t rise.

When fewer firms compete with each other for labor and market share, wages and productivity fall, and the gap between the haves and the have-nots widens. But I think market concentration might have another undesirable effect: It might be partially responsible for the hollowing-out of rural and smalltown America. Although Wal-Mart and CVS might pay their workers more than the local businesses they replace, in doing so, they also replace local owners with distant CEOs. That both weakens the fabric of a community and siphons money from the town where a store is located to wherever the company’s headquarters are. The profits from Brothers Pizza in Altoona, Pennsylvania stay in Altoona, Pennsylvania; the profits from the Domino’s down the street go to Domino’s headquarters in Ann Arbor, Michigan. While the middle of the 20th century saw economic activity becoming more evenly distributed throughout the country, that trend has since reversed: Since the 1980s, economic inequality between geographic regions has increased, and economic activity and wealth have become more concentrated around coasts and cities. Consistent with the hypothesis that skilled workers may benefit from market concentration, educated elites are becoming increasingly concentrated in a few cities — where, perhaps, they manage and service the huge companies that employ their less-skilled former neighbors back in flyover country.

This isn’t just an American problem: Small towns and local businesses are suffering in France as well. And France and the United States have something else in common: The rise of right-wing nativist populist movements drawing strength from regions that have seen good jobs move elsewhere. Donald Trump won 86% of U.S. counties, but those counties produced only 36% of the country’s GDP. Trump’s victory was fueled by people left behind by the geographic concentration of the U.S. economy.

Why are a few huge firms coming to dominate the American and global economies? In the U.S., one explanation is that the Department of Justice changed its approach to antitrust law in the 1980s. Since then, the Department has allowed mergers to proceed even if they hurt competition, so long as they don’t raise prices. But this doesn’t answer the question of why companies push to grow so large in the first place. The answer is economies of scale: Companies can often drive down the cost of production by producing goods in larger numbers. Neoclassical economics simply assumes this problem away by assuming that firms face increasing, not decreasing, marginal costs — that is, that the 1,000th widget costs more to produce than the 10th. If firms do face increasing marginal costs, then markets will reach an efficient, competitive equilibrium, without any monopolies. The thing is, in many industries — if not most — this assumption simply doesn’t hold. Hamburgers, internet searches, cell phone service: All these things are cheaper to mass-produce than to produce on a small scale. Accordingly, the markets for hamburgers, internet searches and cell phone coverage are dominated by one or a few huge firms, because bigger firms will face lower costs, offer lower prices, and undercut their competitors. Any industry with declining marginal costs will not reach a competitive equilibrium, but will tend towards what economist call a “natural monopoly,” or at least oligopoly. Perhaps that’s true of more industries than microeconomics textbooks let on.

Traditionally, governments have responded to this problem with antitrust laws that prevent companies from growing too big. Maybe the U.S. should return to an antitrust regime premised on encouraging competition, not just keeping prices low. Alternatively, we could impose a progressive tax on inputs for firms facing economies of scale: The inputs a company buys to produce the 1,000th widget would be taxed at a higher rate than the inputs for the 10th. There might be a crippling flaw in this idea; admittedly, I just made it up. But we’ve got some time to think, as no one is going to do anything about this problem at a federal level for at least four years.

Some other interesting pieces on the topic:

Advertisements