This post surveys some recent developments in what might be termed the politics of federal labor preemption. In a recently decided case, Michigan Building and Trades Council v. Snyder, the Sixth Circuit held that a Michigan law prohibiting governmental entities from signing Project Labor Agreements with unions was not preempted by the National Labor Relations Act (“NLRA”) because in passing the law Michigan was acting as a market participant, not a regulator. This decision is notable for two reasons: first, it significantly expands the market participant exception to federal labor preemption, as developed by the Supreme Court, to encompass broad legislation containing no temporal or project-specific limitations; second, it heralds a potential increase in the use of the market participant exception—a doctrine mostly developed through cases involving pro-union local and state governments acting as market participants in ways that ultimately supported unions—to further anti-union local and state policy agendas. The coincidence of these doctrinal and political developments is particularly striking.

The post is split into three parts. The first part summarizes the foundations of federal labor preemption and the market participant exception. The second part discusses Michigan Building and Trades Council v. Snyder. Finally, the third part explores some potential implications of Snyder.

Background on NLRA Preemption and the Market Participant Exception

The National Labor Relations Act contains one of the most expansive preemption regimes in American law. While the statute contains no preemption clause, the Supreme Court has read into the Act a Congressional intent to entrust administration of the labor policy for the Nation to a centralized administrative agency, the National Labor Relations Board, foreclosing state and local intervention into the rules of union organizing and bargaining. The Court has carved out an important exception to its robust preemption doctrines, however, premised on the distinction between government as regulator and government as proprietor. Under the market participant exception, where state and local governments act as private participants in the marketplace, and not as regulators, federal labor preemption does not apply. Preemption only applies when the state or local government action is regulatory in nature.

The distinction between regulatory and proprietary state action was first articulated in the 1993 Supreme Court case Building Trades Council v. Associated Builders and Contractors (“Boston Harbor”). In Boston Harbor, the question was whether a Project Labor Agreement (“PLA”) signed between a state government agency and a union was preempted by the NLRA. The PLA in question arose subsequent to a court order to clean up pollution in Boston harbor without delay. The state agency tasked with overseeing the project determined that maintaining labor-management peace was essential to ensuring that the project would not be disrupted by labor disputes. Hence, it negotiated an agreement (the PLA) with a local union which recognized the union as the bargaining agent for project employees, required new employees to join the union, prohibited the union from striking for 10 years, and required all contractors and subcontractors who might bid on the project to agree to be bound by the PLA.

The Boston Harbor Court held that the PLA was not preempted because preemption doctrines apply only to state regulation, while in this case the state acted as a proprietor. The NLRA was intended to supplant only state labor regulation, not all legitimate state activity that affects labor, the Court reasoned. In determining that signing the PLA was a proprietary state action, and not an attempt at regulation, the Court credited the state’s claim that it was attempting to ensure an efficient project that would be completed as quickly and effectively as possible at the lowest cost. The Court found this claim credible in part because the PLA was narrowly tailored to one particular job, and therefore did not extend beyond the state’s proprietary interest in cleaning up Boston harbor. According to the Court, in signing the PLA the state acted just like a private contractor, with the same incentives. The state was participating in the market, not regulating it.

The Supreme Court revisited the distinction between government as regulator and government as proprietor in the 2008 case Chamber of Commerce v. Brown. Brown concerned a California statute that prohibited employers from using state funds to support or oppose unionization and required employers to keep detailed records proving that any money spent supporting or opposing unionization did not come from the state. The Court, in holding that the law was preempted by the NLRA, concluded that the state enacted the law in its capacity as a regulator rather than a market participant. Citing Boston Harbor, the Court noted that the law was not specifically tailored to one particular job. Moreover, the express purpose of the law was to further a labor policy, not to advance a proprietary interest. While the state had a legitimate interest in ensuring state funds are spent appropriately, the law was not tailored to fit this interest, in part because it contained a number of exceptions permitting the use of state funds for select employer advocacy programs that promote unions.

In both Boston Harbor and Brown, the Supreme Court developed the market participant exception by distinguishing between regulatory and proprietary state action. In both cases, a key factor in making this distinction was whether the state action was specifically tailored to one particular project. Moreover, where the state claimed a proprietary interest, it had to show that its actions fit the interest. If the state action deviated from the purported proprietary interest, then the Court inferred a regulatory intent susceptible to federal labor preemption.

Michigan Building and Trades Council v. Snyder

The Michigan legislature passed a law in 2012 prohibiting all government units within the state from entering Project Labor Agreements of the sort used in the Boston Harbor case. The law is particularly broad, affecting all projects developed by every governmental unit in the state. The legislature declared the purpose of the law to be the preservation of fair and open competition. A number of labor organizations challenged the law, arguing in part that it is too broad to be proprietary because it does not consider projects on a case-by-case basis.

In September 2013, the Sixth Circuit upheld the law, holding in Michigan Building and Trades Council v. Snyder that Michigan’s statute advances the proprietary interest of efficient use of resources and is limited enough to advance that interest; it is therefore not regulatory, and not preempted by the NLRA. The Sixth Circuit credited the statute’s statement of purpose as evidence that the legislation was passed in an effort to improve efficiency in government projects, not to regulate. Furthermore, it concluded that the limits on the act demonstrate its proprietary nature. The act has no effect on private projects, and does not forbid private parties from entering PLAs, even on public projects. It just forbids governmental units from entering PLAs.

Responding to the objection raised by labor organizations that the Michigan statute is too broad to be proprietary because it does not consider projects on a case-by-case basis, the Snyder court pointed out that private proprietors often act on an across-the-board basis without becoming regulators—the legislature here just decided to take the PLA decision out of the hands of the government and leave it to private contractors. Just because the legislature acted on all projects at once, unlike the agency in Boston Harbor, does not by itself mean that the action was regulatory.

In holding that the Michigan statute is proprietary despite its use of a blanket rule, the Snyder court relied heavily on a 2002 D.C. Circuit case called Building and Construction Trades Department v. Allbaugh, which involved a challenge to President George W. Bush’s ban on federal agencies’ requiring PLAs for federally funded projects. The Allbaugh court held that the President’s ban was proprietary, reasoning that there is no logical justification for holding that a blanket rule establishing a consistent practice for PLAs is regulatory when the only decisions governed by the rule are those made by the government in a proprietary capacity as a market participant. In other words, where the government crafts a blanket rule that only constrains its decision-making as a proprietor, the blanket rule is proprietary, not regulatory.

The Snyder court thus seemed to consider the breadth of the Michigan statute unproblematic for preemption purposes because it treated the actions of the government as proprietor like the actions of private proprietors. Just as private parties can make general rules guiding their conduct without thereby becoming regulators, so can the government make general rules guiding its proprietary conduct without acting as a regulator.

As the Snyder dissent notes, however, the majority’s approach seems to contradict the approach taken by the Supreme Court in Boston Harbor, where the Court argued that private proprietors can in fact “regulate” in the private sphere when they act on the basis of policy concerns rather than a profit motive—but this “regulation” is unlike governmental regulation because the Supremacy Clause does not require pre-emption of private conduct. Hence, for the Boston Harbor Court, “States have a qualitatively different role to play from private parties. When the State acts as a regulator, it performs a role that is characteristically a governmental rather than a private role.” By ignoring this difference, the Snyder majority threatens to collapse the distinction between proprietary and regulatory governmental action.

Implications of Snyder

Since Boston Harbor, a number of states and cities have used the market participant exception to justify a variety of pro-union labor peace ordinances. These ordinances have generally been upheld by courts where they have contained temporal or project-based limits. For example, in N. Ill. Chapter of Associated Builders & Contractors, Inc. v. Lavin, the Seventh Circuit held that an Illinois statute conditioning the grant of construction subsidies on pre-hire agreements that included union security and no-strike clauses was not regulation because the state “limited its condition to the project financed by the subsidy.” Similarly, in Hotel Emps. & Rest. Emps. Union, Local 57 v. Sage Hospitality Res., LLC, the Third Circuit held that a Pittsburgh ordinance conditioning the receipt of tax increment financing (“TIF”) on signing labor agreements was proprietary because the city “limited its condition to hotels and hospitality projects receiving TIF funds.” For a list of the major labor peace ordinances and laws around the country, see pages 13-15 of this report. Overall, the market participant exception has been effectively and extensively used by pro-union local and state governments to pass limited legislation supporting local unions.

Given the prevalence of labor peace ordinances and laws passed by pro-union local and state governments, it is not surprising that anti-union local and state governments have in recent years sought to use the market participant exception to justify ordinances and laws prohibiting labor peace agreements. For example, in May 2013 Georgia passed a law prohibiting all state government entities from signing labor peace ordinances. Similar legislation passed in Tennessee in April 2013.

Snyder is particularly interesting in the context of this trend because the case combines an anti-union political valence with an expansion of the market participant exception to include broad state laws that contain no temporal or project-based limitations, an expansion that seems to stray from the Supreme Court’s analysis in Boston Harbor and Brown. This combination could spread to other circuit courts, given that the recently passed legislation in Georgia (HB 361) and Tennessee (SB 1017) is just as broad as the Michigan law dealt with in Snyder. It is likely that these laws will be challenged on grounds similar to the challenge in Snyder. The courts reviewing these laws will find it particularly difficult to follow Snyder, however, because unlike the Michigan law in Snyder neither statute states a proprietary governmental purpose based on efficiency or other related concepts—both statutes explicitly identify their purpose as the advancement of certain labor policies.