By Russell Pittman (US Department of Justice)1
The issue of how to provide incentives for a private firm to invest in the knowledge and capabilities of its labor force is one that has intrigued economists since at least the time of Alfred Marshall. Marshall wrote in his Principles of Economics:
A good deal has already been said of the technical training of adults, of the decadence of the old apprenticeship system, and of the difficulty of finding anything to take its place. Here again we meet the difficulty that whoever may incur the expense of investing capital in developing the abilities of the workman, those abilities will be the property of the workman himself: and thus the virtue of those who have aided him must remain for the greater part its own reward.2
Gary Becker advanced this discussion by differentiating between “general” training – “useful in many firms in addition to the firm providing it” – and “specific” training – “training that increases productivity more in firms providing it” or, at the extreme, “has no effect on the productivity of trainees that would be useful in other firms.”3
Becker argued that firms would invest in the “general” training of their employees only if the employees absorbed the cost themselves with a contemporaneous lower wage, since otherwise competing employers could “free ride” on the expenditures of the firm providing the training by hiring away the trained workers. On the other hand, “specific” training, having little or no value to other employers, would not raise the opportunity wage of the workers, and so the firm providing the training could expect to reap the benefits from its expenditures on training in terms of higher worker productivity in the future.
Later authors have found that firms in fact do pay for general training, and that Becker’s strong prediction otherwise rests on an assumption of perfectly operating labor markets.4 In a world of imperfect labor markets, firms seem to pay for a variety of types of worker training, including training in skills potentially useful to their competitors, seeking to discourage ex post opportunistic behavior by trained workers through a variety of behavioral and contractual devices. Two such devices in particular seem to occur frequently and have attracted a good deal of attention by competition agencies and litigants in recent years.
The first is non-compete clauses in agreements between the employee and employer: contractual obligations for current employees to “refrain from accepting employment in a similar line of work for a specific period in a certain geographic area.”5 A variant is a contractual obligation to refrain from soliciting existing customers or clients upon moving to a new employer, also usually for a specific period of time.
Such clauses are a common component of employment contracts in the United States, from the least skilled to the highest paid workers. The most popular justification offered for them is the protection of the employer’s trade secrets as well as customer goodwill and relationships, but the maintenance of employer incentives for training and the development of human capital is also frequently put forth.
The record of the recent USFTC workshop on this topic suggests a number of relevant stylized facts:
- Non-complete clauses are present at least as often in contracts for low-skilled, low-wage workers as for higher-paid, higher-skilled workers;
- These contracts are usually “contracts of adhesion”: standardized contracts agreed to by the worker rather than the outcome of bilateral negotiations;
- In the U.S., the enforceability of such contracts varies a great deal by state, including what is perceived as virtual unenforceability in California;
- Even in states where they are not enforced, such contracts may be imposed on poorly informed workers and used as an effective threat if they consider leaving;
- The likely result is fewer job opportunities and lower wages, particularly for workers in the lower range of skills and wages.
A variety of reform proposals have been put forward, at the FTC workshop and elsewhere, to restrict the ability of firms to impose non-compete clauses on their workers, especially in those sectors and for those workers for which trade secrets, intellectual property, and extensive training are not obviously relevant – for example, fast food franchises.6 Proposals for FTC enforcement actions include those focusing on both the antitrust and consumer protection provisions of the FTC Act.
The second device of apparently widespread use and increasing enforcement interest is anticompetitive agreements among firms competing with each other as hirers (or “buyers”) in labor markets – in particular, wage-fixing and “no-poach” agreements. Wage fixing agreements are what they sound like – price fixing among competitors when the price fixed is the wage rate, at a particular level or within a particular range. No-poach agreements are agreements variously not to approach, solicit, or hire the employees of competitors.7
In this case an early treatment of the issue comes from a source even older than Alfred Marshall. In 1776, Adam Smith wrote:
We rarely hear, it has been said, of the combinations of masters, though frequently of those of workmen. But whoever imagines upon this account that masters rarely combine is as ignorant of the rule as of the subject. Masters are always and everywhere in a sort of tacit, but constant and uniform combination, not to raise the wages of labor above their actual price. To violate this combination is everywhere a most unpopular action, and a sort of reproach to a master among his neighbours and equals. We seldom, indeed, hear of this combination, because it is the usual, and one may say, the natural state of things, which nobody ever hears of. Masters, too, sometimes enter into particular combinations to sink the wages of labour even below this rate. These are always conducted with the utmost silence and secrecy….8
As U.S. enforcers have come to observe and understand an apparently increasing number of anticompetitive agreements of these types, they have made the points strongly that a) the antitrust laws are as concerned with anticompetitive behavior in purchasing (monopsony) as in selling (monopoly), and, in particular, b) the antitrust laws apply to labor markets just as much as they apply to markets for other commodities.9
Box 1: A Note on Monopsony. Some defenders of the use of monopsony power in labor and other markets argue that monopsony, by lowering the price of the input, results in a lower price of output downstream, and thus benefits final customers. This argument reflects a fundamental misunderstanding of microeconomic theory. Upstream market power may be accompanied by either downstream competition or downstream market power. A move to monopsony when downstream markets are competitive cannot affect downstream prices (by definition); its result is a misallocation of resources upstream, as willing suppliers choose to withdraw from this market in response to anticompetitively low prices and too little of the affected input is utilized.10 A move to monopsony by a firm with existing downstream market power results in the same upstream resource misallocation and reduction in quantity used of the affected input, and this generally translates into a downstream quantity reduction, and therefore an additional welfare loss.11
The Antitrust Guidance for Human Resource Professionals issued by the Antitrust Division and the FTC in October 2016 cites three cases brought by one or the other agency in the 1990’s against forms of wage-fixing and six cases brought by the Division since 2007 against forms of either wage-fixing or no-poach agreements. Most recently, as noted by Assistant Attorney General Makan Delrahim at the USDOJ’s workshop on competition issues in labor markets, the Division successfully challenged a no-poach agreement between the world’s two largest rail equipment manufacturers (U.S. v. Knorr-Bremse and Westinghouse Air Brake, 2018) and, bringing merger law into this arena, successfully challenged the proposed merger of two of the largest U.S. health care insurers, partly on the grounds of a likely increase in buying power over the reimbursement rates paid to physicians and other health-care providers (U.S. v. Anthem and Cigna, 2016).
One of the most interesting of the successful challenges to no-poach agreements came in the tech sector.
On September 24, 2010, the Antitrust Division filed a civil antitrust complaint, as well as a proposed settlement, that attacked various bilateral agreements among Adobe, Apple, Google, Intuit, and Pixar not to solicit each other’s employees – “no cold calls” agreements.12 The complaint alleged agreements not to solicit each other’s “specialized computer engineers and scientists” for employment among the following firms: Apple and Google, Apple and Adobe, Apple and Pixar, Google and Intel, and Google and Intuit. The complaint indicated that the agreements “were created and enforced by senior executives of these companies.” The proposed settlement “more broadly prohibits the companies from entering, maintaining or enforcing any agreement that in any way prevents any person from soliciting, cold calling, recruiting, or otherwise competing for employees.”13 The U.S. District Court for the District of Columbia issued its Final Judgment approving the proposed settlement on March 17, 2011.
Several aspects of this case reward a closer look.
First, in its Competitive Impact Statement the Division reaffirmed that Section 1 of the Sherman Act applies to monopsonistic as well as monopolistic behavior.
There is no basis for distinguishing allocation agreements based on whether they involve input or output markets. Anticompetitive agreements in both input and output markets create allocative inefficiencies. Hence, naked restraints on cold calling customers, suppliers, or employees are similarly per se unlawful.
Second, though the complaint claimed a per se violation of Section 1 of the Sherman Act, it was brought as a civil rather than a criminal allegation. On some occasions, the Division has brought civil actions before bringing subsequent criminal actions for similar behavior. For example, the Division’s first criminal prosecutions for price fixing among health care professionals – U.S. v. Alston (1994), against dentists in Tucson, and U.S. v. Lake Country Optometric Society (1994), against optometrists in Texas – followed earlier similar cases brought as civil enforcement actions.14 At the DOJ labor workshop, Assistant Attorney General Delrahim reaffirmed his intention to bring criminal cases for per se violations in this area going forward.
Third, again as made clear in the Competitive Impact Statement, agreements that restrict competition may under some circumstances be evaluated under the “rule of reason” rather than the per se rule.
[A]n agreement that would normally be condemned as a per se unlawful restraint on competition may nonetheless be lawful if it is ancillary to a legitimate procompetitive venture and reasonably necessary to achieve the procompetitive benefits of the collaboration. Ancillary restraints therefore are not per se unlawful, but rather evaluated under the rule of reason, which balances a restraint’s procompetitive benefits against its anticompetitive effects. To be considered “ancillary” under established antitrust law, however, the restraint must be a necessary or intrinsic part of the procompetitive collaboration. Restraints that are broader than reasonably necessary to achieve the efficiencies from a business collaboration are not ancillary and are properly treated as per se unlawful. (Citations omitted)
The Division determined that while various of the Defendant companies did indeed engage in “legitimate collaborative projects,” the no-cold-call agreements were not “tied to any specific collaboration, … narrowly tailored to the scope of any specific collaboration,” or otherwise “ancillary” to those collaborations under the antitrust laws.
Finally, because the Division brought the case under the per se rule, there was no requirement to prove that the Defendant firms individually, bilaterally, or collectively possessed market power in the hiring and employing of “specialized computer engineers and scientists,” nor to determine and demonstrate whether the geographic metes and bounds of such a product market might be local, regional, national, or worldwide, nor to demonstrate the magnitude of the anticompetitive effect alleged on workers. A comparable rule-of-reason case would likely have required such a showing, as would a comparable provision in a merger challenge alleging harm in particular labor markets.15 A follow-on class action suit filed on behalf of 64,000 employees of the Defendant firms resulted in settlement agreements for payments of hundreds of millions of dollars to members of the class.16
As is standard with such judgments and decrees in the modern era, the Final Judgment also enumerates steps that the Defendant firms must take to ensure compliance, and establishes enforcement jurisdiction with the Court for five years.17
Returning to the question with which we began, are there ways for firms to protect their investments in training of skilled workers from the “free riding” of their competitors? The answer appears to be yes, but within carefully defined limits.
Regarding non-compete clauses in employer-employee agreements, there remains serious debate as to what level of “training” merits protection from free riding by competitors. Is the training provided to a high school graduate as a new fast-food employee comparable to that provided to a software engineer at a Silicon Valley firm for these purposes? If the former is indeed worthy of such protection, is there a meaningful difference between free riding by a fellow franchisee of the same franchise and by a franchisee of a rival franchise?
Many of the enforcers, practitioners, and scholars at the FTC conference exhibited some skepticism that, as a general matter, low-skilled workers received training that qualified as a legitimate justification for employer restrictions on their mobility post-job. On the other hand, such restrictions for highly skilled workers – all the way up to top executives – were generally considered legitimate and perhaps even desirable if set with reasonable limits. It was emphasized both that the FTC’s jurisdiction over such agreements – in either a legal challenge or a rule-making – is untested, and that the current situation features great variation in the enforceability of such agreements among U.S. states.
Regarding wage-fixing and no-poach agreements, the law seems clearer. “Naked” agreements among firms not to compete on wages or not to solicit each other’s employees will be treated as per se violations of Section 1 of the Sherman Act and may well result in criminal prosecution in the future. As per se violations, they are not in general subject to defenses arguing that the benefits of such restrictions outweigh the competitive harms. As a caveat, however, note that while trial courts have declined to dismiss per se enforcement actions, none have yet had a full trial on the merits.
Finally, if the defendants establish that the agreements are reasonably necessary to achieving efficiency-enhancing benefits of legitimate cooperative activities among the firms (such as a research joint venture), then they may be treated under the rule of reason rather than as per se violations. In that case, the government or another plaintiff would be required to demonstrate that the firms possess power in a well-defined antitrust labor market, and, if the defendants establish a procompetitive benefit of the agreements, that the benefit could be achieved through less restrictive means or that the anticompetitive effect of the agreement outweighed any procompetitive benefit. Similar detailed economic investigations would likely be required in a challenge to a merger under Section 7 of the Clayton Act that alleged competitive harm in labor markets.
Click here for a PDF version of the article
1 Director of Economic Research, Antitrust Division, U.S. Department of Justice, and Visiting Professor, Kyiv School of Economics. The original publication of the paper is in the International Labor Brief, vol. 18, #12, 2020), in Korean. The author is grateful to Yungshin Jang and the Korean Labor Institute for the kind invitation to address this issue. The author did not work on any of the cases mentioned and has not had access to confidential information concerning them. Helpful comments on an earlier draft by Norm Familant, Patrick Greenlee, Doha Mekki, and Jeff Wilder are gratefully acknowledged. The views expressed do not purport to reflect those of the U.S. Department of Justice, and nothing in this document may be cited in any enforcement proceeding against the Department of Justice.
2 Alfred Marshall, Principles of Economics: An Introductory Volume (8th ed.), London: Macmillan, 1920, at VI.iv.4, p. 565.
3 Gary Becker, “Investment in Human Capital: A Theoretical Analysis,” Journal of Political Economy 70: 5, part 2 (1962), 9-49; Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, New York: Columbia Univ. Press (for NBER), 1964; 3d ed. Chicago: Univ. Chicago Press (for NBER), 1993.
4 Daron Acemoglu and Jörn-Steffen Pischke, “The Structure of Wages and Investment in General Training,” Journal of Political Economy 107 (1999), 539-572.
5 See generally U.S. Federal Trade Commission, Non-Competes in the Workplace: Examining Antitrust and Consumer Protection Issues, Transcript of Workshop, January 9, 2020, https://www.ftc.gov/system/files/documents/public_events/1556256/non-compete-workshop-transcript-full.pdf.
6 See, for example, Sarah Whitten, “Jimmy John’s drops noncompete clauses following settlement,” CNBC, June 22, 2016, https://www.cnbc.com/2016/06/22/jimmy-johns-drops-non-compete-clauses-following-settlement.html.
7 See generally U.S. Department of Justice, Antitrust Division, Competition in Labor Markets, Transcript of Workshop, September 19, 2019, https://www.justice.gov/atr/page/file/1209071/download, and Doha Mekki, Statement before the Subcommittee on Antitrust, Competition and Consumer Rights of the Committee on the Judiciary of the U.S. Senate, for a Hearing on “Antitrust and Economic Opportunity: Competition in Labor Markets,” October 29, 2019, https://docs.house.gov/meetings/JU/JU05/20191029/110152/HHRG-116-JU05-Wstate-MekkiD-20191029.pdf.
8 Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, London: W. Strahan and T. Caddell, 1776, book 1, chapter 8, “On the Wages of Labour”.
9 The Sherman Act “does not confine its protection to consumers, or to purchasers, or to competitors, or to sellers. Nor does it immunize the outlawed acts because they are done by any of these. The Act is comprehensive in its terms and coverage, protecting all who are made victims of the forbidden practices by whomever they may be perpetrated.” Mandeville Island Farms v. Am. Crystal Sugar Co., 334 U.S. 219, 236 (1948) (citations omitted).
10 See, for example, U.S. v. Cargill and Continental Grain (2000), in which the Division blocked a merger of grain processing companies that would have reduced competition in the purchase and elevator storage of grain, despite the assumption that downstream grain markets were worldwide and competitive: https://www.justice.gov/atr/case-document/competitive-impact-statement-57.
11 See, for example, Roger Blair & Jeffrey Harrison, Monopsony: Antitrust Law and Economics, Princeton University Press, 1993.
12 U.S. v. Adobe Systems, et al., Complaint, September 24, 2010, https://www.justice.gov/atr/case-document/complaint-0.
13 Department of Justice, “Justice Department Requires Six High Tech Companies to Stop Entering Into Anticompetitive Employee Solicitation Agreements: Settlement Preserves Competition for High Tech Employees,” press release, September 24, 2010, https://www.justice.gov/opa/pr/justice-department-requires-six-high-tech-companies-stop-entering-anticompetitive-employee.
14 See Department of Justice, Antitrust Division, “Summary of Antitrust Division Health Care Cases (Since August 25, 1983),” https://www.justice.gov/atr/page/file/1077686/download: “The Department alleged that the conspiracy caused the participants of three of the four prepaid dental plans to pay higher copayment fees to the defendants and unnamed coconspirators than they might otherwise have had to pay. The case was the first criminal case the Antitrust Division brought against medical practitioners in over 50 years.” Similarly, the Division challenged movie distribution “split” agreements in civil actions before bringing its first criminal action in 1985: https://www.justice.gov/archive/atr/public/press_releases/1985/325872.pdf.
15 For examples of this type of analysis, see Ioana Marinescu & Herbert Hovenkamp, “Anticompetitive Mergers in Labor Markets,” Indiana Law Journal 94 (2019); José Azar, Ioana Marinescu & Marshall Steinbaum, “Labor Market Concentration,” Journal of Human Resources 55 (2020); and José Azar, Ioana Marinescu, Marshall Steinbaum & Blendi Taska, “Concentration in U.S. Labor Markets: Evidence from Online Vacancy Data,” Labour Economics 66 (2020).
16 In re: High-Tech Employee Antitrust Litigation, https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_Litigation.
17 U.S. District Court for the District of Columbia, Final Judgment in U.S. v. Adobe Systems, et al., March 17, 2011, https://www.justice.gov/atr/case-document/final-judgment-0.
