All Policy Briefs

Policy Brief 12: Antitrust and Labor Market Power

PDF Version

José Azar, Ioana Marinescu, and Marshall Steinbaum

May 2019

Some of the material in this brief draws on proposals made in our other policy-facing work: Ioana Marinescu and Herbert Hovenkamp, “Anticompetitive Mergers in Labor Markets,” Faculty Scholarship at Penn Law, February 20, 2018; Ioana Marinescu and Eric A. Posner, “A Proposal to Enhance Antitrust Protection Against Labor Market Monopsony,” Working Paper (Roosevelt Institute, 2018); Marshall Steinbaum, “A Missing Link: The Role of Antitrust Law in Rectifying Worker Power in Our High-Profit, Low Wage Economy,” Issue Brief (Roosevelt Institute, 2018); Marshall Steinbaum and Maurice E. Stucke, “The Effective Competition Standard: A New Standard for Antitrust,” University of Chicago Law Review, Symposium on the Chicago School of Antitrust, 2019; Marshall Steinbaum, “Principles for Antitrust Legislation in the 116th Congress,” Take Care(blog), January 5, 2019; Marshall Steinbaum, “Antitrust, the Gig Economy, and Labor Market Power,” Law and Contemporary Problems, 2019.We also draw on the paper by Suresh Naidu, Eric A. Posner, and E. Glen Weyl, “Antitrust Remedies for Labor Market Power,” Harvard Law Review, 2018.But as a coherent agenda for antitrust enforcement in labor markets, this brief is original.

 

 

Starting with the Chicago School’s influence in the late 1970s and 1980s, antitrust enforcement has been weakened under the assumption that market power is justified by economic efficiency. While consumers are the main focus of antitrust enforcement, the weakening of antitrust enforcement has likely also adversely impacted workers, thus contributing to increasing inequality.

In this brief, we outline elements of an antitrust reform agenda aimed at reversing the weakening of antitrust enforcement, insofar as it pertains to and has strengthened the power employers have to set wages and working conditions for their workers, without countervailing power on the part of workers, who have limited ability to leave for another job in order to increase their pay.

This brief is organized as follows. We first summarize a theory of labor market monopsony that can explain a number of otherwise-puzzling facts about the labor market and workers’ status in it, including stylized facts such as a negative relationship between employer concentration and earnings, inter-firm earnings inequality, and declining job-to-job mobility. We then outline an antitrust policy agenda that speaks to various aspects of employer power in labor markets: the consumer welfare standard, measuring market power for antitrust purposes, anti-competitive conduct in labor markets such as noncompete clauses and no-poaching agreements, mergers that harm workers as sellers of labor, monopsonization of labor markets as a violation of Section 2 of the Sherman Act, and, lastly, the potential for countervailing collective power on the part of workers.

The Theory and Empirics of Labor Market Monopsony

Monopoly power is the ability of an individual seller to control all or a substantial part of the market and thereby dictate the terms of trade, including charging prices in excess of its marginal cost and imposing disadvantageous non-price provisions on its customers. Monopsony is the mirror image of that—the ability of buyers to dictate prices (wages, in the labor context), without fear that many of their workers will leave for another job, or to dictate working conditions and terms of employment that transfer some of the value created by the employer-employee relationship to the employer.

Monopsony literally refers to a single buyer in a market, but monopsony power in labor markets can and does arise in less stark conditions: when potential employers are few, when the process of finding another job is costly or a worker is tied to his or her current job by family commitments or the need for health insurance or other job-related benefits. Under such circumstances, employers are able to profitably pay their workers less than their contribution to production (marginal productivity): while some workers quit in response to such exploitation, enough workers remain to make wage suppression profitable.

The basic underlying mechanism in a simple model of labor monopsony is that individual firms face an upward-sloping labor supply schedule. This contrasts with the perfectly competitive case, when individual firms face infinitely elastic labor supply. In the latter, a tiny reduction in the wage one firm pays will result in all its workers leaving. Under monopsony, on the other hand, pushing wages down results in less than all of them leaving for competing employers. The equilibrium of a single labor market in which employers have monopsony power will consist of a wage set below workers’ marginal revenue product of labor, since employers can get away with paying workers less than they earn for the firm without having many of those workers depart. It will also lead to lower labor demand, and therefore lower employment, relative to the competitive case. The employer earns a profit on each worker, namely, the difference between the value of what each worker produces and his or her cost in terms of wages. This profit is called the “markdown” or “exploitation” in the monopsony literature. There is also, in general, excess labor supplied to individual firms, and some workers remain unemployed (or work fewer hours than they are willing to).

A more complicated theory of monopsony involves heterogeneous employers with varying productivity per worker. More productive firms tend to both be larger and to pay more. But in a competitive labor market, just as in a competitive product market, the most productive firm in a given market would be expected to employ all workers, and there would be no inter-firm wage inequality for homogeneous workers because they would all be working at the single active firm in that market. Inter-firm wage inequality arises under monopsony if firms in the market all face an imperfectly elastic labor supply. The most productive firm pays more than others can afford to because workers there are more productive and therefore worth more to the employer, but that employer also has the most wage-setting power and therefore the ability to pay wages with the greatest markdown below marginal productivity.[1]This dynamic of more productive firms paying higher wages but also enjoying more monopsony power gives rise to earnings inequality across firms employing similar workers, as well as a firm-size wage premium.[2]

Finally, with both homogeneous and heterogeneous firms, a decreasing arrival rate of job offers (or, alternatively, a more frictional search-and-matching process) will reduce the rate of job transitions for workers. The more difficult it is to obtain an outside offer, the more wage-setting power current employers have, and the greater the markdown of wages below marginal productivity.

In general terms, these theoretical models predict broad labor market patterns that have been documented in the empirical literature: finite, and low, labor supply elasticities to the individual firm,[3]a negative concentration-earnings relationship within a given labor market,[4]inter-firm earnings inequality for similar workers,[5]declining job-to-job transition rates thanks to the infrequency of outside job offers,[6]and a flattening earnings-tenure relationship for individual workers who remain in the same job, since they are unable to obtain the outside job offers that would induce their employers to bid to retain them.[7]These particular findings accompany the time trends in labor market aggregates consistent with declining worker power relative to employers: rising earnings inequality[8], the divergence of the median wage from average productivity per worker[9], and, more recently, the decline of the labor share of GDP.[10]It is these facts that have motivated the debate within and outside academic economics about monopsony power in labor markets, and which motivate the policy agenda we set out below.

The Consumer Welfare Standard

Part of the revolution in antitrust law that took place as a result of the Chicago School was the adoption of the ‘consumer welfare standard,’ namely, the idea that harm to competition within the legal meaning of the antitrust laws corresponds to harm to consumers and their welfare—consumer surplus in the most straightforward economic application. This idea is manifested in the phrase “antitrust protects competition, not competitors.”[11]

What this phrase refers to is the idea that antitrust might itself be anti-competitive, because it has the potential to be put to use by incumbents to suppress rather than to promote competition. In this theory, incumbents might use the legal system to protect their market share from innovative entrants, by claiming that conduct that challenged that market share on the merits, for example by introducing new distribution technologies that reduce the costs of production or eliminated unnecessary middlemen, violate the antitrust laws through exclusion or some other means, when in fact they represent the kind of competition antitrust should be promoting rather than punishing. The Chicago School critique of mid-century antitrust held that many antitrust cases were opportunistic attempts to impede the economy’s natural creative destruction, and thus threatened aggregate welfare by reducing the competition on which economic progress depends. Therefore, we should not measure harm to competition by whether the ostensible victim loses market share or profits, but rather by whether consumersare made worse-off. If they are not, then the presumption is that whatever conduct is being challenged is ‘competition on the merits’ and should not be illegal. Would-be private antitrust plaintiffs have to assert this type of “antitrust injury” in order for their case to survive, and in many instances litigation is decided based on econometric predictions about consumer price effects.

Such a legal apparatus has overlooked harm to workers and monopsony power in labor markets. There has never been a merger challenged premised primarily on harm to competition in labor markets.[12]It was only in 2016 that the Justice Department and the FTC issued “Guidance for Human Resources professionals” that warns against collusion in the form of agreements not to poach workers.[13]Recently, the Justice Department has retreated from that strong enforcement stance by claiming that no-poach agreements in franchising contracts (standardized contracts between a franchisor and each of its many franchisees) are not necessarily illegal, because franchising contracts are vertical restraints and thus subject to a lower standard of legal liability than agreements between competitors, i.e. employers hiring from the same labor market.[14]What the DOJ overlooks in making the case against the automatic illegality of franchising no-poach agreements is that the reasons for weaker enforcement against vertical restraints derive from their ostensible benefits for consumers. There’s no plausible benefit to workerswhose employment options are limited by contractual restrictions on franchisees against hiring them elsewhere in the network where they work, just as there’s no plausible benefit to consumers from contracts that forbid alternative sellers other than the one they currently patronize from selling to them.

This particular question of the legal status and standard for review for franchising no-poach agreements is telling evidence that the existing antitrust enforcement regime, based as it is on the consumer welfare standard, is inadequate to the question of policing anti-competitive structure and conduct in labor markets, which, as the economic evidence recounted in the previous section makes clear, is pervasive. For that reason, we make the following recommendations for amending antitrust laws generally in order to increase enforcement against labor market monopsony:

  • Policy-makers should make clear that the antitrust laws protect competition in both labor markets and product markets, and that documenting increases in consumer prices is **not** necessary to prove harm to competition within the meaning of the antitrust laws.
  • Reductions in wages, wage shares (as a percentage of firm revenue), employment, hiring, or job quality should be prima facieevidence of harm to competition within the meaning of the antitrust laws and cannot be traded off or weighed against price or output effects in antitrust analysis.
  • It has become standard for antitrust analysis to include a component in which defendants can claim that whatever conduct, merger, or market structure is being challenged as harmful to competition has countervailing economic benefits in the form of “efficiencies.” For example, if a merger causes a company to have greater price-setting power in output markets, an offsetting efficiency in the form of a reduction in the cost of production might have a countervailing effect on the final price of output to consumers, and so on net that merger would not be anti-competitive and therefore not illegal.

The scope for such efficiencies claims has been narrowed in some recent cases, for example in the Justice Department’s successful cases against the mergers of the health insurers Aetna and Humana and Anthem and Cigna. But policy-makers should go further:

  • The anti-competitive exercise of additional monopsony power in labor markets is not efficient and should not be considered an “efficiency” for antitrust purposes, even if it leads to a reduction in cost of production.
  • More work needs to be done to distinguish productive efficiencies from the exercise of monopsony power. For example, an efficient consolidation of redundant accounting departments between merged firms might reduce wages if the market for accountants where the merging parties hire is monopsonized. Whether that qualifies as a cognizable efficiency should depend on its welfare effect in the market in question, but how to operationalize that in antitrust practice remains to be investigated by future scholarship.
  • Conduct that is evaluated under the Rule of Reason when assessing whether it illegally harms competition in output markets, such as restrictions on competition in franchising contracts, should not necessarily and automatically be assessed under the Rule of Reason for its effect on labor markets. Many of the economic claims that formed the basis for courts to adopt the Rule of Reason effectively treated perfect competition in labor markets as a given.[15]It is therefore in error to assume the same logic applies in labor markets.

Measuring Market Power

Many mergers and antitrust conduct cases hinge on whether the would-be defendant possesses market power, for the sound reason that actions that would have the effect of reducing competition in markets in which incumbents have market power are likely to have a different motivation, and different impact, in markets where incumbents do not possess significant market power. In antitrust practice, market power has come to be equivalent in most applications to market concentration. The Horizontal Merger Guidelines establish concentration thresholds above which a merger is considered to be likely to reduce competition, and monopolization caselaw has established (different) concentration thresholds for adjudicating market power for the purpose of assessing liability for unilateral conduct. In practice, the assessment of market power becomes an exercise in market definition: how large or small is the relevant antitrust market, and therefore how much market share do the incumbents (whether the would-be defendant or its competitors) enjoy in that market? Define the market expansively enough and no one has market power in that market because no one’s market share is high enough.

This narrow conception of how to measure (and litigate) market power fails to take into account economic evidence that incumbent firms have market power, and in particular, that employers possess market power in labor markets. Concentration in an antitrust market may not imply market power,[16]and conversely, lack of concentration in an antitrust market is quite consistent with incumbents’ possessing market power.[17]For that reason, we propose that policy-makers consider expanding the indicia of market power available for use in antitrust cases pertaining to labor markets:

  • A market share of over 50% of employment (or alternatively, of posted job vacancies) in a well-defined antitrust labor market.[18]
  • The ability to lower wages below what would be charged in a competitive market.
  • The ability to wage-discriminate, that is, to pay similar workers working in the same market significantly different wages.
  • The ability to impose disadvantageous non-wage contractual terms on workers without compensation.

Anti-competitive Conduct in Labor Markets

The 2016 Guidance for Human Resources Professionals is a useful jumping-off point for anti-competitive conduct in labor markets, but it has certain weaknesses deriving from the fact that it operates in the shadow of judicial rulings that constrain enforcers’ ability to crack down. As the recent DOJ Statement of Interest in the franchising no-poach case shows, there’s still ample room for employers to dodge antitrust liability by availing themselves of the legal formalisms already granted deference, such as vertical restraints. That is why noncompete agreements, which are contracts between employers and workers preventing workers from taking alternative employment, have become so pervasive.[19]

In fact, the DOJ’s no-poach case against prominent Silicon Valley employers of software developers showed that those employers likely made use of legally-dangerous no-poach agreements precisely because California employment law took the noncompete option off the table, leaving them with the no-poach option that ultimately brought them into contact with federal antitrust law.

In Congressional testimony in December 2018, FTC Chairman Joseph Simons answered a question from Rep. Jerrold Nadler by saying that his agency was investigating noncompete clauses, but “there’s a lot of circumstances where the company that is imposing the non-compete doesn’t have market power and it would be difficult for us to reach that under the antitrust laws.” This statement is curious, because imposing a noncompete clause without compensation (as is usually the case) is itself evidence that employers have market power in labor markets.

For those reasons, antitrust law should be amended to ensure that employers with labor market power do not further harm competition in the labor market. In particular, for employers that have market power, the following should be illegal[20]:

  • Noncompete clauses and no-poaching agreements.[21]
  • Restrictions on sharing information about wages and working conditions among workers or job applicants.
  • Reclassification of employees as independent contractors. This has been shown to be a mechanism for exercising employers’ market power against workers and thereby reducing wages.[22]
  • Mandatory arbitration clauses and class action waivers in employment contracts.

Merger Review

Labor markets are already highly concentrated. Therefore, we would expect that mergers that by their nature further reduce competition in labor markets might have an adverse impact on workers. And yet, as stated above, the antitrust enforcers have never challenged a merger on the grounds that it would reduce competition in labor markets. In October 2018, the FTC chairman testified to Congress that the agency staff had been instructed to look at labor market impact for every merger they review, but thus far that has not been manifested in any agency enforcement action.[23]

The economic analysis in a typical merger review proceeds by defining antitrust markets likely to be affected by the merger and simulating or otherwise predicting the merger’s effect in those markets. In general terms, this is done by estimating the demand curve facing the merging parties (presuming they compete as sellers in the market) and then predicting how much, if at all, the combined entity (or its competitors) would be able to increase price. Then that estimate is balanced against any merger-specific efficiencies that might serve to reduce cost and therefore exert downward pressure on consumer prices.

Merger review in labor markets could be done in the same way, in broad terms, with the object of predicting downward pressure on wages or the worsening of conditions for workers resulting from increased monopsony power on the part of the merging parties (or their competitors). Therefore, we recommend the following:

  • The agency merger review process should be expanded to include analysis of competitive effects in labor markets, including the augmentation of agency resources in order to staff such an increase in the substance of merger review.
  • For the purpose of merger review, it makes sense for enforcers to begin by defining labor markets by commuting zones and 6-digit Standard Occupational Code. As we show in other work, this market definition is likely to be conservative in that a monopsonist in a labor market defined even more narrowly would likely find it profitable to impose a wage reduction without significant loss of workers—given what we know about low firm-level labor supply elasticities.[24]Therefore, the burden of proof would be on the merging parties to show that the labor markets from which they hire are in fact broader than that standard market definition.

Monopsonization

Although the Sherman Act has been held to pertain to both buyer and seller market power and its abuse, as with mergers, there has never been a monopsonization case focused on control over labor markets. Given the prevalence of monopsonistic conditions in labor markets, we think enforcement in this area is overdue. Moreover, monopolization jurisprudence under Section 2 of the Sherman Act has become unwieldy and over-burdensome to plaintiffs following the US v. Microsoftlitigation, so in order to give life to any monopsonization enforcement regime, procedural burdens need to be streamlined and the definition of market power widened to take account of evidence of prevalent monopsony power in labor markets, as documented in this brief.

To that end, we make the following recommendations:

  • The plain language of the Sherman Act should be augmented to read

It shall be unlawful for any employer engaged in commerce, in the course of such commerce, to monopsonize, attempt to monopsonize, or combine or conspire with any other person or persons to monopsonize, a labor market.

  • For the purpose of monopsonization, just as in merger review, the rule of thumb for labor market definition should be a 6-digit SOC code by commuting zone. This rule of thumb could be modified with evidence that the labor market over which a hypothetical monopsonist could impose a wage reduction is either wider or narrower.
  • Evidence of market power in a labor market would consist either of a significant market share in a market defined as above, or direct evidence that an employer can lower the wages of its employees below what would be charged in a competitive market, impose disadvantageous contractual terms on workers, or wage-discriminate.
  • Proof of monopsonization would consist of both establishing market power (according to any one of the list of indicia of market power outlined above) and anti-competitive acts to extend or maintain that market power. Anti-competitive acts for the purpose of assessing monopsonization liability in an antitrust labor market would include but are not limited to
    • An anti-competitive merger.
    • The use of non-compete clauses or no-poaching agreements.
    • Non-disclosure agreements pertaining to the terms of employment.
    • Unfair labor practices as defined by the National Labor Relations Act.
    • Employment misclassification.
    • Class action waivers and/or mandatory arbitration clauses.
    • Any other action that has the effect of significantly reducing competition in the labor market, for example fixing wages or wage discrimination.
  • Monopsonization damages and remedies should be the same as under the existing Section 2 of the Sherman Act, and workers (both statutory employees and independent contractors) and labor organizations, as well as competing employers victimized by a competitor’s anti-competitive conduct in labor markets, should have standing to litigate, in addition to public enforcers.

Countervailing Power

The prevalence of wage-setting power on the part of employers invites the remedy of countervailing power in the form of worker organizations and collective bargaining. Indeed, antitrust has recognized this since the Clayton Act exempted “the labor of a human being” from the antitrust laws, following the use of antitrust enforcement actions to end strikes in 1892, 1894, and 1908.[25]But the antitrust exemption for labor eventually came to be tied to the statutory employment relationship through legislation and caselaw in the 1930s and early 1940s. What that has meant is that as statutory employment has receded and employers become more adept at placing their workers in the “independent contractor” category,[26]the exemption from antitrust for organizing activity among workers has receded as well.[27]

This was seen most recently in the ongoing antitrust litigation against Seattle for permitting ridesharing drivers to bargain collectively, despite their non-employee status. The Chamber of Commerce sued the city under the Sherman Act, and the DOJ and FTC filed an amicus brief siding with the Chamber and hinting that in the absence of the labor exemption, collective bargaining by ridesharing drivers would be a per seviolation of the Sherman Act.[28]The Ninth Circuit Court of Appeals sided with the Chamber and the federal agencies, forcing the city to revise the ordinance to rule out collective bargaining over wages. That concession was still not sufficient to satisfy the Chamber, which, in renewed filings, declared that the ordinance remains a collective group boycott rendered illegal by the Sherman Act.

Antitrust’s treatment of collective bargaining by ridesharing drivers contrasts with its treatment of ridesharing platforms, which have thus far escaped antitrust scrutiny for price- and wage-fixing behavior that authorities believe to be per-se illegal when undertaken by drivers. The one private action alleging price- and wage-fixing by Uber itself to survive a motion to dismiss was later sent to arbitration thanks to Uber’s mandatory arbitration clause.[29]But public enforcement authorities are not bound by any mandatory arbitration clause.

  • To address these issues, policy-makers could consider extending the antitrust labor exemption to workers who lack traditional employee status under the National Labor Relations Act.
  • Further, public enforcers should consider the antitrust implications of the gig economy platforms’ use of the independent contractor classification.[30]In particular, they should investigate whether business models that consist of coordinating and setting prices and terms of trade for the provision of services by independent contractors violates Section 1’s prohibition on anti-competitive restraints adopted through concerted or joint action among multiple entities.[31]

Conclusion

This brief summarizes the theory and empirics of labor market monopsony and applies the findings from that research agenda to antitrust policy. Under the consumer welfare standard, antitrust has de-prioritized issues of labor market power and anti-competitive conduct and market structures that profit by suppressing wages and worsening working conditions. The recommendations made in this brief would go a long way toward reversing that unjustified imbalance between antitrust enforcement in the product and labor market.

Endnotes

[1]It’s important to note that in a monopsonized labor market with heterogeneous firms, a negative concentration-earnings relationship exists alongside a firm-size wage premium. This contrasts with some naïve criticism of the importance of labor market monopsony, to the effect that a firm-size wage premium is evidence againstemployer power in labor markets. In fact, this evidence is consistent with monopsony power.

[2]A model of this type is presented in Daniel Berger, Kyle Herkenhoff, and Simon Mongey, “Labor Market Power” (Working Paper, 2019).A similar firm-size wage relationship can be generated by ex-ante homogeneous firms in a wage-posting game such as in Kenneth Burdett and Dale Mortensen, “Wage Differentials, Employer Size, and Unemployment,” International Economic Review39, no. 2 (1998): 257–73.

[3]Douglas A. Webber, “Firm Market Power and the Earnings Distribution,” Labour Economics35 (2015): 123–34; Arindrajit Dube et al., “Monopsony in Online Labor Markets,” American Economic Review: Insights1, no. 1 (2019); Arindrajit Dube, Laura Giuliano, and Jonathan Leonard, “Fairness and Frictions: Impact of Unequal Raises on Quit Behavior,” American Economic Review, no. 9149 (2018).

[4]José Azar, Ioana Marinescu, and Marshall Steinbaum, “Labor Market Concentration,” NBER Working Papers, no. 24147 (2017); Efraim Benmelech, Nittai Bergman, and Hyunseob Kim, “Strong Employers and Weak Employees: How Does Employer Concentration Affect Wages?,” NBER Working Papers, no. 24307 (2018); Kevin Rinz, “Labor Market Concentration, Earnings Inequality, and Earnings Mobility,” CARRA Working Paper Series, no. 2018–10 (2018); Yue Qiu and Aaron Sojourner, “Labor Market Concentration and Labor Compensation” (Working Paper, 2019).

[5]Jae Song et al., “Firming Up Inequality,”Quarterly Journal of Economics134, no. 1 (2019): 1–50.

[6]Raven Molloy et al., “Understanding Declining Fluidity in the U.S. Labor Market,” Brookings Papers on Economic Activity, 2016, 183–237; Mike Konczal and Marshall Steinbaum, “Declining Entrepreneurship, Business Dynamism, and Labor Mobility: A Demand-Side Approach” (Roosevelt Institute, 2016); Ryan Decker et al., “Where Has All the Skewness Gone? The Decline in High-Growth (Young) Firms in the U.S.,” NBER Working Papers, no. 21776 (2015).

[7]Henry Hyatt and James Spletzer, “The Shifting Job Tenure Distribution,” Labour Economics41, no. 1 (2016): 363–77.

[8]Thomas Piketty, Emmanuel Saez, and Gabriel Zucman, “Distributional National Accounts: Methods and Estimates for the United States,” Quarterly Journal of Economics133, no. 2 (2018): 553–609.

[9]“The Productivity-Pay Gap,” Economic Policy Institute, August 2018, https://www.epi.org/productivity-pay-gap/.

[10]David Autor et al., “The Fall of the Labor Share and the Rise of Superstar Firms,” NBER Working Papers, no. 23396 (2017).

[11]Eleanor Fox, “We Protect Competition, You Protect Competitors,” World Competition26, no. 2 (2003): 149–65.

[12]Marinescu and Hovenkamp, “Anticompetitive Mergers in Labor Markets.”

[13]Federal Trade Commission and Department of Justice, “Antitrust Guidance for Human Resource Professionals,” October 2016.

[14]Boris Bershteyn et al., “DOJ Is Trying to Rein In Franchise No-Poach Suits,” Law360(blog), February 19, 2019, https://www.law360.com/foodbeverage/articles/1130056/doj-is-trying-to-rein-in-franchise-no-poach-suits.

[15]For example, maximum resale price maintenance has been assumed to benefit consumers by holding down prices of final output goods and preventing retailers from charging an excessive markup. Its effect on workers was never considered. “…We find it difficult to maintain that vertically imposed maximum prices could harm consumers or competition to the extent necessary to justify their per se invalidation.” State Oil Co. v. Khan, 522 U.S. 3 (1997).

[16]Harold Demsetz, “Industry Structure, Market Rivalry, and Public Policy,” The Journal of Law & Economics16, no. 1 (April 1973): 1–9.

[17]José Azar, Ioana Marinescu, and Marshall Steinbaum, “Measuring Labor Market Power Two Ways,” American Economic Review Papers & Proceedings, 2019; Dube et al., “Monopsony in Online Labor Markets.”

[18]See below for a discussion of market definition for labor markets in merger review, as well as Marinescu and Posner, “A Proposal to Enhance Antitrust Protection Against Labor Market Monopsony”; José Azar et al., “Concentration in US Labor Markets: Evidence from Online Vacancy Data,” NBER Working Papers, no. 24395 (2018).

[19]Evan Starr, J.J. Prescott, and Norman Bishara, “Noncompetes in the U.S. Labor Force,” December 24, 2017, https://ssrn.com/abstract=2625714; Sandeep Vaheesan, “Petition for FTC Rulemaking to Prohibit Worker Non-Compete Clauses” (Open Markets Institute, March 20, 2019), https://openmarketsinstitute.org/wp-content/uploads/2019/03/Petition-for-Rulemaking-to-Prohibit-Worker-Non-Compete-Clauses.pdf.

[20]Marinescu and Posner, “A Proposal to Enhance Antitrust Protection Against Labor Market Monopsony,”

[21]Vaheesan, “Petition for FTC Rulemaking to Prohibit Worker Non-Compete Clauses.”is a recent rule-making petition to the FTC made by the Open Markets Institute and co-signed by numerous other organizations and individual experts. It calls for administrative rule-making under Section 5 of the FTC Act to ban all non-compete clauses, regardless of whether the employer has market power.

[22]Arindrajit Dube and Ethan Kaplan, “Does Outsourcing Reduce Wages in the Low-Wage Service Occupations? Evidence from Janitors and Guards,” ILR Review63, no. 2 (2010): 287–306.

[23]Joseph Simons, “Oversight of the Enforcement of the Antitrust Laws,” § Senate Committee on the Judiciary (2018).

[24]Azar et al., “Concentration in US Labor Markets: Evidence from Online Vacancy Data.”

[25]In re Debs, No. 158 (US Supreme Court 1895); United States v. Workingmen’s Amalgamated Council of New Orleans, 54 U.S. 994 (United States Circuit Court for the Eastern District of Louisiana 1893); Loewe v. Lawlor, 208 US 274 (US Supreme Court 1908).

[26]David Weil, The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It(Cambridge, MA: Harvard University Press, 2014); David Weil, “Why We Should Worry About Monopsony,” Institute for New Economic Thinking(blog), September 2, 2018, https://www.ineteconomics.org/perspectives/blog/why-we-should-worry-about-monopsony.

[27]Sanjukta Paul, “The Enduring Ambiguities of Antitrust Liability for Worker Collective Action,” Loyola University Chicago Law Review47 (2016): 969.

[28]Marshall Steinbaum, “The Feds Side Against Alt-Labor,” Next New Deal(blog), November 16, 2017, http://rooseveltinstitute.org/feds-side-against-alt-labor/.

[29]Marshall Steinbaum, “Antitrust Implications of Labor Platforms,” CPI Antitrust Chronicle, 2018; Marshall Steinbaum, “Uber’s Antitrust Problem,” The American Prospect, May 11, 2016.

[30]The federal Department of Labor recently expressed approval of the independent contractor classification that is typical of gig economy labor platforms. Keith E. Sonderling, “US Department of Labor Wage and Hour Division Opinion Letter on FLSA Employment Classification for Gig Economy Workers,” Opinion Letter, April 29, 2019, https://www.dol.gov/whd/opinion/FLSA/2019/2019_04_29_06_FLSA.pdf.

[31]See Sanjukta Paul, “Antitrust as Allocator of Coordination Rights,” UCLA Law Review67, no. 2 (2020) for an extended discussion.

All Policy Briefs