Stricter merger policy guidelines will increase competition, leading to higher wages and welfare for workers, write Kyle Herkenhoff and Simon Mongey. The authors use economic modeling to show that the stricter 2023 Guidelines will improve worker welfare, and that even tighter thresholds can be applied to labor markets to amplify worker welfare gains from antitrust policy.
Preventing mergers that generate “monopsony” power will promote competition and prevent worker harm. Let us begin by defining “monopsony.” Most readers of this article will have spent many hours of their formative years playing the popular board game “Monopoly.” The objective is simple: buy all the productive assets in the economy and then extract all money from your competitors. If one were to invent a game called “Monopsony,” its objective would also be simple: control all of the jobs in town, and then use that dominant position to offer lower wages (where else could the workers go?) to maximize profits.
By now, a large number of top economic publications have argued that monopsony power is pervasive across countries, including the United States (Arnold 2019), Brooks et al (2021), Azar et al (2022), Berger et al (2022a), Yeh et al (2022), Lamadon et al (2022), Berger et al (2023c)). However, existing tools used to quell monopsony, including the federal minimum wage, are unlikely to have much power in the United States (Berger, Herkenhoff, Mongey 2022b). Minimum wages only have “bite” for firms with low profit margins and little monopsony power (e.g. local coffee shops). Small, low-pay firms are, according to our measurements, less monopsonistic to begin with, and yet they are the firms pushed out of business by large minimum wage hikes. As these smaller firms exit the market, greater labor market shares are diverted to larger firms with more market power (e.g. national coffee shops with supply chains etc.).
This leaves scope for alternate tools, including merger policy, to promote competition in the labor market. The greatest obstacle to the application of antitrust law to the labor market is inertia; in the past, scholars and regulators have ignored the labor market implications of mergers (Posner 2021, Hovenkamp 2022). Product markets were the focus and labor markets effects treated as “out of market effects” (Hemphill and Rose 2018). However, the Biden administration in July of 2021 issued an executive order calling on antitrust agencies to devote more efforts to curbing labor market power:
“It is the policy of my Administration to enforce the antitrust laws to combat the excessive concentration of industry, the abuses of market power, and the harmful effects of monopoly and monopsony — especially as these issues arise in labor markets…”
Consequently, in November 2021, the Department of Justice sued to block the merger of publishers Penguin Random House and Simon & Schuster on grounds of “harm to American workers, in this case authors, through consolidation among buyers…referred to as ‘monopsony.’” By November of2022, Simon & Schuster and Penguin Random House abandoned their merger. More recently, in December of 2023, the DOJ and Federal Trade Commission codified their intent to promote labor market competition by introducing guidelines to prevent mergers that are harmful to workers.
Our recent article (Berger, Hasenzagl, Herkenhoff, Mongey, and Posner 2023a) uses economic modeling to assess the worker welfare implications of the new 2023 Merger Guidelines. Merger Guidelines provide concentration thresholds, such as the post-merger Herfindahl-Hirschman Index (HHI),above which a merger is presumed to be anticompetitive. The Agencies (FTC and DOJ) are significantly more likely to challenge a merger when it falls above a “high concentration” threshold. In 1982, a post-merger HHI greater than 1800 was considered “high concentration” and presumed anticompetitive. By 2010, the Guidelines were watered down and the threshold was raised to 2500, thus classifying fewer mergers as anticompetitive. In theory, these Guidelines were also supposed to apply to labor markets, but in practice they were not.
The new 2023 Merger Guidelines revert to the earlier, stricter Guidelines, and also made clear that their application to labor markets is going to be an important factor in antitrust policy going forward. In our public comment (Berger, Hasenzagl, Herkenhoff, Mongey, and Posner 2023b) and accompanying working paper (Berger, Hasenzagl, Herkenhoff, Mongey, and Posner 2023a), we formally compute worker welfare gains under various HHI thresholds. To do so, we build on some of my coauthored earlier work (Berger et al 2022a) and develop a theory of monopsony in which firms potentially have common ownership of plants.
In the model, as in the real world, workers are free to move across markets, but they have different preferences or amenity values for firms (for example, think commute distance). These are akin to mobility or switching costs.
Importantly, we define a market using the NAICS3 industry code within a commuting zone. Think of real estate in Minneapolis as an example.. We also look at data from other countries in which occupation data and industries are available side-by-side (Berger et al 2023c), to show that both occupation- and industry-based market definitions yield similar “leakage rates.” We also show that our industry-based labor market definition satisfies the hypothetical monopsonist test, meaning that a firm that controls the market would impose more than a 5% total compensation cut.
We estimate the model on U.S. Census data and show that it replicates the path of employment and wages post-merger reported by David Arnold. In particular, we find that the post-merger change in employment is -14% in the data and -9% in the model. And we have larger post-merger earnings losses in highly concentrated markets.
We then simulate a set of mergers based on Arnold (2019) and compute worker welfare for various Merger Guidelines. We assume the mergers generate the usual “ad-hoc” 5% merger efficiency gains (i.e. the post-merger firm is 5% more productive).
We then apply the HHI thresholds of the 1982/2023 and 2010 Guidelines to our simulated mergers. First, we simulate the 2010 Guidelines in which mergers are prevented above a threshold of 2500. We find that in markets where mergers are permitted, worker welfare losses are on average $36,000 per affected market, and in markets where mergers are prevented, the Guidelines avoid large million dollar losses to workers. We then apply the 1982/2023 concentration threshold of 1800. We find that workers in labor markets in which mergers are permitted have welfare gains worth $20,000 per market, on average.
The 2023 Guidelines further state that “Labor markets frequently have characteristics that can exacerbate the competitive effects of a merger between competing employers.” We apply our model to test whether tighter thresholds should be applied to labor markets over and above product markets. We find that imposing an even tighter HHI threshold of 1500 results in worker welfare gains of 42,000 per affected market. While we demonstrate the potential gains to workers from tighter thresholds, we do not propose an “optimal” merger review threshold. To find the actual optimum the costs of merger enforcement must be weighed against the gains, and we have little information on the costs.
While labor markets have historically been quite fluid in the U.S., that has changed over the last 30 years, and nearly all metrics for labor market mobility have fallen since the 1990s (see Hyatt and Spletzer 2013), and the trend in work-from-home arrangements is slowly reverting to pre-pandemic levels. Corresponding quantitative work suggests that lower mobility has contributed to greater monopsony power (see Bagga 2023). In our own work, we find that roughly 10% of markets in the U.S. are “single-firm markets,” meaning one company controls the jobs in that region. Additionally, in those single-firm markets, wages are reduced by 70% of what they would be if there was more job competition (Berger, Hasenzagl, Herkenhoff, Mongey, and Posner, 2023a). The large estimated wage markdowns in the U.S. economy suggest that stricter antitrust enforcement may be an effective intervention in labor markets. Our model simulations confirm this by showing that the lower HHI threshold for merger reviews in the 2023 draft guidelines — when applied to labor markets — will benefit workers and prevent monopsony power. Thus, stricter merger policy will promote labor market competition and better wages for workers.
Articles represent the opinions of their writers, not necessarily those of ProMarket, the University of Chicago, the Booth School of Business, or its faculty.
