Are Employers Playing a Game of Monopsony?

“A lot of people had ignored it,” Chicago Booth’s Chad Syverson says of monopsony power. “In the past 10 years, there’s been a concerted effort to measure it, and it does seem to be a big deal.”

If monopsony is more prevalent than economists have traditionally assumed, it may play an important role in suppressing the price of one of the most closely watched and socially important inputs to virtually every product: labor. Economists’ efforts to detect and understand monopsony, therefore, could be pivotal to revealing the causes of workers’ frustration.

A world of wedges

In the free-market nirvana that provides the setting for many theoretical economic models, competition is unfettered and alternatives are always close at hand. So many producers vie to sell goods and services that none can raise prices without sending their customers scurrying to rivals. The result is that the price a company charges for its product equals or is very close to its marginal cost of production—that is, the cost of producing one additional unit.

Conversely, buyers are so plentiful that any who demand discounts from suppliers get turned away. Therefore, the price a company pays for any given element it uses to produce its goods or services—its inputs—should be equal to how much an additional unit of that element increases the company’s output. Economists call this the input’s marginal product.

In the real world, things get complicated. For one, companies don’t just have variable costs—the costs of the labor, materials, and other inputs that go into the product—but also fixed costs such as facilities, equipment, insurance, and other overhead. The need to pay off these costs is one reason companies might charge more than the marginal cost of production.

Profit, of course, is another reason. Competition helps keep profit margins in check, threatening companies with a loss of market share if they try to extract too much profit from their customers. But perfect competition, if it exists at all, is the exception. The number of market participants, geography, and seasonality, among other factors, can tip the competitive balance, undermining the economic ideal.

The question is, if these markups reflect companies’ growing market power, over which markets are they exerting influence: the product markets they’re selling into, or the input markets they’re buying from to fuel production? How much of the widening wedge is due to companies’ ability to hold down what they pay for inputs, and how much of it comes from their power to drive up what they charge their customers?

For economists, answering that question is harder than it may sound. Researchers face a major challenge in gaining access to the data they need to measure markdowns and monopsony power. Neither marginal cost nor marginal product is directly observable, which means economists have to use proxies in an attempt to estimate them. This makes consensus about the exact prevalence of market power difficult to achieve.

The big are getting bigger

The rise in markups that De Loecker, Eeckhout, and Unger trace is not universal. In fact, they find no change in the median markup among all companies. “The rise in the average markup is driven by a few firms at the top of the distribution,” they write. “Most firms see no increase in markups, while a few firms see a large increase.”

This outsize influence by a small number of large companies echoes findings from Patterson, whose research with MIT’s David Autor, University of Zurich’s David Dorn, Harvard’s Lawrence F. Katz, and John Van Reenen of the London School of Economics helped document the rise of “superstar firms.” Their study demonstrates that “industries across the vast bulk of the US private sector”—and in other developed countries—have seen industry-wide sales concentrating in a smaller number of companies over time, a dynamic with macroeconomic implications.

The rise of superstars is disproportionately concentrated in industries experiencing relatively rapid technological change, as measured by the growth of patent intensity and the total productivity of inputs. The winner-takes-most market structure the researchers describe is apparent in parts of the retail and transportation industries, as well as in the high-tech sector.

This trend suggests that technological dynamism, rather than anticompetitive forces, is the main driver, they write. However, it’s also possible that as industry superstars emerge, they’re increasingly able to further their interests by lobbying policymakers and erecting market barriers that foil efforts by smaller companies to grow and new entrants to offer alternative products.

Consider high-tech darling Apple. Steve Jobs turned it into a superstar company by revolutionizing the personal electronics market with computers for the masses and game-changing smartphones. Once it gained a dominant position, it capitalized by introducing a “walled garden” of apps, which prompted independent developers to cry foul and regulators to launch probes into potentially anticompetitive practices.

Brown’s Spencer Yongwook Kwon, Booth’s Yueran Ma, and Frankfurt School of Finance & Management’s Kaspar Zimmermann find evidence that not only are sales concentrating among a smaller pool of companies, but so are assets and net income. What’s more, they find that this process of concentration has been building for at least a century in the US.

To measure concentration, the researchers collected data from 1918 to 2018 using the Internal Revenue Service’s Statistics of Income, which are published annually and cover all corporate businesses. They calculate that since the 1930s, the share of US economic activity conducted by the top 1 percent of businesses, as ranked by assets, has increased from 70 percent to 90 percent. Meanwhile, the portion of assets held by the top 0.1 percent of companies has risen to 88 percent from 47 percent.

Their work indicates that rising concentration was stronger in manufacturing and mining prior to the 1970s, which is consistent with these industries’ migration into mass production. In the following decades, the move toward concentration was stronger elsewhere, as information technology transformed other sectors. More generally, the timing of rising concentration in an industry aligns with the era when research and development and IT have consumed a growing portion of that industry’s total investment, the researchers find.

Workers are falling behind

Ma notes that the trend in which the big are getting even bigger has social as well as economic implications, especially for workers. “How do we distribute surplus? How do we let employees bargain with companies?” she asks.

There’s evidence to suggest that, in fact, workers aren’t sharing in the economic gains accruing to their increasingly large and concentrated employers. In 2014, University of Minnesota’s Loukas Karabarbounis and Chicago Booth’s Brent Neiman found that labor’s share of income—the portion of economic output that goes to workers—had shrunk since 1980, in the US and around the world. The international nature of the phenomenon—Karabarbounis and Neiman found the global decline was about 5 percent—suggests the forces at work extend beyond standard bogeymen such as cheap imports, immigrant labor, and the decline of collective bargaining.

Economists have explored a number of explanations for this decline. Karabarbounis and Neiman argue that the cost of capital fell relative to the cost of labor, which could have encouraged employers to install IT equipment and adopt automation to substitute for human workers.

Patterson and her coresearchers find evidence that it’s the rise of superstar firms that has largely driven the trend, as the winners in various markets are those that have higher markups, which naturally leaves labor with a smaller fraction of income.

But other research provides support for the idea that a power imbalance in the labor market could be contributing to the fall of the labor share, at least in some parts of the economy. In a study of US manufacturing based on Census Bureau microdata, Analysis Group’s Ivan Kirov and New York University’s James Traina find that in 1972, pay for workers was equal to the marginal value of what they contributed to production. By 2014, labor’s slice of that pie had fallen to a half of what they added to the mix.

The researchers characterize this trend as reflecting a substantial rise in manufacturers’ power over their US workers during the period they studied. The widening labor markdowns resulted not from a drop in wage growth, which remained fairly stable. They were instead due to an acceleration in the marginal revenue that manufacturers generated but declined to pass along to employees. The widening of the labor wedge was exceptionally sharp in the early 2000s, they find.

Notably, US manufacturers’ growing leverage over their workers was driven neither by monopoly power nor the concentration of labor among a small number of local employers, the researchers conclude. They point instead to another prominent feature of the modern economy: technology. As production technology has consumed an ever-growing share of total input costs, it has also boosted the productivity of the workers who use it, while enabling their employers to hold on to most of the surplus wealth they create. This increase in producers’ labor-market power, in the researchers’ view, is the dominant factor in aggregate market-power wedges in US manufacturing.

“This result helps explain the sharp fall in the manufacturing labor share,” they write. “Labor market power comes from technology, whereby plants capture most of the surplus from increased labor productivity.”

Given the evidence for a technology-driven rise in employers’ labor-market power, Traina says, policymakers face the challenge of balancing strong growth with equitable gains.

Government officials usually implement consolidation programs in a bid to improve productivity growth, but they can have the opposite effect when producers exert oligopsony power, Rubens writes. This is especially a concern for agricultural markets in developing countries, which often feature internal trade restrictions similar to those in the Chinese tobacco industry.

The World Bank estimated in 2018 that 560 million people worldwide live in extreme poverty in rural areas, with most getting by as subsistence farmers. While there are many reasons for this economic plight, Rubens notes that economists are increasingly paying attention to the role played by monopsony power.

More than wages

While the details of how employers exert monopsony power are still coming into focus, policymakers are broadly aware of the implications for workers. When the US Federal Trade Commission successfully blocked a merger between grocery chains Kroger and Albertsons late last year, it expressed concern about the deal’s potential effects on labor. Rejection would ensure that the supermarket operators “continue to compete for workers through higher wages, better benefits, and improved working conditions,” the commission said in a written statement.

As the FTC suggested, the effects of a power imbalance in the labor market go beyond wages. Companies with leverage over their workforce can dictate the terms of employment. Consider the trend toward outsourcing jobs, which has created an underclass of workers who often receive inferior wages, benefits, and job security than their noncontract colleagues.

In their work on superstar firms, Patterson and her coresearchers suggest such companies are able to flex their muscle by outsourcing, increasing the amount of activity conducted by third-party firms and independent workers. Janitorial and clerical work, food services, and logistics are among the types of jobs affected, they note.

This apparent “fissuring” of the workplace can directly reduce the share of economic activity that goes to labor by excluding a large portion of the workforce from the benefits accorded to high-wage employers. Another possible related effect is to reduce the bargaining power of in-house and outsourced workers and increase labor-market competition, the researchers add.

The measurement challenge

Given the finite amount of research conducted so far and its limitations, the possibility remains that input markdowns are not as prevalent as product markups, Syverson cautions. Instead, markdowns may be restrained by differences among buyers, the cost of switching suppliers, and imperfect information, he reports.

Markdowns, to whatever extent they do exist, effectively act as taxes on the providers of inputs, just as markups are essentially taxes on the products in a given market. They also lower productivity by allocating resources away from how they’d be deployed in more competitive environments. When combined, the two metrics summarize the imperfect nature of real-world markets.

Although some of the details remain sketchy, it’s clear that the effects of market wedges are large and very real. They alter the way an economy distributes wealth, influence workers’ sense of fairness (or unfairness), and may fuel the sort of populist backlash that’s now playing out in many developed economies.