Harvard Law Review Harvard Law Review Harvard Law Review

Sherman Act

Chamber of Commerce v. City of Seattle

Ninth Circuit Holds Collective Bargaining Ordinance Not Entitled to State Action Immunity.

Lochner v. New York,1×1. 198 U.S. 45 (1905). in which the Supreme Court divined a liberty of contract to invalidate economically inefficient regulation, has been repudiated as antidemocratic judicial activism.2×2. See id. at 75 (Holmes, J., dissenting); Jamal Greene, The Anticanon, 125 Harv. L. Rev. 379, 417–22 (2011). Writing in the wake of Lochner, the Court held in Parker v. Brown3×3. 317 U.S. 341 (1943). that the Sherman Antitrust Act,4×4. 15 U.S.C. § 1 (2012). which casts all “restraint of trade” as unlawful,5×5. Id. evinces “no hint that it was intended to restrain state action or official action directed by a state.”6×6. Parker, 317 U.S. at 351; see also Paul R. Verkuil, State Action, Due Process and Antitrust: Reflections on Parker v. Brown, 75 Colum. L. Rev. 328, 330 (1975). Anticompetitive conduct directed by a state evades federal preemption if it satisfies the test set out in California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc.,7×7. 445 U.S. 97 (1980). which requires that the conduct (1) follow from “clearly articulated and affirmatively expressed . . . state policy,”8×8. Id. at 105 (internal quotation marks omitted) (quoting City of Lafayette v. La. Power & Light Co., 435 U.S. 389, 410 (1978) (plurality opinion)). and (2) be “‘actively supervised’ by the State itself.”9×9. Id. (quoting City of Lafayette, 435 U.S. at 410 (plurality opinion)). Recently, in Chamber of Commerce v. City of Seattle,10×10. 890 F.3d 769 (9th Cir. 2018). the Ninth Circuit held that a municipal ordinance authorizing collective bargaining between rideshare drivers and companies flunks both Midcal prongs.11×11. Id. at 782, 787. In narrowly deploying the clear articulation prong, the court raised the specter of municipal Lochnerism.12×12. This piece is not the first to consider antitrust law’s state action exemption in the context of Lochner. See Merrick B. Garland, Antitrust and State Action: Economic Efficiency and the Political Process, 96 Yale L.J. 486, 508–12 (1987).

When the Seattle City Council enacted Ordinance 12496813×13. Seattle, Wash., Municipal Code §§ 6.310.110, .735 (2015) (amended 2019). (Ordinance) on December 14, 2015, it became the first U.S. city to allow rideshare drivers to unionize.14×14. Daniel Beekman, Seattle First U.S. City to Give Uber, Other Contract Drivers Power to Unionize, Seattle Times (June 2, 2016, 12:01 PM), https://www.seattletimes.com/seattle-news/politics/unions-for-taxi-uber-drivers-seattle-council-votes-today/“>https://www.seattletimes.com/seattle-news/politics/unions-for-taxi-uber-drivers-seattle-council-votes-today/”>https://www.seattletimes.com/seattle-news/politics/unions-for-taxi-uber-drivers-seattle-council-votes-today/ [https://perma.cc/KGG9-945F]. Because of their legal status as independent contractors, rideshare drivers do not have a right to collectively bargain as “employees” under the National Labor Relations Act15×15. 29 U.S.C. §§ 151–169 (2012). (NLRA).16×16. Id. § 152. Independent contractor status can present several disadvantages to workers. See Dmitri Iglitzin & Jennifer L. Robbins, The City of Seattle’s Ordinance Providing Collective Bargaining Rights to Independent Contractor For-Hire Drivers: An Analysis of the Major Legal Hurdles, 38 Berkeley J. Emp. & Lab. L. 49, 52 (2017). Under the Ordinance, the City selects a “Driver Representative” that negotiates for drivers on subjects selected by the City, like driver earnings.17×17. Municipal § 6.310.735.B, .H(1). Before any agreement is made final, the City reviews the terms reached to ensure compliance with the Ordinance.18×18. Id. § 6.310.735 (H)(2). The Chamber of Commerce brought suit against the City, claiming that the Ordinance violates antitrust law by facilitating price fixing between independent contractors.19×19. Chamber of Commerce v. City of Seattle, 274 F. Supp. 3d 1155, 1159 (W.D. Wash. 2017). The Chamber also alleged that the Ordinance is preempted by the NLRA and other federal and state laws.20×20. Id.

The district court dismissed the complaint.21×21. Id. at 1176. After establishing that the Chamber had standing and that their claims were ripe, the court located the Ordinance under the protection of Parker.22×22. See id. at 1160–63, 1169. Regarding the first Midcal prong, the court found that Washington statutes “clearly delegate authority for regulating the for-hire transportation industry to local government units and authorize them to use anticompetitive means in furtherance of the goals of safety, reliability, and stability.”23×23. Id. at 1163. Pointing to the City’s administrative role with respect to rideshare drivers, their driver representatives, and the negotiated agreement, the court held that the Ordinance is supervised by the State and therefore satisfies the second Midcal prong.24×24. Id. at 1168–69; see id. at 1167 (noting in the context of the City’s oversight that cities are not required under Parker to show active supervision by the state itself).

Moving on, the court held that the Ordinance was not preempted by the NLRA. First, because both parties agreed that rideshare drivers are independent contractors, and because the court would therefore not resolve the issue of the drivers’ legal status, the Ordinance was not subject to Garmon25×25. San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236 (1959). preemption,26×26. City of Seattle, 274 F. Supp. 3d at 1170–71. which requires first-instance determinations of whether certain activity comes under the NLRA be left to the National Labor Relations Board (NLRB).27×27. See Garmon, 359 U.S. at 244–45. Nor was the Ordinance subject to Machinists28×28. Lodge 76, Int’l Ass’n of Machinists v. Wis. Emp’t Relations Comm’n, 427 U.S. 132 (1976). preemption,29×29. See City of Seattle, 274 F. Supp. 3d at 1171–74. which forbids state regulation of activity Congress intentionally left free from regulation.30×30. See Machinists, 427 U.S. 132 at 146–47. The court determined that independent contractors are less like supervisors — who are intentionally exempted from collective bargaining — and more like agricultural workers — whose state right to collective bargaining was left open by Congress.31×31. See City of Seattle, 274 F. Supp. 3d at 1173–74. The Chamber appealed the decision.32×32. The district court also dismissed the remaining state and federal law claims, but the Chamber did not appeal these holdings. See id. at 1174–76; City of Seattle, 890 F.3d at 775–76.

The Ninth Circuit affirmed in part, reversed in part, and remanded.33×33. City of Seattle, 890 F.3d at 795. Writing for the panel, Judge Smith34×34. Judge Smith was joined by Judges Murguia and Robreno. Judge Robreno was sitting by designation from the Eastern District of Pennsylvania. reversed the district court on Parker immunity, holding that the Ordinance fails both Midcal prongs.35×35. City of Seattle, 890 F.3d at 782, 787. The panel found that the state statutes offered by the City did not “plainly show” either that the legislature contemplated allowing rideshare “drivers to price-fix their compensation,” or that such an “anticompetitive result” was “foreseeable.”36×36. Id. at 783. Regulations are clearly articulated by the state “if the anticompetitive effect was the ‘foreseeable result’ of what the State authorized.” Id. at 782 (internal quotation marks omitted) (quoting FTC v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 227 (2013)). In the Ninth Circuit, the clear articulation prong first requires a plain showing of state authorization, after which foreseeability is used to define “the reach of antitrust immunity.” Id. at 783 (quoting Shames v. Cal. Travel & Tourism Comm’n, 626 F.3d 1079, 1084 (9th Cir. 2010)). The statutes relied upon by Seattle authorized municipal regulation of “for hire transportation services without liability under federal antitrust laws,”37×37. Wash. Rev. Code Ann. § 46.72.001 (West 2019). but the panel distinguished transportation services from the “payment arrangements” between the rideshare companies and their drivers.38×38. City of Seattle, 890 F.3d at 784. But see O’Connor v. Uber Techs., Inc., 82 F. Supp. 3d 1133, 1141 (N.D. Cal. 2015) (“Uber does not simply sell software; it sells rides. Uber is no more a ‘technology company’ than Yellow Cab is a ‘technology company’ because it uses CB radios to dispatch . . . .”); Cotter v. Lyft, Inc., 60 F. Supp. 3d 1067, 1078 (N.D. Cal. 2015). Likewise, the statute authorized the regulation of for-hire vehicles, but not that of the fees charged by rideshare companies through their mobile apps.39×39. City of Seattle, 890 F.3d at 785. To make this distinction, the panel relied on Medic Air Corp. v. Air Ambulance Authority, 843 F.2d 1187, 1189–90 (9th Cir. 1988).

The panel then turned to the state supervision prong, explaining first that state supervision is important because a nonstate entity claiming immunity may have self-dealing interests divergent from the state’s “definition of the public good.”40×40. City of Seattle, 890 F.3d at 787 (quoting N.C. State Bd. of Dental Exam’rs v. FTC, 135 S. Ct. 1101, 1112 (2015)); see id. at 787–88. The panel continued that the Ordinance must meet the supervision prong even though municipalities have been held exempt from state supervision under Town of Hallie v. City of Eau Claire.41×41. 471 U.S. 34, 47 (1985) (“[A]ctive state supervision is not a prerequisite to exemption from the antitrust laws where the actor is a municipality rather than a private party.”); see City of Seattle, 890 F.3d at 788. To the extent that municipalities receive any state supervision exemption under Hallie, it is because of their political accountability and lack of incentive for self-dealing — conditions “eviscerated by the involvement of private parties in this case.”42×42. City of Seattle, 890 F.3d at 790; see id. at 788–90. The panel concluded that the Ordinance was “[c]learly” not supervised by the state.43×43. Id. at 789. On remand, the panel allowed the district court to revisit which mode of antitrust preemption analysis applies: per se or rule of reason.44×44. Id. at 781. For an explanation of the two modes of analysis, see infra pp. 2365–66.

Judge Smith affirmed the district court’s holding on NLRA preemption. Because the Ordinance expressly disclaims the need to determine the legal status of drivers, and the Chamber failed to show that the NLRB would rule in the Chamber’s favor with respect to any such determination, Garmon preemption did not apply.45×45. See City of Seattle, 890 F.3d at 794–95. The panel then cited multiple reasons why Machinists preemption did not apply either, including that legislative history about excluding independent contractors from the NLRB definition of “employees” revealed that Congress intended to return to an earlier definition, not preempt collective bargaining regulation, and that demonstrating the exclusion of independent contractors from “employees,” without more, is insufficient for a showing of Machinists preemption.46×46. Id. at 793.

That the Ninth Circuit could deny Parker immunity in this case reveals the doctrine to be a paper wall separating federal courts from substantive review of city lawmaking. Although the Ordinance failed both Midcal prongs, it is useful to set aside the active supervision prong and focus on the risks presented by the court’s clear articulation analysis.47×47. Relative to the clear articulation requirement, the Court has provided relatively little guidance as to what constitutes active supervision. See Rebecca Haw Allensworth, The New Antitrust Federalism, 102 Va. L. Rev. 1387, 1434–35 (2016). In the most recent Parker decision, North Carolina State Board of Dental Examiners v. FTC, 135 S. Ct. 1101 (2015), the Court signaled interest in further developing this prong. See Allensworth, supra at 1434–35. In this case, doctrine did not compel the panel to interpret the state statute so narrowly. The policy justifications underlying the clear articulation prong did not demand a narrow construction either. Even so, the panel rigidly interpreted the Washington statutes, paving the way for hopeful litigants to do the same. Narrowing the clear articulation prong narrows the Midcal test, opening local regulation to substantive review by federal courts. The tactic deployed by the panel, and its consequences, can make vulnerable large swathes of local regulation.

The panel did not need to interpret the statutes so narrowly; the Supreme Court actually eschews extreme applications of the clear articulation doctrine, offering instead a Goldilocks directive. To the panel’s partial credit, courts should not apply the clear articulation prong “too loosely.”48×48. FTC v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 229 (2013). Broad applications “impede [states’] freedom of action,”49×49. FTC v. Ticor Title Ins. Co., 504 U.S. 621, 635 (1992). forcing them to play legislative whack-a-mole and “disclaim” anticompetitive conduct as it pops up.50×50. Phoebe Putney, 568 U.S. at 236 (“[L]oose application of the clear-articulation test . . . effectively requir[es] States to disclaim any intent to displace competition to avoid inadvertently authorizing anticompetitive conduct.”). Instead, courts should construe statutes narrowly enough to make legislatures take political responsibility for anticompetitive conduct.51×51. See Ticor Title, 504 U.S. at 636. However, courts should not take an exceedingly narrow view of state authorization. Despite the prong’s name, the Court has rejected calls for a clear statement rule.52×52. Town of Hallie v. City of Eau Claire, 471 U.S. 34, 43 (1985); see also City of Seattle, 890 F.3d at 782 (acknowledging that although an express statement is not required, a “plain and clear” showing of specific state authorization is required). Courts that demand an express statement of state authorization adopt “an unrealistic view of how [state] legislatures work,”53×53. Town of Hallie, 471 U.S. at 43. and undermine local authority.54×54. See id. at 43–44.

Nor do the policy goals underlying clear articulation doctrine demand reading state statutes narrowly. The active supervision prong is a strong check against states accidentally authorizing anticompetitive conduct; if a state does not authorize the given conduct, it won’t have a mechanism to supervise it.55×55. See id. at 46 (“[T]he requirement of active state supervision serves essentially an evidentiary function: it is one way of ensuring that the actor is engaging in the challenged conduct pursuant to state policy.”). And it is unclear that federal courts are well positioned to distinguish state statutes that accidentally authorize anticompetitive conduct from those that are intentionally flexible — state drafting procedures and statutory norms differ from their federal counterparts.56×56. See Richard A. Briffault, Beyond Congress: The Study of State and Local Legislatures, 7 N.Y.U. J. Legis. & Pub. Pol’y 23, 24–30 (2003). Moreover, to the extent that political accountability concerns drive the need for states to take responsibility for anticompetitive conduct,57×57. See FTC v. Ticor Title Ins. Co., 504 U.S. 621, 636 (1992). local regulation is doubly accountable: it is sensitive to lobbying pressures and elections at both state and local levels.58×58. State voters can compel the state to contract local power, preempt its exercise, and tinker with the authorization and supervision schemes; city voters can vote directly on the local regulation.

The Washington statutes at issue can be comfortably interpreted under the clear articulation doctrine. The statutes are not a general grant of authority to cities in the way of home rule59×59. See Cmty. Commc’ns Co. v. City of Boulder, 455 U.S. 40, 56 (1982) (finding home rule powers insufficient to authorize anticompetitive conduct). Home rule gives cities powers of self-government. See id. at 43. or corporate powers.60×60. See FTC v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 227–28 (2013) (finding only a general grant of corporate powers and no contemplation of anticompetitive conduct). Grants of corporate power give political subdivisions certain corporate powers and functions for the purpose of participating in the marketplace. See, e.g., id. at 227 n.6, 228. They are directed toward transportation services and enumerate categories of and purposes for local regulation.61×61. The statute sets out six categories of regulation, such as licensing requirements, rates charged, routes and operations, “safety and equipment requirements,” and “[a]ny other requirements adopted to ensure safe and reliable for hire vehicle transportation service.” Wash. Rev. Code Ann. § 46.72.160 (West 2019). The statutes even expressly authorize anticompetitive local regulation.62×62. The legislature authorizes municipal regulation of “for hire transportation services without liability under federal antitrust laws.” Wash. Rev. Code Ann. § 46.72.001 (West 2019); cf. Cantor v. Detroit Edison Co., 428 U.S. 579, 584–85 (1976) (finding that widespread regulation of industry failed to authorize anticompetitive conduct); Medic Air Corp. v. Air Ambulance Auth., 843 F.2d 1187, 1189 (9th Cir. 1988) (finding no contemplation of authorization of anticompetitive conduct). In order to conclude that the Ordinance was not clearly authorized by the state, the panel stepped outside the Goldilocks directive and split hairs between the markets stipulated by the statute; it found that rideshare companies are not really transportation services at all, but vendors in a market of drivers, whereas the statute concerns consumer markets.63×63. See City of Seattle, 890 F.3d at 785. But the statute concerns both driver and consumer regulation,64×64. The statute permits regulation of “entry into the business of providing for hire vehicle transportation services” and “the rates charged for providing for hire vehicle transportation service.” Wash. Rev. Code Ann. § 46.72.160 (1), (3) (West 2019). and this interpretation of rideshare companies elides their role as market coordinators.65×65. See Sanjukta M. Paul, Uber as For-Profit Hiring Hall: A Price-Fixing Paradox and Its Implications, 38 Berkeley J. Emp’t & Lab. L. 233, 237–39 (2017); see also sources cited supra note 38 (rejecting arguments that rideshare companies are not in the business of selling rides). With this decision, hopeful plaintiffs are given a roadmap to challenge local regulation: if you cannot prevail on the statutory text, recharacterize the anticompetitive conduct until no statute clearly authorizes the action.

By taking an especially narrow view of clear articulation, the panel opens the door to municipal Lochnerism. After a federal court denies a city Parker immunity, it judges whether the local regulation is preempted by the Sherman Act.66×66. See Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 885–86 (2007). See generally City of Columbia v. Omni Outdoor Advert., Inc. 499 U.S. 365, 370 (1991); Bus. Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 723, 733–34 (1988); Rice v. Norman Williams Co., 458 U.S. 654, 661 (1982). Some regulation can be categorically preempted if it constitutes a per se antitrust violation, conduct like horizontal price fixing that has been previously identified by courts as always anticompetitive.67×67. See Leegin Creative Leather Prods., Inc., 551 U.S. at 885–86. Everything else is judged under a “rule of reason” test, by which courts weigh the restraint on trade against consumer interests.68×68. See id. Under either analysis, but especially under the rule of reason test, judges can act as “superlegislature” over politically accountable local economic regulation.69×69. See Garland, supra note 12, at 510 (quoting Ferguson v. Skrupa, 372 U.S. 726, 731 (1963)). The Sherman Act’s breadth70×70. See 15 U.S.C. § 1 (2012) (prohibiting all restraints on trade). empowers federal courts in two respects reminiscent of Lochner. First, its broad pro-competitive mandate allows unelected judges to evaluate the wisdom of policy choices against a statutory liberty of contract.71×71. See Lochner v. New York, 198 U.S. 45, 57 (1905) (arguing that the regulation’s “end itself must be appropriate and legitimate”); see also 324 Liquor Corp. v. Duffy, 479 U.S. 335, 359–60 (1987) (O’Connor, J., dissenting) (“[I]n a manner reminiscent of the long-repudiated Lochner . . . , the Court strikes down the ABC Law because it concludes that the law was not ‘effective’ . . . .” (citation omitted)). Second, the imperial scope of the Act subjects large tracts of local regulation to judicial scrutiny.72×72. See William H. Page, Antitrust, Federalism, and the Regulatory Process: A Reconstruction and Critique of the State Action Exemption After Midcal Aluminum, 61 B.U. L. Rev. 1099, 1107 (1981) (“[G]iven the virtually unlimited jurisdictional reach of the Sherman Act, [allowing it full preemptive effect] would effectively forbid all state regulation of price and entry.”). Before he took the bench, Judge Garland, comparing proposals for a narrowed Parker immunity to Lochner, warned that “regulations as disparate as zoning and occupational licensing, exclusive franchises and rent control, minimum wages and minimum hours could all be overturned.”73×73. Garland, supra note 12, at 510.

City of Columbia v. Omni Outdoor Advertising, Inc.74×74. 499 U.S. 365 (1991). can illustrate both the application and the consequences of the panel’s narrowing technique. There, the Court was asked whether zoning powers granted by the state authorized the anticompetitive effects of a city ordinance restricting billboards.75×75. See id. at 368–72. Justice Scalia answered in the affirmative.76×76. Id. at 384. He wrote that a narrow clear articulation prong would have “unacceptable consequences,”77×77. Id. at 371. and agreed with Professors Philip Areeda and Herbert Hovenkamp, who reasoned that demands for “unqualified ‘author-ity’ . . . inevitably [transform the antitrust court into] the standard reviewer . . . of state and local activity.”78×78. Id. at 371–72 (quoting Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law 145 (Supp. 1989)). Imagine the Ninth Circuit panel encountered these facts. Using the narrowing technique, the panel could find the statute lacking. While the state may foresee anticompetitive land use decisions with respect to businesses, billboards do not displace businesses, but instead restrict consumers’ market knowledge. Denying Parker immunity, the panel could review the policy’s merits.

After Seattle’s petition for rehearing en banc was denied,79×79. City of Seattle, 890 F.3d 769 (9th Cir. 2018), reh’g en banc denied, No. 17-35640 (9th Cir. Sep. 14, 2018). the City blinked. At the beginning of 2019 and in the midst of the remand, the City Council amended the Ordinance to remove driver earnings from the collective bargaining topics.80×80. Seattle, Wash., Municipal Code § 6.310.735 (2015) (amended 2019). This was a missed opportunity. Localism and “federalism all the way down”81×81. Heather K. Gerken, The Supreme Court 2009 Term — Foreword: Federalism All the Way Down, 124 Harv. L. Rev. 4, 8 (2010). too can promote the free competition values expressed by federal antitrust law.82×82. See FTC v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 225 (2013); Frank H. Easterbrook, Antitrust and the Economics of Federalism, 26 J.L. & Econ. 23, 28–29, 32–33, 36–37, 41 (1983); see also New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting) (“[A] single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”). To the extent that a great deal of regulation in our system of dual sovereignty is broadly authorized for local implementation, a narrow Parker immunity can create a backdoor for federal scrutiny of state affairs.83×83. Broad state authorization of local regulation is a practical necessity; the United States had over 89,000 local governments as of 2012. See Census Bureau Reports There Are 89,004 Local Governments in the United States, U.S. Census Bureau (Aug. 30, 2012), https://www.census.gov/newsroom/releases/archives/governments/cb12-161.html [https://perma.cc/7QQ9-YSR3]. Moreover, the doctrinal move here shares its context with the broader neoliberalization of antitrust.84×84. See, e.g., Lina M. Khan, Amazon’s Antitrust Paradox, 126 Yale L.J. 710, 716 (2017); Lina Khan & Sandeep Vaheesan, Market Power and Inequality: The Antitrust Counterrevolution and Its Discontents, 11 Harv. L. & Pol’y Rev. 235, 236–37 (2017); Sanjukta Paul, Antitrust as Allocator of Coordination Rights, 67 UCLA L. Rev. (forthcoming 2020) (manuscript at 6–7) (on file with the Harvard Law School Library), available at https://ssrn.com/abstract=3337861,”>https://ssrn.com/abstract=3337861″>https://ssrn.com/abstract=3337861, [https://perma.cc/FV5U-RHV4]. Although Lochner was once firmly anticanon, it is regaining currency in scholarly interest and judicial practice.85×85. See, e.g., Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2381–83 (2018) (Breyer, J., dissenting) (charging the majority with using the First Amendment to strike down economic and social laws of the same kind that were struck down in Lochner); Thomas B. Colby & Peter J. Smith, The Return of Lochner, 100 Cornell L. Rev. 527, 529–32 (2015). At bottom, this decision stands as another rejection of the unique and valuable role cities play in the order of American governance.86×86. Cf. David J. Barron, The Promise of Cooley’s City: Traces of Local Constitutionalism, 147 U. Pa. L. Rev. 487, 491 (1999) (contrasting the important functions of cities and towns with their treatment under federal and state law). Other courts would do well to decide otherwise.