Emory Law Journal

Untangling the Market and the State
Wentong Zheng  *Professor of Law & University Term Professor, University of Florida Levin College of Law. I thank Tom Ankersen, Darren Hutchinson, Lars Noah, Bill Page, Laura Rosenbury, Sharon Rush, D. Daniel Sokol, and Michael Wolf for helpful comments and discussions. The research for this Article was supported by a summer research grant from the University of Florida Levin College of Law.

Abstract

The government plays increasingly active and diversified roles in the modern economy. How to draw the boundary between the market and the state has emerged as a contentious issue in various areas of law, including constitutional law, antitrust, and international trade. This Article surveys and critiques the law’s current approaches to the market-versus-state divide, embodied in four tests based on ownership, control, function, and role, respectively. This Article proposes an alternative market-versus-state test based on the nature of the power being exercised in the challenged action. This power-based test not only better distinguishes between the market and the state, but also illuminates why the market-versus-state distinction needs to be made in the first place. Applying this power-based test would bring much needed logic and clarity to many market-versus-state issues in various legal contexts.

Introduction

The government is on the march. Having grown out of the pre- progressive mode of governance where all commercial disputes were resolved through private litigation, the tentacles of the government are reaching far and wide, extending regulatory oversight over a plethora of social and economic activities, including competition, railroad pricing, food and drug safety, and many others.  1Edward L. Glaeser & Andrei Shleifer, The Rise of the Regulatory State, 41 J. Econ. Literature 401, 401 (2003). Supported by a rapidly growing number of regulatory agencies, the government has assumed greater control over the economy, expanding its role from correcting market failures to administering social and economic justice.  2See Karen Yeung, The Regulatory State, in The Oxford Handbook of Regulation 72–73 (Robert Baldwin et al. eds., 2010).

Aside from its traditional role as a market regulator, the government is also emerging as a major participant in market activities. Among other things, governments own corporations,  3State-owned corporations are common even in Western countries. For a history of state-owned enterprises in Western countries, see Pier Angelo Toninelli, The Rise and Fall of Public Enterprise: The Framework, in The Rise and Fall of State-Owned Enterprise in the Western World 3 (Pier Angelo Toninelli ed., 2000). employ workers,  4In the United States, the federal government is the largest employer in the nation, with more than 2.1 million civilian workers and 1.3 million active duty military who serve in all fifty states and around the world. Office of Mgmt. & Budget, Fiscal Year 2016 Analytical Perspectives of the U.S. Government 75 (2015). and buy large amounts of goods and services.  5For example, “[t]he U.S. [federal] government is the largest single purchaser of goods and services in the world, awarding approximately $500 billion in contracts every year.” SBA’s Role in Government Contracting, U.S. Small Bus. Admin., https://www.sba.gov/contracting/what-government-contracting/sbas-role-government-contracting (last visited Aug. 28, 2017). The footprint of the government has grown larger particularly because of the 2008–2009 global financial crisis, which sparked unprecedented state intervention in the marketplace.  6In the wake of the global financial crisis, the U.S. government seized control of American International Group, one of the world’s largest insurers, to prevent the company from falling into bankruptcy. See Matthew Karnitschnig et al., U.S. to Take Over AIG in $85 Billion Bailout; Central Banks Inject Cash as Credit Dries Up, Wall St. J. (Sept. 16, 2008, 11:59 PM), https://www.wsj.com/articles/SB122156561931242905?mg=prod/accounts-wsj. The U.S. government also took 60% ownership of General Motors as part of the latter’s government-orchestrated bankruptcy process. See Neil King Jr. & Sharon Terlep, GM Collapses Into Government’s Arms, Wall St. J. (June 2, 2009, 12:01 AM), https://www.wsj.com/articles/SB124385428627671889.

The rise of the government is even more dramatic if one’s horizon is broadened to include emerging economies on the world stage. Powerful state-owned enterprises from countries like China, Russia, and Saudi Arabia are ushering in a new form of doing business, generally dubbed as “state capitalism.”  7See The Visible Hand, Economist (Jan. 21, 2012), http://www.economist.com/node/21542931. As one indication of the ascendancy of state capitalism, the thirteen biggest oil firms in the world, which control more than three-quarters of the world’s oil reserves, are all state-backed.  8Id. The government has been so successful in the new era of capitalism that one influential commentator has pronounced “the end of the free market.”  9Ian Bremmer, The End of the Free Market: Who Wins the War Between States and Corporations? (2010).

The changing dynamics between the government and the market have far-reaching implications not just for the economy, but for the law as well. Courts and dispute settlement tribunals around the world are grappling with the increasingly active and diversified roles of the government. Since 2007, the U.S. Supreme Court has ruled on whether the government could require waste hauling firms to bring waste to publicly owned waste transfer and processing facilities,  10United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330 (2007) (holding that county ordinances mandating the delivery of waste to designated waste transfer and processing facilities did not violate the dormant Commerce Clause of the U.S. Constitution). whether a state-owned hospital should be required to answer antitrust complaints when it attempted to acquire the only competing hospital in the region,  11Fed. Trade Comm’n v. Phoebe Putney Health Sys., Inc., 568 U.S. 216 (2013) (holding that a state hospital authority is not entitled to state-action immunity in its acquisition of existing hospitals). and whether a state dental licensing board should be accorded immunity from antitrust law when it prohibited nondental practitioners from providing teeth-whitening services.  12N.C. State Bd. of Dental Exam’rs v. Fed. Trade Comm’n, 135 S. Ct. 1101 (2015) (holding that a dental licensing board is not eligible for state-action immunity because its action was not actively supervised by the state). In the meantime, the World Trade Organization (WTO) opined on whether Chinese state-owned enterprises should be treated as public bodies akin to government agencies  13Panel Report, United States—Countervailing Duty Measures on Certain Products from China, ¶ 7.75, WTO Doc. WT/DS437/R (adopted July 14, 2014) [hereinafter DS437 Panel Report] (holding that the United States acted inconsistently with its WTO obligations when it found that Chinese SOEs were public bodies based only on the grounds that they were majority owned or controlled by the Chinese government). and whether a Canadian provincial government provided a subsidy to renewable energy producers when it purchased electricity from them at above-market rates.  14 Appellate Body Report, Canada—Certain Measures Affecting the Renewable Energy Generation Sector/Canada—Measures Relating to the Feed-in Tariff Program, WTO Doc. WT/DS412/AB/R, WT/DS426/AB/R (adopted May 6, 2013) [hereinafter DS412/DS426 AB Report] (holding that it was unable to determine whether Ontario’s feed-in-tariff program conferred a benefit because the relevant market for the benefit analysis should be the separate competitive markets for wind- and solar-generated electricity).

This Article is an attempt to systematically examine how the government interacts with the market and how the law treats such interactions. For ease of reference, this Article labels the various issues arising from the government-market interactions as “market-versus-state”  15Unless otherwise noted, the term “state” in this Article refers to the government in the abstract sense, not any particular political division or subdivision. issues. While the market-versus-state dynamics have shifted in leaps and bounds, legal scholarship has not kept up with the changes. Scholarly discussions of the market-versus-state divide date mostly back to several decades ago, when the state was still playing a static, confined role in the economy.  16See, e.g., Dan T. Coenen, Untangling the Market-Participant Exemption to the Dormant Commerce Clause, 88 Mich. L. Rev. 395 (1989) (discussing the market-participant exception to the dormant Commerce Clause in U.S. constitutional law); Einer Richard Elhauge, The Scope of Antitrust Process, 104 Harv. L. Rev. 667 (1991) (proposing a process view of the antitrust state action doctrine); Peter D. Enrich, Saving the States from Themselves: Commerce Clause Constraints on State Tax Incentives for Business, 110 Harv. L. Rev. 377 (1996) (proposing an antidistortion standard for determining whether a state tax incentive violates the Commerce Clause); Mark P. Gergen, The Selfish State and the Market, 66 Tex. L. Rev. 1097 (1988) (discussing principles for differentiating between permissible and impermissible state preferences for its own citizens); Walter Hellerstein & Dan T. Coenen, Commerce Clause Restraints on State Business Development Incentives, 81 Cornell L. Rev. 789 (1996) (discussing the tax-subsidy distinction in Commerce Clause jurisprudence); Thomas M. Jorde, Antitrust and the New State Action Doctrine: A Return to Deferential Economic Federalism, 75 Calif. L. Rev. 227 (1987) (discussing the state action doctrine in antitrust law and its relationship to economic federalism); Michael Wells & Walter Hellerstein, The Governmental-Proprietary Distinction in Constitutional Law, 66 Va. L. Rev. 1073 (1980) (discussing the governmental-proprietary distinction in various constitutional law contexts). And the existing academic literature approaches market-versus-state issues in isolated manners, paying essentially no attention to the systemic implications of the different ways of handling market-versus-state issues in different areas of law.  17Constitutional law scholars, for example, have not drawn connections between various market-versus-state issues in constitutional law with their counterparts in other areas of law. See, e.g., Hellerstein & Coenen, supra note 16 (discussing the tax-subsidy distinction in Commerce Clause jurisprudence but not whether and how such a distinction is made in international trade law).

This Article sets out to narrow the gap in understanding how the law should deal with market-versus-state interactions. It starts with a survey of how market-versus-state interactions are being treated in three distinct areas of law: constitutional law, antitrust, and international trade.  18See infra Part I. This Article discusses how the law in those areas employs four tests, based on ownership, control, function, and role, respectively, to draw the boundary between the market and the state. These tests, however, suffer serious limitations. They either presume that all conduct by the government is governmental in nature, or base legal outcomes on factors that do not lend themselves to objective determinations. Alternatively, they are implicitly predicated on a value judgment that needs support on a case-by-case basis.  19See infra Part II.

This Article contends that a more reliable way of distinguishing between the market and the state is to look at the nature of the power being exercised in the challenged action. If the power is a coercive one, backed up by state-sanctioned violence with no recognized basis in property rights, then the state is exercising governmental power and is acting in a governmental capacity. But if the state exercises coercive power by virtue of its control of economic resources, then it is exercising market power that could have been exercised by private actors, and therefore is acting in a proprietary or market capacity.   20See infra Section III.A. This power-based test, coupled with an inquiry as to the fundamental purpose of the legal regime at issue, not only better distinguishes between the market and the state, but also illuminates why the market-versus-state distinction needs to be made in the first place.  21See infra Section III.B. This Article argues that applying this power-based test would bring much needed logic and clarity to many market-versus-state issues in constitutional law, antitrust, and international trade.  22See infra Part IV.

This Article proceeds as follows. Part I provides an overview of the major market-versus-state issues that have come up in the law. Part II discusses the law’s current approaches to the market-versus-state divide, embodied in four tests based on ownership, control, function, and role, respectively. Part III proposes an alternative market-versus-state test based on power and an analytical framework within which the power-based test will be situated. Part IV discusses the application of the power-based test in various legal contexts.

I. The Market Versus the State: The Legal Landscape

The interaction between the market and the state figures prominently in many areas of law, including constitutional law, antitrust, and international trade. In those areas of law, the legal consequences of an action by the state often depend on whether the state acts in a governmental or proprietary capacity, whether the individual or entity carrying out the disputed action is considered to be acting in a public or private capacity, and whether state participation in the market is considered to distort the market. In this Part, this Article surveys the legal treatment of these market-versus-state interactions. This survey revolves around three major legal issues that play prominent roles in constitutional law, antitrust law, and international trade law: the governmental-proprietary distinction, the public-private distinction, and market distortion by the state. The following discussions illustrate how important the market-versus-state distinction has become and, in some cases, how detached the distinction has grown from the actual purpose of the law.

A. The Governmental-Proprietary Distinction

The governmental-proprietary distinction is a “cluster of rules” used by courts to distinguish the state’s core governmental functions from its proprietary or market activities in a variety of contexts.  23Wells & Hellerstein, supra note 16, at 1075. This distinction queries whether the state acts in a governmental or proprietary capacity, with divergent legal consequences flowing from that designation. This distinction is utilized in diverse legal contexts, ranging from constitutional law to antitrust law.

1. Dormant Commerce Clause

The Commerce Clause of the U.S. Constitution stipulates that “Congress shall have power . . . [t]o regulate Commerce . . . among the several States.”  24U.S. Const. art. I, § 8, cl. 3. Courts have interpreted this positive grant of legislative power to Congress as carrying a negative, or dormant, implication that states are prohibited from discriminating against or otherwise burdening interstate commerce.  25See Gen. Motors Corp. v. Tracy, 519 U.S. 278, 287 (1997). For an overview of the dormant Commerce Clause jurisprudence, see Richard B. Collins, Economic Union as a Constitutional Value, 63 N.Y.U. L. Rev. 43, 47–60 (1988). When the state is considered to be acting as a participant in proprietary or market activities, however, courts have held such actions immune from dormant Commerce Clause challenges.  26For background on this “market-participant” or “proprietary-activities” exception to the dormant Commerce Clause, see Gergen, supra note 16, at 1139–43.

The Supreme Court first laid out this market-participant exception to the dormant Commerce Clause in the seminal case of Hughes v. Alexandria Scrap Corp., which involved a Maryland program that imposed stricter documentation requirements for out-of-state scrap processors than for in-state scrap processors for receiving state subsidies for destroying abandoned automobiles.  27426 U.S. 794 (1976). The lower court found the Maryland program unconstitutional, citing its “substantial burdens upon the free flow of interstate commerce.” See Alexandria Scrap Corp. v. Hughes, 391 F. Supp. 46, 62 (D. Md. 1975), rev’d, 426 U.S. 794 (1976). The Supreme Court, however, distinguished the instant case from previous dormant Commerce Clause cases, arguing that Maryland in this case was merely acting as “a purchaser, in effect, of a potential article of interstate commerce.” Alexandria Scrap, 426 U.S. at 808. The Court upheld the Maryland program because “[n]othing in the purposes animating the Commerce Clause prohibits a State, in the absence of congressional action, from participating in the market and exercising the right to favor its own citizens over others.”  28Alexandria Scrap, 426 U.S. at 810.

After Alexandria Scrap, the Court extended the market-participant exception to two other scenarios. In Reeves, Inc. v. Stake, the Court upheld a South Dakota policy of limiting sales of cement produced by a state-owned plant to state residents in times of cement shortage.  29447 U.S. 429 (1980). In White v. Massachusetts Council of Construction Employers, Inc., the Court rejected a Commerce Clause challenge to an executive order of the mayor of Boston requiring all city-funded construction projects be performed by a work force at least half of which were bona fide city residents.  30460 U.S. 204 (1983). In both cases, the Court affirmed “[t]he basic distinction drawn in Alexandria Scrap between States as market participants and States as market regulators.”  31See Reeves, 447 U.S. at 436; White, 460 U.S. at 207 (quoting Reeves, 447 U.S. at 436).

In three other cases, however, the Supreme Court limited the scope of the market-participant exception. In South Central Timber Development, Inc. v. Wunnicke, the Court took up the question of whether the market-participant exception would exempt from Commerce Clause scrutiny an Alaska requirement that timber taken from state lands be processed within the state prior to export.  32467 U.S. 82 (1984). The Court answered the question in the negative, holding that “the State may not avail itself of the market-participant doctrine to immunize its downstream regulation of the timber-processing market in which it [was] not a participant.”  33Id. at 99. In New Energy Co. of Indiana v. Limbach, the Court again declined to apply the market-participant exception.  34486 U.S. 269 (1988). In that case, the Court struck down an Ohio statute that provided tax credit for ethanol producers from Ohio or from states that granted reciprocal tax credit, exemption, or refund for Ohio-produced ethanol.  35Id. at 271, 280. The Court argued that the market-participant exception was inapplicable because “the Ohio action . . . at issue [was] neither its purchase nor its sale of ethanol, but its assessment and computation of taxes—a primeval governmental activity.”  36Id. at 277. Finally, in Camps Newfound/Owatonna, Inc. v. Town of Harrison, the Court once again held that a preferential state tax exemption did not qualify for immunity under the market-participant exception.  37520 U.S. 564, 593 (1997) (“A tax exemption is not the sort of direct state involvement in the market that falls within the market-participation doctrine.”).

2. Positive Commerce Clause

The governmental-proprietary distinction also matters when it comes to the positive commerce power. In National League of Cities v. Usery, the Supreme Court struck down the application of the Fair Labor Standard Act’s minimum wages and maximum hours provisions to local and state government “insofar as [they] operate to directly displace the States’ freedom to structure integral operations in areas of traditional governmental functions.”  38426 U.S. 833, 836, 852 (1976). According to the Court, the exercise of congressional authorities in those areas would impair the states’ ability to function effectively in a federal system, and therefore was not within the authority granted by the Commerce Clause.  39Id. at 852.

A decade later, however, the Court repudiated the governmental-proprietary distinction it made in National League of Cities. In Garcia v. San Antonio Metropolitan Transit Authority, the Court held that the defendant, a municipally owned and operated mass-transit system, was not immune from the minimum-wage and overtime requirements of the Fair Labor Standards Act, despite that it provided a traditional governmental function.  40469 U.S. 528, 530–31 (1985). In a broad ruling, the Court declared that “the attempt to draw the boundaries of state regulatory immunity in terms of ‘traditional governmental function’ is not only unworkable but is also inconsistent with established principles of federalism.”  41Id. at 531. Instead, the Court held, “[T]he principal means chosen by the Framers to ensure the role of the States in the federal system lies in the structure of the Federal Government itself.”  42Id. at 550. The Court implied, however, that a constitutional challenge might remain open if “the internal safeguards of the political process have [not] performed as intended.” Id. at 556.

3. Antitrust

The governmental-proprietary distinction also plays a potential role in antitrust law under the so-called state-action immunity or exemption. Hailed as the “Magna Carta of free enterprise,”  43United States v. Topco Assocs., Inc., 405 U.S. 596, 610 (1972). antitrust law promotes competition and consumer welfare through prohibitions of restraints of trade.  44See Donald F. Turner, The Durability, Relevance, and Future of American Antitrust Policy, 75 Calif. L. Rev. 797, 798 (1987) (“Antitrust law is a pro-competition policy. The economic goal of such a policy is to promote consumer welfare through the efficient use and allocation of resources, the development of new and improved products, and the introduction of new production, distribution, and organizational techniques for putting economic resources to beneficial use.”). But see Robert H. Lande, Consumer Choice as the Ultimate Goal of Antitrust, 62 U. Pitt. L. Rev. 503 (2001) (arguing that the ultimate goal of antitrust law is to promote consumer choice); Thomas B. Nachbar, The Antitrust Constitution, 99 Iowa L. Rev. 57 (2013) (arguing that antitrust law serves a constitutional function of preventing the assertion of control over the conduct of others outside the sphere of one’s own property interests). Certain restraints of trade, however, are exempted from antitrust liability if they are imposed by the state. In Parker v. Brown, the Supreme Court held that a raisin marketing program approved by producers under the California Agricultural Prorate Act did not constitute an illegal restraint of trade within the meaning of the Sherman Act since the restraint was imposed by the state “as an act of government.”  45Parker v. Brown, 317 U.S. 341, 344, 352 (1943) (“The state in adopting and enforcing the prorate program made no contract or agreement and entered into no conspiracy in restraint of trade or to establish monopoly but, as sovereign, imposed the restraint as an act of government . . . .”).

Under the case law, however, a question lingers as to whether the Parker state-action immunity extends to the state’s proprietary activities. The seed of uncertainty was planted in Parker itself. While recognizing the state’s power to impose a restraint of trade “as sovereign,”  46Id. at 352. the Court in Parker made it clear that it was not dealing with the “question of the state or its municipality becoming a participant in a private agreement or combination by others for restraint of trade.”  47Id. at 351–52. Later, in City of Columbia v. Omni Outdoor Advertising, Inc., the Court interpreted this language to mean that the state-action immunity “does not necessarily obtain where the State acts not in a regulatory capacity but as a commercial participant in a given market.”  48499 U.S. 365, 374–75 (1991). The Court later interpreted this statement to be “leaving open the possibility of a market participant exception” to state-action immunity. Fed. Trade Comm’n v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 226 n.4 (2013).

However, the Court has not explicitly opined on whether the state-action immunity recognizes an exception for market or proprietary activities. Most recently, in Federal Trade Commission v. Phoebe Putney Health System, Inc., the Federal Trade Commission challenged a county hospital authority’s acquisition of an existing hospital in the county as anticompetitive.  49See 568 U.S. at 221–22. The proposed transaction would have allowed the county hospital authority to control 86% of the market for acute-care hospital services provided to commercial health care plans and their customers. Id. The Court raised the question of whether there should be a market-participant exception to the state-action immunity, but did not answer it.  50The Court did not consider this question in its opinion because “this argument was not raised by the parties or passed on by the lower courts.” Id. at 226 n.4. Instead, the Court relied on an alternative ground in holding that the county hospital authority was not eligible for the state-action immunity.  51The Court applied the “clear-articulation” test under California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97 (1980), and concluded that the county hospital authority was not eligible for state-action immunity because “there is no evidence the State affirmatively contemplated that hospital authorities would displace competition by consolidating hospital ownership.” Id. at 226.

B. The Public-Private Distinction

Besides the governmental-proprietary distinction, courts often utilize a public-private distinction when approaching the market-versus-state divide in various areas of law. Although the exact form of the public-private distinction varies from context to context, the distinction generally inquires whether the beneficiary of the disputed action or the individual or entity carrying out the disputed action should be classified as public or private. Such distinction leads to the application of different legal standards in constitutional law, antitrust, and international trade, to name just a few examples.

1. Dormant Commerce Clause

The public-private distinction figures prominently in the Supreme Court’s jurisprudence on the dormant Commerce Clause. Over the years, the Court has developed a two-tiered analytical framework to determine whether a state statute violates the dormant Commerce Clause. When a state statute discriminates against interstate commerce, either on its face or in its effect, courts apply strict scrutiny and hold the statute virtually per se invalid.  52See, e.g., Or. Waste Sys., Inc. v. Dep’t of Envtl. Quality, 511 U.S. 93 (1994) (holding that a surge Oregon imposed on in-state disposal of waste generated in other states was facially invalid under the dormant Commerce Clause); City of Philadelphia v. New Jersey, 437 U.S. 617 (1978) (holding that a New Jersey statute prohibiting the importation of solid or liquid waste that originated outside the territorial limits of the state violated the dormant Commerce Clause); Dean Milk Co. v. City of Madison, 340 U.S. 349 (1951) (holding that a city ordinance prohibiting the sale of milk unless bottled within five miles from the center of the city unduly burdened interstate commerce). By contrast, an even-handed state statute with only incidental effects on interstate commerce is outlawed only if its burden on interstate commerce is determined to clearly exceed its local benefits, under what has become known as the Pike balancing test.  53See, e.g., Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 458, 472 (1981) (holding that a Minnesota statute “banning the retail sale of milk in plastic nonreturnable, nonrefillable containers, but permitting such sale in other nonreturnable, nonrefillable containers, such as paperboard milk cartons” did not violate the dormant Commerce Clause because its incidental burden on interstate commerce was not clearly excessive in relation to its putative local benefits); Kassel v. Consol. Freightways Corp., 450 U.S. 662 (1981) (holding that an Iowa statute barring the use of trucks longer than sixty feet on Iowa’s interstate highways substantially burdened interstate commerce in violation of the dormant Commerce Clause); Pike v. Bruce Church, Inc., 397 U.S. 137 (1970) (holding that an Arizona order prohibiting a fruit grower from transporting uncrated cantaloupes from its Arizona ranch to a nearby California city for packing and processing constituted an unlawful burden on interstate commerce).

In a line of cases dealing with local flow-control laws, the Court has distinguished between public and private beneficiaries of a disputed state action in applying this two-tiered analytical framework. In C & A Carbone, Inc. v. Town of Clarkstown, the Court invalidated a town’s flow-control ordinance requiring all solid waste within the town to be deposited at a privately operated transfer station.  54511 U.S. 383 (1994). The town guaranteed a minimum waste flow to the transfer station as a way of financing the cost of the transfer station, which was to be sold to the town in five years for one dollar.  55Id. at 387. The Court struck down the ordinance as “[d]iscrimination against interstate commerce in favor of local business or investment.”  56Id. at 392. However, in a subsequent case, United Haulers Ass’n v. Oneida-Herkimer Solid Waste Management Authority, the Court considered the constitutionality of several counties’ flow-control ordinances requiring businesses hauling waste in the counties to bring waste to waste transfer and processing facilities owned and operated by a public benefit corporation.  57550 U.S. 330 (2007). The Court concluded that the challenged ordinances did not violate the dormant Commerce Clause because they “benefit[ed] a clearly public facility, while treating all private companies exactly the same.”  58Id. at 342. According to the Court, laws favoring public entities should be analyzed under the more deferential Pike balancing test and the laws at dispute in this case passed that test.  59Id. at 346. The Court upheld the laws because they had no “disparate impact on out-of-state as opposed to in-state businesses.” Id. The Court also suggested that the laws in dispute should survive Commerce Clause challenges because they carried out a traditional governmental function. Id. at 344 (“We should be particularly hesitant to interfere with the Counties’ efforts under the guise of the Commerce Clause because ‘[w]aste disposal is both typically and traditionally a local government function.’” (citation omitted)).

2. Antitrust

The public-private distinction also plays an important role in the state-action immunity in antitrust law, where different legal standards apply depending on whether the individual or entity carrying out the disputed action is classified as public or private.

In Parker, the Supreme Court declared that “[w]e find nothing in the language of the Sherman Act or in its history which suggests that its purpose was to restrain a state or its officers or agents from activities directed by its legislature.”  60Parker v. Brown, 317 U.S. 341, 350–51 (1943). The Court later clarified that the Parker immunity automatically applies when the anticompetitive conduct is by the state itself through the state legislature or the state supreme court.  61See, e.g., Hoover v. Ronwin, 466 U.S. 558, 567–68 (1984) (“Thus, under the Court’s rationale in Parker, when a state legislature adopts legislation, its actions constitute those of the State, and ipso facto are exempt from the operation of the antitrust laws.” (citation omitted)); Bates v. State Bar of Ariz., 433 U.S. 350, 359–60 (1977) (holding that the state bar of Arizona automatically received Parker immunity because the anticompetitive conduct was pursuant to the affirmative command of the Arizona Supreme Court).

When a private entity seeks the state-action immunity, however, the Court requires the private entity to show that its conduct was pursuant to a “clearly articulated” state policy and that it was “actively supervised” by the state in carrying out the challenged conduct.  62Cal. Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97, 105 (1980) (quoting City of Lafayette v. La. Power & Light Co., 435 U.S. 389, 410 (1978)). Applying this two-pronged test, the Court in California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc. invalidated a California statute requiring wholesale wine producers to file resale price schedules with the state.  63Id. at 105–06. The Court concluded that although the California wine pricing system satisfied the “clear articulation” requirement, it did not meet the “active state supervision” requirement because “[t]he State simply authorize[d] price setting and enforce[d] the prices established by private parties.”  64Id. at 105. Most recently, in North Carolina State Board of Dental Examiners v. Federal Trade Commission, the Court denied state-action immunity to certain anticompetitive conduct by the North Carolina State Board of Dental Examiners on grounds that the board was comprised of active market participants and carried out the conduct with no active supervision by the state.  65See N.C. State Bd. of Dental Exam’rs v. Fed. Trade Comm’n, 135 S. Ct. 1101, 1116–17 (2015).

3. International Trade

The public-private distinction is also made in international trade law in identifying and measuring subsidies. By way of background, Article VI of the General Agreement on Tariffs and Trade (GATT) authorizes contracting states to impose extra tariffs—known as “countervailing duties”—on imports to offset certain subsidies conferred upon the imports when the imports cause or threaten to cause material injury to the importing country’s domestic industries.  66See General Agreement on Tariffs and Trade 1994, Marrakesh Agreement Establishing the World Trade Organization, art. VI, Apr. 15, 1994, 1867 U.N.T.S. 190 [hereinafter GATT 1994]. Under the WTO Agreement on Subsidies and Countervailing Measures (“SCM Agreement”), for a subsidy to be actionable within the WTO framework, it must be a “financial contribution by a government or any public body.”  67See Agreement on Subsidies and Countervailing Measures, art. 1.1(a)(1), Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1869 U.N.T.S. 14 [hereinafter SCM Agreement]. The subsidy must also confer a benefit on the recipient and be “specific to an enterprise or industry or a group of enterprises or industries.” Id. at arts. 1.1(b), 2.1. If the entity accused of giving a subsidy is a private body, the investigating authority must demonstrate that a government “entrusts or directs” the private body to carry out the subsidy-giving activity.  68Id. at art. 1.1(a)(1)(iv).

The question, however, is whether state-owned entities—either state-owned enterprises (SOEs) or state-owned commercial banks—should be considered public or private bodies for purposes of subsidy identification. The policy dilemmas surrounding this question are obvious. On one hand, SOEs or state-owned commercial banks may be operated in the same manner as private businesses.  69See infra note 101 and accompanying text. But on the other hand, they may pursue governmental goals, making their conduct more akin to that of a government.  70“[A] principal justification for the existence of SOEs is to accomplish [governmental objectives, or] objectives that, due to market failure, would otherwise be impossible.” See Curtis J. Milhaupt & Wentong Zheng, Beyond Ownership: State Capitalism and the Chinese Firm, 103 Geo. L.J. 665, 681 (2015).

The classification of SOEs or state-owned commercial banks as public bodies has been extensively debated and litigated following the initiation of countervailing duty investigations by the United States into imports from China.  71The United States started investigating subsidies conferred on imports from China in 2007, in a reversal of its long-standing policy of not imposing countervailing duties on imports from what it considered to be “non-market economies.” See Memorandum from Shauna Lee-Alaia & Lawrence Norton, Office of Policy, Imp. Admin., to David M. Spooner, Assistant Sec’y, Imp. Admin., Countervailing Duty Investigation of Coated Free Sheet Paper from the People’s Republic of China–Whether the Analytical Elements of the Georgetown Steel Opinion are Applicable to China’s Present-Day Economy (Mar. 29, 2007), http://enforcement.trade.gov/frn/summary/PRC/E7-21046-1.pdf. The U.S. Department of Commerce (USDOC), the agency responsible for countervailing duty investigations in the United States, initially took a formalistic approach, treating all Chinese SOEs and state-owned commercial banks as public bodies.  72In Light-Walled Rectangular Pipe and Tube from People’s Republic of China in 2008, the USDOC investigated certain Chinese state-owned steel producers for allegedly providing subsidies to downstream steel pipe and tube producers by selling steel products to them for less than adequate remuneration. See Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Issues and Decision Memorandum for the Final Determination in the Countervailing Duty Investigation of Light-Walled Rectangular Pipe and Tube from the People’s Republic of China 8–9 (June 13, 2008) [hereinafter China Light-Walled RPT I&D Memo], http://enforcement.trade.gov/frn/summary/prc/E8-14250-1.pdf. The USDOC adopted a “majority ownership” rule, under which any SOEs that were majority owned by the Chinese government were considered public bodies. See id. at 29. In Coated Free Sheet Paper from the People’s Republic of China in 2007, the USDOC similarly treated Chinese state-owned commercial banks as public bodies, on grounds that the Chinese government maintained near complete ownership of the banking sector in China and exercised extensive control and influences over the operations of the state-owned commercial banks. See Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Issues and Decision Memorandum on the Final Determination in the Countervailing Duty Investigation of Coated Free Sheet Paper from the People’s Republic of China 55–61 (Oct. 17, 2007) [hereinafter China CFS Paper I&D Memo], http://enforcement.trade.gov/frn/summary/PRC/E7-21046-1.pdf. A WTO dispute settlement panel upheld this approach in a WTO dispute settlement proceeding, DS379.  73See Panel Report, United States–Definitive Anti-Dumping and Countervailing Duties on Certain Products from China, ¶¶ 8.138, 8.143, WT/DS379/R (adopted Oct. 22, 2010) [hereinafter DS379 Panel Report]. On appeal, the WTO Appellate Body  74The WTO Appellate Body is “a standing body of seven persons that hears appeals from reports issued by panels in disputes brought by WTO Members. . . . [O]nce adopted by the [WTO] Dispute Settlement Body, [Appellate Body] reports must be accepted by the parties to the dispute.” See Dispute Settlement: Appellate Body, World Trade Org., https://www.wto.org/english/tratop_e/dispu_e/appellate_body_e.htm (last visited Sept. 14, 2017). agreed with the panel’s finding in DS379 as to the Chinese state-owned commercial banks,  75The Appellate Body noted that the USDOC considered “extensive evidence relating to the relationship between the [state owned commercial banks] and the Chinese Government, including evidence that the [state owned commercial banks] are meaningfully controlled by the government in the exercise of their functions.” See Appellate Body Report, United States–Definitive Anti-Dumping and Countervailing Duties on Certain Products from China, ¶ 355, WT/DS379/AB/R (adopted Mar. 11, 2011) [hereinafter DS379 AB Report]. The Appellate Body concluded that “these considerations, taken together, demonstrate that the USDOC’s public body determination in respect of [state owned commercial banks] was supported by evidence on the record that these [state owned commercial banks] exercise governmental functions on behalf of the Chinese Government.” Id. but rejected its finding as to the Chinese SOEs.  76The Appellate Body found that the USDOC’s reliance on government ownership was not sufficient because “evidence of government ownership, in itself, is not evidence of meaningful control of an entity by government and cannot, without more, serve as a basis for establishing that the entity is vested with authority to perform a governmental function.” Id. at ¶ 346. In another dispute settlement proceeding, DS437, a dispute settlement panel again rejected the practice of treating all Chinese SOEs as public bodies.   77The Panel concluded that the USDOC acted inconsistently with Article 1.1(a)(1) of the SCM Agreement when it found that Chinese SOEs were public bodies “based solely on the grounds that these enterprises were (majority) owned, or otherwise controlled, by the Government of China.” See DS437 Panel Report, supra note 13, ¶ 7.75.

C. Market Distortion by the State

Besides the governmental-proprietary distinction and the public-private distinction, another market-state interaction that has intrigued the law, primarily international trade law, is how the presence of the government in the market affects the integrity of the market. Under GATT Article VI, an importing country is allowed only to impose countervailing duties not in excess of an amount equal to the estimated subsidy.  78GATT 1994, supra note 66, art. VI:3. The SCM Agreement requires that the amount of the subsidy be calculated by comparing the treatment the subsidy recipient receives from the government to the treatment the subsidy recipient would receive or could have received on the private market.  79Article 14 of the SCM Agreement provides guidelines on how to calculate the amount of a subsidy in four scenarios, involving the government provision of equity capital, government-provided loans, government-provided loan guarantees, and the provision of goods or services or purchase of goods by a government. SCM Agreement, supra note 67, at art. 14. In each of the scenarios, Article 14 requires a comparison between the government subsidized outcome and the market outcome. Id. The WTO Appellate Body has interpreted this requirement to embody a market benchmark—a benchmark that identifies and measures a subsidy by determining “whether the recipient has received a ‘financial contribution’ on terms more favourable than those available to the recipient in the market.”  80Appellate Body Report, Canada—Measures Affecting the Export of Civilian Aircraft, ¶ 157, WT/DS70/AB/R (adopted Aug. 2, 1999). For a discussion of the embrace of market benchmarks under the SCM Agreement, see Wentong Zheng, The Pitfalls of the (Perfect) Market Benchmark: The Case of Countervailing Duty Law, 19 Minn. J. Int’l L. 1, 15–18 (2010).

One question arising from the use of market benchmarks for subsidies is what to do with markets that are substantially influenced by the government. In United States–Final Countervailing Duty Determination With Respect to Certain Softwood Lumber from Canada, the Appellate Body addressed whether the United States could reject private stumpage prices in Canada as a benchmark for measuring the subsidies Canada allegedly conferred upon softwood lumber producers through below-market stumpage rates.  81Appellate Body Report, United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada, ¶ 45(b), WT/DS257/AB/R (adopted Jan. 19, 2004). The Appellate Body held that an investigating authority “may use a benchmark other than private prices in the country of provision . . . if it is first established that private prices in that country are distorted because of the government’s predominant role in providing those goods.”  82Id. at ¶ 90. After the Appellate Body recognized the possibility of using benchmarks other than in-country private market prices, the United States rejected private market prices for loans, inputs, and lands in China as subsidy benchmarks in several countervailing duty investigations involving imports from China, on grounds that such in-country private market prices were distorted by the Chinese government.  83In China CFS Paper I&D Memo in 2007, the USDOC refused to use the interest rates for loans made by private and foreign banks in China as the benchmark for measuring whether Chinese policy banks and state-owned commercial banks provided loans at below-market interest rates, because “[the Chinese government]’s intervention in the banking sector creates significant distortions, even restricting and influencing private and foreign banks within the PRC.” China CFS Paper I&D Memo, supra note 72, at 5. In Circular Welded Carbon Quality Steel Pipe from the People’s Republic of China, the USDOC rejected Chinese domestic steel prices as the benchmark for determining whether Chinese SOEs sold steel inputs to downstream producers at below-market prices, because where “the government provides the majority, or a substantial portion of the market for a good or service, prices for such goods and services in the country will be considered significantly distorted and will not be an appropriate basis of comparison for determining whether there is a benefit.” Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Final Determination in the Countervailing Duty Investigation of Circular Welded Carbon Quality Steel Pipe from the People’s Republic of China 64 (May 29, 2008) [hereinafter China CWP I&D Memo] http://ia.ita.doc.gov/frn/summary/PRC/E8-12606-1.pdf.(final determination). Similarly, in Laminated Woven Sacks from the People’s Republic of China, the USDOC rejected Chinese domestic land prices as the benchmark for determining whether the Chinese government granted land-use rights to Chinese producers at below-market prices, because “Chinese land prices are distorted by the significant government role in the market.” Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Final Affirmative Countervailing Duty Determination: Laminated Woven Sacks from the People’s Republic of China 15 (June 16, 2008) [hereinafter China Sacks I&D Memo], http://ia.ita.doc.gov/frn/summary/PRC/E8-14256-1.pdf (final determination).

China challenged the USDOC’s market-distortion analysis before the WTO. In DS379, the Appellate Body sided with the USDOC and upheld its market-distortion analysis for Chinese input subsidies.  84See DS379 AB Report, supra note 75, at ¶¶ 448–58. The Appellate Body reasoned that the Chinese government’s 96.1% market share “makes it likely that the government as the predominant supplier has the market power to affect through its own pricing strategy the pricing by private providers for the same goods, and induce them to align with government prices.”  85Id. at ¶ 455. The Appellate Body also upheld the USDOC’s market-distortion analysis for Chinese loan subsidies.  86See id. at ¶¶ 471–90.

In DS437, however, the Appellate Body reversed itself on Chinese input subsidies. According to the Appellate Body in DS437, the USDOC could not base its rejection of Chinese domestic prices as the subsidy benchmark merely on the fact that government-related entities were the predominant suppliers of the relevant goods.  87See Appellate Body Report, United States—Countervailing Duty Measures on Certain Products from China, ¶ 4.95, WT/DS437/AB/R (adopted Dec. 18, 2014) [hereinafter DS437 AB Report]. The Appellate Body suggested that to find market distortion, the USDOC needed to explain whether and how government-related suppliers “possessed and exerted market power such that other in-country prices were distorted” and whether the prices of “government-related [suppliers] themselves were market determined.”  88Id. at ¶ 4.96.

II. The Existing Market-Versus-State Tests

As discussed above, the modern state interacts with the market in myriad ways, and the law is often called upon to weigh the legal significances of such interactions. However, because market-versus-state interactions take place in markedly different contexts, attempts to arrive at a unified theory of how the law treats the market-state interaction are all but guaranteed to fail.

That said, one can glean valuable insights into how the law approaches the market-state interaction by examining the analytical methods employed by the law in this effort. In this Part, this Article discusses how current law relies on four tests, based on ownership, control, function, and role, respectively, to evaluate market-versus-state interactions. Many of these tests, however, reflect assumptions that have been called into question by changing roles of the state and more nuanced understandings of the market-versus-state dynamics. The ensuing discussions examine the four tests and their limitations.

A. Ownership-Based Test

In drawing the boundary between the market and the state, the law has often used the ownership of the entity that carries out the disputed conduct as the demarcation line. Under this ownership-based test, the ultimate factor that would dictate the legal outcome in a specific case is whether the entity that carries out the conduct is publicly or privately owned.

A subtle form of the ownership-based test is the one used in constitutional law in cases dealing with local flow-control statutes. In 2007, the Supreme Court in United Haulers held that certain local flow-control ordinances requiring waste haulers to bring waste to publicly owned transfer and processing facilities did not violate the dormant Commerce Clause.   89United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth. 550 U.S. 330, 345 (2007). This holding stands in stark contrast with the Court’s holding in the earlier case of C & A Carbone, in which the Court struck down similar flow-control ordinances as per se discrimination against out-of-state waste haulers.  90C & A Carbone, Inc. v. Town of Clarkstown, 511 U.S. 383, 392–94 (1994). The only difference between United Haulers and C & A Carbone is that the former involved waste transfer and processing facilities owned by the government while the latter involved privately owned facilities.  91United Haulers, 550 U.S. at 334. The Court found this difference constitutionally significant, holding that “it does not make sense to regard laws favoring local government and laws favoring private industry with equal skepticism.”  92Id. at 343. According to the Court, laws favoring in-state business over out-of-state competition are often the product of “simple economic protectionism,” while “[l]aws favoring local government . . . may be directed toward any number of legitimate goals unrelated to protectionism.”  93Id. The legality of the local flow-control ordinances, therefore, entirely hinged upon whether the government owned the facilities in question.

Leaving aside the question of whether discrimination in favor of government-owned businesses is more likely than discrimination in favor of private businesses to serve legitimate public interest,  94In his dissenting opinion in United Haulers, Justice Alito argued that laws favoring government-owned businesses were no less discriminatory than laws favoring private businesses, as they “inur[ed] to the benefit of local residents who are employed at the facility, local businesses that supply the facility with goods and services, and local workers employed by such businesses.” Id. at 364 (Alito, J., dissenting). He observed that “[e]xperience in other countries, where state ownership is more common than it is in this country, teaches that governments often discriminate in favor of state-owned businesses (by shielding them from international competition) precisely for the purpose of protecting those who derive economic benefits from those businesses, including their employees.” Id. “Such discrimination,” according to the Justice, “amounts to economic protectionism in any realistic sense of the term.” Id. a fatal flaw of the ownership-based test is that it is overly formalistic and easily circumventable. The facility in C & A Carbone was built by a private contractor for the city free of charge, in return for a guaranteed minimum waste flow for five years for which it was allowed to charge above-market tipping fees.  95C & A Carbone, 511 U.S. at 387. At the end of the five-year period, the contractor would sell the facility to the city for a nominal amount—one dollar.  96Id. This arrangement was essentially a way of financing the cost of constructing the facility, with the private contractor paying the construction cost up front and getting reimbursed by the city through guaranteed higher tipping fees.  97Id. at 393. The city could simply take out a loan and use the loan proceeds to acquire the ownership of the facility from the contractor.  98The city would presumably have little difficulty obtaining the loan, as it could use the future revenues of the facility, backed by flow-control ordinances, as collateral for the loan. With the facility being a government-owned one, the city would be able to bypass the constitutional restriction erected under C & A Carbone. The bottom line here is that ownership can be easily purchased and sold. A legal test whose outcome depends on something as ephemeral as ownership is inherently arbitrary.

A more overt form of the ownership-based test is often utilized in international trade law. This ownership-based test treats state-owned enterprises as public bodies, and their mere presence in the marketplace as market distorting. In several countervailing duty proceedings involving imports from China, the USDOC adopted a “majority ownership” rule, under which any Chinese SOEs that were majority owned by the Chinese government were considered public bodies for subsidy determination purposes.  99See supra note 72 and accompanying text. The normal rule followed by the USDOC was to apply a five-factor test, which inquires about the government ownership of an entity, the government’s presence on the entity’s board of directors, the government’s control over the entity’s activities, the entity’s pursuit of governmental policies or interests, and whether the entity is created by statute. See China Light-Walled RPT I&D Memo, supra note 72, at 27. The USDOC refused to apply the five-factor test to Chinese SOEs on the grounds that the Chinese government failed to provide sufficient information on factors other than the government ownership of the SOEs. Id. at 28–30. Similarly, the USDOC equated government ownership with market distortion in its subsidy benchmark analysis, when it determined that “where . . . the government provides the majority, or a substantial portion of the market for a good or service, prices for such goods and services in the country will be considered significantly distorted and will not be an appropriate basis of comparison for determining whether there is a benefit.”  100China CWP I&D Memo, supra note 83.

The problem with this overt form of the ownership-based test is that it goes against a principal tenet of the modern views of the government. At least from a theoretical perspective, governments could manage SOEs in a hands-off, commercially oriented manner, and ensure a level-playing field between SOEs and private businesses in the marketplace.  101When the U.S. government took over General Motors at the depth of the 2008–2009 financial crisis, it issued guidelines stating that it would manage General Motors “‘in a hands-off, commercial manner’ and not get involved in issuing day-to-day directives to GM.” Neil King Jr. et al., Potential Conflicts Abound in Government Role, Wall St. J. (June 1, 2009, 11:59 PM), http://www.wsj.com/articles/SB124381322189670529. Indeed, such even-handed treatment of state-owned businesses is considered highly desirable, and is promoted as a model for government participation in the economy.  102See OECD, OECD Guidelines on Corporate Governance of State-Owned Enterprises 20 (2015) (recommending that “the legal and regulatory framework for SOEs . . . ensure a level playing field and fair competition in the marketplace when SOEs undertake economic activities”); OECD, Competitive Neutrality: Maintaining a Level Playing Field Between Public and Private Businesses 29–82 (2012) (advocating competitive neutrality between SOEs and private businesses). Equating government ownership with governmental conduct essentially denies this theoretical possibility and presumes that the government plays a distorting role whenever it participates in the economy.

The ownership-based test has been discredited in international trade law. In DS379, the WTO Appellate Body held that “evidence of government ownership, in itself, is not evidence of meaningful control of an entity by government and cannot, without more, serve as a basis for establishing that the entity is vested with authority to perform a governmental function.”  103DS379 AB Report, supra note 75, at ¶ 346. The Appellate Body also rejected “the application of a per se rule, according to which an investigating authority could properly conclude in every case, and regardless of any other evidence, that the fact that the government is the predominant supplier means that private prices are distorted” for the use of out-of-country benchmarks.  104Id. at ¶ 443. In DS437, the Appellate Body again stated that it was inappropriate to infer from the government’s dominant presence in the market that the market price is distorted.  105DS437 AB Report, supra note 87, at ¶ 4.95.

B. Control-Based Test

Another market-versus-state test is to equate government control, rather than government ownership, with governmental conduct. Under this control-based test, if the government exercises effective control of an entity, then the entity will be deemed a governmental entity and its conduct will be considered governmental in nature.

The control-based test has been used primarily in international trade law. Asked to determine the legality of the USDOC’s majority-ownership rule, which treated Chinese SOEs as public bodies based on government ownership alone,  106See supra note 72 and accompanying text. the WTO dispute settlement panel in DS379 interpreted the term “public body” in Article 1.1(a)(1) of the SCM Agreement as “any entity controlled by a government.”   107DS379 Panel Report, supra note 73, at ¶ 8.94. In defining government control, the panel relied on “the everyday financial concept of a ‘controlling interest’ in a company.”  108Id. at ¶ 8.134. What is needed for governments to exercise a controlling interest is “a maximum of 50 per cent plus one share of the voting stock of a company, with the possibility that a much smaller voting block can be controlling, depending on how dispersed the ownership of the remaining shares is, and the extent to which the other shareholders participate in voting.”  109Id. (emphasis omitted). Government ownership, according to the panel, is “highly relevant (indeed potentially dispositive) evidence of government control.”  110Id. The panel thus found “no legal error . . . in giving primacy to evidence of majority government-ownership.”  111Id. at ¶ 8.136.

The WTO Appellate Body, however, rejected the DS379 panel’s singular focus on government control. The Appellate Body acknowledged that “evidence that a government exercises meaningful control over an entity and its conduct may serve, in certain circumstances, as evidence that the relevant entity possesses governmental authority and exercises such authority in the performance of governmental functions.”  112DS379 AB Report, supra note 75, at ¶ 318. But, “apart from an express delegation of authority in a legal instrument, the existence of mere formal links between an entity and government in the narrow sense is unlikely to suffice to establish the necessary possession of governmental authority.”  113Id. at ¶ 318. The Appellate Body finally concluded that “control of an entity by a government, in itself, is not sufficient to establish that an entity is a public body.”  114Id. at ¶ 320.

The Appellate Body’s skepticism of government control is well founded. A control-based market-versus-state test suffers from the same fundamental flaw as the ownership-based test: it ignores the possibility that a government-controlled entity may carry out market activities in the same manner as a private entity.  115Presumably, this possibility underlies the need to differentiate the state’s governmental conduct from its market conduct in the first place. In other words, even if the government has a controlling interest in an entity, it may operate the entity in accordance with market principles.  116Again, such market-oriented management is actively promoted as a best practice for state-owned enterprises. See supra note 102 and accompanying text. Equating government control with governmental conduct, like equating government ownership with governmental conduct, wrongly presumes that all conduct of an entity owned or controlled by the government is governmental in nature.

Another problem with the control-based test is that the concept of government control is often elusive, particularly in countries with weak legal institutions. When the government does not scrupulously observe the boundary between public and private properties, government control extends beyond entities in which the government has a controlling interest in the financial sense.  117See Milhaupt & Zheng, supra note 70, at 685. Take China, a country with respect to which the market-versus-state issue often comes up, for example.  118Note that almost all of the market-versus-state issues in international trade law arise in the context of China. See supra Sections I.B–C. In China, the government could exercise control over firms in which it holds only a minority ownership interest or even no ownership interest at all.  119See Marshall W. Meyer & Changqi Wu, Paulson Inst., Making Ownership Matter: Prospects for China’s Mixed Ownership Economy 9 (2014). The picture becomes even more blurred when the concept of control is broadened from control in the financial sense to control in the residual rights sense, as the Chinese government arguably exercises ultimate control over all firms, whether state-owned or privately owned, through extra-legal means.  120See Milhaupt & Zheng, supra note 70, at 685. But on the other side of the coin, government control of SOEs in China is diluted because of delegation of control to SOE executives and capture of state power by SOEs.  121Id. at 676–83. As a result, “firms in China exhibit substantial similarities” in terms of their practical relationships with the government, regardless of their formal ownership and control structures.  122Id. at 669. In light of these realities, a control-based test for distinguishing between the market and the state will unduly strain the institutional capacity of the authority charged with making that determination.

C. Function-Based Test

Yet another test that has been employed to delineate the boundary between the market and the state is a function-based test, which considers the state to be acting in a governmental capacity if and when the state is performing a governmental function. As discussed below, this function-based test has emerged as a frequently used market-versus-state test in multiple areas of law.

In constitutional law, jurists have proposed to use the function-based test to determine whether states and local government entities should be immune from dormant Commerce Clause scrutiny. In Reeves, where the U.S. Supreme Court used the market-participant exception to uphold South Dakota’s sale of cement produced by a state-owned plant to state residents only,  123See Reeves, Inc. v. Stake, 447 U.S. 429 (1980). Justice Powell dissented and rejected the application of the market-participant exception in this situation.  124Id. at 452–53 (Powell, J., dissenting). According to Justice Powell, while South Dakota entered the private market for cement as a market participant, the dormant Commerce Clause still applied because South Dakota’s marketing policy “cut off interstate trade”  125Id. at 452. and “favor[ed] private, in-state customers over out-of-state customers.”  126Id.at 452–53. The threshold question, Justice Powell argued, was not whether the state entered the private market per se,  127Justice Powell would distinguish the scenario in which the state enters the private market to supply its own needs from the scenario in which the state enters the private market for the advantage of its private citizens. In the former scenario, the state may “act without regard to the private marketplace and remove itself from the reach of the Commerce Clause.” Id. at 450. But in the latter scenario, the state “may not evade the constitutional policy against economic Balkanization.” Id. but “the nature of the government activity involved.”  128Id. at 449. Justice Powell argued for a function-based test to determine whether the Commerce Clause is implicated: “If a public enterprise undertakes an integral operatio[n] in areas of traditional governmental functions, the Commerce Clause is not directly relevant.”  129Id. (alteration in original) (internal quotations and citations omitted).

Following Justice Powell’s initial backing of the function-based test in Reeves, the test gained wider acceptance in subsequent Commerce Clause cases. In United Haulers, the Court stated that it was “particularly hesitant” to use the Commerce Clause to strike down the waste flow-control laws at dispute because “[w]aste disposal is both typically and traditionally a local government function.”  130United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 344 (2007) (alteration in original). In Davis, the nature of the function performed by the government was elevated to be the key factor in disposing of Commerce Clause challenges. Writing for the majority of the Court, Justice Souter in Davis argued that a governmental function is “likely motivate[ed] by legitimate objectives distinct from . . . simple economic protectionism” to such an extent that the standard Commerce Clause does not apply.  131Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 341 (2008). In a footnote, Justice Souter further argued that the point of the function-based test was “not to draw fine distinctions among governmental functions, but to find out whether the preference was for the benefit of a government fulfilling governmental obligations or for the benefit of private interests.” Id. at 341 n.9. This reasoning, however, is rather circular, as whether a preference is for the benefit of the government obviously depends on the classification of the governmental function in question. The Court upheld Kentucky’s tax scheme favoring in-state municipal bonds because “the issuance of debt securities to pay for public projects is a quintessentially public function.”  132Id. at 341–42.

The function-based test has also played a role in deciding whether states and local government entities should be made subject to federal antitrust law. In Jefferson County Pharmaceutical Ass’n v. Abbott Laboratories, the Court addressed whether pharmacies owned and operated by a state university and a county hospital should be required to defend antitrust claims under the Robinson-Patman Act.  133See 460 U.S. 150 (1983). The Court answered the question affirmatively, holding that the issue before it was not state purchases “for use in traditional governmental functions.”  134Id. at 153–54. Instead, the issue was “limited to state purchases for the purpose of competing against private enterprises.”  135Id. at 154. The Court held that the defendants’ acts at issue, the retail sale of pharmaceutical drugs, was not “‘indisputably’ an attribute of state sovereignty.” Id. at 154 n.6.

Finally, international trade law has also embraced the function-based test as the preferred test for determining whether an entity is a public body for subsidy identification purposes. In DS379, after rejecting the control-based test for public bodies,  136See supra notes 112–14 and accompanying text. the WTO Appellate Body held that the assessment of whether an entity is a public body “must focus on evidence relevant to the question of whether the entity is vested with or exercises government authority.”  137DS379 AB Report, supra note 75, at ¶ 345. The panel in DS437 further interpreted the Appellate Body’s holding in DS379 to mean that “the critical consideration in identifying a public body is the question of authority to perform governmental functions.”  138DS437 Panel Report, supra note 13, at ¶ 7.66.

While widely used, the function-based test does not withstand scrutiny. As Justice Alito argued in his dissent in United Haulers, “any standard that turns on a judicial appraisal of whether a particular governmental function is ‘integral’ or ‘traditional’ is unsound in principle and unworkable in practice.’’  139United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 368–69 (2007) (Alito, J., dissenting) (internal quotations omitted). Whether a function is a traditional governmental one obviously depends on whether it is traditionally performed by the government. But traditions evolve over time, rendering the determination of what constitutes traditional governmental functions not subject to objective criteria. For example, the governmental function at dispute in United Haulers—waste disposal—is traditionally performed by local governments, but most of the waste produced in the United States today is managed by private businesses.  140Id. at 369. Some other functions, such as social security and health care, were not traditionally performed by governments, but became quintessential governmental functions as governments expanded their roles over time.  141Historical Background and Development of Social Security, Soc. Security Admin., https://www.ssa.gov/history/briefhistory3.html (last visited Sept. 1, 2017). As a federal district court put it, a traditional governmental function is identified “in the same way pornography is sometimes identified: someone knows it when they see it, but they can’t describe it.”  142San Antonio Metro. Transit Auth. v. Donovan, 557 F. Supp. 445, 453 (W.D. Tex. 1983).

For these reasons, the Court explicitly abandoned the function-based test in Garcia, in which the Court upheld the application of the Fair Labor Standards Act to a municipally owned mass-transit system despite that mass transportation was considered a traditional governmental function.  143Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 531 (1985). The Court rejected the historical approach to state immunity because “it prevents a court from accommodating changes in the historical functions of the States, changes that have resulted in a number of once-private functions like education being assumed by the States and their subdivisions.”  144Id. at 543–44. The Court further rejected nonhistorical standards for selecting immune governmental functions such as those seeking to identify “uniquely” or “necessary” governmental functions.  145Id. at 545.

D. Role-Based Test

Yet another test that has been used to distinguish between the market and the state is to look at the role the government plays in the market-creation process. The focus of this test is whether the government created the market in question in the first place or merely participates in an existing market. This test would give the government special treatment in the former, but not the latter, scenario.

In constitutional law, a role-based test has been offered as a basis for absolving states of dormant Commerce Clause violations. In his concurring opinion in Alexandria Scrap, Justice Stevens argued that Maryland should be allowed to subsidize in-state businesses—and in-state businesses only—for recycling abandoned automobiles because the market for recycling abandoned automobiles was created by the Maryland subsidy program in the first place.  146Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 815–16 (1976) (Stevens, J., concurring). According to Justice Stevens, the interstate commerce said to be burdened by the Maryland subsidy program—the movement of abandoned automobiles from Maryland to out-of-state scrapping plants—“would never have existed if in the first instance Maryland had decided to confine its subsidy to operators of Maryland plants.”  147Id. at 815. The subsidy program Maryland implemented in 1969 initially included subsidies for scrapping plants in Virginia and Pennsylvania. Id. The majority in Alexandria Scrap, however, refused to adopt Justice Stevens’ reasoning because it found that “the record contain[ed] no details of the hulk market prior to the bounty scheme.”  148Id. at 809 n.18 (majority opinion). Indeed, in his dissent, Justice Brennan presented evidence from the record that interstate commerce in abandoned automobiles existed prior to the Maryland subsidy program. Id. at 824 n.6 (Brennan, J., dissenting). But the majority in Alexandria Scrap was receptive to the idea that if the government was indeed the creator of the interstate commerce, it should be able to reduce or eliminate the commerce without violating the dormant Commerce Clause.  149Id. at 809 n.18 (majority opinion) (“We would hesitate to hold that the Commerce Clause forbids state action reducing or eliminating a flow of commerce dependent for its existence upon state subsidy instead of private market forces.”).

The role-based test in the Commerce Clause context found a strong advocate in Professor Laurence Tribe, who attempted to justify the market-participant exception to the dormant Commerce Clause on market-creation grounds.  150See Laurence H. Tribe, Constitutional Choices 146 (1985). Professor Tribe argued that “[t]he principle that necessarily underlies the market participant-market regulator distinction is that, when the state is creating commerce that would not otherwise exist, it has greater freedom to shape that commerce than when it is intruding into a previously existing private market.”  151Id.

A similar role-based test has been adopted in international trade law as well. In Canada–Renewable Energy/Canada–Feed-In Tariff Program, Japan and the European Union argued that Ontario’s feed-in-tariff (FIT) program, which offered to purchase electricity from producers of wind power and solar photovoltaic energy at rates above those accorded to producers of conventional power, constituted a subsidy within the meaning of the SCM Agreement.  152 See DS412/DS426 AB Report, supra note 14, at ¶¶ 1.6–1.7. The Appellate Body rejected this argument, holding that it was unable to determine whether the FIT program conferred a benefit, an element necessary for a finding of a subsidy.  153Id. at ¶ 5.245. The Appellate Body determined that it was unable to complete its analysis because the record from the panel proceeding below contained insufficient factual findings and uncontested evidence that would allow it to draw a conclusion. See id. The Appellate Body held that it could not simply compare the rates offered to renewable energy producers to rates prevailing in the competitive wholesale electricity market as a whole because the relevant markets for the benefit analysis should be the separate competitive markets for wind- and solar-generated electricity.  154Id. at ¶ 5.178. To hold otherwise, the Appellate Body suggested, would mean that the decision by the government to develop renewable energy would “in and of itself” be considered as conferring a benefit.  155Id. at ¶ 5.188 (emphasis omitted). If the comparison is between rates offered to renewable energy producers and rates in the wholesale electricity market, a benefit will necessarily exist because the former are higher than the latter. See Rajib Pal, Has the Appellate Body’s Decision in Canada—Renewable Energy/Canada—Feed-in Tariff Program Opened the Door for Production Subsidies?, 17 J. Int’l Econ. L. 125, 126 (2014). That outcome would have been insensible because, according to the Appellate Body, “a distinction should be drawn between, on the one hand, government interventions that create markets that would otherwise not exist and, on the other hand, other types of government interventions in support of certain players in markets that already exist, or to correct market distortions therein.”  156DS412/DS426 AB Report, supra note 14, at ¶ 5.188. The Appellate Body further explained that “[w]here a government creates a market, it cannot be said that the government intervention distorts the market, as there would not be a market if the government had not created it.”  157Id.

The logic of the role-based test appears to be that when the government creates the market in the first place, it should not be held liable for deviations from the normal rules of the marketplace. In constitutional law, the normal rule is non-discrimination against out-of-state residents or businesses.  158See CTS Corp. v. Dynamics Corp. of Am., 481 U.S. 69, 87 (1987) (“[P]rincipal objects of dormant Commerce Clause scrutiny are statutes that discriminate against interstate commerce.”). In international trade law, the normal rule is competitive pricing decided by market forces.  159This norm is embodied in the use of market benchmarks for identifying and measuring subsidies. See supra Section I.C. But when the government is the creator of the market, it can ignore these normal rules because, as the argument goes, the market would not have existed but for the government’s action.  160Tribe, supra note 150.

The problems with this role-based test are twofold. First, it is not at all clear how to ascertain whether the government creates the market in the first place. When the government supplies cement to in-state residents to replace supply from out-of-state sources, does that constitute the creation of market?  161This is the factual set-up in Reeves, where South Dakota restricted the sale of cement produced by a state-owned cement plant to in-state residents only. See supra note 29 and accompanying text. Professor Tribe believes so, because “lack of supply blocked purchases of cement by state residents [and] the state lessened the obstacle by providing additional supply.”  162Tribe, supra note 150. But the creation of the additional supply creates a market only in the sense that the commerce between the source of the additional supply and the buyers would not have existed but for the additional supply. From the perspective of the pre-existing suppliers, the creation of the additional supply only means intrusion into a pre-existing market.  163Prior to the construction of the state-owned cement plant, producers outside of South Dakota were supplying all the cement used in the state and were making substantial profits. Reeves, Inc. v. Stake, 447 U.S. 429, 431 n.1 (1980). Professor Dan Coenen argued that while South Dakota satisfied more buyers’ needs at lower prices by generating an alternative and additional source of supply, the additional supply displaced out-of-state sellers in a pre-existing market. See Coenen, supra note 16, at 410–11.

The second problem with the role-based test proves more insurmountable. Even assuming that it is straightforward to distinguish between creating a new market and intruding into a pre-existing market, the role-based test is implicitly predicated upon a judgment that the creation of new markets is more valuable than adhering to the normal rules of the marketplace.  164In a similar fashion, Professor Coenen asked: “Why . . . does the ‘creation’ of ‘commerce’ necessarily raise a sufficiently powerful equity to justify state discrimination against interstate commerce?” Coenen, supra note 16, at 411. The application of the role-based test would mean that the government’s creation of new markets—however it is defined and ascertained—is so worthy that it warrants the rejection of any principles at issue, be it nondiscrimination in constitutional law or competitive pricing in international trade law. At least in constitutional law, this judgment goes against court precedent requiring compliance with the nondiscrimination rule even in interstate commerce created by the government.  165See H.P. Hood & Sons, Inc. v. Du Mond, 336 U.S. 525 (1949) (holding that the State of New York was not allowed to deny an out-of-state milk distributor license to purchase milk from within the state for supply to Boston). But even if there is a strong case for the judgment, the case needs to be made, not merely assumed. Arguing for the moral superiority of market creation is inherently context-specific; it requires case-by-case analysis of the value of market creation in comparison to the value of the competing goals. This task could not, by any means, be disposed of through an across-the-board role-based test.

III. Toward a Power-Based Market-Versus-State Test

The foregoing analysis outlines four tests, based on ownership, control, function, and role, respectively, that have been advanced in different areas of law to distinguish between the market and the state. These market-versus-state tests, however, suffer serious limitations. They either presume that all conduct by the government is governmental in nature, as is the case under the ownership-based and control-based tests.  166See supra notes 101–02, 115–16 and accompanying text. Or they base legal outcomes on factors that do not lend themselves to objective determinations, as is the case under the function-based and role-based tests.  167See supra notes 140–42, 161–63 and accompanying text. Or they are implicitly predicated on a value judgment that needs support on a case-by-case basis, as is the case with the role-based test.  168See supra notes 164–65 and accompanying text.

This Part argues that in the search for a reliable market-versus-state test, one defining characteristic of the government has been ignored: the coercive power that only the government, as the sovereign, possesses. This Part proposes a power-based test that draws the boundary between the market and the state based on whether the state is exercising what could be referred to as the governmental power, or coercive power to exert control through the operation of law with no basis in recognized property rights. This governmental power is distinguishable from another coercive power possessed by the government, namely, coercive market power by virtue of the government’s property rights in economic resources. As detailed below, this power-based test avoids the flaws plaguing other market-versus-state tests while, above all, capturing the essence of the government.

This Part first discusses the power-based test and how it differs from the approach in current case law. This Part then argues that the power-based test should be preceded by an inquiry as to the fundamental purpose of the legal regime at issue and whether that purpose implicates the distinction between governmental power and market power. This purpose inquiry is crucial as it determines whether the power-based test should apply in the first place.

A. Governmental Power Versus Market Power

One test that has been ignored by the prevailing market-versus-state analysis is to look at the nature of the power that is being exercised by the state in a specific state action. Under this test, if the power is a coercive one backed by state-sanctioned violence, then the state action will be considered governmental in nature. If the state only exercises power stemming from its market position, then it will be considered to be acting in a market or proprietary capacity.

As sovereigns, governments exercise coercive power over the people they govern.  169See Town of Graham v. Karpark Corp., 194 F.2d 616, 620–21 (4th Cir. 1952) (“A city has two classes of powers, the one legislative or governmental, by virtue of which it controls its people as their sovereign, the other proprietary or business, by means of which it acts and contracts for the private advantage of the inhabitants of the city and of the city itself.” (quoting Omaha Water Co. v. City of Omaha, 162 F. 225 (8th Cir. 1908))). This sovereign coercive power, sometimes referred to as regulatory power, is the power to exert control through the operation of law with no basis in recognized property rights.  170Nachbar, supra note 44, at 70–73 (distinguishing between regulatory control and property rights). One prominent example of this sovereign coercive power is the government’s power to impose taxes.  171See Donald H. Regan, The Supreme Court and State Protectionism: Making Sense of the Dormant Commerce Clause, 84 Mich. L. Rev. 1091, 1194 (1986) (describing taxation as one of the more coercive government powers). The sovereign coercive power of the government has animated many legal principles designed to keep the government in check.  172See, e.g., Free v. Miles, 333 F.3d 550, 554 (5th Cir. 2003) (per curiam) (“The limited function of [the common law rule that a prisoner is entitled to credit for time served when he is incarcerated through no fault of his own] is to prevent the government from abusing its coercive power to imprison a person by artificially extending the duration of his sentence through releases and re-incarcerations.”); Flint v. Mullen, 499 F.2d 100, 105 (1st Cir. 1974) (Coffin, C.J., dissenting) (“The safeguards which courts have created around the exercise of the Fifth Amendment’s privilege against self-incrimination are essentially bulwarks against abuse of the government’s coercive power.”); Milwaukee Deputy Sheriffs Ass’n v. Clarke, 513 F. Supp. 2d 1014, 1022 (E.D. Wis. 2007) (holding that the actions of the county sheriff in inviting representatives of a religious organization to speak at department leadership conference and roll calls, which police deputies were required to attend, was an unconstitutional endorsement of religion through use of coercive power of government). In a similar vein, the possession and exercise of sovereign coercive power has been considered the hallmark of governmental conduct.  173In constitutional law, for example, a private actor may be considered to act as an instrument or agent of the government for purposes of Fourth and Fifth Amendments if “‘the government exercise[s] such coercive power or such significant encouragement that it is responsible’ for the [private actor’s] conduct.” See United States v. Garlock, 19 F.3d 441, 443 (8th Cir. 1994) (quoting Fidelity Fin. Corp. v. Fed. Home Loan Bank, 792 F.2d 1432, 1435 (9th Cir. 1986)).

The government’s sovereign coercive power needs to be differentiated from another kind of coercive power the government possesses: coercive power by virtue of the government’s property rights in economic resources. For example, when the government refuses to hire contractors whose workforce does not include a sufficient percentage of local residents, the government is essentially attempting to coerce potential contractors to hire local workers up to the set percentage.  174This is the factual setup presented in White v. Massachusetts Council of Construction Employers, Inc., in which the city of Boston required all city-funded construction projects “be performed by a work force consisting of at least half bona fide [city] residents.” 460 U.S. 204, 205–06 (1983). But this coercion is backed not by threats of state-sanctioned violence, but by the government’s willingness to turn away noncompliant contractors from government projects. To the extent that the government projects account for a significant share of the projects available on the market, this willingness to say no on the part of the government could rise to the level of coercion.  175Even if the government is just one of many employers on the market, the government’s willingness to reject noncompliant contractors is still coercive with respect to the specific government projects in question. For the sake of convenience, this Article will refer to coercion by virtue of the government’s sovereign power as “government coercion” and coercion by virtue of the government’s market power as “market coercion.”

Government coercion and market coercion share a major commonality. From the perspective of the parties being coerced, whether the coercion is a government one or a market one makes little practical difference. For instance, when a contractor is compelled to hire 50% of its employees from local residents, the contractor would not care whether it is compelled by a government mandate or by the needs to secure government contracts. Professor Donald Regan argues that government-spending programs are “less coercive than regulatory programs or taxes with similar purposes,” and therefore, “they seem to interfere less, or less objectionably, with the ordinary workings of the market economy.”  176Regan, supra note 171. Regan put forward this argument as one of the differential treatments of spending versus regulation or tax under the dormant Commerce Clause. See id. But in terms of the impact on the parties being coerced, the degree of coerciveness is substantially similar between government coercion and market coercion.  177Some may argue that market coercion is not coercion in the true sense because one always has the choice of foregoing the market opportunity and thus not being coerced. But as Professor Coenen argues, the same can be said of government coercion, as one always has the choice of not doing business in a specific jurisdiction. See Coenen, supra note 16, at 415.

Despite their similarities, what sets government coercion apart from market coercion is that the former could only be exercised by the government, while the latter could be exercised by both the government and private actors.  178By definition, private actors could only exercise market coercion by virtue of their property rights in economic resources, as they are not sovereigns and lack the ability to exercise regulatory power. For this reason, this Article refers to the power of government coercion as the “governmental power” and the power to exercise market coercion as the “market power.” A power-based test takes advantage of this distinction and draws the boundary between the market and the state by inquiring about the nature of the power being exercised by the government. Under this power-based test, the government acts in a governmental capacity if and only if it exercises the governmental power—power that could not be exercised by private actors.

The power-based test proposed in this Article differs from the approach in current case law that distinguishes between the market and the state by analogizing the government to private actors. This approach, also known as the “private trader analogy”  179See Treg A. Julander, State Resident Preference Statutes and the Market Participant Exception to the Dormant Commerce Clause, 24 Whittier L. Rev. 541, 558 (2002). or “private-actor analogy,”  180See Stephanie Landry, Comment, State Immunity from the Dormant Commerce Clause: Extension of the Market-Participant Doctrine from State Purchase and Sale of Goods and Services to Natural Resources, 25 Nat. Resources J. 515, 523 (1985). focuses on whether the specific state action in question is something that a private actor could or would have done. In Reeves, for example, Justice Powell in his dissent suggested that in shutting off cement sales to out-of-state customers, South Dakota was not really behaving like a private cement producer because it was responding to “political concerns that likely would be inconsequential to a private cement producer.”  181Reeves, Inc. v. Stake, 447 U.S. 429, 453 (1980) (Powell, J., dissenting). Similarly, in Wisconsin Department of Industry, Labor and Human Relations v. Gould, Inc., the Supreme Court held that Wisconsin was “not functioning as a private purchaser” of products when it prohibited state purchases from repeat labor law violators.  182475 U.S. 282, 289 (1986) (quoting Gould, Inc. v. Wis. Dep’t of Indus., Labor and Human Relations, 750 F.2d 608, 614 (7th Cir. 1984)).

The problem with the private-actor analogy is that it requires the government to behave exactly like private actors to be considered acting in a private capacity.  183For the government to be considered to be acting in a private capacity under this private actor analogy, the government could not respond to concerns or factors that would not be taken into account by private actors. See supra note 183 and accompanying text. But by definition, the government responds to incentives or considerations distinct from those facing private actors.  184As a social planner, the government in conducting its affairs necessarily considers factors that go beyond the realms of consideration of the individual members of the society. See Jean Hindriks & Gareth D. Myles, Intermediate Public Economics 425–26 (2d ed. 2013) (describing the difference between the social planner’s welfare function and the welfare functions of individual members of the society). So the private-actor analogy essentially amounts to a per se rule under which a government action is considered governmental in nature simply because it is being conducted by the government.  185This per se rule is not unlike the ownership- and control-based tests, which treat conduct by entities owned or controlled by the government as governmental in nature. See supra Sections II.A–B.

By contrast, the power-based test proposed in this Article would focus on the nature of the power being exercised by the government in conducting the action in question. In Reeves, the power-based test would have treated South Dakota’s cement sales as a market action, because South Dakota was exercising the power of deciding to whom it will sell its products  186See Reeves, Inc. v. Stake, 447 U.S. 429, 432–33 (1980). —a power that every private actor possesses.  187A private actor’s power to choose the party with which to deal is subject to an important limitation under antitrust law. When the private actor is a dominant firm, it can exercise the right to refuse to deal with a competitor only if there are legitimate competitive reasons for the refusal. See Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985). Similarly, in Gould, Wisconsin was exercising the power of deciding from whom to source its supplies,  188See Wis. Dep’t of Indus., Labor and Human Relations v. Gould, 475 U.S. 282, 283–84 (1986). and the power-based test would have treated the state purchases in question as a market action despite Wisconsin’s non-market motive in imposing the purchase restriction.  189See id. at 287.

This distinction between the power-based test and the prevailing private-actor analogy is crucial. The emphasis of the power-based test is whether a private actor possesses the power that is being exercised by the government, not whether a private actor would want to exercise that power. A good example of this difference can be found in Alexandria Scrap, in which Maryland exercised its prerogative as a market participant to bid up the price of abandoned automobiles through subsidies, but limited the availability of the subsidies to in-state businesses through stricter documentation requirements for out-of-state businesses.  190See Hughes v. Alexandria Scrap Corp., 426 U.S. 794 (1976). The inquiry under the power-based test would be whether a private actor possesses the same kind of power being exercised by Maryland—buying products at higher-than-market prices and choosing from whom to buy the products. The answer to that question is obviously yes. So although a rational private actor would not have wanted to exercise that power, the fact that it possesses the power leads to the conclusion that Maryland was acting in a proprietary or market capacity.  191In Davis, Justice Souter stated that “in Alexandria Scrap, Maryland employed the tools of regulation to invigorate its participation in the market for automobile hulks.” Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 346 (2008). Justice Souter was mistaken in this conclusion, as Maryland was imposing stricter documentation requirements for out-of-state businesses in connection with its subsidy spending. The power exercised by Maryland was not regulatory in nature, as a private actor looking to purchase products at higher-than-market prices also has the power to impose stricter documentation requirements on potential sellers.

Besides capturing the fundamental distinction between the market and the state, the power-based test overcomes the weaknesses of the other market-versus-state tests that are based on ownership, control, function, and role. The power-based test recognizes that the government could act in a market capacity when it is not exercising government coercion.  192Failure to recognize this possibility is the main problem with the ownership- and control-based tests. See supra Sections II.A–B. The power-based test is not susceptible to evolving standards, as it is rooted in the nature of government power, which does not change over time.  193By contrast, the standard for the function-based test is ever-changing depending on the prevailing conceptions of the government’s functions. See supra Section II.C. And it is not implicitly based on a value judgment that needs to be justified.  194Unlike the role-based test, which is based on an implicit value judgment that market creation trumps the statutory goals in question, the power-based test does not judge the relative values of the different kinds of government power. For discussions of the value dependency of the role-based test, see supra Section II.D. In sum, the power-based test provides an objective, time-invariant, and value-free way of distinguishing between the state’s governmental role and market role.

B. A Purpose Inquiry

The power-based test proposed in this Article comes with an important caveat. Because the state interacts with the market in numerous settings, not all of which implicate the distinction between governmental power and market power, the power-based test may not be apposite in all settings where the market-versus-state distinction is being made. Therefore, the power-based test needs to be preceded by an inquiry as to the fundamental purpose of the legal regime at issue and whether that purpose implicates the distinction between governmental power and market power. Not only is this purpose-inquiry critical to the application of the power-based test, but it also helps reveal the meaning and purpose behind the market-versus-state distinction.

The need for a purpose inquiry can be illustrated using constitutional law as an example. Under the market-participant exception to the dormant Commerce Clause, the state will be exempted from normal constitutional discipline if it is acting in a proprietary capacity.  195See supra section I.A.1. But the question is why. Answering this question requires an analysis of the fundamental purpose of the dormant Commerce Clause. As Professors Michael Wells and Walter Hellerstein argue, “[I]f economic Balkanization is the evil that the commerce clause was designed to prevent, what difference does it make whether the evil is brought about by states acting in their governmental or proprietary capacities?”  196Wells & Hellerstein, supra note 16, at 1125. But if the Commerce Clause is intended to address only the ills brought about by government coercion through the exercise of the government’s regulatory power, then the distinction between the state’s governmental and proprietary capacities will become constitutionally meaningful.  197For further discussions of the purpose of the Commerce Clause, see infra Section IV.A.

This analytical framework—the power-based test preceded by a purpose inquiry—can be applied to any legal regimes that raise market-versus-state issues. In antitrust law, for example, the application of this analytical framework turns the debate on whether there should be a market-participant exception to the state-action immunity into a question about the purpose of the state-action immunity. Is the purpose of the immunity to take cognizance of federalism concerns arising from the states’ exercise of regulatory power only,  198John F. Hart, “Sovereign” State Policy and State Action Antitrust Immunity, 56 Fordham L. Rev. 535, 536 (1988). or is it to make federal antitrust law yield to all conduct by states, in both governmental and proprietary capacities? The answer to this question is of utmost importance to discerning the scope of the state-action immunity.  199For more discussions of the purpose of the state-action immunity in antitrust law, see infra Section IV.B.

For some legal regimes, the purpose inquiry may lead to the conclusion that both governmental power and market power are implicated. As explained below, this is the case in international trade law, which restricts governments’ ability to confer subsidies in both governmental and proprietary capacities.  200See infra Section IV.C. The market-state distinction, however, may still be relevant to such legal regimes if different legal standards apply to the different roles of the government. In that case, the power-based test can be used to determine in what role the government acts and consequently what legal standard applies.

IV. Applying the Power-Based Market-Versus-State Test

The power-based test, coupled with an inquiry as to the purpose of the underlying legal regime in relation to state power, yields valuable insights into the market-versus-state distinction. As detailed below, this analytical framework not only distinguishes the market and the state, but also answers the question of why there needs to be a market-versus-state distinction in the first place. The application of the power-based test sheds light on thorny market-versus-state issues in constitutional law, antitrust law, and international trade law.

A. Constitutional Law

Despite the Supreme Court’s insistence on a state-versus-market distinction for dormant Commerce Clause purposes, scholars have questioned the validity of such distinction. Professors Wells and Hellerstein, for example, argue that the state-versus-market distinction has no legitimate place in dormant Commerce Clause inquiries because the state’s market activities result in the same economic Balkanization as do the state’s governmental activities.  201See Wells & Hellerstein, supra note 16, at 1125–26. Other scholars also reject prevailing justifications for the market-participant exception on numerous policy grounds.  202See, e.g., Karl Manheim, New-Age Federalism and the Market Participant Doctrine, 22 Ariz. St. L.J. 559, 582–601 (1990); Jonathan D. Varat, State “Citizenship” and Interstate Equality, 48 U. Chi. L. Rev. 487, 505–07 (1981).

Under the power-based test, it becomes clear that different conceptualizations of the purpose of the dormant Commerce Clause motivate the different approaches to the market-participant exception to the clause. In Alexandria Scrap, the district court adopted an expansive interpretation of the Commerce Clause, based on the premise that “this Nation is a common market in which state lines cannot be made barriers to the free flow of both raw materials and finished goods in response to the economic laws of supply and demand.”  203See Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 803 (1976). Under this interpretation, any state actions that inhibit the national common market are within the purview of the dormant Commerce Clause, regardless of whether the states inhibit the national common market through governmental power or market power.  204Id. at 805 (“This line of reasoning is not without force if its basic premise is accepted. That premise is that every action by a State that has the effect of reducing in some manner the flow of goods in interstate commerce is potentially an impermissible burden.”). Rejecting the district court’s reasoning, the Supreme Court in Alexandria Scrap narrowed the reach of the dormant Commerce Clause to include only state interference with “the natural functioning of the interstate market either through prohibition or through burdensome regulation.”  205Id. at 806. In other words, according to the Court, only when the states are exercising government coercion will the dormant Commerce Clause be implicated. Based on this narrower interpretation of the dormant Commerce Clause, the Court in Alexandria Scrap went on to recognize a market-participant exception. Similarly, in Reeves, the Court stated that “the Commerce Clause responds principally to state taxes and regulatory measures impeding free private trade in the national marketplace. There is no indication of a constitutional plan to limit the ability of the States themselves to operate freely in the free market.”  206Reeves, Inc. v. Stake, 447 U.S. 429, 436–37 (1980) (internal citation omitted).

If the purpose of the dormant Commerce Clause is to only reign in governmental power, then whether the state exercises that power in the specific action at issue becomes the ultimate factor in determining whether the dormant Commerce Clause applies. In other words, when the state exercises market power and market power only, it acts in a proprietary capacity and will be out of the purview of the dormant Commerce Clause.

This insight would inject much-needed logic into the case law on the market-participant exception to the dormant Commerce Clause. In Alexandria Scrap, Maryland was exercising the power of spending money to subsidize the recycling of abandoned automobiles.  207Alexandria Scrap, 426 U.S. at 797. Although Maryland imposed a documentation requirement on the subsidy recipients, that requirement was a condition for receiving the subsidies, something in which the state had property rights.  208Id. at 798. So at most Maryland was exercising market power, a power that private actors possess.  209Although a private actor would not want to give out money for free to encourage the recycling of abandoned automobiles, it does have the power to do so and to impose documentation requirements as a condition. In Reeves, South Dakota was exercising the power of choosing to whom to sell its products, a quintessential private power.  210See Reeves, 447 U.S. at 432–33. Similarly, in White, the City of Boston was exercising market power, not governmental power, when it required contractors to hire at least 50% of their labor force from local residents before they were allowed to work on city-funded projects.  211See White v. Mass. Council of Constr. Emp’rs, Inc., 460 U.S. 204 (1983). The City of Boston would have exercised the power of government coercion had it required all businesses, not just businesses working on city-funded projects, to hire at least 50% of their labor force from local residents. The Court correctly held, therefore, that the government in all these cases qualified for the market-participant exception.

The power-based test does indicate that the Court was wrong in not applying the market-participant exception in Wunnicke, where Alaska required that timber taken from state lands be processed within the state prior to export.  212South-Central Timber Dev., Inc. v. Wunnicke, 467 U.S. 82, 84 (1984). The Court declined to apply the market-participant exception to the Alaska requirement because it was a downstream restriction with regulatory effects.  213Id. at 99. The Court stated that “[t]he limit of the market-participant doctrine must be that it allows a State to impose burdens on commerce within the market in which it is a participant, but allows it to go no further.”  214Id. at 97. Viewed in light of the power-based test, this characterization of the market-participant exception is both overly broad and overly narrow. It is overly broad because the market-participant exception does not always allow a state to impose burdens on commerce within the market in which it is a participant—it only allows a state to impose burdens that could be imposed by private actors.  215In United Haulers, for example, the government imposed restrictions with regulatory effects in the market in which it was a participant—the waste transfer and processing market. See United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 334 (2007). It is overly narrow because when a state exercises market power only, it is accorded the market-participant status even if the restriction extends beyond the immediate market in which the state is a participant. This was the case in White, in which the Court upheld the City of Boston’s labor force requirement for government contractors despite the fact that the city itself was not a participant in the labor market at issue.  216White, 460 U.S. at 205–06. The city was a participant in the market for construction projects, not in the labor market for workers who would work on those projects.

The power-based test would point to a different outcome in Wunnicke, because Alaska in that case was merely imposing a restriction on the processing of timber taken from state-owned lands in which Alaska had property rights.   217Wunnicke, 467 U.S. at 84. In so doing, Alaska was exercising a power possessed by private actors, who could impose similar processing requirements for timber taken from their own lands. The power-based test, therefore, would exempt the Alaska requirement from scrutiny under the dormant Commerce Clause.

The power-based test would also help solve another puzzle in the dormant Commerce Clause jurisprudence. Under current case law, while the Court draws a distinction between impermissible market regulation and permissible market participation,  218See, e.g., id. at 93 (plurality opinion) (“Our cases make clear that if a State is acting as a market participant, rather than as a market regulator, the dormant Commerce Clause places no limitation on its activities.”). it has not delineated a clear boundary between the two. For instance, the Court treats tax credits and exemptions as market regulation on the grounds that taxation is inherently governmental.  219See, e.g., New Energy Co. of Ind. v. Limbach, 486 U.S. 269, 277 (1988) (“The market-participant doctrine has no application here. The Ohio action ultimately at issue is neither its purchase nor its sale of ethanol, but its assessment and computation of taxes—a primeval governmental activity.”). But it fails to explain why subsidies, which it upholds as market participation under the market-participant rule,  220Recall that Alexandria Scrap involved a state subsidy program. See supra note 27 and accompanying text. are any less governmental than taxation.  221Government spending on economic and social programs accounts for a large percentage of the government’s activities. The U.S. federal government, for example, spends almost half of its budget on the so-called entitlement programs, including Social Security, Medicaid, and Medicare. See Romina Boccia, Heritage Found., Federal Spending by the Numbers, 2014: Government Spending Trends in Graphics, Tables, and Key Points (Including 51 Examples of Government Waste) (2014). One economist has referred to government spending in the form of subsidies one of the “three major sets of measures” by which the government “alter[s] the distribution of income.” Milton H. Spencer, Contemporary Economics 72 (2d ed. 1974). Commentators offered several justifications for the more relaxed treatment of subsidies under the dormant Commerce Clause on the basis of policy and history.  222See Coenen, supra note 16, at 479–81; Collins, supra note 25, at 98–103; Enrich, supra note 16, at 442–43; Gergen, supra note 16, at 1134–53; Hellerstein & Coenen, supra note 16, at 846–48; Paul S. Kline, Publicly-Owned Landfills and Local Preferences: A Study of the Market Participant Doctrine, 96 Dick. L. Rev. 331, 370–91 (1992); Regan, supra note 171, at 1193–95. One difference noted by commentators between taxation and subsidies is that the latter is relatively expensive, making it less likely to cause serious damage to the economy.  223See Regan, supra note 171; see also Hellerstein & Coenen, supra note 16, at 846–47. Another frequently made argument is that the “conscious funding” required of subsidies operates as a check against their proliferation.  224Saul Levmore, Interstate Exploitation and Judicial Intervention, 69 Va. L. Rev. 563, 585 (1983); see also Coenen, supra note 16, at 479; Collins, supra note 25, at 102–03; Enrich, supra note 16, at 442–43; Kline, supra note 222, at 374. Commentators also resorted to formalism and tradition as explanations.  225See Dan T. Coenen, Business Subsidies and the Dormant Commerce Clause, 107 Yale L.J. 965, 980–81 (1998); Coenen, supra note 16, at 480; Gergen, supra note 16, at 1136–37. But however convincing these justifications might be,  226Commentators have questioned some of those justifications. For instance, Coenen argues that it is debatable whether subsidy programs are inherently more expensive than discriminatory state regulatory and tax programs. See Coenen, supra note 16, at 434. they have no bearings on the question of why the state is a market regulator with respect to taxation, but a market participant with respect to subsidies.  227These justifications either do not address the market-participation question at all, or just circle back to the argument that taxation is inherently governmental. See Enrich, supra note 16, at 442.

The power-based test provides a straightforward justification for the differential treatment of taxation and subsidies under the dormant Commerce Clause. In Limbach, the Court held that Ohio was not acting as a market participant when it provided tax credit to ethanol producers from Ohio or from states that granted reciprocal tax credit, exemption, or refund for Ohio-produced ethanol.  228New Energy Co. of Ind. v. Limbach, 486 U.S. 269, 271, 277 (1988). The Court reasoned that the state action at issue was the “assessment and computation of taxes—a primeval governmental activity.”  229Id. at 277. In Camps Newfound/Owatonna, Inc. v. Town of Harrison, the Court held that Maine was not engaging in a proprietary activity when it provided property tax exemptions only to charities that catered principally to state residents, on grounds that “[a] tax exemption is not the sort of direct state involvement in the market that falls within the market-participation doctrine.”  230520 U.S. 564, 593 (1997). In neither case, however, did the Court specify why state actions involving taxation are inherently governmental. The power-based test supplies the missing rationale. Taxation is inherently governmental because it entails the exercise of governmental power, a power that no private actors possess. Seen in this light, state actions involving taxation are governmental in nature and should not be eligible for the market-participant exception.  231There is a potential fault line in the Court’s treatment of taxation cases. While the Court in Limbach and Camps Newfound/Owatonna struck down the discriminatory tax credit or exemption at issue as violations of the dormant Commerce Clause, it upheld, in Davis, a Kentucky tax scheme in which the state exempted interest income on state municipal bonds, but not interest income on out-of-state municipal bonds, from state income taxes. See Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 333–34, 341–42 (2008). The nature of the power exercised by the government in Davis, however, was arguably indistinguishable from the nature of power exercised by the government in Limbach and Camps Newfound/Owatonna. So although Kentucky could be said to be participating in the market for municipal bonds by making in-state municipal bonds more attractive, the fact that it was exercising the power of government coercion makes it clear that it was acting in a governmental capacity. The power-based test, therefore, would squarely place the state action in Davis subject to the dormant Commerce Clause.To the extent that Davis needs to be distinguished from Limbach and Camp Newfounds/Owatonna, such distinctions need to come from elsewhere. One likely basis on which Davis could be distinguished from Limbach and Camp Newfounds/Owatonna is that in Davis, the ultimate beneficiary of the state action was the state itself, while in Limbach and Camp Newfounds/Owatonna, the ultimate beneficiaries of the state actions were individuals or entities other than the state. In Davis, the ultimate beneficiary of the state action was the state itself because the discriminatory tax exemption made it easier for the political subdivisions of the state to issue bonds. In Limbach, the ultimate beneficiaries of the state action were all those who would benefit from the local production of ethanol—farmers, distributors, and suppliers, etc. In Camp Newfounds/Owatonna, the ultimate beneficiaries of the state action were all residents who would stand to benefit from the services provided by the charities eligible for the tax exemption. It could be argued that when the ultimate beneficiary of the state action is the state itself, it does not raise the same level of concerns about discrimination as when the ultimate beneficiary of the state action is someone other than the state. Note that this distinction is not based on the government ownership or control of the ultimate beneficiary of the state action and thus can avoid the problems associated with the ownership- and control-based tests. For discussions of the ownership- and control-based tests, see supra Sections II.A–B.

By contrast, when the state gives subsidies, it exercises a power that every private actor possesses: giving out money for free without receiving things of equal value in return.  232A private business, for example, could decide to subsidize a particular group of customers using revenues collected from other customers. Turbotax, the popular tax software, charges $0 for customers who file the most basic tax returns but $54.99–$114.99 for customers who file more complicated returns. See Compare Turbotax Online Products, TurboTax, https://turbotax.intuit.com/personal-taxes/compare/online/ (last visited Aug. 23, 2017). The only case in which the Court struck down a state subsidy program was West Lynn Creamery, Inc. v. Healy, in which Massachusetts imposed a tax on fluid milk sold by dealers to Massachusetts retailers and distributed the assessment to Massachusetts dairy farmers.  233512 U.S. 186, 188 (1994). The Court held that the Massachusetts program unconstitutionally discriminated against interstate commerce even though the two components of the program—the subsidy component and the tax component—would be valid separately.  234Id. at 198–202. It was the coupling of a subsidy with a simultaneously enacted tax, according to the Court, that raised constitutional difficulties.  235Id. at 200–01. Viewed against the backdrop of the power-based test, the Court’s objection to an otherwise legal subsidy scheme in West Lynn makes perfect sense, as the state was exercising governmental power as well as market power. It was the coupling of the two powers that sets West Lynn apart from other subsidy cases.  236The coupling of governmental and market power of spending money takes West Lynn out of the realm of the market-participant exception. The Court in West Lynn could have brought more consistency to the case law by explicitly noting that the market-participant exception did not apply, rather than being silent about the market-participant exception at all. For discussions of the Court’s inconsistency in using the market-participant exception as the framework for analyzing dormant Commerce Clause issues, see supra Section II.A.1.

The power-based test also elucidates another line of cases involving local flow-control ordinances. In C & A Carbone and United Haulers, the Court was confronted with the question of whether flow-control ordinances requiring waste haulers to bring waste to designated waste transfer and processing facilities violated the dormant Commerce Clause. The Court answered yes in C & A Carbone but no in United Haulers, with the only difference between the two cases being that the designated facility was privately owned in the former case but publicly owned in the latter.  237For discussions of C & A Carbone and United Haulers, see supra Section II.A.1. It is abundantly clear, however, that the government was exercising governmental power in both cases, as the government was mandating the disposal of waste in which it had no property rights.  238In Department of Revenue of Kentucky v. Davis, Justice Souter suggested that United Haulers “may also be seen under the broader rubric of the market participant doctrine.” 553 U.S. 328, 343 (2008) (plurality opinion). According to Justice Souter, the fact that the state simultaneously exercises regulatory power while participating in market activities does not make the market-participant rule inapposite; the dispositive factor is the government’s market activities. Id. at 346 (“Not only did the public authority acting in [United Haulers] process trash, but its governmental superiors forbade trash haulers to deal with any other processors. This latter fact did not determine the outcome, however; the dispositive fact was the government’s own activity in processing trash.”). Justice Souter would have upheld the flow-control ordinances at issue in United Haulers under the market-participant exception because the counties in that case participated in the market for trash processing. Id. at 346–47 (“We upheld the government’s decision to shut down the old market for trash processing only because it created a new one all by itself, and thereby became a participant in a market with just one supplier of a necessary service.”). Justice Souter also characterized two other market-participant cases, Alexandria Scrap and White, as examples of the government’s commercial activities being joined by its regulatory efforts. Id. at 345–46. But obviously, Justice Souter’s expansive view of the market-participant exception was not shared by the rest of the majority. And as previously discussed, the government ownership of the waste transfer and processing facilities at issue in C & A Carbone and United Haulers was a rather fictitious way of distinguishing the two cases.  239See supra notes 95–98 and accompanying text. The power-based test, therefore, would demand equal treatment of the flow-control ordinances in those two cases.

B. Antitrust Law

In antitrust law, complex market-versus-state issues revolve not so much around how to draw the market-versus-state distinction as around whether the market-versus-state distinction is relevant at all. In cases involving the state-action immunity doctrine, it is relatively clear whether the state was acting in a regulatory capacity or a proprietary capacity. In Town of Hallie v. City of Eau Claire, for example, the action being challenged was the defendant city’s monopoly over the provision of sewage treatment services and its attempt to tie the provision of such services to the provision of sewage collection and transportation services.  240471 U.S. 34, 36–37 (1985). In Federal Trade Commission v. Phoebe Putney Health System, Inc., for another example, the action being challenged was the defendant hospital authority’s acquisition of another hospital.  241568 U.S. 216, 222 (2013). In both cases, the defendants were engaged in market activities that were otherwise indistinguishable from conduct by private businesses. By contrast, in North Carolina State Board of Dental Examiners v. Federal Trade Commission, the action being challenged was the issuing of cease-and-desist letters by the North Carolina State Board of Dental Examiners to non-dentist teeth whitening service providers.  242135 S. Ct. 1101, 1108–09 (2015). It is obvious that in issuing the letters, the board was exercising governmental power, as no private actors possess the power to force other private actors to stop providing services that they are legally entitled to provide.  243Whether the board was exercising the power of government coercion is a question separate from whether the board was a governmental body. Even if the board was a private body, that distinction does not prevent the conclusion that it was exercising governmental power.

While it is relatively straightforward to distinguish between the market and the state in the antitrust context, the market-versus-state distinction has not really mattered much under current case law. Under the state-action immunity doctrine, what matters is whether the action at issue reflects the state’s policy, regardless of the capacity in which the state carries out the action.  244See supra note 62 and accompanying text. So even if the state acts in a proprietary capacity, as in Phoebe Putney, as long as the action is clearly articulated by state policy and actively supervised by the state, it would be eligible for the state-action immunity.  245In Phoebe Putney, the Court denied state-action immunity to the defendant on the basis that there was not a sufficient articulation of state policy to authorize the anticompetitive conduct. 568 U.S. at 228.

But had there been a market-participant exception to the state-action immunity, it would have mattered whether the state acts in a governmental or proprietary capacity. Under such an exception, when the state acts in a proprietary capacity, the state will be subject to antitrust scrutiny regardless of whether it authorizes the action. In Phoebe Putney, an amicus curiae argued that the Court should recognize such an exception, although its argument was based, erroneously, on the notion that the state’s commercial activities are not a traditional government function.  246See Brief for National Federation of Independent Business as Amicus Curiae in Support of the Petitioner at 6–24, Phoebe Putney, 568 U.S. 216 (No. 11-1160). The amicus curiae argued that applying state-action immunity to the state’s commercial conduct would exceed the state-action immunity doctrine’s purpose because “[m]arket-participant state conduct is not an ‘integral operation in an area of traditional government functions.’” Id. at 15 (quoting City of Lafayette v. La. Power & Light Co., 435 U.S. 389, 424 (1978) (Burger, C.J., concurring)). This argument resembles the function-based test that distinguishes between the regulatory and market roles of the state based on whether the function at issue is a traditional government function. See supra Section II.C. The Court, however, declined to take up this question, because it “was not raised by the parties or passed on by the lower courts.”  247See Phoebe Putney, 568 U.S. at 226 n.4.

Should there be a market-participant exception to the state-action immunity in antitrust law? While the Court has not explicitly endorsed a market-participant exception, its holding in Abbott Laboratories provides some support for such an exception.  248Jefferson Cty. Pharm. Ass’n v. Abbott Labs., 460 U.S. 150 (1983). In that case, plaintiff pharmacies sued defendant state university and county hospital pharmacies for inducing pharmaceutical manufacturers to sell their products to the defendants at prices lower than those charged to the plaintiffs in violation of the Robinson-Patman Act.  249Id. at 151–52. The Robinson-Patman Act provides: “It shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality . . . where the effect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them . . . .” Id. at 152 n.1 (quoting 15 U.S.C. § 13(a) (1982)). The defendants sought to dismiss the case “on the ground that state purchases are exempt [from the Robinson-Patman Act] as a matter of law.”  250Id. at 153. The Court rejected this argument, holding that “the exemption does not apply where a State has chosen to compete in the private retail market.”  251Id. at 154. The Court observed that “[o]n numerous occasions, this Court has affirmed the comprehensive coverage of the antitrust laws and has recognized that these laws represent ‘a carefully studied attempt to bring within [them] every person engaged in business whose activities might restrain or monopolize commercial intercourse among the states.’”  252Id. at 157. The Court held that the term “person” used in the Robinson-Patman Act is sufficiently broad to cover governmental bodies. Id. at 155. It is unclear, however, whether this sweeping statement implies a general market-participant exception to state-action immunity, given that the Court in this case predicated its holding on its analysis of the text and legislative history of the specific statute at dispute, the Robinson-Patman Act.  253The Court concluded that “the plain language of the [Robinson-Patman] Act strongly suggests that there is no exemption for state purchases to compete with private enterprise.” Id. at 156–57. The Court then found that “[t]he legislative history [of the Robinson-Patman Act] falls far short of supporting respondents’ contention that there is an exemption for state purchases of ‘commodities’ for ‘resale.’” Id. at 159.

In another case, City of Lafayette v. Louisiana Power & Light Co., the Court also sent mixed signals on the market-participant exception to the state-action immunity.  254435 U.S. 389 (1978). In that case, the Court addressed whether defendant cities that owned and operated electric utility systems should be allowed to dismiss antitrust counterclaims filed against them simply by reason of their status as state agencies or subdivisions of a state.  255Id. at 392. The counterclaims alleged that, among other things, one of the defendant cities contracted to provide gas and water service to the plaintiff’s electric customers only on the condition that the customers purchase electricity from the city. Id. at 403–04. On one hand, the Court rejected defendants’ argument that “the antitrust laws are intended to protect the public only from abuses of private power and not from actions of municipalities that exist to serve the public weal.”  256Id. at 403. Instead, the Court stated that “the economic choices made by public corporations in the conduct of their business affairs . . . are not inherently more likely to comport with the broader interests of national economic well-being than are those of private corporations.”  257Id. But on the other hand, the Court went on to hold that “the fact that municipalities, simply by their status as such, are not within the Parker doctrine, does not necessarily mean that all of their anticompetitive activities are subject to antitrust restraints.”  258Id. at 413. According to the Court, municipalities are still eligible for immunity from antitrust laws if their activities are directed by the state.  259Id. at 416. The Court agreed with the lower court that further inquiries should be made to determine whether the defendants’ actions were directed by the state. See id. at 413–15. Therefore, despite its assertion of the broad coverage of antitrust laws, the Court in City of Lafayette indeed implied that antitrust laws did not apply to the state’s actions as a market participant.  260Recall that the defendant cities’ actions at issue in City of Lafayette were commercial in nature. See supra note 255 and accompanying text.

The power-based market-versus-state test would bring clarity to the debate on whether there should be a market-participant exception to the state-action immunity. Under the power-based test, whether a state’s proprietary activities should be outside of the purview of a specific legal regime depends on whether the legal regime is implicated only when the state exercises governmental power. In the context of the state-action immunity, this inquiry questions the fundamental purpose of this immunity.

It is a near consensus that the state-action immunity is based on the concept of federalism.  261See, e.g., Robert P. Inman & Daniel L. Rubinfeld, Making Sense of the Antitrust State-Action Doctrine: Balancing Political Participation and Economic Efficiency in Regulatory Federalism, 75 Tex. L. Rev. 1203, 1250–71 (1997) (examining the antitrust state-action doctrine from the perspective of federalism); Jorde, supra note 16, at 230 (arguing that the Parker Court grounded the state action doctrine and its own judicial restraint on the structure of economic federalism); 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, 1105 (1981) (“[T]he Parker Court did not base its construction of the Sherman Act on legislative history, but instead derived a presumption of congressional intent from the concept of federalism.”). Federalism promotes “citizen participation in government, efficiency in government, creative experimentation, and diffusion of power.”  262Jorde, supra note 16, at 230–31. The Court’s willingness to defer to “a dual system of government in which . . . the states are sovereign” forms the basis of the state-action immunity.  263Parker v. Brown, 317 U.S. 341, 351 (1943). But the question is whether federalism concerns arise to the same degree when a state is exercising governmental power and when it is exercising market power. In other words, the question is whether the nature of the power being exercised by the state is relevant to the value of federalism.

While this question has no easy answers, the weight of the arguments tilts in favor of no market-participant exception to the state-action immunity. It could certainly be argued that when a state exercises market power, its prerogatives as a sovereign are not compromised by the application of federal antitrust law to the same degree as when it exercises governmental power. But no matter whether a state exercises governmental or market power, deference to state actions serves the same goals of federalism, namely promoting citizen participation, efficiency in government, creative experimentation, and diffusion of power.  264For instance, it could be argued that when citizens of a state authorize an SOE to engage in anticompetitive conduct, they are participating in government to the same degree as when they impose coercive requirements. It could also be argued that they are addressing the special needs of the local population, thereby promoting government efficiency. And such participation equally promotes policy experimentation and diffusion of power. Furthermore, as Professor Coenen argued, restrictions on the state’s exercise of market activities are a greater intrusion into state sovereignty than are restrictions on the state’s regulatory activities, as “state resources are the state’s ‘own’ in a way that the state’s regulatory powers are not.”  265See Coenen, supra note 16, at 427. The power-based test, therefore, would support shielding states’ market or proprietary activities from federal antitrust law.

C. International Trade Law

Applying the power-based test to international trade law reveals that international trade law regulates both governmental power and market power. As discussed below, this conclusion is dictated by the purpose of international trade law in general and international subsidy law in particular.

A core principle of international trade law is the “liberal economic doctrine,” which recognizes the benefits of free trade to all countries participating in international trade.  266Arie Reich, From Diplomacy to Law: The Juridicization of International Trade Relations, 17 Nw. J. Int’l L. & Bus. 775, 781 (1997). Consistent with this principle, the primary purpose of the GATT and the WTO is to dismantle barriers to trade.  267Id. at 780; see also Jan Hoogmartens, EC Trade Law Following China’s Accession to the WTO 10 (2004). As a general matter, international trade law seeks to reduce or eliminate trade barriers that arise from both the regulatory and market conduct of the states. A typical trade barrier erected by the states in their regulatory capacities is tariffs,  268Tariffs are akin to taxes in that their imposition requires the power of government coercion, as no private actors possess the power to collect money without providing goods or services of equivalent values. which the GATT and the WTO have successfully reduced over the years.  269Average tariffs in industrial countries plummeted from 40% in 1947, when the GATT entered into force, to 6.3% in 1994 just prior to the entry into force of the Uruguay Round agreements, and further to 3.9% as a result of the Uruguay Round agreements that led to the establishment of the WTO. See Raj Bhala, Rethinking Antidumping Law, 29 Geo. Wash. J. Int’l L. & Econ. 1, 3 (1995). International trade law also imposes disciplines on the states when they undertake market activities through “state trading enterprises.”  270GATT 1994, supra note 66, at art. XVII. The states are required, among other things, to operate such enterprises in accordance with commercial considerations.  271Article XVII of GATT 1994 requires: “[S]uch enterprises shall, having due regard to the other provisions of this Agreement, make any such purchases or sales solely in accordance with commercial considerations, including price, quality, availability, marketability, transportation and other conditions of purchase or sale, and shall afford the enterprises of the other contracting parties adequate opportunity, in accordance with customary business practice, to compete for participation in such purchases or sales.” Id. at art. XVII:1(b). These disciplines on the states’ market activities are based on the understanding that the states may operate such businesses “so as to create serious obstacles to trade.”  272Id. at art. XVII:3.

That international trade law is concerned about the states’ market conduct can also be seen in the very existence of international subsidy law. As previously discussed, the conferral of subsidies generally involves no governmental power; any private actors could give out money without receiving a quid pro quo.  273See supra note 232 and accompanying text. In legal regimes that are intended to only curb the state’s exercise of governmental power—the dormant Commerce Clause under constitutional law, for example  274For a discussion of the purpose of the dormant Commerce Clause, see supra Section IV.A. —the granting of subsidies by the states poses no threats to the goal of the legal regimes. That explains why state actions involving subsidies are generally exempted from scrutiny in such legal regimes.  275The only exception is when the granting of subsidies involves the power of government coercion, in which case the subsidy-granting action will be made subject to the discipline of the state’s regulatory actions. See, e.g., West Lynn Creamery, Inc. v. Healy, 512 U.S. 186 (1994). The fact that strict subsidy rules exist under international trade law indicates a fundamental difference between the purpose of international trade law and that of constitutional law.

Although international trade law implicates both governmental power and market power, it still makes a distinction between the market and the state. In international subsidy law, an important question is when an entity owned or controlled by the state is a “public body.”  276For a background discussion on the “public body” issue, see supra Section I.B.3. This question matters because the legal rules for identifying a countervailable subsidy differ depending on whether the entity in question is a public or private body.  277See supra notes 67–68 and accompanying text. As discussed above, the WTO experimented with ownership-based and control-based definitions of public bodies before settling on a function-based definition.  278See supra Sections II.A–C. According to the WTO dispute settlement panel in DS437, “the critical consideration in identifying a public body is the question of authority to perform governmental functions.”  279DS437 Panel Report, supra note 13, at ¶ 7.66. This definition, however, begs the question of what constitutes a governmental function.  280For more discussion of the problems with the function-based definition of public bodies, see supra Section II.C. The logic of the power-based test points to a straightforward answer to this question. Under the power-based test, a public body should be defined as an entity that possesses and exercises governmental power. So when an SOE is allegedly providing a subsidy, whether the SOE should be deemed a public body will depend on whether the SOE is equipped with coercive power with no basis in recognized property rights.  281In DS437, China argued that “[a] public body, like government in the narrow sense, thus must itself possess the authority to ‘regulate, control, supervise or restrain’ the conduct of others.” DS437 Panel Report, supra note 13, at ¶ 7.67 (alteration in original). The panel, however, rejected this argument on textual grounds. Id. at ¶ 7.68. This power-based definition would avoid the problems associated with other definitions based on ownership, control, function, or role.

Conclusion

What is the boundary between the market and the state? The prevailing approaches focus on the government’s ownership, control, function, or role to answer this question. This Article challenges these approaches and proposes an alternative market-versus-state test that focuses on the nature of the power being exercised in the challenged action. This power-based test captures the essence of governmental conduct and, when coupled with inquiries as to the fundamental purpose of the legal regime at issue, sheds light on the more important question of why the market-versus-state distinction matters at all. The insights from the power-based test lay the foundation for a coherent methodology for tackling complex legal issues surrounding the market-versus-state distinction.

Footnotes

*Professor of Law & University Term Professor, University of Florida Levin College of Law. I thank Tom Ankersen, Darren Hutchinson, Lars Noah, Bill Page, Laura Rosenbury, Sharon Rush, D. Daniel Sokol, and Michael Wolf for helpful comments and discussions. The research for this Article was supported by a summer research grant from the University of Florida Levin College of Law.

1Edward L. Glaeser & Andrei Shleifer, The Rise of the Regulatory State, 41 J. Econ. Literature 401, 401 (2003).

2See Karen Yeung, The Regulatory State, in The Oxford Handbook of Regulation 72–73 (Robert Baldwin et al. eds., 2010).

3State-owned corporations are common even in Western countries. For a history of state-owned enterprises in Western countries, see Pier Angelo Toninelli, The Rise and Fall of Public Enterprise: The Framework, in The Rise and Fall of State-Owned Enterprise in the Western World 3 (Pier Angelo Toninelli ed., 2000).

4In the United States, the federal government is the largest employer in the nation, with more than 2.1 million civilian workers and 1.3 million active duty military who serve in all fifty states and around the world. Office of Mgmt. & Budget, Fiscal Year 2016 Analytical Perspectives of the U.S. Government 75 (2015).

5For example, “[t]he U.S. [federal] government is the largest single purchaser of goods and services in the world, awarding approximately $500 billion in contracts every year.” SBA’s Role in Government Contracting, U.S. Small Bus. Admin., https://www.sba.gov/contracting/what-government-contracting/sbas-role-government-contracting (last visited Aug. 28, 2017).

6In the wake of the global financial crisis, the U.S. government seized control of American International Group, one of the world’s largest insurers, to prevent the company from falling into bankruptcy. See Matthew Karnitschnig et al., U.S. to Take Over AIG in $85 Billion Bailout; Central Banks Inject Cash as Credit Dries Up, Wall St. J. (Sept. 16, 2008, 11:59 PM), https://www.wsj.com/articles/SB122156561931242905?mg=prod/accounts-wsj. The U.S. government also took 60% ownership of General Motors as part of the latter’s government-orchestrated bankruptcy process. See Neil King Jr. & Sharon Terlep, GM Collapses Into Government’s Arms, Wall St. J. (June 2, 2009, 12:01 AM), https://www.wsj.com/articles/SB124385428627671889.

7See The Visible Hand, Economist (Jan. 21, 2012), http://www.economist.com/node/21542931.

8Id.

9Ian Bremmer, The End of the Free Market: Who Wins the War Between States and Corporations? (2010).

10United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330 (2007) (holding that county ordinances mandating the delivery of waste to designated waste transfer and processing facilities did not violate the dormant Commerce Clause of the U.S. Constitution).

11Fed. Trade Comm’n v. Phoebe Putney Health Sys., Inc., 568 U.S. 216 (2013) (holding that a state hospital authority is not entitled to state-action immunity in its acquisition of existing hospitals).

12N.C. State Bd. of Dental Exam’rs v. Fed. Trade Comm’n, 135 S. Ct. 1101 (2015) (holding that a dental licensing board is not eligible for state-action immunity because its action was not actively supervised by the state).

13Panel Report, United States—Countervailing Duty Measures on Certain Products from China, ¶ 7.75, WTO Doc. WT/DS437/R (adopted July 14, 2014) [hereinafter DS437 Panel Report] (holding that the United States acted inconsistently with its WTO obligations when it found that Chinese SOEs were public bodies based only on the grounds that they were majority owned or controlled by the Chinese government).

14 Appellate Body Report, Canada—Certain Measures Affecting the Renewable Energy Generation Sector/Canada—Measures Relating to the Feed-in Tariff Program, WTO Doc. WT/DS412/AB/R, WT/DS426/AB/R (adopted May 6, 2013) [hereinafter DS412/DS426 AB Report] (holding that it was unable to determine whether Ontario’s feed-in-tariff program conferred a benefit because the relevant market for the benefit analysis should be the separate competitive markets for wind- and solar-generated electricity).

15Unless otherwise noted, the term “state” in this Article refers to the government in the abstract sense, not any particular political division or subdivision.

16See, e.g., Dan T. Coenen, Untangling the Market-Participant Exemption to the Dormant Commerce Clause, 88 Mich. L. Rev. 395 (1989) (discussing the market-participant exception to the dormant Commerce Clause in U.S. constitutional law); Einer Richard Elhauge, The Scope of Antitrust Process, 104 Harv. L. Rev. 667 (1991) (proposing a process view of the antitrust state action doctrine); Peter D. Enrich, Saving the States from Themselves: Commerce Clause Constraints on State Tax Incentives for Business, 110 Harv. L. Rev. 377 (1996) (proposing an antidistortion standard for determining whether a state tax incentive violates the Commerce Clause); Mark P. Gergen, The Selfish State and the Market, 66 Tex. L. Rev. 1097 (1988) (discussing principles for differentiating between permissible and impermissible state preferences for its own citizens); Walter Hellerstein & Dan T. Coenen, Commerce Clause Restraints on State Business Development Incentives, 81 Cornell L. Rev. 789 (1996) (discussing the tax-subsidy distinction in Commerce Clause jurisprudence); Thomas M. Jorde, Antitrust and the New State Action Doctrine: A Return to Deferential Economic Federalism, 75 Calif. L. Rev. 227 (1987) (discussing the state action doctrine in antitrust law and its relationship to economic federalism); Michael Wells & Walter Hellerstein, The Governmental-Proprietary Distinction in Constitutional Law, 66 Va. L. Rev. 1073 (1980) (discussing the governmental-proprietary distinction in various constitutional law contexts).

17Constitutional law scholars, for example, have not drawn connections between various market-versus-state issues in constitutional law with their counterparts in other areas of law. See, e.g., Hellerstein & Coenen, supra note 16 (discussing the tax-subsidy distinction in Commerce Clause jurisprudence but not whether and how such a distinction is made in international trade law).

18See infra Part I.

19See infra Part II.

20See infra Section III.A.

21See infra Section III.B.

22See infra Part IV.

23Wells & Hellerstein, supra note 16, at 1075.

24U.S. Const. art. I, § 8, cl. 3.

25See Gen. Motors Corp. v. Tracy, 519 U.S. 278, 287 (1997). For an overview of the dormant Commerce Clause jurisprudence, see Richard B. Collins, Economic Union as a Constitutional Value, 63 N.Y.U. L. Rev. 43, 47–60 (1988).

26For background on this “market-participant” or “proprietary-activities” exception to the dormant Commerce Clause, see Gergen, supra note 16, at 1139–43.

27426 U.S. 794 (1976). The lower court found the Maryland program unconstitutional, citing its “substantial burdens upon the free flow of interstate commerce.” See Alexandria Scrap Corp. v. Hughes, 391 F. Supp. 46, 62 (D. Md. 1975), rev’d, 426 U.S. 794 (1976). The Supreme Court, however, distinguished the instant case from previous dormant Commerce Clause cases, arguing that Maryland in this case was merely acting as “a purchaser, in effect, of a potential article of interstate commerce.” Alexandria Scrap, 426 U.S. at 808.

28Alexandria Scrap, 426 U.S. at 810.

29447 U.S. 429 (1980).

30460 U.S. 204 (1983).

31See Reeves, 447 U.S. at 436; White, 460 U.S. at 207 (quoting Reeves, 447 U.S. at 436).

32467 U.S. 82 (1984).

33Id. at 99.

34486 U.S. 269 (1988).

35Id. at 271, 280.

36Id. at 277.

37520 U.S. 564, 593 (1997) (“A tax exemption is not the sort of direct state involvement in the market that falls within the market-participation doctrine.”).

38426 U.S. 833, 836, 852 (1976).

39Id. at 852.

40469 U.S. 528, 530–31 (1985).

41Id. at 531.

42Id. at 550. The Court implied, however, that a constitutional challenge might remain open if “the internal safeguards of the political process have [not] performed as intended.” Id. at 556.

43United States v. Topco Assocs., Inc., 405 U.S. 596, 610 (1972).

44See Donald F. Turner, The Durability, Relevance, and Future of American Antitrust Policy, 75 Calif. L. Rev. 797, 798 (1987) (“Antitrust law is a pro-competition policy. The economic goal of such a policy is to promote consumer welfare through the efficient use and allocation of resources, the development of new and improved products, and the introduction of new production, distribution, and organizational techniques for putting economic resources to beneficial use.”). But see Robert H. Lande, Consumer Choice as the Ultimate Goal of Antitrust, 62 U. Pitt. L. Rev. 503 (2001) (arguing that the ultimate goal of antitrust law is to promote consumer choice); Thomas B. Nachbar, The Antitrust Constitution, 99 Iowa L. Rev. 57 (2013) (arguing that antitrust law serves a constitutional function of preventing the assertion of control over the conduct of others outside the sphere of one’s own property interests).

45Parker v. Brown, 317 U.S. 341, 344, 352 (1943) (“The state in adopting and enforcing the prorate program made no contract or agreement and entered into no conspiracy in restraint of trade or to establish monopoly but, as sovereign, imposed the restraint as an act of government . . . .”).

46Id. at 352.

47Id. at 351–52.

48499 U.S. 365, 374–75 (1991). The Court later interpreted this statement to be “leaving open the possibility of a market participant exception” to state-action immunity. Fed. Trade Comm’n v. Phoebe Putney Health Sys., Inc., 568 U.S. 216, 226 n.4 (2013).

49See 568 U.S. at 221–22. The proposed transaction would have allowed the county hospital authority to control 86% of the market for acute-care hospital services provided to commercial health care plans and their customers. Id.

50The Court did not consider this question in its opinion because “this argument was not raised by the parties or passed on by the lower courts.” Id. at 226 n.4.

51The Court applied the “clear-articulation” test under California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97 (1980), and concluded that the county hospital authority was not eligible for state-action immunity because “there is no evidence the State affirmatively contemplated that hospital authorities would displace competition by consolidating hospital ownership.” Id. at 226.

52See, e.g., Or. Waste Sys., Inc. v. Dep’t of Envtl. Quality, 511 U.S. 93 (1994) (holding that a surge Oregon imposed on in-state disposal of waste generated in other states was facially invalid under the dormant Commerce Clause); City of Philadelphia v. New Jersey, 437 U.S. 617 (1978) (holding that a New Jersey statute prohibiting the importation of solid or liquid waste that originated outside the territorial limits of the state violated the dormant Commerce Clause); Dean Milk Co. v. City of Madison, 340 U.S. 349 (1951) (holding that a city ordinance prohibiting the sale of milk unless bottled within five miles from the center of the city unduly burdened interstate commerce).

53See, e.g., Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 458, 472 (1981) (holding that a Minnesota statute “banning the retail sale of milk in plastic nonreturnable, nonrefillable containers, but permitting such sale in other nonreturnable, nonrefillable containers, such as paperboard milk cartons” did not violate the dormant Commerce Clause because its incidental burden on interstate commerce was not clearly excessive in relation to its putative local benefits); Kassel v. Consol. Freightways Corp., 450 U.S. 662 (1981) (holding that an Iowa statute barring the use of trucks longer than sixty feet on Iowa’s interstate highways substantially burdened interstate commerce in violation of the dormant Commerce Clause); Pike v. Bruce Church, Inc., 397 U.S. 137 (1970) (holding that an Arizona order prohibiting a fruit grower from transporting uncrated cantaloupes from its Arizona ranch to a nearby California city for packing and processing constituted an unlawful burden on interstate commerce).

54511 U.S. 383 (1994).

55Id. at 387.

56Id. at 392.

57550 U.S. 330 (2007).

58Id. at 342.

59Id. at 346. The Court upheld the laws because they had no “disparate impact on out-of-state as opposed to in-state businesses.” Id. The Court also suggested that the laws in dispute should survive Commerce Clause challenges because they carried out a traditional governmental function. Id. at 344 (“We should be particularly hesitant to interfere with the Counties’ efforts under the guise of the Commerce Clause because ‘[w]aste disposal is both typically and traditionally a local government function.’” (citation omitted)).

60Parker v. Brown, 317 U.S. 341, 350–51 (1943).

61See, e.g., Hoover v. Ronwin, 466 U.S. 558, 567–68 (1984) (“Thus, under the Court’s rationale in Parker, when a state legislature adopts legislation, its actions constitute those of the State, and ipso facto are exempt from the operation of the antitrust laws.” (citation omitted)); Bates v. State Bar of Ariz., 433 U.S. 350, 359–60 (1977) (holding that the state bar of Arizona automatically received Parker immunity because the anticompetitive conduct was pursuant to the affirmative command of the Arizona Supreme Court).

62Cal. Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97, 105 (1980) (quoting City of Lafayette v. La. Power & Light Co., 435 U.S. 389, 410 (1978)).

63Id. at 105–06.

64Id. at 105.

65See N.C. State Bd. of Dental Exam’rs v. Fed. Trade Comm’n, 135 S. Ct. 1101, 1116–17 (2015).

66See General Agreement on Tariffs and Trade 1994, Marrakesh Agreement Establishing the World Trade Organization, art. VI, Apr. 15, 1994, 1867 U.N.T.S. 190 [hereinafter GATT 1994].

67See Agreement on Subsidies and Countervailing Measures, art. 1.1(a)(1), Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1869 U.N.T.S. 14 [hereinafter SCM Agreement]. The subsidy must also confer a benefit on the recipient and be “specific to an enterprise or industry or a group of enterprises or industries.” Id. at arts. 1.1(b), 2.1.

68Id. at art. 1.1(a)(1)(iv).

69See infra note 101 and accompanying text.

70“[A] principal justification for the existence of SOEs is to accomplish [governmental objectives, or] objectives that, due to market failure, would otherwise be impossible.” See Curtis J. Milhaupt & Wentong Zheng, Beyond Ownership: State Capitalism and the Chinese Firm, 103 Geo. L.J. 665, 681 (2015).

71The United States started investigating subsidies conferred on imports from China in 2007, in a reversal of its long-standing policy of not imposing countervailing duties on imports from what it considered to be “non-market economies.” See Memorandum from Shauna Lee-Alaia & Lawrence Norton, Office of Policy, Imp. Admin., to David M. Spooner, Assistant Sec’y, Imp. Admin., Countervailing Duty Investigation of Coated Free Sheet Paper from the People’s Republic of China–Whether the Analytical Elements of the Georgetown Steel Opinion are Applicable to China’s Present-Day Economy (Mar. 29, 2007), http://enforcement.trade.gov/frn/summary/PRC/E7-21046-1.pdf.

72In Light-Walled Rectangular Pipe and Tube from People’s Republic of China in 2008, the USDOC investigated certain Chinese state-owned steel producers for allegedly providing subsidies to downstream steel pipe and tube producers by selling steel products to them for less than adequate remuneration. See Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Issues and Decision Memorandum for the Final Determination in the Countervailing Duty Investigation of Light-Walled Rectangular Pipe and Tube from the People’s Republic of China 8–9 (June 13, 2008) [hereinafter China Light-Walled RPT I&D Memo], http://enforcement.trade.gov/frn/summary/prc/E8-14250-1.pdf. The USDOC adopted a “majority ownership” rule, under which any SOEs that were majority owned by the Chinese government were considered public bodies. See id. at 29. In Coated Free Sheet Paper from the People’s Republic of China in 2007, the USDOC similarly treated Chinese state-owned commercial banks as public bodies, on grounds that the Chinese government maintained near complete ownership of the banking sector in China and exercised extensive control and influences over the operations of the state-owned commercial banks. See Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Issues and Decision Memorandum on the Final Determination in the Countervailing Duty Investigation of Coated Free Sheet Paper from the People’s Republic of China 55–61 (Oct. 17, 2007) [hereinafter China CFS Paper I&D Memo], http://enforcement.trade.gov/frn/summary/PRC/E7-21046-1.pdf.

73See Panel Report, United States–Definitive Anti-Dumping and Countervailing Duties on Certain Products from China, ¶¶ 8.138, 8.143, WT/DS379/R (adopted Oct. 22, 2010) [hereinafter DS379 Panel Report].

74The WTO Appellate Body is “a standing body of seven persons that hears appeals from reports issued by panels in disputes brought by WTO Members. . . . [O]nce adopted by the [WTO] Dispute Settlement Body, [Appellate Body] reports must be accepted by the parties to the dispute.” See Dispute Settlement: Appellate Body, World Trade Org., https://www.wto.org/english/tratop_e/dispu_e/appellate_body_e.htm (last visited Sept. 14, 2017).

75The Appellate Body noted that the USDOC considered “extensive evidence relating to the relationship between the [state owned commercial banks] and the Chinese Government, including evidence that the [state owned commercial banks] are meaningfully controlled by the government in the exercise of their functions.” See Appellate Body Report, United States–Definitive Anti-Dumping and Countervailing Duties on Certain Products from China, ¶ 355, WT/DS379/AB/R (adopted Mar. 11, 2011) [hereinafter DS379 AB Report]. The Appellate Body concluded that “these considerations, taken together, demonstrate that the USDOC’s public body determination in respect of [state owned commercial banks] was supported by evidence on the record that these [state owned commercial banks] exercise governmental functions on behalf of the Chinese Government.” Id.

76The Appellate Body found that the USDOC’s reliance on government ownership was not sufficient because “evidence of government ownership, in itself, is not evidence of meaningful control of an entity by government and cannot, without more, serve as a basis for establishing that the entity is vested with authority to perform a governmental function.” Id. at ¶ 346.

77The Panel concluded that the USDOC acted inconsistently with Article 1.1(a)(1) of the SCM Agreement when it found that Chinese SOEs were public bodies “based solely on the grounds that these enterprises were (majority) owned, or otherwise controlled, by the Government of China.” See DS437 Panel Report, supra note 13, ¶ 7.75.

78GATT 1994, supra note 66, art. VI:3.

79Article 14 of the SCM Agreement provides guidelines on how to calculate the amount of a subsidy in four scenarios, involving the government provision of equity capital, government-provided loans, government-provided loan guarantees, and the provision of goods or services or purchase of goods by a government. SCM Agreement, supra note 67, at art. 14. In each of the scenarios, Article 14 requires a comparison between the government subsidized outcome and the market outcome. Id.

80Appellate Body Report, Canada—Measures Affecting the Export of Civilian Aircraft, ¶ 157, WT/DS70/AB/R (adopted Aug. 2, 1999). For a discussion of the embrace of market benchmarks under the SCM Agreement, see Wentong Zheng, The Pitfalls of the (Perfect) Market Benchmark: The Case of Countervailing Duty Law, 19 Minn. J. Int’l L. 1, 15–18 (2010).

81Appellate Body Report, United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada, ¶ 45(b), WT/DS257/AB/R (adopted Jan. 19, 2004).

82Id. at ¶ 90.

83In China CFS Paper I&D Memo in 2007, the USDOC refused to use the interest rates for loans made by private and foreign banks in China as the benchmark for measuring whether Chinese policy banks and state-owned commercial banks provided loans at below-market interest rates, because “[the Chinese government]’s intervention in the banking sector creates significant distortions, even restricting and influencing private and foreign banks within the PRC.” China CFS Paper I&D Memo, supra note 72, at 5. In Circular Welded Carbon Quality Steel Pipe from the People’s Republic of China, the USDOC rejected Chinese domestic steel prices as the benchmark for determining whether Chinese SOEs sold steel inputs to downstream producers at below-market prices, because where “the government provides the majority, or a substantial portion of the market for a good or service, prices for such goods and services in the country will be considered significantly distorted and will not be an appropriate basis of comparison for determining whether there is a benefit.” Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Final Determination in the Countervailing Duty Investigation of Circular Welded Carbon Quality Steel Pipe from the People’s Republic of China 64 (May 29, 2008) [hereinafter China CWP I&D Memo] http://ia.ita.doc.gov/frn/summary/PRC/E8-12606-1.pdf.(final determination). Similarly, in Laminated Woven Sacks from the People’s Republic of China, the USDOC rejected Chinese domestic land prices as the benchmark for determining whether the Chinese government granted land-use rights to Chinese producers at below-market prices, because “Chinese land prices are distorted by the significant government role in the market.” Memorandum from Stephen J. Claeys, Deputy Assistant Sec’y for Imp. Admin., to David M. Spooner, Assistant Sec’y for Imp. Admin., on the Final Affirmative Countervailing Duty Determination: Laminated Woven Sacks from the People’s Republic of China 15 (June 16, 2008) [hereinafter China Sacks I&D Memo], http://ia.ita.doc.gov/frn/summary/PRC/E8-14256-1.pdf (final determination).

84See DS379 AB Report, supra note 75, at ¶¶ 448–58.

85Id. at ¶ 455.

86See id. at ¶¶ 471–90.

87See Appellate Body Report, United States—Countervailing Duty Measures on Certain Products from China, ¶ 4.95, WT/DS437/AB/R (adopted Dec. 18, 2014) [hereinafter DS437 AB Report].

88Id. at ¶ 4.96.

89United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth. 550 U.S. 330, 345 (2007).

90C & A Carbone, Inc. v. Town of Clarkstown, 511 U.S. 383, 392–94 (1994).

91United Haulers, 550 U.S. at 334.

92Id. at 343.

93Id.

94In his dissenting opinion in United Haulers, Justice Alito argued that laws favoring government-owned businesses were no less discriminatory than laws favoring private businesses, as they “inur[ed] to the benefit of local residents who are employed at the facility, local businesses that supply the facility with goods and services, and local workers employed by such businesses.” Id. at 364 (Alito, J., dissenting). He observed that “[e]xperience in other countries, where state ownership is more common than it is in this country, teaches that governments often discriminate in favor of state-owned businesses (by shielding them from international competition) precisely for the purpose of protecting those who derive economic benefits from those businesses, including their employees.” Id. “Such discrimination,” according to the Justice, “amounts to economic protectionism in any realistic sense of the term.” Id.

95C & A Carbone, 511 U.S. at 387.

96Id.

97Id. at 393.

98The city would presumably have little difficulty obtaining the loan, as it could use the future revenues of the facility, backed by flow-control ordinances, as collateral for the loan.

99See supra note 72 and accompanying text. The normal rule followed by the USDOC was to apply a five-factor test, which inquires about the government ownership of an entity, the government’s presence on the entity’s board of directors, the government’s control over the entity’s activities, the entity’s pursuit of governmental policies or interests, and whether the entity is created by statute. See China Light-Walled RPT I&D Memo, supra note 72, at 27. The USDOC refused to apply the five-factor test to Chinese SOEs on the grounds that the Chinese government failed to provide sufficient information on factors other than the government ownership of the SOEs. Id. at 28–30.

100China CWP I&D Memo, supra note 83.

101When the U.S. government took over General Motors at the depth of the 2008–2009 financial crisis, it issued guidelines stating that it would manage General Motors “‘in a hands-off, commercial manner’ and not get involved in issuing day-to-day directives to GM.” Neil King Jr. et al., Potential Conflicts Abound in Government Role, Wall St. J. (June 1, 2009, 11:59 PM), http://www.wsj.com/articles/SB124381322189670529.

102See OECD, OECD Guidelines on Corporate Governance of State-Owned Enterprises 20 (2015) (recommending that “the legal and regulatory framework for SOEs . . . ensure a level playing field and fair competition in the marketplace when SOEs undertake economic activities”); OECD, Competitive Neutrality: Maintaining a Level Playing Field Between Public and Private Businesses 29–82 (2012) (advocating competitive neutrality between SOEs and private businesses).

103DS379 AB Report, supra note 75, at ¶ 346.

104Id. at ¶ 443.

105DS437 AB Report, supra note 87, at ¶ 4.95.

106See supra note 72 and accompanying text.

107DS379 Panel Report, supra note 73, at ¶ 8.94.

108Id. at ¶ 8.134.

109Id. (emphasis omitted).

110Id.

111Id. at ¶ 8.136.

112DS379 AB Report, supra note 75, at ¶ 318.

113Id. at ¶ 318.

114Id. at ¶ 320.

115Presumably, this possibility underlies the need to differentiate the state’s governmental conduct from its market conduct in the first place.

116Again, such market-oriented management is actively promoted as a best practice for state-owned enterprises. See supra note 102 and accompanying text.

117See Milhaupt & Zheng, supra note 70, at 685.

118Note that almost all of the market-versus-state issues in international trade law arise in the context of China. See supra Sections I.B–C.

119See Marshall W. Meyer & Changqi Wu, Paulson Inst., Making Ownership Matter: Prospects for China’s Mixed Ownership Economy 9 (2014).

120See Milhaupt & Zheng, supra note 70, at 685.

121Id. at 676–83.

122Id. at 669.

123See Reeves, Inc. v. Stake, 447 U.S. 429 (1980).

124Id. at 452–53 (Powell, J., dissenting).

125Id. at 452.

126Id.at 452–53.

127Justice Powell would distinguish the scenario in which the state enters the private market to supply its own needs from the scenario in which the state enters the private market for the advantage of its private citizens. In the former scenario, the state may “act without regard to the private marketplace and remove itself from the reach of the Commerce Clause.” Id. at 450. But in the latter scenario, the state “may not evade the constitutional policy against economic Balkanization.” Id.

128Id. at 449.

129Id. (alteration in original) (internal quotations and citations omitted).

130United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 344 (2007) (alteration in original).

131Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 341 (2008). In a footnote, Justice Souter further argued that the point of the function-based test was “not to draw fine distinctions among governmental functions, but to find out whether the preference was for the benefit of a government fulfilling governmental obligations or for the benefit of private interests.” Id. at 341 n.9. This reasoning, however, is rather circular, as whether a preference is for the benefit of the government obviously depends on the classification of the governmental function in question.

132Id. at 341–42.

133See 460 U.S. 150 (1983).

134Id. at 153–54.

135Id. at 154. The Court held that the defendants’ acts at issue, the retail sale of pharmaceutical drugs, was not “‘indisputably’ an attribute of state sovereignty.” Id. at 154 n.6.

136See supra notes 112–14 and accompanying text.

137DS379 AB Report, supra note 75, at ¶ 345.

138DS437 Panel Report, supra note 13, at ¶ 7.66.

139United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 368–69 (2007) (Alito, J., dissenting) (internal quotations omitted).

140Id. at 369.

141Historical Background and Development of Social Security, Soc. Security Admin., https://www.ssa.gov/history/briefhistory3.html (last visited Sept. 1, 2017).

142San Antonio Metro. Transit Auth. v. Donovan, 557 F. Supp. 445, 453 (W.D. Tex. 1983).

143Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 531 (1985).

144Id. at 543–44.

145Id. at 545.

146Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 815–16 (1976) (Stevens, J., concurring).

147Id. at 815. The subsidy program Maryland implemented in 1969 initially included subsidies for scrapping plants in Virginia and Pennsylvania. Id.

148Id. at 809 n.18 (majority opinion). Indeed, in his dissent, Justice Brennan presented evidence from the record that interstate commerce in abandoned automobiles existed prior to the Maryland subsidy program. Id. at 824 n.6 (Brennan, J., dissenting).

149Id. at 809 n.18 (majority opinion) (“We would hesitate to hold that the Commerce Clause forbids state action reducing or eliminating a flow of commerce dependent for its existence upon state subsidy instead of private market forces.”).

150See Laurence H. Tribe, Constitutional Choices 146 (1985).

151Id.

152 See DS412/DS426 AB Report, supra note 14, at ¶¶ 1.6–1.7.

153Id. at ¶ 5.245. The Appellate Body determined that it was unable to complete its analysis because the record from the panel proceeding below contained insufficient factual findings and uncontested evidence that would allow it to draw a conclusion. See id.

154Id. at ¶ 5.178.

155Id. at ¶ 5.188 (emphasis omitted). If the comparison is between rates offered to renewable energy producers and rates in the wholesale electricity market, a benefit will necessarily exist because the former are higher than the latter. See Rajib Pal, Has the Appellate Body’s Decision in Canada—Renewable Energy/Canada—Feed-in Tariff Program Opened the Door for Production Subsidies?, 17 J. Int’l Econ. L. 125, 126 (2014).

156DS412/DS426 AB Report, supra note 14, at ¶ 5.188.

157Id.

158See CTS Corp. v. Dynamics Corp. of Am., 481 U.S. 69, 87 (1987) (“[P]rincipal objects of dormant Commerce Clause scrutiny are statutes that discriminate against interstate commerce.”).

159This norm is embodied in the use of market benchmarks for identifying and measuring subsidies. See supra Section I.C.

160Tribe, supra note 150.

161This is the factual set-up in Reeves, where South Dakota restricted the sale of cement produced by a state-owned cement plant to in-state residents only. See supra note 29 and accompanying text.

162Tribe, supra note 150.

163Prior to the construction of the state-owned cement plant, producers outside of South Dakota were supplying all the cement used in the state and were making substantial profits. Reeves, Inc. v. Stake, 447 U.S. 429, 431 n.1 (1980). Professor Dan Coenen argued that while South Dakota satisfied more buyers’ needs at lower prices by generating an alternative and additional source of supply, the additional supply displaced out-of-state sellers in a pre-existing market. See Coenen, supra note 16, at 410–11.

164In a similar fashion, Professor Coenen asked: “Why . . . does the ‘creation’ of ‘commerce’ necessarily raise a sufficiently powerful equity to justify state discrimination against interstate commerce?” Coenen, supra note 16, at 411.

165See H.P. Hood & Sons, Inc. v. Du Mond, 336 U.S. 525 (1949) (holding that the State of New York was not allowed to deny an out-of-state milk distributor license to purchase milk from within the state for supply to Boston).

166See supra notes 101–02, 115–16 and accompanying text.

167See supra notes 140–42, 161–63 and accompanying text.

168See supra notes 164–65 and accompanying text.

169See Town of Graham v. Karpark Corp., 194 F.2d 616, 620–21 (4th Cir. 1952) (“A city has two classes of powers, the one legislative or governmental, by virtue of which it controls its people as their sovereign, the other proprietary or business, by means of which it acts and contracts for the private advantage of the inhabitants of the city and of the city itself.” (quoting Omaha Water Co. v. City of Omaha, 162 F. 225 (8th Cir. 1908))).

170Nachbar, supra note 44, at 70–73 (distinguishing between regulatory control and property rights).

171See Donald H. Regan, The Supreme Court and State Protectionism: Making Sense of the Dormant Commerce Clause, 84 Mich. L. Rev. 1091, 1194 (1986) (describing taxation as one of the more coercive government powers).

172See, e.g., Free v. Miles, 333 F.3d 550, 554 (5th Cir. 2003) (per curiam) (“The limited function of [the common law rule that a prisoner is entitled to credit for time served when he is incarcerated through no fault of his own] is to prevent the government from abusing its coercive power to imprison a person by artificially extending the duration of his sentence through releases and re-incarcerations.”); Flint v. Mullen, 499 F.2d 100, 105 (1st Cir. 1974) (Coffin, C.J., dissenting) (“The safeguards which courts have created around the exercise of the Fifth Amendment’s privilege against self-incrimination are essentially bulwarks against abuse of the government’s coercive power.”); Milwaukee Deputy Sheriffs Ass’n v. Clarke, 513 F. Supp. 2d 1014, 1022 (E.D. Wis. 2007) (holding that the actions of the county sheriff in inviting representatives of a religious organization to speak at department leadership conference and roll calls, which police deputies were required to attend, was an unconstitutional endorsement of religion through use of coercive power of government).

173In constitutional law, for example, a private actor may be considered to act as an instrument or agent of the government for purposes of Fourth and Fifth Amendments if “‘the government exercise[s] such coercive power or such significant encouragement that it is responsible’ for the [private actor’s] conduct.” See United States v. Garlock, 19 F.3d 441, 443 (8th Cir. 1994) (quoting Fidelity Fin. Corp. v. Fed. Home Loan Bank, 792 F.2d 1432, 1435 (9th Cir. 1986)).

174This is the factual setup presented in White v. Massachusetts Council of Construction Employers, Inc., in which the city of Boston required all city-funded construction projects “be performed by a work force consisting of at least half bona fide [city] residents.” 460 U.S. 204, 205–06 (1983).

175Even if the government is just one of many employers on the market, the government’s willingness to reject noncompliant contractors is still coercive with respect to the specific government projects in question.

176Regan, supra note 171. Regan put forward this argument as one of the differential treatments of spending versus regulation or tax under the dormant Commerce Clause. See id.

177Some may argue that market coercion is not coercion in the true sense because one always has the choice of foregoing the market opportunity and thus not being coerced. But as Professor Coenen argues, the same can be said of government coercion, as one always has the choice of not doing business in a specific jurisdiction. See Coenen, supra note 16, at 415.

178By definition, private actors could only exercise market coercion by virtue of their property rights in economic resources, as they are not sovereigns and lack the ability to exercise regulatory power.

179See Treg A. Julander, State Resident Preference Statutes and the Market Participant Exception to the Dormant Commerce Clause, 24 Whittier L. Rev. 541, 558 (2002).

180See Stephanie Landry, Comment, State Immunity from the Dormant Commerce Clause: Extension of the Market-Participant Doctrine from State Purchase and Sale of Goods and Services to Natural Resources, 25 Nat. Resources J. 515, 523 (1985).

181Reeves, Inc. v. Stake, 447 U.S. 429, 453 (1980) (Powell, J., dissenting).

182475 U.S. 282, 289 (1986) (quoting Gould, Inc. v. Wis. Dep’t of Indus., Labor and Human Relations, 750 F.2d 608, 614 (7th Cir. 1984)).

183For the government to be considered to be acting in a private capacity under this private actor analogy, the government could not respond to concerns or factors that would not be taken into account by private actors. See supra note 183 and accompanying text.

184As a social planner, the government in conducting its affairs necessarily considers factors that go beyond the realms of consideration of the individual members of the society. See Jean Hindriks & Gareth D. Myles, Intermediate Public Economics 425–26 (2d ed. 2013) (describing the difference between the social planner’s welfare function and the welfare functions of individual members of the society).

185This per se rule is not unlike the ownership- and control-based tests, which treat conduct by entities owned or controlled by the government as governmental in nature. See supra Sections II.A–B.

186See Reeves, Inc. v. Stake, 447 U.S. 429, 432–33 (1980).

187A private actor’s power to choose the party with which to deal is subject to an important limitation under antitrust law. When the private actor is a dominant firm, it can exercise the right to refuse to deal with a competitor only if there are legitimate competitive reasons for the refusal. See Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985).

188See Wis. Dep’t of Indus., Labor and Human Relations v. Gould, 475 U.S. 282, 283–84 (1986).

189See id. at 287.

190See Hughes v. Alexandria Scrap Corp., 426 U.S. 794 (1976).

191In Davis, Justice Souter stated that “in Alexandria Scrap, Maryland employed the tools of regulation to invigorate its participation in the market for automobile hulks.” Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 346 (2008). Justice Souter was mistaken in this conclusion, as Maryland was imposing stricter documentation requirements for out-of-state businesses in connection with its subsidy spending. The power exercised by Maryland was not regulatory in nature, as a private actor looking to purchase products at higher-than-market prices also has the power to impose stricter documentation requirements on potential sellers.

192Failure to recognize this possibility is the main problem with the ownership- and control-based tests. See supra Sections II.A–B.

193By contrast, the standard for the function-based test is ever-changing depending on the prevailing conceptions of the government’s functions. See supra Section II.C.

194Unlike the role-based test, which is based on an implicit value judgment that market creation trumps the statutory goals in question, the power-based test does not judge the relative values of the different kinds of government power. For discussions of the value dependency of the role-based test, see supra Section II.D.

195See supra section I.A.1.

196Wells & Hellerstein, supra note 16, at 1125.

197For further discussions of the purpose of the Commerce Clause, see infra Section IV.A.

198John F. Hart, “Sovereign” State Policy and State Action Antitrust Immunity, 56 Fordham L. Rev. 535, 536 (1988).

199For more discussions of the purpose of the state-action immunity in antitrust law, see infra Section IV.B.

200See infra Section IV.C.

201See Wells & Hellerstein, supra note 16, at 1125–26.

202See, e.g., Karl Manheim, New-Age Federalism and the Market Participant Doctrine, 22 Ariz. St. L.J. 559, 582–601 (1990); Jonathan D. Varat, State “Citizenship” and Interstate Equality, 48 U. Chi. L. Rev. 487, 505–07 (1981).

203See Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 803 (1976).

204Id. at 805 (“This line of reasoning is not without force if its basic premise is accepted. That premise is that every action by a State that has the effect of reducing in some manner the flow of goods in interstate commerce is potentially an impermissible burden.”).

205Id. at 806.

206Reeves, Inc. v. Stake, 447 U.S. 429, 436–37 (1980) (internal citation omitted).

207Alexandria Scrap, 426 U.S. at 797.

208Id. at 798.

209Although a private actor would not want to give out money for free to encourage the recycling of abandoned automobiles, it does have the power to do so and to impose documentation requirements as a condition.

210See Reeves, 447 U.S. at 432–33.

211See White v. Mass. Council of Constr. Emp’rs, Inc., 460 U.S. 204 (1983). The City of Boston would have exercised the power of government coercion had it required all businesses, not just businesses working on city-funded projects, to hire at least 50% of their labor force from local residents.

212South-Central Timber Dev., Inc. v. Wunnicke, 467 U.S. 82, 84 (1984).

213Id. at 99.

214Id. at 97.

215In United Haulers, for example, the government imposed restrictions with regulatory effects in the market in which it was a participant—the waste transfer and processing market. See United Haulers Ass’n v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 334 (2007).

216White, 460 U.S. at 205–06. The city was a participant in the market for construction projects, not in the labor market for workers who would work on those projects.

217Wunnicke, 467 U.S. at 84.

218See, e.g., id. at 93 (plurality opinion) (“Our cases make clear that if a State is acting as a market participant, rather than as a market regulator, the dormant Commerce Clause places no limitation on its activities.”).

219See, e.g., New Energy Co. of Ind. v. Limbach, 486 U.S. 269, 277 (1988) (“The market-participant doctrine has no application here. The Ohio action ultimately at issue is neither its purchase nor its sale of ethanol, but its assessment and computation of taxes—a primeval governmental activity.”).

220Recall that Alexandria Scrap involved a state subsidy program. See supra note 27 and accompanying text.

221Government spending on economic and social programs accounts for a large percentage of the government’s activities. The U.S. federal government, for example, spends almost half of its budget on the so-called entitlement programs, including Social Security, Medicaid, and Medicare. See Romina Boccia, Heritage Found., Federal Spending by the Numbers, 2014: Government Spending Trends in Graphics, Tables, and Key Points (Including 51 Examples of Government Waste) (2014). One economist has referred to government spending in the form of subsidies one of the “three major sets of measures” by which the government “alter[s] the distribution of income.” Milton H. Spencer, Contemporary Economics 72 (2d ed. 1974).

222See Coenen, supra note 16, at 479–81; Collins, supra note 25, at 98–103; Enrich, supra note 16, at 442–43; Gergen, supra note 16, at 1134–53; Hellerstein & Coenen, supra note 16, at 846–48; Paul S. Kline, Publicly-Owned Landfills and Local Preferences: A Study of the Market Participant Doctrine, 96 Dick. L. Rev. 331, 370–91 (1992); Regan, supra note 171, at 1193–95.

223See Regan, supra note 171; see also Hellerstein & Coenen, supra note 16, at 846–47.

224Saul Levmore, Interstate Exploitation and Judicial Intervention, 69 Va. L. Rev. 563, 585 (1983); see also Coenen, supra note 16, at 479; Collins, supra note 25, at 102–03; Enrich, supra note 16, at 442–43; Kline, supra note 222, at 374.

225See Dan T. Coenen, Business Subsidies and the Dormant Commerce Clause, 107 Yale L.J. 965, 980–81 (1998); Coenen, supra note 16, at 480; Gergen, supra note 16, at 1136–37.

226Commentators have questioned some of those justifications. For instance, Coenen argues that it is debatable whether subsidy programs are inherently more expensive than discriminatory state regulatory and tax programs. See Coenen, supra note 16, at 434.

227These justifications either do not address the market-participation question at all, or just circle back to the argument that taxation is inherently governmental. See Enrich, supra note 16, at 442.

228New Energy Co. of Ind. v. Limbach, 486 U.S. 269, 271, 277 (1988).

229Id. at 277.

230520 U.S. 564, 593 (1997).

231There is a potential fault line in the Court’s treatment of taxation cases. While the Court in Limbach and Camps Newfound/Owatonna struck down the discriminatory tax credit or exemption at issue as violations of the dormant Commerce Clause, it upheld, in Davis, a Kentucky tax scheme in which the state exempted interest income on state municipal bonds, but not interest income on out-of-state municipal bonds, from state income taxes. See Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 333–34, 341–42 (2008). The nature of the power exercised by the government in Davis, however, was arguably indistinguishable from the nature of power exercised by the government in Limbach and Camps Newfound/Owatonna. So although Kentucky could be said to be participating in the market for municipal bonds by making in-state municipal bonds more attractive, the fact that it was exercising the power of government coercion makes it clear that it was acting in a governmental capacity. The power-based test, therefore, would squarely place the state action in Davis subject to the dormant Commerce Clause.To the extent that Davis needs to be distinguished from Limbach and Camp Newfounds/Owatonna, such distinctions need to come from elsewhere. One likely basis on which Davis could be distinguished from Limbach and Camp Newfounds/Owatonna is that in Davis, the ultimate beneficiary of the state action was the state itself, while in Limbach and Camp Newfounds/Owatonna, the ultimate beneficiaries of the state actions were individuals or entities other than the state. In Davis, the ultimate beneficiary of the state action was the state itself because the discriminatory tax exemption made it easier for the political subdivisions of the state to issue bonds. In Limbach, the ultimate beneficiaries of the state action were all those who would benefit from the local production of ethanol—farmers, distributors, and suppliers, etc. In Camp Newfounds/Owatonna, the ultimate beneficiaries of the state action were all residents who would stand to benefit from the services provided by the charities eligible for the tax exemption. It could be argued that when the ultimate beneficiary of the state action is the state itself, it does not raise the same level of concerns about discrimination as when the ultimate beneficiary of the state action is someone other than the state. Note that this distinction is not based on the government ownership or control of the ultimate beneficiary of the state action and thus can avoid the problems associated with the ownership- and control-based tests. For discussions of the ownership- and control-based tests, see supra Sections II.A–B.

232A private business, for example, could decide to subsidize a particular group of customers using revenues collected from other customers. Turbotax, the popular tax software, charges $0 for customers who file the most basic tax returns but $54.99–$114.99 for customers who file more complicated returns. See Compare Turbotax Online Products, TurboTax, https://turbotax.intuit.com/personal-taxes/compare/online/ (last visited Aug. 23, 2017).

233512 U.S. 186, 188 (1994).

234Id. at 198–202.

235Id. at 200–01.

236The coupling of governmental and market power of spending money takes West Lynn out of the realm of the market-participant exception. The Court in West Lynn could have brought more consistency to the case law by explicitly noting that the market-participant exception did not apply, rather than being silent about the market-participant exception at all. For discussions of the Court’s inconsistency in using the market-participant exception as the framework for analyzing dormant Commerce Clause issues, see supra Section II.A.1.

237For discussions of C & A Carbone and United Haulers, see supra Section II.A.1.

238In Department of Revenue of Kentucky v. Davis, Justice Souter suggested that United Haulers “may also be seen under the broader rubric of the market participant doctrine.” 553 U.S. 328, 343 (2008) (plurality opinion). According to Justice Souter, the fact that the state simultaneously exercises regulatory power while participating in market activities does not make the market-participant rule inapposite; the dispositive factor is the government’s market activities. Id. at 346 (“Not only did the public authority acting in [United Haulers] process trash, but its governmental superiors forbade trash haulers to deal with any other processors. This latter fact did not determine the outcome, however; the dispositive fact was the government’s own activity in processing trash.”). Justice Souter would have upheld the flow-control ordinances at issue in United Haulers under the market-participant exception because the counties in that case participated in the market for trash processing. Id. at 346–47 (“We upheld the government’s decision to shut down the old market for trash processing only because it created a new one all by itself, and thereby became a participant in a market with just one supplier of a necessary service.”). Justice Souter also characterized two other market-participant cases, Alexandria Scrap and White, as examples of the government’s commercial activities being joined by its regulatory efforts. Id. at 345–46. But obviously, Justice Souter’s expansive view of the market-participant exception was not shared by the rest of the majority.

239See supra notes 95–98 and accompanying text.

240471 U.S. 34, 36–37 (1985).

241568 U.S. 216, 222 (2013).

242135 S. Ct. 1101, 1108–09 (2015).

243Whether the board was exercising the power of government coercion is a question separate from whether the board was a governmental body. Even if the board was a private body, that distinction does not prevent the conclusion that it was exercising governmental power.

244See supra note 62 and accompanying text.

245In Phoebe Putney, the Court denied state-action immunity to the defendant on the basis that there was not a sufficient articulation of state policy to authorize the anticompetitive conduct. 568 U.S. at 228.

246See Brief for National Federation of Independent Business as Amicus Curiae in Support of the Petitioner at 6–24, Phoebe Putney, 568 U.S. 216 (No. 11-1160). The amicus curiae argued that applying state-action immunity to the state’s commercial conduct would exceed the state-action immunity doctrine’s purpose because “[m]arket-participant state conduct is not an ‘integral operation in an area of traditional government functions.’” Id. at 15 (quoting City of Lafayette v. La. Power & Light Co., 435 U.S. 389, 424 (1978) (Burger, C.J., concurring)). This argument resembles the function-based test that distinguishes between the regulatory and market roles of the state based on whether the function at issue is a traditional government function. See supra Section II.C.

247See Phoebe Putney, 568 U.S. at 226 n.4.

248Jefferson Cty. Pharm. Ass’n v. Abbott Labs., 460 U.S. 150 (1983).

249Id. at 151–52. The Robinson-Patman Act provides: “It shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality . . . where the effect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them . . . .” Id. at 152 n.1 (quoting 15 U.S.C. § 13(a) (1982)).

250Id. at 153.

251Id. at 154.

252Id. at 157. The Court held that the term “person” used in the Robinson-Patman Act is sufficiently broad to cover governmental bodies. Id. at 155.

253The Court concluded that “the plain language of the [Robinson-Patman] Act strongly suggests that there is no exemption for state purchases to compete with private enterprise.” Id. at 156–57. The Court then found that “[t]he legislative history [of the Robinson-Patman Act] falls far short of supporting respondents’ contention that there is an exemption for state purchases of ‘commodities’ for ‘resale.’” Id. at 159.

254435 U.S. 389 (1978).

255Id. at 392. The counterclaims alleged that, among other things, one of the defendant cities contracted to provide gas and water service to the plaintiff’s electric customers only on the condition that the customers purchase electricity from the city. Id. at 403–04.

256Id. at 403.

257Id.

258Id. at 413.

259Id. at 416. The Court agreed with the lower court that further inquiries should be made to determine whether the defendants’ actions were directed by the state. See id. at 413–15.

260Recall that the defendant cities’ actions at issue in City of Lafayette were commercial in nature. See supra note 255 and accompanying text.

261See, e.g., Robert P. Inman & Daniel L. Rubinfeld, Making Sense of the Antitrust State-Action Doctrine: Balancing Political Participation and Economic Efficiency in Regulatory Federalism, 75 Tex. L. Rev. 1203, 1250–71 (1997) (examining the antitrust state-action doctrine from the perspective of federalism); Jorde, supra note 16, at 230 (arguing that the Parker Court grounded the state action doctrine and its own judicial restraint on the structure of economic federalism); 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, 1105 (1981) (“[T]he Parker Court did not base its construction of the Sherman Act on legislative history, but instead derived a presumption of congressional intent from the concept of federalism.”).

262Jorde, supra note 16, at 230–31.

263Parker v. Brown, 317 U.S. 341, 351 (1943).

264For instance, it could be argued that when citizens of a state authorize an SOE to engage in anticompetitive conduct, they are participating in government to the same degree as when they impose coercive requirements. It could also be argued that they are addressing the special needs of the local population, thereby promoting government efficiency. And such participation equally promotes policy experimentation and diffusion of power.

265See Coenen, supra note 16, at 427.

266Arie Reich, From Diplomacy to Law: The Juridicization of International Trade Relations, 17 Nw. J. Int’l L. & Bus. 775, 781 (1997).

267Id. at 780; see also Jan Hoogmartens, EC Trade Law Following China’s Accession to the WTO 10 (2004).

268Tariffs are akin to taxes in that their imposition requires the power of government coercion, as no private actors possess the power to collect money without providing goods or services of equivalent values.

269Average tariffs in industrial countries plummeted from 40% in 1947, when the GATT entered into force, to 6.3% in 1994 just prior to the entry into force of the Uruguay Round agreements, and further to 3.9% as a result of the Uruguay Round agreements that led to the establishment of the WTO. See Raj Bhala, Rethinking Antidumping Law, 29 Geo. Wash. J. Int’l L. & Econ. 1, 3 (1995).

270GATT 1994, supra note 66, at art. XVII.

271Article XVII of GATT 1994 requires: “[S]uch enterprises shall, having due regard to the other provisions of this Agreement, make any such purchases or sales solely in accordance with commercial considerations, including price, quality, availability, marketability, transportation and other conditions of purchase or sale, and shall afford the enterprises of the other contracting parties adequate opportunity, in accordance with customary business practice, to compete for participation in such purchases or sales.” Id. at art. XVII:1(b).

272Id. at art. XVII:3.

273See supra note 232 and accompanying text.

274For a discussion of the purpose of the dormant Commerce Clause, see supra Section IV.A.

275The only exception is when the granting of subsidies involves the power of government coercion, in which case the subsidy-granting action will be made subject to the discipline of the state’s regulatory actions. See, e.g., West Lynn Creamery, Inc. v. Healy, 512 U.S. 186 (1994).

276For a background discussion on the “public body” issue, see supra Section I.B.3.

277See supra notes 67–68 and accompanying text.

278See supra Sections II.A–C.

279DS437 Panel Report, supra note 13, at ¶ 7.66.

280For more discussion of the problems with the function-based definition of public bodies, see supra Section II.C.

281In DS437, China argued that “[a] public body, like government in the narrow sense, thus must itself possess the authority to ‘regulate, control, supervise or restrain’ the conduct of others.” DS437 Panel Report, supra note 13, at ¶ 7.67 (alteration in original). The panel, however, rejected this argument on textual grounds. Id. at ¶ 7.68.