Harvard Law Review Harvard Law Review Harvard Law Review

Antitrust

Antitrust Federalism, Preemption, and Judge-Made Law

Both the United States government and the governments of the fifty states use antitrust principles to regulate firms. A collection of federal statutes, first and foremost the Sherman Act,1×1. 15 U.S.C. §§ 1–7 (2018). outlaws anticompetitive behavior under federal law. The federal executive branch, through the Federal Trade Commission (FTC) and the Department of Justice’s Antitrust Division (DOJ), enforces the federal statutes.2×2. 1 Louis Altman & Malla Pollack, Callmann on Unfair Competition, Trademarks and Monopolies § 4:51 (4th ed., update 2019). Meanwhile, each state has its own antitrust statutes outlawing anticompetitive behavior.3×3. E.g., Donnelly Act, N.Y. Gen. Bus. Law §§ 340–347 (West 2020); see 1 Altman & Pollack, supra note 2, § 4:52. Kansas was the first state to pass an antitrust law; it did so in 1889, a year before Congress passed the Sherman Act. David Millon, The First Antitrust Statute, 29 Washburn L.J. 141, 141 (1990). The states’ agencies enforce their own antitrust laws, and they can enforce federal antitrust law as parens patriae4×4. Parens patriae is “[a] doctrine by which a government has standing to prosecute a lawsuit on behalf of a citizen.” Parens Patriae, Black’s Law Dictionary (11th ed. 2019). for full treble damages thanks to the Hart-Scott-Rodino Antitrust Improvements Act of 19765×5. Pub. L. No. 94-435, 90 Stat. 1383 (codified in scattered sections of 15 U.S.C.); see id. at 1394. (Hart-Scott-Rodino). However, when state legislation itself produces anticompetitive effects that seem to violate federal antitrust principles, the state gets a free pass: “[A]nticompetitive restraints are immune from antitrust scrutiny if they are attributable to an act of ‘the State as sovereign.’”6×6. Einer Richard Elhauge, The Scope of Antitrust Process, 104 Harv. L. Rev. 667, 673 (1991) (quoting S. Motor Carriers Rate Conference, Inc. v. United States, 471 U.S. 48, 64 (1985)). The states’ immunity is often labeled “Parker immunity” after the seminal case Parker v. Brown, 317 U.S. 341 (1943). See, e.g., Jean Wegman Burns, Embracing Both Faces of Antitrust Federalism: Parker and ARC America Corp., 68 Antitrust L.J. 29, 43 (2000).

Wherever the federal and state governments share regulatory authority, federalism concerns naturally follow. Federalism refers to the division, overlap, and balance of power between the federal and state governments in our federal system.7×7. See Barry Friedman, Valuing Federalism, 82 Minn. L. Rev. 317, 323 (1997). The emergence of a strong national government since the New Deal has turned federalism into a state-centric concept about protecting the states’ role in that balance.8×8. See Erwin Chemerinsky, The Federalism Revolution, 31 N.M. L. Rev. 7, 8 (2001). This state-centric federalism is partially baked into the Constitution: for example, the Tenth Amendment confirms that the Constitution reserves powers not delegated to the United States for the fifty states,9×9. U.S. Const. amend. X. Under current constitutional jurisprudence, the Tenth Amendment confirms, but does not grant, this reservation. See Murphy v. NCAA, 138 S. Ct. 1461, 1476 (2018). and some scholars have attributed a state-centric view of federalism to the Guarantee Clause.10×10. U.S. Const. art. IV, § 4; see, e.g., Deborah Jones Merritt, The Guarantee Clause and State Autonomy: Federalism for a Third Century, 88 Colum. L. Rev. 1, 22–36 (1988). However, when, as with antitrust, the federal and state governments share concurrent regulatory authority, the Constitution alone cannot resolve the federalism-nationalism balancing act.

Even when it is not a constitutional hurdle, federalism is still a relevant constitutional value. The Framers embraced federalism for its policy virtues,11×11. See, e.g., The Federalist No. 46 (James Madison). and contemporary judges and scholars laud federalism for its modern-day policy perks.12×12. See, e.g., Murphy, 138 S. Ct. at 1477 (describing the policy benefits of anticommandeering as a federalism tool); Jessica Bulman-Pozen & Heather K. Gerken, Essay, Uncooperative Federalism, 118 Yale L.J. 1256, 1284–94 (2009) (describing the value of federalism in the modern United States). The Supreme Court often invokes federalism in the form of a presumption that Congress does not lightly intrude on state sovereignty.13×13. See, e.g., Gregory v. Ashcroft, 501 U.S. 452, 460–61 (1991) (applying a clear statement rule based on this presumption). One example is the Court’s presumption against preemption: a party alleging federal preemption of state law faces a judicial presumption that Congress did not intend to make that choice.14×14. See Maryland v. Louisiana, 451 U.S. 725, 746 (1981) (stating that, in the context of preemption of state laws, “[c]onsideration . . . starts with the basic assumption that Congress did not intend to displace state law”). That presumption is validated by Congress’s choice to refrain from preempting state law in the antitrust arena: state and federal antitrust laws have coexisted since the federal government’s first steps into the arena in 1890.15×15. See James May, Antitrust Practice and Procedure in the Formative Era: The Constitutional and Conceptual Reach of State Antitrust Law, 1880–1918, 135 U. Pa. L. Rev. 495, 499 (1987). In fact, some state antitrust laws preceded the Sherman Act. See id.

This congressional restraint is controversial, and likely to grow more so. Some scholars have argued powerfully that Congress should preempt state antitrust laws.16×16. See, e.g., Robert W. Hahn & Anne Layne-Farrar, Federalism in Antitrust, 26 Harv. J.L. & Pub. Pol’y 877, 884–907 (2003); Richard A. Posner, Address, Federalism and the Enforcement of Antitrust Laws by State Attorneys General, 2 Geo. J.L. & Pub. Pol’y 5, 12–13 (2004). These arguments may gain renewed prominence, as antitrust as a whole has recently achieved greater political salience than it has enjoyed in a century.17×17. Carl Shapiro, Antitrust in a Time of Populism, 61 Int’l J. Indus. Org. 714, 714–15 (2018) (“Antitrust is sexy again. . . . Not since 1912, when Teddy Roosevelt ran for President emphasizing the need to control corporate power, have antitrust issues had such political salience.”). In the state context, attorneys general have increasingly taken antitrust action in high-profile matters the federal government has declined to pursue. In 2019, states opposed the merger between Sprint and T-Mobile,18×18. See Victoria Graham, States Keep Door Open for Better Antitrust Deal from T-Mobile, Bloomberg L. (Dec. 3, 2019, 5:30 AM), https://news.bloomberglaw.com/mergers-and-antitrust/states-keep-door-open-for-better-antitrust-deal-from-t-mobile [https://perma.cc/866M-NT6J]. and many began to investigate potential antitrust violations in Big Tech.19×19. Leah Nylen et al., State Big Tech Antitrust Probes Picking Up Steam, MLex (July 31, 2019), https://mlexmarketinsight.com/insights-center/editors-picks/antitrust/north-america/state-big-tech-antitrust-probes-picking-up-steam [https://perma.cc/M7UL-LEZU]. While some recent, high-profile state antitrust actions have been brought under federal antitrust laws,20×20. See, e.g., Redacted Second Amended Complaint at 34, New York v. Deutsche Telekom AG, No. 19-cv-05434 (S.D.N.Y. Feb. 11, 2020) (bringing claims under federal law only). others have been brought under state law.21×21. See, e.g., Complaint for Violations of the Cartwright Act (Bus. & Prof. Code § 16720 et seq.) at 41–45, People ex rel. Becerra v. Sutter Health, No. CGC-18-565398 (Cal. Super. Ct. Mar. 29, 2018) (bringing claims under state law only). Moreover, a number of the current state antitrust actions are at the investigatory stage22×22. See Nylen et al., supra note 19.  — states could potentially bring federal claims, state claims, or both. Newsworthy state involvement in antitrust policing is bringing attention to the states’ antitrust role more generally, and that attention will likely bring scrutiny to the oddity of America’s competing antitrust systems.

This Note argues that, in considering its position within this debate, Congress should grapple with federal antitrust law’s peculiar status as a largely judicially created regulatory regime. Congress should be wary of allowing federal judge-made law to preempt state legislative power. Even when the federal government preempts state legislation, the federalism balance is partially preserved by democratic checks on federal power. Yet, when a nondemocratic branch is making the law, those checks disappear. Moreover, the federal judiciary is a uniquely poor policymaking body; its lack of policymaking chops does not support overriding states’ policy choices. These factors highlight the need for Congress to account for the degree to which current antitrust law is largely judge made.

Part I outlines the general landscape of antitrust federalism. It first describes antitrust federalism’s three components and then surveys arguments for and against maintaining one of those components: the coexistence of state and federal antitrust laws. Following this survey, Part II offers a new defense of the current system: federal antitrust law’s judge-made status makes it particularly unsuitable to preemption. Finally, Part III compares antitrust’s judge-made law to other preemptive federal common law, concluding that federal antitrust preemption would be uniquely susceptible to Part II’s criticism.

I. The Antitrust Federalism Landscape

Antitrust federalism, meaning the space carved out for the states in the more generally federal antitrust arena, can be thought of as made up of two “swords” — the first the states’ ability to bring suit under federal antitrust law and the second their ability to enact and enforce their own state antitrust laws — and one “shield” — immunity from federal antitrust law for state actions.23×23. Burns, supra note 6, at 29. The swords allow states to attack antitrust offenders, while the shield allows states to defend against federal antitrust action.

All three elements of antitrust federalism find their roots in con-gressional action or the courts’ interpretation of congressional inaction. The power to enforce federal antitrust law as parens patriae for full treble damages — the first sword — was granted to the states by Congress in Hart-Scott-Rodino.24×24. See 15 U.S.C. § 15c (2018). States could sue as parens patriae before Hart-Scott-Rodino, but not for treble damages. See Susan Beth Farmer, More Lessons from the Laboratories: Cy Pres Distributions in Parens Patriae Antitrust Actions Brought by State Attorneys General, 68 Fordham L. Rev. 361, 372–73, 376 (1999). On the judicial front, the Supreme Court acknowledged state immunity from federal antitrust actions — the shield — in Parker v. Brown,25×25. 317 U.S. 341 (1943). Although this Note deals exclusively with federalism concerns surrounding the preemption of state antitrust laws, for an interesting discussion of the federalism concerns surrounding Parker immunity, see Elhauge, supra note 6. noting that the Sherman Act did not explicitly mention its application to state action.26×26. Parker, 317 U.S. at 350–52. Finally, when the Court confirmed that states’ ability to make their own antitrust laws — the second sword and the one discussed in this Note — was not preempted in California v. ARC America Corp.,27×27. 490 U.S. 93 (1989). it considered the same Sherman Act silence.28×28. Id. at 101–02. ARC America did not foreclose the possibility that some other state antitrust laws would be federally preempted, and courts have sometimes found preemption of state laws applied to interstate commerce when that application was inconsistent with federal antitrust law. See 1 Altman & Pollack, supra note 2, § 4:53 (collecting cases). Moreover, “[d]ismissal of a federal antitrust claim on the merits precludes, on res judicata grounds, subsequent refiling of the same claim under state antitrust law.” Id. Finally, some federal law outside the antitrust arena has preempted state antitrust law. See, e.g., Myers v. Pension Fund, Local One Amalgamated Lithographers of Am., 704 F. Supp. 343, 345 (E.D.N.Y. 1989) (holding that the Employee Retirement Income Security Act of 1974 (ERISA) preempted a New York antitrust law).

This is all to say that antitrust’s federalism tools are congressionally, not constitutionally, given rights and are therefore congressionally rescindable. Congress could amend Hart-Scott-Rodino or make explicit that the Sherman Act applies to state action.29×29. See Alan J. Meese, Essay, Antitrust Federalism and State Restraints of Interstate Commerce: An Essay for Professor Hovenkamp, 100 Iowa L. Rev. 2161, 2165 & n.19 (2015). For the rare view that Parker immunity is constitutionally granted and could not be congressionally overridden, see Burns, supra note 6, at 38–39. But see Meese, supra, at 2165 n.19. In any case, there are apparently no scholars who argue that the two swords could not be rescinded. See Douglas H. Ginsburg & Scott H. Angstreich, Multinational Merger Review: Lessons from Our Federalism, 68 Antitrust L.J. 219, 228 n.32 (2000). And, crucially for this Note’s discussion, although state antitrust law is not judicially preempted, Congress could choose to expressly preempt it in the future.30×30. See Meese, supra note 29, at 2165.

There are strong policy arguments for express congressional preemption of state antitrust law. The remainder of this Part attempts to outline the general pros and cons of congressional antitrust preemption but is not meant to be exhaustive or to cover new ground. The intent is to situate Part II’s argument about federalism and preemption by judge-made law within the broader policy landscape.

A. The Patchwork Regime Problem

First, critics of the status quo argue that a patchwork regime of state antitrust laws can make it expensive for companies that operate across state borders to comply. State and federal regimes share similar philosophies regarding most of antitrust law.31×31. See Burns, supra note 6, at 35. But state antitrust laws do not perfectly mirror their federal counterparts — and the antitrust laws of the different states are heterogeneous themselves.32×32. See Donald L. Flexner & Mark A. Racanelli, State and Federal Antitrust Enforcement in the United States: Collision or Harmony?, 9 Conn. J. Int’l L. 501, 510–11 (1994). Disputes are concentrated in a few areas of the doctrine, like vertical restraints and mergers.33×33. See Burns, supra note 6, at 35–36. For example, states often focus on damage to intrabrand competition when enforcing limits on vertical restraints, whereas federal antitrust law focuses primarily on interbrand competition.34×34. Id. at 36. Additionally, state merger guidelines often materially differ from federal guidelines,35×35. See Ginsburg & Angstreich, supra note 29, at 227–28. But see id. at 228 n.29 (noting that state merger guidelines have moved closer to federal guidelines in recent decades). and states are likelier to define markets “more narrowly,” “refus[e] to consider efficiencies” favored by federal agencies, and show a concern for local jobs and competitors that does not “enter . . . the [federal] calculus.”36×36. Id. at 228. An inconsistent antitrust regime that may conflict between states could cause economic inefficiency, for example by discouraging companies from undertaking what might otherwise be an economically efficient merger.37×37. See Hahn & Layne-Farrar, supra note 16, at 891; see also Robert W. Hahn & Anne Layne-Farrar, Perspective, The Case for Federal Preemption in Antitrust Enforcement, Antitrust, Spring 2004, at 79, 80–81.

This critique relies in part on the federal government having a better approach to vertical restraints and mergers, and that is anything but clear. The classic federalism argument that states function as laboratories of democracy38×38. New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting).  applies here: antitrust law is far from settled, and having multiple regimes allows for testing different theories. For example, some scholars argue that the states are correct to consider intrabrand competition’s effects on price, especially in certain markets.39×39. See, e.g., Warren S. Grimes, The Life Cycle of a Venerable Precedent: GTE Sylvania and the Future of Vertical Restraints Law, Antitrust, Fall 2002, at 27, 27–28. Similarly, in the merger context, there is support for both the states’ refusal to consider only economic efficiency40×40. See, e.g., Walter Adams & James W. Brock, Antitrust, Ideology, and the Arabesques of Economic Theory, 66 U. Colo. L. Rev. 257, 302–09, 316–25 (1995). and their push for heightened antimerger enforcement.41×41. See, e.g., Jonathan B. Baker & Carl Shapiro, Reinvigorating Horizontal Merger Enforcement, in How the Chicago School Overshot the Mark 235, 257–66 (Robert Pitofsky ed., 2008). Of course, the laboratories of democracy might not work so well in the antitrust context: because of the interwoven economic effects of federal and state antitrust laws and enforcement in an interconnected national economy, determining the effects of one state’s slightly different antitrust regime would be difficult.42×42. See Hahn & Layne-Farrar, supra note 16, at 892. But federalism can still offer benefits by breaking the antitrust orthodoxy: by putting different policies on the table, a multilevel regime reminds us both that there are different possible “best” antitrust policies and that antitrust law has a variety of potential goals.43×43. See Burns, supra note 6, at 41–43.

B. The One-State Dominator Problem

Closely related to the patchwork regime problem is the one-state dominator problem: because national firms may not always be able to maintain different business practices in each state, firms could be forced to follow the law of whichever state has the strictest antitrust policy nationwide. For example, a single state could use its own antitrust laws to “challenge the largest nationwide transactions so long as it can show that the state itself, its citizens, or its economy is affected in a way that provides standing.”44×44. Robert H. Lande, When Should States Challenge Mergers: A Proposed Federal/State Balance, 35 N.Y.L. Sch. L. Rev. 1047, 1047 (1990); see id. at 1050, 1052. If a nationwide merger is illegal under one state’s laws, it may not be worth it for the firm to restructure the transaction in order to merge in all but one jurisdiction. This reality could allow for the state with the strictest antitrust policy to dominate the policy decisions of every other state and of the federal government.45×45. See id. at 1061–62.

The one-state dominator problem is exacerbated by unrecognized interstate externalities: in making its antitrust laws, a state is not forced to consider the harm or benefit to businesses based outside of its borders.46×46. Cf. Richard L. Revesz, Rehabilitating Interstate Competition: Rethinking the “Race-to-the-Bottom” Rationale for Federal Environmental Regulation, 67 N.Y.U. L. Rev. 1210, 1222 (1992) (discussing interstate externalities in the context of environmental regulation). These uninternalized externalities make it more likely that a state will overregulate. The laboratory-of-democracy defenses to the patchwork regime problem, with their variety-is-the-spice-of-life flair, fail to explain why an individual state’s antitrust regime should be allowed to dominate the policy of the entire nation.

Consider a recently passed Maryland law regulating wholesale pharmaceutical prices. The law prohibited manufacturers or wholesalers from “price gouging,” defined as “an unconscionable increase in the price of” certain drugs.47×47. Ass’n for Accessible Meds. v. Frosh, 887 F.3d 664, 666 (4th Cir. 2018) (first quoting Md. Code Ann., HealthGen. § 2-802(a) (LexisNexis 2019), invalidated by Frosh, 887 F.3d 664; and then quoting id. § 2-801(c)). Federal antitrust law does not prevent monopolists from receiving the reward of monopoly prices, under the theory that potential future monopoly profits encourage present investment.48×48. Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 407 (2004). The Maryland law can be viewed as a limit on this monopolist tolerance in the pharmaceutical space, preventing pharmaceutical companies from taking advantage of their dominant market position in the treatment of certain diseases. Not all states had decided to regulate drug prices, with most hewing more closely to the general rule of monopoly tolerance.49×49. See Recent Case, Association for Accessible Medicines v. Frosh, 887 F.3d 664 (4th Cir. 2018), 132 Harv. L. Rev. 1748, 1748 & n.2 (2019). Based on its drafting, however, Maryland’s law could have had significant implications nationwide: even assuming the law required some sort of connection to an eventual consumer sale in Maryland,50×50. The majority and the dissent in Association for Accessible Medicines v. Frosh, 887 F.3d 664, disagreed on whether the law as written regulated even transactions with no nexus to Maryland whatsoever. Compare id. at 670–71, with id. at 679–80 (Wynn, J., dissenting). the law regulated a wholesaler’s initial sale, whether or not that sale occurred in Maryland, so long as the drug was eventually resold in Maryland.51×51. Id. at 671–72 (majority opinion). As such, any manufacturer who sold drugs to a Maryland retailer would have to set their initial prices in consideration of Maryland’s law. Pricing is a core antitrust issue; why should Maryland be able to set the nation’s pricing policy?

Or consider the ability of indirect purchasers to sue under antitrust laws. In Illinois Brick Co. v. Illinois,52×52. 431 U.S. 720 (1977). the Supreme Court held that only direct purchasers of a price-fixed good or service, not subsequent indirect purchasers, could sue for treble damages under the Clayton Act.53×53. 15 U.S.C. §§ 12–27 (2018); 29 U.S.C. §§ 52–53 (2018); see Ill. Brick, 431 U.S. at 746. In response, twenty-six states passed “‘Illinois Brick–repealer laws’ authorizing indirect purchasers to bring damages suits under state antitrust law.”54×54. Opening Brief of Defendant-Appellant Qualcomm Inc. at 58, Stromberg v. Qualcomm Inc., No. 19-15159 (9th Cir. June 3, 2019) [hereinafter Qualcomm Brief]. In ARC America, the Court held the Illinois Brick–repealer laws valid. 490 U.S. 93, 97, 101–02 (1989). But these twenty-six states have an impact even on the residents of nonrepealer states. In a class action currently on appeal in the Ninth Circuit, a district court applied California antitrust law — including California’s repealer law — to a nationwide class that included class members from nonrepealer states.55×55. See Qualcomm Brief, supra note 54, at 1, 5, 63. The defendant-appellant has argued that this application undermines the nonrepealer states’ interest in choosing their own consumer-business balance.56×56. Id. at 63–64. Last Term, the Supreme Court endorsed a victim-friendly definition of “direct purchaser” in Apple Inc. v. Pepper, 139 S. Ct. 1514 (2019). See The Supreme Court, 2018 Term — Leading Cases, 133 Harv. L. Rev. 242, 384–85, 390–91 (2019). After Apple, Illinois Brick’s prohibition is less of a boon to defendants, and there may be less room between federal law and the law of Illinois Brick–repealer states.

The Maryland and Ninth Circuit examples may be more bogeymen than real threats to federalism. First, alternate doctrines aside from antitrust preemption work to keep individual state interests in check. For example, the Fourth Circuit enjoined enforcement of the Maryland law on dormant commerce clause grounds.57×57. Ass’n for Accessible Meds. v. Frosh, 887 F.3d 664, 674 (4th Cir. 2018). Where one state intrudes too much on other states’ ability to regulate antitrust — where “[t]he potential for ‘the kind of competing and interlocking local economic regulation that the Commerce Clause was meant to preclude’ is . . . both real and significant”58×58. Id. (quoting Healy v. Beer Inst., 491 U.S. 324, 337 (1989)). — the Constitution, rather than Congress, can prevent the one-state dominator problem’s greatest harms. Dormant commerce clause challenges are not limited to the Maryland case’s facts. In fact, the Fourth Circuit dissent complained that the majority’s logic would invalidate other state antitrust laws, including Illinois Brick–repealer laws.59×59. Id. at 688 (Wynn, J., dissenting).

Second, the trouncing of federalism in cases like these is often overstated. Take the defendant-appellant’s depiction of the interests in the Ninth Circuit case as an example of exaggerated federalism costs. The district court found that the nonrepealer states had no interest in having their laws applied because the defendant-appellant was a California company; California’s more consumer-friendly law would only help nonrepealer-state residents, not hurt nonrepealer-state businesses.60×60. Qualcomm Brief, supra note 54, at 65–66. If the nonrepealer states have an interest in denying their own consumers access to relief when there is no benefit to their own businesses, it seems tangential to an interest in striking their own consumer-business balances. Instead, a choice to prioritize foreign defendants over in-state consumers appears more like an attempt to govern the national consumer-business balance, a choice imbued with far less federalism oomph than the defendant-appellant portrayed.

Whether exaggerated or not, a worry that antitrust federalism allows one state to dominate national antitrust policy weighs in favor of congressional antitrust preemption. This problem, however, is not unique to antitrust. Any area of law in which states fail to internalize the harms of overregulation, meaning any law that regulates businesses with a national footprint, could be dominated by one state.61×61. The interstate externalities that the one-state dominator problem implicates are separate from the “race-to-the-bottom” problems that Professor Richard Revesz discusses in the environmental arena. See Revesz, supra note 46, at 1222–23. In a race to the bottom, the concern is that states will compete with each other for business by lowering regulation below the otherwise-optimal level. Id. at 1213–19. The concern of a dominated antitrust regime, on the other hand, is that one state will overregulate, and, because a national business cannot easily exit a single state, will thereby drag other states upward. In the environmental context, the one-state dominator problem is more akin to a critique of California’s ability to set national automobile standards because of its major market for automobiles — although that ability is congressionally condoned. See id. at 1228 & n.61; Coral Davenport, Trump to Revoke California’s Authority to Set Stricter Auto Emissions Rules, N.Y. Times (Sept. 20, 2019), https://nyti.ms/305luT3 [https://perma.cc/C8KE-9JFF]. If Congress were to take the one-state dominator problem too seriously, it would swallow up huge swaths of state regulation, excluding states from their traditional role in consumer protection, at least where the largest (and potentially most worrisome) industries are implicated.

C. The Overdeterrence Problem

Third, critics argue that a multilevel antitrust regime threatens to overdeter procompetitive conduct. The policy behind much of preemption is to prevent state law from interfering with detailed, well-balanced federal regulation: obstacle preemption exists to prevent states from “stand[ing] as . . . obstacle[s] to the accomplishment and execution of the full purposes and objectives of Congress,”62×62. Hines v. Davidowitz, 312 U.S. 52, 67 (1941). and field preemption exists to prevent state interference where Congress “left no room for lower-level regulation.”63×63. Hannah J. Wiseman, Disaggregating Preemption in Energy Law, 40 Harv. Envtl. L. Rev. 293, 312 (2016). Although it is not field or obstacle preempted,64×64. But see supra note 28 (discussing rare circumstances where antitrust law is obstacle preempted). antitrust law exhibits the type of detailed regulatory balance that the preemption doctrines attempt to prevent states from damaging. Much of antitrust law is built on finding the perfect balance of standards and remedies: the law must properly deter anticompetitive acts without deterring healthy competition.65×65. Compare, e.g., Am. Soc’y of Mech. Eng’rs, Inc. v. Hydrolevel Corp., 456 U.S. 556, 572 (1982) (“A principal purpose of the antitrust private cause of action . . . [is] to deter anticompetitive practices.”), with id. at 593–94 (Powell, J., dissenting) (lamenting that overzealous antitrust law may overdeter private action that promotes public welfare). A state law that shifts remedies or standards can upset this careful balancing, thus overdeterring desirable private action.

Critics can point directly to ARC America as evidence of this overdeterrence threat. The Court’s decision in Illinois Brick, which limited suits by indirect purchasers, relied in large part on a belief that concentrating suits in direct purchasers would avoid overdeterrence.66×66. See Gregory J. Werden & Marius Schwartz, Illinois Brick and the Deterrence of Antitrust Violations — An Economic Analysis, 35 Hastings L.J. 629, 634–35 (1984). By allowing for additional suits, ARC America created extra deterrence not envisioned by the federal antitrust scheme.67×67. See Posner, supra note 16, at 13.

Like the patchwork regime critique, the overdeterrence critique is weakened if the federal regime has failed to achieve proper balancing. Many antitrust regimes around the globe adopt different balances than the United States does. The European Union, for example, differs from the United States on remedial structure, the standard for illegal unilateral conduct, and market definition, among other issues.68×68. See Einer Elhauge & Damien Geradin, Global Antitrust Law and Economics 11, 306, 369 (3d ed. 2018). Moreover, many scholars argue that the U.S. antitrust balance is off and that more enforcement is needed.69×69. See, e.g., Tim Wu, The Curse of Bigness 128–29, 131–33 (2018); Shapiro, supra note 17, at 737–41. Even if U.S. antitrust policies are getting the balance generally right, it is unlikely that the federal regime is so finely tuned that any added deterrence will destroy the balance.

D. The Misaligned Incentives Problem70×70. The final two problems focus on issues in state antitrust enforcement, and so they apply to both “swords” of antitrust federalism — state enforcement of federal antitrust law as well as state enforcement of state antitrust law.

Fourth, in the misaligned incentives problem, critics argue that states do not have proper incentives when they enforce state antitrust laws. Although state antitrust laws are supposed to mainly target intrastate antitrust violations, courts have refused to police that limit too strictly.71×71. See 1 Altman & Pollack, supra note 2, § 4:53. In an interconnected economy where seemingly hyperlocal activity can have national implications,72×72. See, e.g., Wickard v. Filburn, 317 U.S. 111, 127–28 (1942). courts have admitted that limiting state antitrust laws to cases that do not touch the national economy would “fence[] off” “a very large area . . . in which the States w[ould] be practically helpless to protect their citizens.”73×73. Commonwealth v. McHugh, 93 N.E.2d 751, 762 (Mass. 1950); see also 1 Altman & Pollack, supra note 2, § 4:53. But, even though suits under state laws may have nationwide consequences, state attorneys general lack nationwide incentives. Critics of the status quo worry that elected attorneys general are more susceptible to lobbying by state interests than are appointed federal enforcers and that a cost-benefit analysis is flawed where a state can attack a company headquartered out of state in order to protect one headquartered in state.74×74. See, e.g., Hahn & Layne-Farrar, supra note 16, at 885; Richard A. Posner, Antitrust in the New Economy, 68 Antitrust L.J. 925, 942 (2001).

These fears seem mostly imagined. The idea that elected attorneys general are bringing antitrust suits to hurt competitors of state businesses“appears to [have] little empirical support[,] . . . and none has been provided by the advocates of this position.”75×75. Katherine Mason Jones, Federalism and Concurrent Jurisdiction in Global Markets: Why a Combination of National and State Antitrust Enforcement Is a Model for Effective Economic Regulation, 30 Nw. J. Int’l L. & Bus. 285, 331 (2010); see also Harry First, The Role of States, 69 Geo. Wash. L. Rev. 1004, 1031 (2001); Lande, supra note 44, at 1067–68. Past state antitrust enforcers have stated that, while they considered state-specific factors when deciding where to spend their limited resources, those factors would be used only to choose “from among those cases that also made sense on traditional economic grounds.”76×76. Lande, supra note 44, at 1068 n.98 (citing a statement to the author by “Lloyd Constantine, while head of the NAAG Antitrust Task Force and the Chief of New York’s Antitrust Section”); see also, e.g., Jones, supra note 75, at 331–32 (discussing the criteria cited by “Patricia Conners, the head of the NAAG Multistate Task Force,” id. at 331).

And there is reason to believe that these enforcers are telling the truth. For one thing, states often make antitrust decisions that seem to go against the interests of major state employers. For example, New York antitrust enforcers have taken antitrust positions adverse to both Verizon and IBM, top New York employers.77×77. Jones, supra note 75, at 332. For another, a state that is only minutely affected by an antitrust action is unlikely to bring that action alone. If a state is only trivially affected by allegedly anticompetitive conduct, “that state is very unlikely as a practical and political matter to spend the enormous sums of money required to sustain a challenge.”78×78. Lande, supra note 44, at 1069. If a state is majorly affected but is the only state affected, then the misaligned incentives critique does not apply because there is no competing set of national incentives. And in a case that actually has major impacts in multiple states, it is unlikely that one state could act without other states wanting to join in on the enforcement.79×79. First, supra note 75, at 1031 (“Litigation against anticompetitive behavior that has multistate effects will likely be brought on a multistate basis.”). When states work together on antitrust enforcement, they tend to cooperate closely with one another, especially through the National Association of Attorneys General’s (NAAG) antitrust group.80×80. See Lande, supra note 44, at 1069. Even if an individual state might be swayed by state-specific concerns, it is unlikely that it could convince a multistate coalition to act on those concerns — the group would be forced to evaluate the action on its more national merits.81×81. First, supra note 75, at 1031.

E. The Incompetent States Problem

Finally, critics argue that state enforcers will make error-ridden antitrust choices due to a lack of resources, experience, and expertise. Whereas federal enforcers have significant budgets for antitrust enforcement, the percentage of funding set aside for antitrust enforcement by state attorneys general is minute.82×82. See, e.g., Hahn & Layne-Farrar, supra note 16, at 889 & tbl.1. Because of this lack of resources, state enforcers have been accused of staffing antitrust cases with senior attorneys who, while experienced in civil litigation generally, are antitrust novices.83×83. See id. at 887–88; Lande, supra note 44, at 1064. These factors have led critics to argue that state attorneys general handle antitrust suits poorly, clogging the judicial pipeline with questionable suits.84×84. See Posner, supra note 74, at 941. State attorneys general are accused of acting as free riders on federal actions and of making settlements more difficult rather than undertaking useful enforcement.85×85. See Hahn & Layne-Farrar, supra note 16, at 890; Lande, supra note 44, at 1063–64; Posner, supra note 74, at 940.

But there is reason to dispute critics’ claims. The critique of individual attorneys general ignores the states’ ability to work in unison. Cooperating through NAAG, states are able to build on each other’s experiences in antitrust enforcement.86×86. See First, supra note 75, at 1014–15. Thus, worries about inexperienced antitrust divisions working alone may be overstated. Although interstate coordination may weaken their point, critics can retort that most state actions are not coordinated: according to NAAG’s State Antitrust Litigation Database, only nineteen of the fifty-six civil antitrust actions brought by states between 2014 and 2019 were brought by multiple states working together,87×87. Antitrust Multistate Litigation Database, NAAG, http://app3.naag.org/antitrust/search/advSearchCivil.php [https://perma.cc/3FV7-TG2F]. The data was gathered by searching the database’s civil litigation records for all suits filed since 2014 and then checking the attached documents for lawsuit information. although many of the noncooperative suits regarded intrastate anticompetitive conduct.88×88. See, e.g., Consent Decree, State v. Tesoro Alaska Co., No. 3AN-16-CIV (Alaska Super. Ct. June 10, 2016); Assurance of Discontinuance Pursuant to G.L. Ch. 93 § 9 and G.L. Ch. 93A § 5, Commonwealth v. Nantucket Ass’n of Real Estate Brokers, No. 16-3466B (Mass. Super. Ct. Nov. 14, 2016). This same dataset, however, also undermines the critics’ argument that states act only as free riders: only nineteen of the fifty-six suits included federal participation.89×89. Antitrust Multistate Litigation Database, supra note 87. Finally, much of the criticism leveled at state attorneys general occurred before a renaissance in state law enforcement. Since Judge Posner derided the skill of state attorneys general in 2001,90×90. See Posner, supra note 74, at 941. lawyers and judges, including Chief Justice Roberts, have recognized a marked improvement in state attorney offices’ advocacy.91×91. See Kevin C. Newsom, The State Solicitor General Boom, ABA (Mar. 14, 2013), https://www.americanbar.org/groups/litigation/committees/appellate-practice/articles/2013/winter2013-0313-state-solicitor-general-boom [https://perma.cc/Z3G8-XUYT]. Whether or not Judge Posner’s critiques were valid at the turn of the century, it is unclear that the landscape remains the same today. Finally, this critique undermines the arguments, noted earlier, that state law enforcement is overdeterring competition or creating a patchwork of antitrust law. If states are nothing but free riders, then we need not worry about overdeterrence.

II. Concerns with Preemption by Judge-Made Law

In this jumble of critiques and retorts, it is not obvious that one side wins the day. In addition to the above considerations, Congress should take into account the oddity of preempting state law with federal common law in deciding whether to preempt state antitrust regimes. Even though federal antitrust is based in congressionally passed statutes, the judiciary’s interpretations of those statutes have shaped the law far more than has any statutory text. Unlike a legislature, the federal judiciary is both largely undemocratic and poorly suited to be a policymaker. These two distinctions counsel against the preemption of state antitrust laws by federal judge-made law and should enter Congress’s calculus along with Part I’s considerations.

Federal antitrust is uniquely judicial. For such a complex area of law, antitrust’s statutory grounding is rather simple. The Sherman Act, the backbone of federal antitrust, is shockingly short.92×92. John Shepard Wiley Jr., A Capture Theory of Antitrust Federalism, 99 Harv. L. Rev. 713, 713 (1986) (“Its brevity is charming — and significant.”). Substantive antitrust law stems mainly from the Sherman Act, the Clayton Act (including amendments from the Robinson-Patman Act, 15 U.S.C. § 13 (2018)), the Federal Trade Commission Act, id. §§ 41–58 (FTC Act), and Hart-Scott-Rodino. And both the Sherman Act and the Clayton Act offer only broad, vague legal standards to the courts.93×93. Thomas W. Merrill, The Common Law Powers of Federal Courts, 52 U. Chi. L. Rev. 1, 44 (1985). Courts and scholars have viewed these open-ended statutes as a legislative delegation to the judiciary.94×94. See, e.g., Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 899 (2007) (“In antitrust, the federal courts . . . act more as common-law courts than in other areas governed by federal statute.” (omission in original) (quoting Nw. Airlines, Inc. v. Transp. Workers Union of Am., 451 U.S. 77, 98 n.42 (1981))); Dan M. Kahan, Lenity and Federal Common Law Crimes, 1994 Sup. Ct. Rev. 345, 347. But cf. Alexander Volokh, Judicial Non-delegation, the Inherent-Powers Corollary, and Federal Common Law, 66 Emory L.J. 1391, 1453–56 (2017) (arguing that Congress actually meant the Sherman Act to incorporate only prior restraint-of-trade case law and that, if “modern-day doctrine can’t really be justified as a series of steps since 1890 developing the original common-law standard . . . , [then] that’s the courts’ fault, not Congress’s . . . [;] a problem of judges acting ultra vires, not of Congress delegating away its legislative power,” id. at 1456). The judiciary has accepted this delegation fully. For example, section 1 of the Sherman Act’s text explicitly outlaws combinations “in restraint of trade or commerce.”95×95. 15 U.S.C. § 1. Courts have run with this phrase, defining it to include many different types of restraints, including horizontal price-fixing, market divisions, output limitations, refusals to deal, tying, and bundling agreements, all of which are then judged under either a judicially created per se standard or a judicially created rule of reason standard.96×96. See William Holmes & Melissa Mangiaracina, Antitrust Law Handbook §§ 2:1–2:22 (2019). As then-Justice Rehnquist described federal securities law, so too is federal antitrust law “a judicial oak which has grown from little more than a legislative acorn.”97×97. Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 737 (1975).

To be clear, the problem with preemption based on a regulatory regime created through congressional delegation to the judiciary is not that that delegation is unconstitutional. Some scholars do argue that the judicial creation of federal antitrust common law breaches the Constitution’s separation of powers.98×98. See, e.g., Andrew S. Oldham, Sherman’s March (in)to the Sea, 74 Tenn. L. Rev. 319, 379–80 (2007). But see, e.g., Volokh, supra note 94, at 1453–56 (arguing that the Sherman Act does not violate the nondelegation doctrine). The Supreme Court, however, has not been persuaded by this argument. After all, the Sherman Act still reigns supreme after over a century of fleshing out by the federal judiciary.99×99. See Oldham, supra note 98, at 321–23.

As a policy matter, however, the undemocratic nature of the federal judiciary makes preemption by federal judge-made law more troubling than legislative preemption. In most instances of preemption, state interests are protected through process federalism. Process federalism protects the states from federal encroachment through political and procedural safeguards.100×100. See Ernest A. Young, Two Cheers for Process Federalism, 46 Vill. L. Rev. 1349, 1358–59 (2001). The states’ political safeguard is that state representatives pervade the federal system: “States are represented in Congress; states participate in the election of the President through the Electoral College;” and, in a cooperative-federalist manner, “state officials may . . . participate in the administration of federal programs.”101×101. Id. at 1355 (footnote omitted); see also Bulman-Pozen & Gerken, supra note 12, at 1262–63. The hope is that the states’ defenders in Washington will prevent the degradation of state interests.102×102. See Young, supra note 100, at 1359. The states’ procedural safeguards add to their political safeguard: “Even if members of Congress have the will to encroach upon or displace state prerogatives, . . . the legislative gauntlet makes it difficult for Congress to do so.”103×103. David S. Rubenstein, The Paradox of Administrative Preemption, 38 Harv. J.L. & Pub. Pol’y 267, 326 (2015). Any attempt at preemption would require that legislation survive bicameralism, presentment, and Congress’s internal rules.104×104. Id. Some scholars have heavily critiqued process federalism, arguing that the procedural and political safeguards are not all they are cracked up to be. See, e.g., Saikrishna B. Prakash & John C. Yoo, The Puzzling Persistence of Process-Based Federalism Theories, 79 Tex. L. Rev. 1459, 1461 (2001). But these criticisms have largely been that process federalism cannot be all of federalism — or, in other words, that process federalism’s safeguards without more do not do enough to safeguard the states’ rights in the federalist-nationalist balance. See, e.g., id. (“We do not doubt that the political safeguards may, on occasion, safeguard the states’ constitutional powers and prerogatives. . . . We challenge, however, the conclusion . . . that the political safeguards represent the only protections for federalism. In our view, the exclusive theory is simply inconsistent with the Constitution’s text, structure, and original understanding.”). This Note takes no stance on whether process federalism is enough; whether it is enough when coupled with the safeguards of cooperative federalism, see Bulman-Pozen & Gerken, supra note 12, at 1262–63, and uncooperative federalism, see id. at 1265–71; or whether maintenance of the proper balance requires judicial review, Prakash & Yoo, supra, at 1462. Instead, this Part’s argument is solely that process federalism does present some federalist safeguards — safeguards that are extinguished when state lawmaking power is preempted by judge-made law.

Preemption by judge-made law, however, avoids these constraints.105×105. See Young, supra note 100, at 1359, 1363; cf. Rubenstein, supra note 103, at 323–25. On the political front, it is far easier for a state to lobby Congress than it is for a state to lobby a federal judge. Moreover, it is certain that the Supreme Court will not have representatives from all fifty states. Finally, whereas voters displeased with federal encroachment into the state sphere can vote out their congressional delegation, they are stuck with federal judges for life.106×106. Eric Rasmusen, Judicial Legitimacy as a Repeated Game, 10 J.L. Econ. & Org. 63, 64 (1994). On the procedural front, a judicial opinion has no checks aside from appeal.107×107. See id. In fact, the ability of judges to subvert procedural safeguards could encourage Congress to delegate more to the judiciary.108×108. Cf. Rubenstein, supra note 103, at 326.

It is true that, in order to expressly preempt state antitrust law in favor of federal judge-made antitrust law, Congress would first have to overcome process federalism’s dual safeguards. There is reason to believe that the safeguards have worked so far. Even in the face of influential detractors like Judge Posner, Congress has only expanded the states’ antitrust role.109×109. See Farmer, supra note 24, at 376 (explaining Hart-Scott-Rodino’s expansion of state power). However, whereas purely legislative preemption would require Congress to pass through the safeguards every time it decided to meddle with state antitrust policies, preemption by judge-made law would require Congress to overcome the safeguards only once. After express preemption passed through, the courts could go about frequently changing federal antitrust law without political or procedural checks, and states would have little recourse.

While these process federalism concerns also apply to federal agency delegation, the concerns are greater in the face of judicial delegation. Agency delegation avoids process federalism’s political safeguards because “[a]gencies are not beholden to states in any politically thick sense.”110×110. Rubenstein, supra note 103, at 323–24. It avoids procedural safeguards because an agency “bypasses the legislative dam.”111×111. Id. at 326. But the executive branch is more democratic than is the judiciary. In the antitrust context, DOJ enforcement policy shifts when a new President takes over.112×112. See, e.g., Flexner & Racanelli, supra note 32, at 503–04 (noting the shift in federal antitrust focus upon the election of President Ronald Reagan). States also have more influence on agencies than they do on the judiciary. Federal regulatory regimes often involve state implementation that can “generate dynamics of mutual dependency that may make federal agencies receptive to state interests,”113×113. Rubenstein, supra note 103, at 324. a type of relationship that has been labeled “cooperative federalism.”114×114. E.g., Susan Rose-Ackerman, Comment, Cooperative Federalism and Co-optation, 92 Yale L.J. 1344, 1344 (1983) (internal quotation marks omitted). States can also affect agency policy by challenging it, a type of federalism known as “uncooperative federalism.”115×115. E.g., Bulman-Pozen & Gerken, supra note 12, at 1260. See generally id. at 1260–84. Although process federalism’s safeguards are weakened when Congress delegates to agencies, that weakening is at least softened by the above factors — factors that do not help redeem judicial delegation.

The judiciary’s ill fit as policymaker is also a cause for federalism concerns. After all, judiciaries are not built for policymaking, as even Supreme Court Justices regularly admit.116×116. See, e.g., Carpenter v. United States, 138 S. Ct. 2206, 2265 (2018) (Gorsuch, J., dissenting) (“Legislators are responsive to their constituents and have institutional resources designed to help them discern and enact majoritarian preferences. Politically insulated judges come armed with only the attorneys’ briefs, a few law clerks, and their own idiosyncratic experiences.”). Although judges are educated, they may not have expertise in antitrust economics117×117. Lina M. Khan, The End of Antitrust History Revisited, 133 Harv. L. Rev. 1655, 1679 (2020) (book review). and they are even less likely than the political branches to be able to gather that expertise through others.118×118. Cf. Paul H. Robinson & Barbara A. Spellman, Sentencing Decisions: Matching the Decisionmaker to the Decision Nature, 105 Colum. L. Rev. 1124, 1132–34 (2005). It may be easy for an antitrust expert like Judge Posner to favor the federal judiciary’s policymaking over that of state legislative and executive branches, but his level of antitrust knowledge is the exception, not the rule.119×119. See Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 5–6 (1984). Even outside of the legal bubble, Judge Posner has been recognized as uniquely expert in antitrust law. See Richard Posner, Distinguished Fellow 2018, Am. Econ. Ass’n, https://www.aeaweb.org/about-aea/honors-awards/distinguished-fellows/richard-posner [https://perma.cc/794Z-5839] (“Richard Posner is the most significant pioneer in the economic analysis of law. . . . He was among the leaders in interpreting antitrust jurisprudence and regulation through the lens of economic efficiency — contributions that fundamentally changed the practice and analysis of antitrust in the United States and around the world.”). Moreover, to the extent that deciding antitrust policy means deciding what values antitrust law should herald,120×120. See Adams & Brock, supra note 40, at 326 (“[I]t is critically important to recognize that in choosing among [antitrust] theories, ‘courts are not selecting tools from a kit, but making important social judgments that should be made with full awareness.’” (quoting Lawrence A. Sullivan,Antitrust, Microeconomics, and Politics: Reflections on Some Recent Relationships, 68 Calif. L. Rev. 1, 12 (1980))). judges are poor arbiters of the electorate’s views and may replace society’s values with their own.121×121. See Carpenter, 138 S. Ct. at 2265 (Gorsuch, J., dissenting); Robinson & Spellman, supra note 118, at 1133. Moreover, because they are limited to adjudicating cases and controversies,122×122. See Martin H. Redish & Sopan Joshi, Litigating Article III Standing: A Proposed Solution to the Serious (but Unrecognized) Separation of Powers Problem, 162 U. Pa. L. Rev. 1373, 1380 n.31 (2014) (“While the phrase ‘case or controversy’ does not actually appear in the Constitution, it is a shorthand used to describe the limits of justiciability.”).  courts lack the ability to tackle legislative problems head-on and instead must wait for problems to come to them.123×123. Id. at 1381. The very purpose of the case-or-controversy requirement was to cement the judicial branch’s judicial, rather than legislative, function.124×124. Id. at 1380–82. This requirement can explain large blind spots in current federal antitrust law. For example, because merging parties will often drop an agency-opposed merger rather than take the issue to court, “the Supreme Court has not decided a case about . . . substantive merger standards since . . . [before] Hart-Scott-Rodino.”125×125. Elhauge & Geradin, supra note 68, at 991; see also Khan, supra note 117, at 1679. Especially in comparison to agencies, with their myriad policymaking tools,126×126. See M. Elizabeth Magill, Agency Choice of Policymaking Form, 71 U. Chi. L. Rev. 1383, 1386–89 (2004). the judiciary has policymaking limitations, making the courts look anemic in that sphere. Where Congress has abdicated its role as policymaker in favor of the ill-equipped judiciary (rather than the better-equipped executive), it would be folly to force the state legislature to abdicate its policymaking role as well.

Of course, state judiciaries often play a role in molding antitrust policy, and one could argue that there would be no problem with federal judge-made law preempting state judge-made law. Key to the benefit of states as laboratories of democracy is the states’ ability to experiment legislatively. To say that states can “experiment[]” and act as “scientific policymakers”127×127. Akhil Reed Amar, Five Views of Federalism: “Converse-1983” in Context, 47 Vand. L. Rev. 1229, 1234 (1994). or “innovate” through a collective “evolutionary process”128×128. Friedman, supra note 7, at 399. makes most sense if we are actually referring to state legislatures and agencies. So what is the issue with overriding state courts?

For one thing, much (although not all) of what state judiciaries do in the antitrust arena is actually decided by state legislatures. Some state legislatures, for example, have explicitly directed their courts to apply federal case law to corresponding state statutes.129×129. See Herbert Hovenkamp, State Antitrust in the Federal Scheme, 58 Ind. L.J. 375, 377 n.10 (1983). Moreover, most state judiciaries are elected.130×130. Judicial Selection: Significant Figures, Brennan Ctr. for Just. (May 8, 2015), https://www.brennancenter.org/rethinking-judicial-selection/significant-figures [https://perma.cc/5UL3-GPJL]. As noted above, sensitivity to the electorate is key to the values-based part of policymaking, so state courts may actually be better policymakers than federal courts are.

Even if all state antitrust law were formed by unelected state judges, there would still be a basic sovereignty rationale for preferring state court decisions to federal court decisions. Congress should recognize that the federal judiciary is a poor policymaking body, but whether the states believe their judiciaries are up to the policymaking challenge is for the states to decide. Consider the secondary holding of Erie Railroad Co. v. Tompkins.131×131. 304 U.S. 64 (1938). Of course, Justice Brandeis famously and primarily held that the concept of a “transcendental body of law” was a “fallacy” and that, therefore, the federal courts should not create a general federal common law.132×132. Id. at 79 (quoting Black & White Taxicab & Transfer Co. v. Brown & Yellow Taxicab & Transfer Co., 276 U.S. 518, 533 (1928) (Holmes, J., dissenting)). This holding may not have had complete staying power: Justice Brandeis’s critique of common law’s underpinnings has not prevented the development of a post-Erie federal common law. 19 Arthur R. Miller, Federal Practice & Procedure § 4514 (3d ed., update 2019) (“Since Erie, it has become clear that this statement is not completely accurate.”). But, secondarily and more relevant here, Justice Brandeis did not prevent the states from continuing to hand down common law from on high: “[W]hether the law of the State shall be declared by its Legislature in a statute or by its highest court in a decision is not a matter of federal concern.”133×133. Erie, 304 U.S. at 78. Justice Brandeis’s willingness to critique common law’s “fallacy” and yet still allow the states to embrace it came from a concern with state sovereignty, an understanding that the states have the right to structure themselves in a manner that the federal government, a separate sovereign, opposes.134×134. See id. at 78–79. That same sovereignty rationale should allow states to decide that their judiciaries are sufficiently good at policymaking, even if the federal government should realize that its judiciary is not.

III. Antitrust Preemption Compared

Preemption by federal statutory common law would be especially problematic in the antitrust arena. Antitrust is not unique in its quasi–common law status,135×135. See Oldham, supra note 98, at 322 n.19 (collecting statutes under which the “[Supreme] Court has . . . asserted common law powers”). and Congress has expressly preempted state law in some areas where judge-made law exists. For example, courts have created a large amount of federal common law around the Employee Retirement Income Security Act of 1974136×136. Pub. L. No. 93–406, 88 Stat. 829 (codified as amended in scattered sections of 26 and 29 U.S.C.). (ERISA), and Congress expressly preempted certain state laws within ERISA’s purview.137×137. See Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 48, 56 (1987). Unlike federal antitrust law, however, less is left to the courts in ERISA and other federal common law areas.

Two qualities of federal antitrust law make it an extreme case of judicial delegation and therefore more susceptible to federalism critiques in the case of preemption: its statutory brevity and its minimal agency intervention. On the first point, the substantive federal antitrust statutes are exceedingly brief. The Sherman Act, federal antitrust law’s seminal statute, took up two pages of the Statutes at Large.138×138. See Sherman Act, ch. 647, 26 Stat. 209 (1890) (codified as amended at 15 U.S.C. §§ 1–7 (2018)); see also Wiley, supra note 92, at 713 (“Its brevity is charming — and significant.”). The Clayton Act is also relatively brief at eleven pages.139×139. See Clayton Act, ch. 323, 38 Stat. 730 (1914) (codified as amended at 15 U.S.C. §§ 12–27 and 29 U.S.C. §§ 52–53 (2018)). ERISA, on the other hand, digs into the specifics. At 207 pages, it is less legislative acorn and more legislative oak tree.140×140. See 88 Stat. 829; cf. Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 737 (1975). While word count may not be a perfect substitute for thoroughness, the disparity here is too large to be ignored.

The length distinction matters because it helps define how much has really been delegated to the judiciary. When crafting ERISA’s common law, courts can pull from a large source evincing Congress’s intent. Due to the brevity of Congress’s statutory antitrust commands, however, courts have little congressional guidance in their antitrust policy pronouncements. Almost everything is left up to the judiciary.

On the second point, the executive branch has played only a small role in fleshing out the antitrust statutes, leaving the lion’s share to judges. The FTC and DOJ are responsible for federal enforcement of antitrust law.141×141. 1 Altman & Pollack, supra note 2, § 4:51. But, unlike many regulatory areas, antitrust has not seen the promulgation of many binding regulations. Neither the Sherman Act nor the Clayton Act authorizes either agency to promulgate rules defining vague terms. Whether or not the FTC Act authorizes the FTC to promulgate rules defining unfair methods of competition (the presence of such authority is far from clear), the agency has not done so.142×142. See Elhauge & Geradin, supra note 68, at 13 & n.10.; see also Khan, supra note 117, at 1680–81 (acknowledging the FTC’s neglect of its “norm-creating role,” id. at 1680, but advocating “[r]eorienting [it] to serve an administrative function” to take power “away from the courts,” id. at 1681).

This is not to say that the FTC and DOJ do not have significant impacts on the development of federal antitrust policy. For one thing, the agencies choose when to enforce antitrust laws, thus affecting the kinds of cases and controversies that the judiciary sees.143×143. See Flexner & Racanelli, supra note 32, at 501. But the Sherman Act and the Clayton Act offer private rights of action,144×144. 15 U.S.C. § 15 (2018); Chatanooga Foundry & Pipe Works v. City of Atlanta, 203 U.S. 390, 396 (1906). and states may sue as parens patriae under Hart-Scott-Rodino, so the agencies’ enforcement decisions cannot completely control the law’s development. Many seminal antitrust cases were decided without an agency party.145×145. See, e.g., Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 882 (2007). Granted, even in suits not brought by an agency, the agencies can have significant sway: the executive branch often acts as amicus curiae,146×146. See, e.g., Brief for the United States as Amicus Curiae Supporting Petitioner, Leegin, 551 U.S. 877 (No. 06-480). and courts sometimes base their decisions directly on the government’s amicus briefs.147×147. See, e.g., Columbia Steel Casting Co. v. Portland Gen. Elec. Co., 111 F.3d 1427, 1443 (9th Cir. 1996) (“We are persuaded by . . . an amicus curiae brief filed by the Antitrust Division of the Department of Justice that we erred . . . .”). And the FTC and DOJ publish enforcement guidelines that courts have found highly persuasive, even though those guidelines are nonbinding and meant mainly as an expression of enforcement policy.148×148. See Hillary Greene, Guideline Institutionalization: The Role of Merger Guidelines inAntitrust Discourse, 48 Wm. & Mary L. Rev. 771, 781–809 (2006). For example, the “Merger Guidelines are for all practical purposes the law on merger liability.”149×149. Rebecca Haw Allensworth, Essay, The Influence of the Areeda-Hovenkamp Treatise in the Lower Courts and What It Means for Institutional Reform in Antitrust, 100 Iowa L. Rev. 1919, 1929 (2015). That being said, other enforcement guidelines have been far less influential — some “have been cited only a handful of times each.”150×150. Id. In the end, although amicus briefs and guidelines may often win over judges, it is judges who make final decisions, and who sometimes ignore agencies.151×151. See id.; Greene, supra note 148, at 822–23.

On the other hand, the executive branch plays a larger role in defining ERISA’s intricacies. ERISA invites the Department of Labor (DOL) to promulgate regulations defining many of its terms.152×152. See, e.g., 29 U.S.C. § 1002(2), (14), (16), (18), (26), (28), (37), (40) (2018). Unlike the FTC with its (potential) authority to promulgate regulations under the FTC Act, the DOL actively takes advantage of its authority under ERISA: in 2018 and 2019, the Employee Benefits Security Administration, an agency within the DOL, published nine final rules in the Federal Register implementing ERISA.153×153. See Employee Benefits Security Administration, Fed. Register, https://www.federalregister.gov/agencies/employee-benefits-security-administration [https://perma.cc/9TVN-4TTW]. Unlike the antitrust guidelines, DOL regulations are binding law that courts must, not might, follow.

Just like the length of the statute, the degree of executive delegation affects the true degree of judicial delegation. Whereas agencies help fill in the gaps of ERISA, thus replacing some judicial delegation with executive delegation, agencies leave all antitrust delegation to the courts. While preemption by agency-made law has some of the same federalism and separation of powers concerns as preemption by judge-made law,154×154. Cf. Rubenstein, supra note 103, at 268, 320–26. agencies are more democratically accountable than are courts and are better equipped to make policy. Notwithstanding the influence of the Merger Guidelines, antitrust executive action has not compensated for the deficiencies of preemption by statutory common law to the same extent as have DOL’s regulations. Promulgated regulations, like DOL’s, must go through notice and comment under the Administrative Procedure Act.155×155. 5 U.S.C. §§ 551, 553–559, 701–706 (2018). Notice and comment promotes democratic accountability: it “compels the agency to consult each of its stakeholder ‘constituents,’ ensuring that the ultimate agency action will in some way be responsive to the concerns of all interested parties.”156×156. Travis O. Brandon, Fearful Asymmetry: How the Absence of Public Participation in Section 7 of the ESA Can Make the “Best Available Science” Unavailable for Judicial Review, 39 Harv. Envtl. L. Rev. 311, 321 (2015). Additionally, some scholars argue that notice and comment might lead to better policymaking: it could be “a forum for democratic deliberation, in which agencies and citizens alike will change their views in response to reasoning of others.”157×157. Robert Knowles, National Security Rulemaking, 41 Fla. St. U. L. Rev. 883, 898 (2014); see id. at 897–98. As such, notice and comment can help executive branch action overcome both the (un)democratic and policymaking critiques of judge-made law. The Merger Guidelines, on the other hand, nonbinding as they are, may be created without any public engagement, and often have been.158×158. Greene, supra note 148, at 818 n.170 (“None of the merger guidelines were promulgated under a grant of rule-making authority, and only recently have some guidelines been promulgated in a process consistent with the APA. In fact, the guidelines were frequently promulgated with no adherence to rule-like procedures such as public notice and comment.”). So, even where federal antitrust law does have executive branch participation, that participation does not provide as meaningful a democratic check as does DOL participation for ERISA.

None of this is to say that ERISA preemption fails to raise federalism concerns or that the concerns addressed in Part II are unique to antitrust. In its decision to expressly preempt state law in ERISA, a wise Congress should have considered the difficulties of preemption via judge-made law. Part II’s concerns with preemption via judge-made law could be applied to any delegation to the judiciary that overrides the states’ will. But given the brevity of federal antitrust statutes and the relative lack of executive branch involvement, Congress should be even more wary if it decides to preempt state antitrust law.159×159. Complaints about brevity and lack of executive branch involvement land an even stronger blow against preemption via statute-independent federal common law. A grant of federal common lawmaking power does not have to be statutory. All that is needed to support the development of federal common law is “some expressed or implied affirmative grant of power to the national government by the Constitution or a statute enacted pursuant to it.” 19 Miller, supra note 132, § 4514. When courts make law from a constitutional grant, there may not even be a brief statute. See, e.g., Clearfield Tr. Co. v. United States, 318 U.S. 363, 366–67 (1943) (“When the United States disburses its funds . . . it is exercising a constitutional function or power. . . . In absence of an applicable Act of Congress it is for the federal courts to fashion the governing rule of law according to their own standards.”); see also 19 Miller, supra note 132, § 4515; Volokh, supra note 94, at 1429 (discussing courts’ “statute-independent federal common-lawmaking powers”). Because statute-independent common law is created completely by the courts, preemption via statute-independent common law will preempt the states while also excluding the federal executive branch. Part II’s critique then undermines statute-independent common law preemption even more than it undermines a preemptive Sherman Act.
But Part II proffers only an argument that weighs against preemption; that argument must be balanced against the various pro-preemption critiques of Part I. When it comes to statute-independent common law, the pro-preemption arguments may simply be greater than they are in the antitrust arena. After all, such statute-independent common-lawmaking power exists only “in suits implicating a sufficiently strong interest of the national government.” 19 Miller, supra note 132, § 4515. And it makes sense that common law grounded in the Constitution has more sway than does common law grounded in statute. Although antitrust law has sometimes been likened to the Constitution or other founding documents, see United States v. Topco Assocs., Inc., 405 U.S. 596, 610 (1972) (“Antitrust laws . . . are the Magna Carta of free enterprise.”); Thomas B. Nachbar, The Antitrust Constitution, 99 Iowa L. Rev. 57, 69 (2013), courts simply give its commands less weight than those of the Constitution. Compare, for example, the (limited) deference given to professionals in the antitrust sphere, see Nat’l Soc’y of Prof’l Eng’rs v. United States, 435 U.S. 679, 696 (1978) (analyzing agreements by professionals under the rule of reason), to the zero deference given to professionals under the First Amendment, see Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2371–72 (2018). Even if statute-independent common law’s complete lack of input from the democratic branches increases the power of the federalism critique, that increase is rebutted by an increase in the power of the pro-preemption arguments.

Conclusion

There is little doubt that Congress could decide to preempt state antitrust law. However, although the merits of avoiding a patchwork antitrust regime are compelling, Congress would trigger federalism pitfalls if it were to reform antitrust law by expressly preempting state antitrust law. A preempting Congress should weigh any benefits against the complications of federalism’s procedural and political safeguards and of the judiciary’s weak policymaking ability.

Of course, there is reason to believe that if Congress were to expressly preempt state antitrust law, it would do so as part of a more major antitrust reform effort. Recently, federal antitrust policy has been the subject of critique. Fed up with the seeming omnipresence of corporate giants, some scholarly160×160. See, e.g., Wu, supra note 69, at 127–41; Lina M. Khan, Note, Amazon’s Antitrust Paradox, 126 Yale L.J. 710, 710 (2017). and journalistic161×161. See, e.g., Editorial, Why Does Apple Control Its Competitors?, N.Y. Times (May 2, 2019), https://nyti.ms/2GXCHSP [https://perma.cc/3EHB-NLY7]. discourse has turned on the federal government’s antitrust policies. As things stand, if Congress decides to preempt state antitrust law with current federal antitrust jurisprudence, it would have to decide that the pros of preemption mentioned in Part I outweigh the federalism cons of Part II. But if Congress were to reform antitrust law by creating a new, detailed antitrust regime for courts to interpret, preemption of state antitrust law could avoid the perils of preemption via judge-made law.