While there is general agreement that buyer collusion can be as anticompetitive (and illegal) as seller collusion, the sparse caselaw on buyer cartels leaves many specific doctrinal questions unanswered. However, while buyer cartel cases have been relatively rare, one particular subtype of buyer power has been making consistent antitrust headlines. Employers are “buyers” of their employees’ labor, and cases involving employer wage-fixing and no-poach agreements can shed light on buyer cartel doctrine more generally. In this article, we analyze recent developments in labor antitrust and trace their implications for other buyer power cases, focusing in particular on three doctrinal questions: whether a case falls under the rule of reason or the per se rubric, what kind of competitive harm the plaintiff must show, and how to define the relevant market.
By Sergei Zaslavsky & Laura Kaufmann1
Courts and commentators agree that just as seller cartels are illegal, buyer collusion to fix price or divide markets likewise violates the law.2 But while there is plentiful caselaw on seller cartels and the doctrine is fairly well settled, the caselaw on buyer cartels is relatively sparse. Courts offer general guidance that the same principles apply in cases involving seller power and buyer power,3 but the general guidance leaves many specific questions unanswered.
However, while buyer cartel cases are generally rare, one type of buyer has come under particular scrutiny from antitrust enforcers in recent years. Employers (who “buy” their employees’ labor) have increasingly been in the antitrust spotlight. Over the past 20 years, a wave of wage-fixing suits have hit a wide spectrum of different industries, including nursing,4 energy,5 animation,6 au pair services,7 shepherding,8 professional sports,9 and agriculture,10 and a case alleging that many high-tech companies agreed not to solicit each other’s workers shook up Silicon Valley in the early 2010s.11 Attention to labor antitrust spiked even further in October 2016, when the antitrust agencies published Antitrust Guidance for Human Resource Professionals (“HR Guidelines”),12 which clarified that not only were naked wage-fixing and no-poach agreements per se illegal, but the Department of Justice (“DOJ”) intended to proceed criminally against such agreements.13 Since 2016, DOJ provided additional clarification on employer agreements by filing statements of interest in private cases involving fast food franchises,14 railway equipment companies,15 and medical schools,16] and then followed through on its HR Guidelines by filing the first criminal wage-fixing case in December 202017 (with a superseding indictment involving an additional defendant following in April 2021),18 and the first criminal no-poach case in January 2021.19 In a recent interview with the American Bar Association, Acting Assistant Attorney General for the Antitrust Division Richard Powers indicated that DOJ intends to increase criminal enforcement against employer agreements further,20 so all signs point to this area of law continuing to garner attention.
The result is that while the caselaw on buyer cartels is fairly sparse, the doctrine on employer agreements is increasingly coming into clearer view.21] That is not to say that all important questions have been answered, but courts and enforcers have already weighed in on many important aspects of employer collusion, and with both private and government enforcement continuing at a high level, the body of applicable common law will continue to grow and refine the doctrine. What can this rapidly developing doctrine pertaining to one specialized type of buyer (the employer) tell us about buyer cartel doctrine more generally? This article proceeds by considering three important aspects of buyer cartel law where our rapidly accumulating experience with labor antitrust cases may be particularly illuminating: the dividing line between rule of reason and per se, the type of harm plaintiffs must show, and market definition.
II. RULE OF REASON v. PER SE
As any practitioner knows, one of the most consequential questions in antitrust litigation is whether the claim proceeds under the per se or rule of reason rubric. The article presumes the reader’s familiarity with these concepts, but suffice it to say that the plaintiff’s path to victory is far less direct in a rule of reason case, with market power, competitive effects, and procompetitive justifications all at issue. Perhaps even more importantly, as DOJ’s recent actions show,22 per se price-fixing and market-allocation agreements may subject defendants to criminal as well as civil liability.
What do labor antitrust cases have to teach us about the dividing line between rule of reason and per se violations? One key question is whether the challenged agreement is naked or ancillary to a broader procompetitive collaboration. As the DOJ explained in a Statement of Interest filed in a no-poach case involving medical faculty, “[i]f a no-poach agreement is reasonably necessary to a separate, legitimate business transaction or collaboration among the employers, it is not per se unlawful as a naked restraint, but instead judged under the rule of reason.”23 The DOJ synthesized the existing caselaw and explained its application in the labor antitrust context: “To be ancillary, an agreement eliminating competition must be subordinate and collateral to a separate, legitimate transaction, and reasonably necessary to make the main transaction more effective in accomplishing its purpose.”24
Of course, where the rubber really hits the road is figuring out what makes a restraint ancillary. Is an agreement among competing purchasers (of labor, or a different input) always going to be subject to rule of reason if it is part of some broader collaboration? How searching is the inquiry into whether the challenged agreement is “reasonably necessary” to the broader collaboration? A case involving temporary nurses in Arizona offers an instructive illustration. A class of per diem and travel nurses sued the Arizona Hospital and Healthcare Association (“AzHHA”), alleging that the organization’s “nursing Registry Program, which contracts with various agencies representing temporary nurses, operated as an illegal and anticompetitive buyers’ cartel” that “jointly set—and wrongfully suppressed—the wages and compensation” of temporary nursing staff.”25 The hospitals accused the plaintiffs of “attempting to ‘cherry pick’ one aspect of AzHHA’s role, without considering the other, pro-competitive activities of the organization,” and even the plaintiffs conceded that the organization “serve[d] some legitimate ends, such as quality assurance.”26 This was not enough to take the case out of per se territory.27 Rather, the court agreed with the nurses that “[t]here was no need for Defendants to begin price-fixing to permit AzHHA to continue providing other benefits to the hospitals.”28 The case suggests that invocation of ancillarity is no free pass, and courts are likely to question whether the challenged conduct is really a necessary part of the broader procompetitive collaboration or agreement.
The contours of the inquiry into whether the supposedly ancillary restraint is really “reasonably necessary” is likely to come into sharper view in the coming years. The vehicle for this doctrinal clarification just may be a case featuring your favorite burger spot. Non-solicitation clauses in franchise agreements (such as the ones commonly used in the fast food industry) have been the subject of much scrutiny recently. The Trump-era DOJ took the position that most franchisor-franchisee restraints are subject to the rule of reason, at least in part because such restraints are likely to be reasonably necessary to the broader franchise agreement.29 But other enforcers are skeptical that ancillary restraint doctrine applies to these agreements. Speaking on an American Bar Association podcast in April 2020, Washington Assistant Attorney General Rahul Rao described DOJ’s position as “somewhat misguided,” arguing that 35-40% of franchise systems do not use no-poach restraints and “the lack of a no-poach provision with those systems has not hindered their growth or success” and that “it’s not clear how a single labor market allocation agreement is organically or plausibly connected to the main transaction.”30 With enforcement and litigation in this sector continuing apace, it will be fascinating to track whether the Biden DOJ will maintain the prior administration’s position or take a more skeptical view of ancillary restraints in line with Washington’s position, and how courts resolve the inevitable controversies over whether challenged employer restraints were “reasonably necessary” to a broader agreement. The answers to these questions will provide a lot of clarity not only for labor antitrust, but for buyer cartel doctrine more generally.
III. DOES HARM TO SELLERS/EMPLOYEES SUFFICE?
While many within the antitrust world question the consumer welfare standard, it remains the touchstone for antitrust analysis, at least for the time being. The conventional thinking, applicable in cases involving seller power, is that the challenged antitrust conduct must harm consumers in order to be illegal. But how do we think about buyer power cases, where the immediate victim is the seller (or the employee — the seller of labor), not the consumer? What kind of competitive harm must the plaintiff show?
Again, the labor antitrust cases, and academic discussion over such cases, are the vanguard that may provide cases for buyer cartel cases more generally. In the AzHHA case, the court treated the claim as per se and thus had no occasion to analyze competitive effects, but did weigh in on whether “the nature of the plaintiff’s alleged injury…was the type the antitrust laws were intended to forestall” in the course of determining whether the nurses had standing to bring suit. The court’s view was that being paid wages below the competitive level was “an example of the type of injury the antitrust laws are meant to protect against.”31 In other words, showing harm to the sellers (employees) was enough.
There are two distinct ways to reach the conclusion that harm to sellers / employees suffices. One is to interpret the consumer welfare standard as applying not literally only to consumers, but to trading partners more generally, be they buyers, sellers, employees, or other economic actors. Interestingly, the clearest legal support for this proposition comes from an old Supreme Court case from the 1940s that predates the rise of the consumer welfare standard. In holding that farmers could state a cause of action against sugar refiners who allegedly “agree[d] among themselves to pay a uniform price for sugar beets grown in California,” the Supreme Court proclaimed that the Sherman Act “does not confine its protection to consumers, or to purchasers, or to competitors, or to sellers,” and “is comprehensive in its terms and coverage, protecting all who are made victims of the forbidden practices by whomever they may be perpetrated.”32 One may wonder whether the rise of the consumer welfare standard creates tension with the Mandeville Island Farms holding. Many labor antitrust scholars would say it does not. In their view, depressing wages below the competitive level ends up hurting consumers and not just employees, and so protecting employees is not at all inconsistent with the antitrust law’s focus on the consumer. As Suresh Naidu, Eric A. Posner, and Glen Weyl explain, if a firm depresses wages it will hire fewer workers, and “if firms employ fewer workers, they will produce less output, resulting in higher prices.”33 Therefore, “monopoly and monopsony are two sides of the same coin, and both harm labor and product markets.”34
As labor antitrust litigation continues to work its way through the courts, it will become clearer whether courts converge on the expansive Mandeville Island Farms position that the Sherman Act protects all victims and not just consumers, the scholars’ view that restraints that hurt employees hurt consumers as well, or some other position. Whatever consensus emerges will have a profound impact on buyer power cases more generally, not just labor antitrust.
IV. MARKET DEFINITION
Market definition is vital in most antitrust cases, and buyer power cases are no exception. Particularly in purchaser cases that fall under the rule of reason, market share of the purchasing group will be a key question in the assessment of competitive effects,35 and to calculate market share one must first identify the correct market.
Here, again, labor antitrust cases provide useful guidance. In particular, then-Judge Sotamayor provided a detailed tutorial on market definition in labor antitrust cases (applicable to buyer cartel cases more generally) in the 2001 Second Circuit opinion Todd v. Exxon.36 Plaintiffs, complaining “that defendants violated § 1 of the Sherman Act by regularly sharing detailed information regarding compensation… and using this information in setting the salaries,” alleged the relevant market to be “the services of experienced, salaried, non-union, managerial, professional and technical (MPT) employees in the oil and petrochemical industry.”37 The district court applied traditional “seller power” market definition framework, and asked whether “the ‘products’ at issue here—MPT employees—are reasonably interchangeable or [whether] there is a cross-elasticity of demand for potential substitutes.”38
Judge Sotamayor explained that “the district court looked through the wrong end of the telescope”: in a buyer-power case, “market is comprised of buyers who are seen by sellers as being reasonably good substitutes.”39 Therefore, “[t]he question [was] not the interchangeability of, for example, lawyers with engineers… [but rather] the interchangeability, from the perspective of an MPT employee, of a job opportunity in the oil industry with, for example, one in the pharmaceutical industry.”40
The lesson certainly carries over into the buyer cartel context. Under Sotomayor’s logic, for example, if there is an alleged cartel of gold purchasers, it us plausible for makers of gold watches to be in the same market as makers of gold earrings. It does not matter whether gold watches are substitutes for gold earrings; the question is whether the watch maker and the earring maker are reasonable substitutes for gold sellers looking to sell their raw material. Geographic market is another issue where labor cases will likely provide useful lessons. In employer collusion cases, the relevant market can be either local or national, depending on whether the employee views jobs in a particular city, state, or the entire nation as plausible substitutes. While geographic market definition is disputed less frequently than product market definition in antitrust litigation, this is an area where we may see more action in the near future in labor antitrust, as the proliferation of video-conferencing technology and growing acceptance of remote work expands the geographic area where one may look for employment. Any elaboration on the law of geographic markets in employer cases would likely have a significant impact on buyer cartel doctrine, and is an area to watch going forward.
While labor antitrust cases have already provided much useful guidance for buyer cartel doctrine, there are certainly remaining questions that will be worth monitoring. With all signs pointing to wage-fixing and no-poach litigation continuing to keep the courts busy for the foreseeable future, the contours of buyer cartel doctrine will continue to be filled in by this particular subtype of “buyer power” case.
1 Sergei Zaslavsky is a counsel in the O’Melveny & Myers Anitrust & Competition Group. Laura Kaufmann is an associate at O’Melveny & Myers.
2 See, e.g. Vogel v. Am. Soc’y of Appraisers, 744 F.2d 598, 601 (7th Cir. 1984) (“[B]uyer cartels, the object of which is to force the prices that suppliers charge the members of the cartel below the competitive level, are illegal per se. Just as a sellers’ cartel enables the charging of monopoly prices, a buyers’ cartel enables the charging of monopsony prices; and monopoly and monopsony are symmetrical distortions of competition from an economic standpoint.”).
3 See, e.g. Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., 549 U.S. 312, 317–18, 322–23 (2007) (stating that test that applied to predatory pricing on the sell-side also applied to predatory bidding on the buy-side and that “similar legal standards should apply to claims of monopolization and to claims of monopsonization”); Buccaneer Energy (USA) Inc. v. Gunnison Energy Corp., 846 F.3d 1297, 1315 (10th Cir. 2017) (“The same general framework for assessing market power applies to monopsony and monopoly situations alike.”); Presque Isle Colon & Rectal Surgery v. Highmark Health, 391 F. Supp. 3d 485, 502 n.12 (W.D. Pa. 2019) (“The Court applies the same standard as a claim of monopoly to this situation of an allege monopsony based on the close “kinship between monopoly and monopsony,” which suggests that “similar legal standards should apply to claims of monopolization and to claims of monopsonization.”).
4 See, e.g. Fleischman v. Albany Med. Ctr., 728 F. Supp. 2d 130 (N.D.N.Y. 2010); Maderazo v. Vanguard Health Sys., 241 F.R.D. 597 (W.D. Tex. 2007); Doe v. Arizona Hosp. & Healthcare Ass’n, No. CV07-1292-PHX-SRB, 2009 WL 1423378 (D. Ariz. Mar. 19, 2009); Reed v. Advoc. Health Care, No. 06 C 3337, 2007 WL 967932 (N.D. Ill. Mar. 28, 2007).
5 See Todd v. Exxon Corp., 275 F.3d 191 (2d Cir. 2001).
6 See In re Animation Workers Antitrust Litig., 123 F. Supp. 3d 1175 (N.D. Cal. 2015).
7 See Beltran v. InterExchange, Inc., 176 F. Supp. 3d 1066, 1073 (D. Colo. 2016).
8 See Llacua v. W. Range Ass’n, 930 F.3d 1161, 1179 (10th Cir. 2019).
9 See Kelsey K. v. NFL Enterprises LLC, No. C 17-00496 WHA, 2017 WL 3115169 (N.D. Cal. July 21, 2017), aff’d, 757 F. App’x 524 (9th Cir. 2018).
10 See Jien v. Perdue Farms, Inc., No. 1:19-CV-2521-SAG, 2020 WL 5544183, at *2 (D. Md. Sept. 16, 2020).
11 See United States v. eBay, Inc., 968 F.Supp.2d 1030 (N.D. Cal. 2013).
13 Id. at 3-4.
14 See Stigar v. Dough Dough, Inc., Case No. 2:18-cv-002440-SAB, Dep’t of Justice Statement of Interest (Mar. 8, 2019), ECF No. 34 (“DOJ Stigar Statement of Interest”).
15 See In Re Railway Industry Employee No-Poach Antitrust Litigation, Case No. 2:18-MC-00798-JFC Dep’t of Justice Statement of Interest (Feb. 8, 2019), ECF No. 158.
16 Seaman v. Duke University, No. 1:15-CV-462, Dep’t of Justice Statement of Interest (Mar. 7, 2019), ECF No. 325 (“DOJ Duke Statement of Interest”).
17 See United States v. Jindal, Case No. 4:20-cr-00358-ALM-KPJ (Dec. 9, 2020), ECF No. 1.
18 See United States v. Jindal, Case No. 4:20-cr-00358-ALM-KPJ (April 15, 2021), ECF No. 21.
19 See United States v. Surgical Care Affiliates, LLC, Case No. 3:21-cr-00011-L (Jan. 5, 2021), ECF No. 1.
20 Richard Powers, Acting Assistant Attorney General, U.S. Department of Justice Antitrust Division, American Bar Association Antitrust in Healthcare Virtual Conference (Feb. 12, 2021).
21 This is particularly the case for wage-fixing agreements (which have been analogized to price-fixing) and no-poach agreements (which have been analogized to market allocation).
22 See Jindal, Case No. 4:20-cr-00358-ALM-KPJ (Dec. 9, 2020), ECF No. 1; Surgical Care Affiliates, LLC, Case No. 3:21-cr-00011-L (Jan. 5, 2021), ECF No. 1.
23 DOJ Duke Statement of Interest at 24.
24 Id. Courts apply a similar framework in no-poach cases. See, e.g. In re Ry. Indus. Emp. No-Poach Antitrust Litig., 395 F. Supp. 3d 464, 480–81 (W.D. Pa. 2019) (“[T]he court first asks whether the challenged conduct truly involves a ‘horizontal’ restraint between actual or potential competitors allocating markets or customers, reducing output, or otherwise restricting competition between them, and, if so, whether the restraint is devoid of plausible procompetitive justification (i.e., is ‘naked’) or is instead plausibly ancillary and necessary to some larger, procompetitive integrative activity.”).
25 Arizona Hosp. & Healthcare Ass’n, 2009 WL 1423378 at *1, 3.
26 Id. at *3.
27 It also bears noting that the fact that the hospitals joined together as part of an industry organization did not prevent the court from treating the alleged wage-fixing as a horizontal agreement among competing hospitals, rather than a hub-and-spoke conspiracy with a series of vertical agreements between the industry organization and the participating hospitals.
28 Id. at *3. (internal square brackets omitted).
29 See DOJ Stigar Statement of Interest at 11, 16.
30 Am. Bar Ass’n, Our Curious Amalgam (Apr. 27, 2020), https://podcast.ourcuriousamalgam.com/episode/43-no-poach-assessing-risk-in-uncertain-seas/.
31 Arizona Hosp. & Healthcare Ass’n, 2009 WL 1423378 at *4.
32 Mandeville Island Farms v. Am. Crystal Sugar Co., 334 U.S. 219, 221, 236 (1948).
33 Antitrust Remedies for Labor Market Power, 132 Harv. L. Rev. 536, 559 (2018).
34 Id. See also Ioana Marinescu & Herbert Hovenkamp, Anticompetitive Mergers in Labor Markets, 94 Ind. L.J. 1031, 1062 (2019) (explaining that when a “firm has market power on both sides of the market… exercising market power on the labor side will entail the purchase of less labor… [and] less labor will lead to less output on the product side”).
35 E.g. Agency guidance for the healthcare industry states that “[t]he Agencies will not challenge, absent extraordinary circumstances, any joint purchasing arrangement among health care providers where… the purchases account for less than 35 percent of the total sales of the purchased product or service in the relevant market.” United States Dep’t of Justice and Fed. Trade Comm’n, Statement of Antitrust Enforcement Policy in Health Care 54-55 (August 1996).
36 275 F.3d 191 (2d Cir. 2001).
37 Id. at 195, 199.
38 Id. at 201.
39 Id. at 201, 202.
40 Id. at 202.