Monopsony Takes Center Stage

Bringing the powerful weapons the federal competition authorities have to bear on the problem of monopsony would be a substantial, but necessary departure from recent practice.



Marshall Steinbaum
Marshall Steinbaum

In October, the Council of Economic Advisors released a report about monopsony in the labor market. That alone was rather astonishing—employer power and its consequences for labor market outcomes has been a distinctly minority concern in the economics profession for quite a while, notwithstanding mounting evidence of its importance coming from a number of subfields.


For that agenda to gain a hearing at the apex of economic policy-making is evidence of the shifting ground in matters of public economic debate. It is also reminiscent of the last time inequality was so high: then, as now, it sparked a sea change in the economics profession, including both the mainstreaming of labor exploitation as a subject of economic research and the founding of the American Economic Association.


The key arguments in the CEA’s paper draw on the evidence of rising inequality in firm-specific wages I referred to in my last ProMarket post. A key determinant of earnings is which firm a worker can gain access to, and a strategy for consolidating earnings at the top of the income distribution is outsourcing labor to subordinate firms so the circle of workers who get paid by the most profitable ones narrows to high earners, who are thus earning more than they have in a century. Interfirm inequality and firm-specific wages are canonical evidence of an uncompetitive labor market. After all, if workers were able to move freely between firms, that should equalize the pay workers of similar experience and education obtain across firms within an industry or geography. 


And yet, the further into the labor market you drill down, to workers in narrowly defined skill and experience groups, the more residual inequality is apparent—until you take into account the firms where they work.


Declining labor mobility is further evidence of monopsony, especially given that it seems to be driven by the declining arrival rate of outside job offers and flattening earnings-tenure profiles. In my article on the subject with Mike Konczal, we showed that declining ‘dynamism’ is not driven by restrictions on labor supply like occupational licensing, but rather by slack labor demand—the classic symptom of monopsony power. And the consequences are concerning: not just rising inequality, but stagnant earnings over the course of careers and declining entrepreneurship and employment growth at young firms.


The CEA brief focuses on the most obvious manifestations of unequal power in the labor market to come to light in recent years: noncompete clauses, which have extended their reach into sectors where they once were unknown. To teachers in charter schools, for example, which otherwise pride themselves on introducing competition into public education. Evidence suggests noncompete clauses deter worker job search, which is critical for wage gains over the course of a career. And they are imposed by employers even where they are legally unenforceable—in other words, the threat is what matters, and unlike in a competitive labor market, these restrictions on worker autonomy are not compensated by higher wages or any other compensating differential. The very fact that they are used and affect behavior where they have no legal basis is evidence for monopsony, because that implies wage and compensation indeterminacy within an employment match.


The CEA brief also draws attention to the mere fact of concentration in the economy in general and in the share of employment in the largest firms. But those summary statistics in fact bely the reality of labor market monopsony, because there is a wage premium attached to the largest firms—though that premium has been declining for the lowest-paid workers, suggesting that the threat of outsourcing labor is disciplining the wage demands of workers at the greatest risk of being outsourced. Some suggest the fact of a large-firm wage premium is prima facie evidence against monopsony, but that interpretation ignores the strategic behavior around who gets to be part of the firm and who is pushed out. Studies of labor outsourcing events show that they do little other than reduce wages—thus employer power manifests precisely by excluding workers from firm-earnings premia.


The brief also brings up the issue of occupational licensing, which the CEA covered in a previous report and which adds a similar anticompetitive flavor to the labor market as monopsony. The problem is that it goes in the opposite direction: the supposed threat of occupational licensing is that it accrues too much power to incumbent, licensed workers, increasing their wages at the expense of employers and would-be competitors. My aforementioned paper presents evidence against occupational licensing as having a major impact on the labor market—very possibly because the larger problem of monopsony counteracts its effects.


So what is to be done? The CEA has already called for a broad inter-agency competition review, but that tends to be sector specific, not focused on the lack of competition in the labor market. In response to the paper, Adam Ozimek suggested that unionization be made a condition for merger approval as a means of balancing employer and worker power—a radical proposal relative to existing competition policy. But that points to the latent power of antitrust law to ensure competition with a wide remit for interpreting harm and imposing remedies. Renata Hesse, the acting Assistant Attorney General for Antitrust, said as much in a recent speech:


The legislative history of the Sherman Act makes it clear that the antitrust laws were intended to benefit participants in the American economy broadly—not just in their capacity as consumers of goods and services.  Senator Sherman said in promoting the Act that one of the problems with monopoly is that: “[i]t commands the price of labor without fear of strikes, for in its field it allows no competitors.” So a merger that gives a company the power to depress wages or salaries or to reduce the prices it pays for inputs is illegal whether or not it also gives that company the power to increase prices downstream.


So while mandating unionization would be a major departure, mitigating the threat of monopsony should absolutely be incorporated in merger review, where, for example, the merged entity could be enjoined from outsourcing labor. And beyond that, several specific ideas related to bringing competition policy to bear in a monopsonistic, fissured labor market follow:


  • Noncompete clauses would be viewed by competition authorities as vertical restraints, unlike the horizontal agreement between large tech-sector employers in Silicon Valley that was the subject of a Justice Department suit in 2010. But a dense enough concentration of noncompete clauses among workers in a given sector, market, or job description could amount to a horizontal agreement among employers to close off hiring opportunities in the relevant labor market. Federal authorities have thus far been deferential to them as a matter for state regulation, but with mounting evidence of their aggregate impact and the strong remedies available under federal statute, there’s no reason to continue with that hands-off policy.


  • Mandatory arbitration agreements and class action waivers may themselves be evidence of a monopsonistic labor market, because like noncompete clauses, they remove legal rights and market options from workers without a compensating differential, and overall, they serve to transfer wealth up the wealth distribution. But the most immediately actionable aspect of them is that they close off the option of private antitrust enforcement through civil litigation, which is a critical component of U.S. competition policy.

Economists tend to view arbitration as instantiating the Coasian preference for bilateral negotiation as opposed to regulation, notwithstanding that the Coase Theorem manifestly does not apply to non-price contract terms in the presence of unequal bargaining power—exactly the context at issue in evaluating monopsony. The Supreme Court has nonetheless increasingly read the Federal Arbitration Act, a relatively minor piece of 1920s legislation in its own time, as a super-statute that precedes rights guaranteed by other laws, like labor organizing, non-discrimination, or, in this case, competition.


The policy proposal here is, first of all, to roll back those judicial opinions. But in coordination with other relevant agencies with statutory authority to regulate bilateral contracts, like the National Labor Relations Board and the Consumer Financial Protection Bureau, the competition authorities could participate in a regulatory procedure to challenge the use of such agreements throughout the economy, on the understanding that they are both anti-competitive themselves and they preclude the important private litigation channel for the enforcement of competition policy.


  • Competition law could also play a role in regulating the fissured workplace, as indicated by the private antitrust lawsuit against Uber and its CEO. The idea is that classifying workers as independent contractors moves that vertical relationship into the purview of antitrust and the law of vertical restraints, where restrictions like requiring contractors to accept orders, fixing the prices they can charge, and mediating the terms on which they are paid by customers become anti-competitive. The core issue here is that fissuring is a strategy for avoiding labor law, but that should not create a regulatory black hole. Instead, the competition authorities should enforce the guarantee of market access on fair terms contained in the antitrust laws as they were first conceived.


Bringing the powerful weapons the federal competition authorities have to bear on the problem of monopsony would be a substantial departure from recent practice. Insofar as the competition authorities care about the labor market at all, they tend to focus on restrictions on supply rather than demand, and at a recent FTC microeconomics conference, panelists were far more comfortable with the idea of reviewing state-level occupational licensing rules than they were confronting the other, more empirically grounded issues in the CEA’s monopsony brief. But as Hesse said in her speech, the antitrust laws were meant to oppose market power wherever it may reside. For the most part, we have the legal structures in place to make the labor market more competitive—the only question is whether we have the political will.


(Note: Marshall Steinbaum is Senior Economist and Fellow at the Roosevelt Institute researching the labor market, inequality, higher education, and student debt.)


Disclaimer: The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.