Skip to content

What If California Assembly Bill 5 Protected Collective Bargaining? An Antitrust Analysis

By: Sara Spaur

March 29, 2020

Introduction

In September 2019, California passed Assembly Bill 5 (“AB-5”), which codifies a 2018 case in which the California Supreme Court adopted the “ABC” test to determine employee status for unemployment insurance, workers’ compensation and wage orders.1Dynamex Operations W. v. Super. Ct., 416 P.3d 1, 40 (Cal. 2018). Under AB-5, a worker is presumptively an employee (instead of an independent contractor) unless the hiring entity shows that the worker meets three conditions.2To prove the worker is an independent contractor, the hiring entity must show the worker: A) is free from the right to control and actual control of the hiring entity; B) performs work outside the hiring entity’s usual business; and C) “is customarily engaged in an independently established trade, occupation, or business.” 2019 Cal. Legis. Serv. Ch. 296 (A.B. 5). This paper poses a hypothetical: what if AB-5 granted all workers that met its definition of employee, such as Uber drivers,3Uber has been predicted to not pass the ABC test. Uber drivers are not “free” from the control and direction of Uber. They lack control over prices, customers (“they are penalized for rejecting trips”), and routes, and “are ‘supervised’ by semi-automated and algorithmic systems that track their acceptance rates, time on trips, customer ratings, and other factors, and drivers can be ‘deactivated’ based on these factors.” Lawrence Mishel & Celine McNicholas, Uber Drivers Are Not EntrepreneursEcon. Pol’y Inst. (Sept. 20, 2019), https://www.epi.org/publication/uber-drivers-are-not-entrepreneurs-nlrb-general-counsel-ignores-the-realities-of-driving-for-uber/. the right to collectively bargain?4Collective bargaining is understood as a group of employees negotiating (or intending to negotiate) for “a larger share of the economic rents associated with a particular transaction . . . .” Prof. Daniel L. Rubinfeld, Testimony before the Antitrust Modernization Commission, Economics Roundtable, at 6 (Jan. 19, 2006), https://pdfs.semanticscholar.org/2810/e41b6b6b89437cc1651a4a051719b15879f3.pdf.   AB-5’s definition of employee is broader than the National Labor Relations Act’s (NLRA) definition of employee, which means a set of workers that would be covered by “hypothetical AB-5”5Hypothetical AB-5 is the same as AB-5, but adds a provision granting bargaining rights to all workers that meet AB-5’s definition of employee. (the hypothetical set) would not be covered by the NLRA. Because AB-5 exempts “licensed insurance agents, certain licensed health care professionals, registered securities broker-dealers or investment advisers . . . and others performing work under a contract for professional services” from its ABC test,62019 Cal. Legis. Serv. Ch. 296 (A.B. 5). the hypothetical set would largely include low-skilled and low-wage employees.

This thought experiment is motivated by United States Chamber of Commerce v. City of Seattle (Seattle), where the Court of Appeals for the Ninth Circuit held that a Seattle ordinance which permitted independent-contractor “for-hire drivers” and “driver coordinators” (such as Uber) to agree on the “nature and amount of payments” was not preempted by the NLRA and was subject to federal antitrust laws.7Chamber of Commerce of the United States v. City of Seattle, 890 F.3d 769, 791–94 (9th Cir. 2018). Since Seattle had not challenged plaintiffs’ claim that the ordinance authorized a per se violation of section 1 of the Sherman Act (“Section 1”),8Id. at 781. the Ninth Circuit accepted such claim without deciding it on the merits, and held that the ordinance did not satisfy the Midcal test for state-action immunity.9Id. at 782.

This paper examines whether bargaining efforts authorized by hypothetical AB-5 could withstand antitrust scrutiny, in the hopes that such analysis may be useful in understanding the increasing intersectionality of labor law and antitrust law in today’s gig economy. Suppose California Uber drivers around Lake Tahoe (near the Nevada border) agreed to bargain for a minimum “price[] they will accept for their services in the market.”10Id. at 781.  If such agreement was challenged as a violation of Section 1, a court could find that a contract or conspiracy existed and it affected interstate commerce. The issue would be whether “the agreement unreasonably restrained trade under either a per se rule of illegality or a rule of reason analysis.”11Tanaka v. Univ. of S. Cal., 252 F.3d 1059, 1062 (9th Cir. 2001). Under Section 1, “a plaintiff must demonstrate (1) that there was a contract, combination, or conspiracy; (2) that the agreement unreasonably restrained trade under either a per se rule of illegality or a rule of reason analysis; and (3) that the restraint affected interstate commerce.” Id.(quotations and citations omitted). The agreement should not violate Section 1 because it falls within the labor exemption, and even if it does not, it could create socially desirable benefits and serve as a countervailing procompetitive force against employer monopsony power.

Part I provides background on the labor exemption and argues that the labor exemption should remove the challenged conduct from antitrust liability. In case the Uber drivers were not found exempt from antitrust laws, Part II examines their coordination under the per se rule of illegality, and rejects the appropriateness of such rule given the conduct’s potentially significant benefits and “the economic impact of [such] practices is not immediately obvious.”12FTC v. Ind. Fed’n of Dentists, 476 U.S. 447, 459 (1986). Part III applies a rule of reason analysis to posit that the challenged conduct does not harm competition, and their association could be procompetitive such that the conduct should withstand Section 1 scrutiny.

I. The Labor Exemption

This paper first considers whether the labor exemption in the Clayton Antitrust Act13Section 6 of the Clayton Antitrust Act states,

Nothing contained in the antitrust laws shall be construed to forbid the existence and operation of labor . . . instituted for the purposes of mutual help, and not having capital stock or conducted for profit . . . nor shall such organizations, or the members thereof, be held or construed to be illegal combinations or conspiracies in restraint of trade, under the antitrust laws.

15 U.S.C. § 17 (2012).
 is a source of antitrust immunity for hypothetical AB-5 and its authorized conduct, which could remove the need for further analysis. A court could determine the labor exemption applies.

The Uber drivers would argue they satisfy the labor exemption’s plain language because they are associated for purposes of mutual aid, lack capital stock and are not conducted for profit.14According to Professor Estreicher,Uber drivers “sell only their own personal services, without any significant capital investment . . . . The provision of a general-purpose automobile, which is also available for personal use, does not reflect the kind of special-purpose investment . . . .”Brief for Professor Samuel Estreicher as Amicus Curiae Supporting Defendants-Appellees at 5, Chamber of Commerce of the United States v. City of Seattle, 890 F.3d 769 (9th Cir. 2018)(No. 17-35640) [hereinafter Brief for Professor Estreicher].  Uber might argue the labor exemption is inapplicable because Uber drivers are not employees under federal labor law and Uber does not satisfy the NLRA’s definition of employer, but is a technology company supplying a ride-referral app.15Appellees’ Response Brief at 23, Seattle, 890 F.3d 769 (9th Cir. 2018) (No. 17-35371) [hereinafter Appellees’ Response Brief].  Yet the case law dismisses the argument that only entities covered by the NLRA fall within the antitrust labor exemption.16For example, as noted in Seattle, “the fact that a group of workers is excluded from the definition of ‘employee’ in § 152(3), without more, does not compel a finding of Machinists preemption . . . § 152(2)–(3) excludes agricultural laborers, domestic workers, and public employees, all of which have been subject to state regulation.” Seattle, 890 F.3d at 793. While section 14(a) of the NLRA excludes supervisors from NLRAcoverage, independent contractors are not explicitly excluded.Id.at 794. But even supervisors “can form unions and seek collective bargaining free of antitrust liability under the Clayton exemption.” Brief for Professor Estreicher, supra note 14, at 7. Therefore, the labor exemption could apply. This paper proceeds in case it does not.

II. Per Se Rule of Illegality Analysis

The first restraint of trade claim this paper anticipates and rejects is one historically asserted against groups of professionals concerned with the price received for their services: such groups constitute horizontal price-fixing arrangements and are per se unlawful.17E.g., Brief of Plaintiffs-Appellees at 8, Chamber of Commerce of the U.S. v. City of Seattle, 890 F.3d 769 (9th Cir. 2018) (No. 17-35371) [hereinafter Brief of Plaintiffs-Appellees] (“The Chamber initially challenged the Ordinance in March 2016, alleging that the Sherman Antitrust Act preempts the Ordinance because the collective-bargaining scheme amounts to price fixing among horizontal competitors.”). Goldfarb v. Virginia State Bar provides an example. In Goldfarb, the alleged restraint was a county bar association’s schedule of “recommended minimum prices for common legal services” that was purportedly voluntary, but state bar ethics opinions found habitual charges below the minimum created a presumption of attorney misconduct.18Goldfarb v. Va. State Bar, 421 U.S. 773, 776–78 (1975). The Supreme Court held the schedule created a “fixed, rigid price floor” because “every lawyer who responded to petitioners’ inquiries [about a title examination] adhered to the fee schedule, and no lawyer asked for additional information in order to set an individualized fee . . . in terms of restraining competition and harming consumers like petitioners the price-fixing activities found here are unusually damaging.”19Id. at 781.

At first glance, an agreement to only accept higher wages would raise the price customers pay to use the employee-drivers’ services, which would decrease output (the number of rides) if the price was above consumers’ willingness to pay and result in a determination that the agreement was per se unlawful.20In other words, this “practice facially appears to be one that would always or almost always tend to restrict competition and decrease output.” Nat’l Collegiate Athletic Ass’n v. Bd. of Regents of Univ. of Okla., 468 U.S. 85, 100 (1984) (citation omitted). The per se rule should be deemed inapplicable for several reasons. First, “finding that there exists ease of entry into the relevant product market can be sufficient to offset the government’s prima facie case of anti-competitiveness.”21FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 55 (D.D.C. 1998) (merger analysis context). Uber drivers, like others in the hypothetical set, face low barriers to entry, so an agreement among some would not reduce competition. Second, growing evidence indicates labor markets for low-wage workers are monopsonistic, and collective bargaining can be a countervailing force to return wages to a competitive level. Third, collective bargaining does not necessarily restrict output because labor does not want to reduce “the size of the pie,” but receive a larger slice of it. Fourth, “judicial inexperience with a particular arrangement counsels against extending the reach of per se rules.”22Bd. of Regents of Univ. of Okla., 468 U.S. at 100 n.21. Judges are inexperienced with arrangements between gig workers. 

For all these reasons any antitrust scrutiny should proceed under a rule of reason analysis. Part III incorporates the first two factors, low barriers to entry and employer monopsony power, into this analysis.

III. Rule of Reason Analysis

The plaintiff (such as Uber) must first show “the restraint produces significant anticompetitive effects within a relevant market”; the burden shifts to defendant Uber drivers to provide evidence of “the restraint’s procompetitive effects”; if defendants’ burden is met, the plaintiff must show there exists a less restrictive alternative.23Tanaka v. Univ. of S. Cal., 252 F.3d 1059, 1063 (9th Cir. 2001) (quotations omitted). As explained below, the hypothetical claim should not prevail under the rule of reason. 

A. The Drivers’ Conduct Would Not Produce Significant Anticompetitive Effects

Because the Ninth Circuit in Seattle did not conduct a rule of reason analysis, it is unclear whether a plaintiff would be able to show significant anticompetitive effects. Within the relevant labor market, which we assume is the market for app-based for-hire driver services,24I assume the market for driver services like Uber’s is distinct from the market for taxicab servicesThis assumption is supported by briefs in the Seattle litigation. E.g., Brief of Plaintiffs-Appellees, supra note 17, at 4; Brief for the United States and the FTC as Amici Curiae Supporting Appellants at 10, Chamber of Commerce of the United States v. City of Seattle, 890 F.3d 769 (9th Cir. 2018) (No. 17-35640). If it is necessary to define the market geographically, a potential way to define it is by commuting zone. See José Azar, Ioana Marinescu & Marshall Steinbaum, Labor Market Concentration 7–8 (Nat’l Bureau of Econ. Research, Working Paper No. 24147, 2019).  the hypothetical agreement would not create significant effects because the Uber drivers lack market power. Market power is the power to control price by restraining one’s own output,25See generally G. Stigler, The Organization of Industry (1968). or the ability to charge higher than the prevailing price in a competitive market.26Suresh Naidu, Eric Posner & E. Glen Weyl, Antitrust Remedies for Labor Market Power, 132 Harv. L. Rev. 537, 538 (2018).  Monopsony power is not necessarily anticompetitive,27As noted in In re Blue Cross Blue Shield Antitrust Litigation, cooperative purchasing agreements “can be used by purchasers to exploit the productive efficiencies that result from conducting business on a larger scale.” 308 F. Supp. 3d 1241, 1275 (N.D. Ala. 2018). but “an agreement to fix prices among buyers rather than sellers . . . would violate § 1 of the Sherman Act just as a price-fixing agreement among sellers would.”28O’Bannon v. Nat’l Collegiate Athletic Ass’n, 7 F. Supp. 3d 955, 991 (N.D. Cal. 2014), aff’d in part, vacated in part, 802 F.3d 1049 (9th Cir. 2015). This Section focuses on low barriers to entry and employer monopsony power as sources of the employee-drivers’ lack of market power, but acknowledges that additional factors including two-sided markets and vertical restraints may also have an effect.

1. Low Barriers to Entry

One reason why the Uber drivers do not have the power to control price by restricting output is low barriers to entry: if one driver refuses to work for a certain wage, another will quickly take her place. An individual can become an Uber driver if they have 1) a car that “is less than 10 years old and in good condition,” 2) a U.S. driver’s license, 3) one year of driving experience, 4) car insurance, 5) “clean vehicle title and vehicle registration,” and 6) passed Uber’s background check.29Brett Helling, How to Become an Uber Driver in 2020Ridester, https://www.ridester.com/how-become-uber-driver/ (last updated Feb. 12, 2020). Given that over ninety percent of American households own a car30Robin Chase, U.S. Dep’t of State, Does Everyone in America Own a Car? (2010), https://static.america.gov/uploads/sites/8/2016/04/You-Asked-Series_Does-Everyone-in-America-Own-a-Car_English_Lo-Res_508.pdf. and Uber drivers have no education requirements besides those that attach to a driver’s license, such as drivers ed, the barriers to entry appear quite low.

It is clear that there are low barriers to entry when the hypothetical collaboration between Uber drivers is compared to cases with “associations of professionals who seek to constrain entry,” including associations of lawyers, physicians and dentists.31Marshall Steinbaum, Antitrust, the Gig Economy, and Labor Market Power, 82 Law & Contemp. Probs. 45, 58 (2019); Fed. Trade Comm’n, Competition in Professional Services in the United States 5–6, 10 (2012), https://www.ftc.gov/system/files/attachments/us-submissions-oecd-2010-present-other-international-competition-fora/iberto-professional-services.pdf. Entry into the market for legal or medical professional services is restricted by the years of specialized education and training required; such requirements do not exist for Uber drivers (or other low-wage workers generally).

2. Employer Monopsony Power

Uber drivers also could not create anticompetitive effects through an agreement because Uber has monopsonypower in the relevant labor market.32As monopoly power has been defined as “substantial” market power, see Bacchus Indus., Inc. v. Arvin Indus., Inc., 939 F.2d 887, 894 (10th Cir. 1991), monopsony power is also defined as “substantial” market power. Monopsony power in this context may be understood as the buyer-employer’s ability to purchase a good—driver services—for less than the prevailing market price. Under a dynamic monopsony model, a firm’s labor supply elasticity is the recruitment elasticity minus the quit elasticity (and “the recruitment elasticity is the negative of the quit rate elasticity”).33Alan B. Krueger & Orley Ashenfelter, Theory and Evidence on Employer Collusion in the Franchise Sector 15 (Nat’l Bureau of Econ. Research, Working Paper No. 24831, 2018). As Naidu et al. stated, “If workers do not quit even if the firm lowers wages significantly (elasticity is low), then the firm enjoys significant market power over the workers . . . If it is low, workers need protection.” Naidu, Posner & Weyl, supra note 26, at 557.  This Section first discusses sources of employer monopsony power and why it is harmful to competition, then describes studies that have estimated the degree of monopsony power or indicators of such power in labor markets.34While evidence of monopsony power or collusion among employers with monopsony power could be grounds for a claim under section 2 of the Sherman Act, this paper leaves such analysis to future study. Assuming that the analysis on the buyer-side is symmetrical to the seller-side under the merger guidelines, the high market shares of buyer-employers, no countervailing seller power, and low elasticity of supply suggest that employer coordination is unlikely to be beneficial.

It is important to distinguish bargaining power, which is created or enhanced by hypothetical AB-5, from monopsony power. An entity with bargaining power can “bargain to get a larger share of the economic rents associated with a particular transaction, without any necessary efficiency consequences.”35Rubinfeld, supra note 4. If there are efficiency consequences, the entity is a buyer with monopsony power: “In the absence of seller market power, a buyer merger that increased monopsony power could be anticompetitive.”36Id.

Sources of monopsony power include product differentiation (workplace-specific amenities or “dis-amenities”), search frictions, and concentration in the relevant labor market.37Naidu, Posner & Weyl, supra note 26, at 553; see also Krueger & Ashenfelter, supra note 33, at 18. In Todd v. Exxon Corporation, the Second Circuit found that the market for nonunion managerial, professional, and technical employees was “susceptible to tacit market coordination” by defendant companies “in the integrated oil and petrochemical industry,” in part because “the supply of labor has an inherently inelastic quality.”38Todd v. Exxon Corp., 275 F.3d 191, 195, 211 (2d Cir. 2001). There are several sources of this inelasticity: “substantial frictions [are] associated with transitioning between jobs,” “workers [may] confine their job searches to an education- or skills-delimited segment of available jobs,” and many individuals are constrained geographically by where they live.39Azar, Marinescu & Steinbaum, supra note 24, at 8. These characteristics allow employers to “exploit their monopsony power by driving prices [i.e. wages] below competitive levels.”40In re Blue Cross Blue Shield Antitrust Litig., 308 F. Supp. 3d 1241, 1275 (N.D. Ala. 2018) (citations omitted). Naidu et al. also note that “labor markets are more vulnerable to monopsony than products markets,” because unlike a product market where only the buyer cares about product features, “the employer cares about the identity and characteristics of the employee and the employee cares about the identity and characteristics of the employer.” Naidu, Posner & Weyl, supra note 26, at 554–55.

Todd also noted that “[a] greater availability of substitute buyers indicates a smaller quantum of market power on the part of the buyers in question.”41Todd, 275 F.3d at 202. As of October 2019, Uber had 69.7% of the U.S. ridesharing market, and its next closest competitor, Lyft, had 28.8% market share.42E. Mazareanu, Leading Ride-Hailing Companies in U.S. by Market Share 2017-2019Statista (Dec. 2, 2019), https://www.statista.com/statistics/910704/market-share-of-rideshare-companies-united-states/. Considering that Uber and Lyft together have a market share of 98.5%, the market is highly concentrated; there are few buyers that an Uber driver could take her labor to in search of a higher wage-price.43Using the Herfindahl-Hirschman Index, the HHI would be 5,689.78, which is above the 2,500 threshold and indicates a highly concentrated market. Such concentration is a strong indication of Uber’s monopsony power.

Even if the workers at issue were not Uber drivers but other low-wage employees in the hypothetical set, studies have found economy-wide increases in the market concentration of firms and decreases in the labor share.44Azar, Marinescu & Steinbaum, supra note 24, at 1 (citations omitted). José Azar, Ioana Marinescu and Marshall Steinbaum studied the effects of labor market concentration on wages by using data on posted vacancies for the “most frequent occupations” in over 8,000 geographic-occupational labor markets from CareerBuilder.com, the largest online U.S. job platform.45Id. at 2. Azar et al. define the labor markets by commuting zone (geographic) and by occupation (line of commerce),46Id. at 5–8. and find that labor market “concentration is high, and increasing concentration is associated with lower wages.”47Id. at 22. The most concentrated industries with an average HHI above 5,000 were “Farm equipment mechanics, Rail car repairers, and Light truck or delivery services drivers.” Id. at 12 (quotations omitted). Empirical evidence of high concentration and market power among buyer-employers supports the argument that seller-employees, in contrast, effectively lack the market power to create anticompetitive effects.48As Azar et al. concludes, “Our results suggest that the anti-competitive effects of concentration on the labor market could be important.” Id. at 22.

Increases in employer concentration have been significantly associated with decreases in wages. A 2018 study by Efraim Benmelech et al. used data from the Longitudinal Business Database to construct Herfindahl-Hirschman Index (HHI) measures of employer concentration at the “county-industry-year level,” and compared them with “measures of average wages and productivity at the establishment level constructed from the Census of Manufacturers (CMF) and the Annual Survey of Manufacturers (ASM).”49Efraim Benmelech et al., Strong Employers and Weak Employees: How Does Employer Concentration Affect Wages? 3 (Nat’l Bureau of Econ. Research, Working Paper No. 24,307, 2018). Benmelech et al. estimate that from 2002 to 2009, increases in HHI measures (higher levels of monopsony power) were associated with decreasing wages, and “the link between wage growth and productivity growth is significantly larger when local-level employer concentration is small.”50Id. at 3–5. Benmelech et al. conclude their results resembled what one would find in monopsonistic labor markets.51Id. at 23.

High labor market concentration is reinforced by non-compete and no-poaching clauses. “58 percent of major franchise chains include ‘noncompetitive clauses’ in their franchise contract that restrict the recruitment and hiring of workers . . . by other units affiliated with the franchisor.”52Krueger & Ashenfelter, supra note 33, at 4. This effectively “reduces the number of competitive employers in a market to no more than the number of franchise companies.”53Id. at 9. Alan Krueger and Orley Ashenfelter observed this among “quick service restaurants” (QSR) in Rhode Island: “Instead of 261 employers competing for QSR workers, this calculation suggests that there are effectively six employers (of equal size).”54Id. at 13. If a franchise set wages at the minimum wage, the clause will restrict employees’ ability to compete for higher wages within the franchise. A 2016 White House report acknowledged that non-compete clauses can depress wages, limit mobility, inhibit innovation, and “may also have a detrimental effect on consumer well-being by restricting consumer choice.”55White House, Non-Compete Agreements: Analysis of the Usage, Potential Issues, and State Responses 2, 5 (2016), https://perma.cc/SGJ9-9UJP (emphasis added).

While an employee without a non-compete clause could theoretically just switch employers to obtain a higher wage, higher wages are not achievable if employers collude to not offer a better wage-price. As Ashenfelter and Krueger note, “the potential for broader collusion is clearly enhanced when [within-franchise] no-poaching agreements are in place,” due to high concentration (“fewer firms need to agree in a more concentrated market”), the substitutability of franchise workers, and the minimum wage serves as a “natural and easily enforced focal point for collusive behavior.”56Krueger & Ashenfelter, supra note 33, at 13–14. Enforcement actions have already been brought for such collusion.57The Department of Justice (DOJ) and FTC broughtthree civil enforcement actions resulting in consent judgments “against technology companies (eBay and Intuit, Lucasfilm and Pixar, and Adobe, Apple, Google, Intel, Intuit, and Pixar) that entered into ‘no poach’ agreements with competitors . . . In two cases, at least one company also agreed to limit its hiring of employees who currently worked at a competitor.” U.S. Dep’t of Justice, Antitrust Div. & Fed. Trade Comm’n, Antitrust Guidance for Human Resource Professionals 3–4 (2016), https://www.justice.gov/atr/file/903511/download [hereinafter DOJ & FTC Guidance].

In sum, the alleged restraint does not harm competition because Uber drivers and others in the hypothetical set lack market power, in contrast to employer monopsony power which is strong and growing, and reducing such power could help raise the wages of low-skilled and low-wage employees.

3. A Final Consideration: Two-Sided Markets

This paper mentions the consideration of two-sided markets because it may affect the Section 1 analysis.58Uber has been recognized as operating two-sided markets. See Miranda Bogen, Uber and “The Taking Economy”: The Dynamics of Two-sided Markets and Algorithmic ExploitationMedium (Mar. 16, 2017), https://medium.com/equal-future/uber-and-the-taking-economy-b75d0b978bf8. Uber would argue it facilitates a single, simultaneous transaction between rider and driver and increases transaction volume by charging the riders less,59See Ohio v. Am. Express Co., 138 S. Ct. 2274, 2286 (2018). such that employee-driver wage coordination would harm the two-sided market. While it may benefit buyers on one side of the market—riders—such low fares are not beneficial to sellers on the other side—drivers. Both riders and drivers are consumers in the economy for other goods and services. As Sanjukta Paul notes, “[t]he app could match riders in space and time without setting prices, and that would exhaust all the efficiencies [that] have been claimed for it.”60Sanjukta Paul, Fissuring and the Firm Exemption, 82 Law & Contemp. Probs. 65, 75 (2019).

B. Procompetitive Benefits and No Less Restrictive Alternative

If a court finds an insufficient showing of anticompetitive effects within the relevant market, the burden shifts to defendant Uber drivers to show procompetitive benefits: 1) increasing bargaining power can transfer a greater share of the rents to labor such that it approaches the allocation that would be observed in a competitive market and decreases employer monopsony power, and 2) higher (competitive) wages could motivate increases in the quality of the driver and driving services. Both benefits are identified in Law v. National Collegiate Athletic Association (NCAA): “Lower prices cannot justify a cartel’s control of prices charged by suppliers, because the cartel ultimately robs the suppliers of the normal fruits of their enterprises. Further, setting maximum prices [salaries] reduces the incentive among suppliers to improve their products.”61134 F.3d 1010, 1022 (10th Cir. 1998) (citation omitted).

The first benefit finds support in a line of cases that hold employer restraints on competition within a labor market may violate Section 1.62See, e.g., Agnew v. Nat’l Collegiate Athletic Ass’n, 683 F.3d 328 (7th Cir. 2012); Todd v. Exxon Corp., 275 F.3d 191 (2d Cir. 2001); Ostrofe v. H.S. Crocker Co., Inc., 740 F.2d 739, 740 (9th Cir. 1984). In Appalachian Coals, Inc. v. United States, the Supreme Court held that a cooperative enterprise was not an undue restraint “merely because it may effect a change in market conditions, where the change would be in mitigation of recognized evils and would not impair, but rather foster, fair competitive opportunities.”63Appalachian Coals v. United States, 288 U.S. 344, 374 (1933), overruled by Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752 (1984). Though Appalachian Coals was later overruled,64See Copperweld Corp., 467 U.S. 752. this case illustrates antitrust law once took a holistic approach and suggests a return to such approach is appropriate.

Second, because “improving product quality[] may be a legitimate procompetitive justification,”65O’Bannon v. Nat’l Collegiate Athletic Ass’n, 7 F. Supp. 3d 955, 1003l (N.D. Cal. 2014), aff’d in part, vacated in part, 802 F.3d 1049 (9th Cir. 2015). so should improving service quality. The O’Bannon v. NCAA court rejected the NCAA’s quality improvement argument because the NCAA did not show the challenged restraints were necessary to obtain the benefits of improved quality of educational services. However, in this hypothetical case, such restraint is necessary in light of employer monopsony power and the lack of an existing regulatory framework to combat such power.

Assuming defendant employee-drivers’ burden is met, a plaintiff must show that “these procompetitive goals can be achieved in other and better ways,’’ which “should either be based on actual experience in analogous situations elsewhere or else be fairly obvious.”66Id. at 1004–05 (internal quotations and citations omitted). No such situations come to mind. While employers may be tempted to create employee groups that can work through employment-related concerns, they risk violating section 8(a)(2) of the NLRA, which makes it “an unfair labor practice for an employer . . . to dominate or interfere with the formation or administration of any labor organization or contribute financial or other support to it.”6729 U.S.C. § 158(a)(2) (2012). Based on the above analysis, a court should determine hypothetical AB-5 withstands Section 1 scrutiny.

Conclusion 

As noted by the Supreme Court in American Needle, Inc. v. National Football League, the “central evil addressed by Sherman Act § 1 is the elimin[ation of] competition that would otherwise exist.”68Am. Needle, Inc. v. Nat’l Football League, 560 U.S. 183, 195 (2010) (internal quotations and citations omitted). If antitrust law is concerned with abuses of market power, it must consider harms to a range of constituents—including employees. Even under a “consumer welfare” standard,69Under the Chicago School’s consumer welfare standard, output should be as high as possible and prices as low possible. Ianni Drivas, Reassessing the Chicago School of Antitrust Law, U. Chi. L. Sch. (June 4, 2019), https://www.law.uchicago.edu/news/reassessing-chicago-school-antitrust-law. antitrust law should not disregard harms to worker competition in labor markets because consumer welfare is affected by consumers’ ability to pay70As posited by Naidu et al., the “markdown on wages caused by monopsony and the markup on prices caused by monopoly are akin to taxes.” See supra note 26, at 560; see supra note 55.—which for many is determined by how much one earns.

In light of weak employee market power, hypothetical AB-5 and the conduct it authorizes would attempt to address the market failures of employer monopsony power and below-competitive wages for low-skilled employees, which is procompetitive and welfare-enhancing and should allow it to withstand antitrust scrutiny.


Sara Spaur, J.D. Class of 2020, N.Y.U. School of Law.

Suggested Citation: Sara Spaur, What if California Assembly Bill 5 Protected Collective Bargaining? An Antitrust Analysis, N.Y.U. J. Legis. & Pub. Pol’y Quorum (2020).