Note from the Editor: The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. We welcome responses to the views presented here. To join the debate, please email us at [email protected].

In early April, a California legislative committee convened to consider AB 288, a bill that would transfer much responsibility over labor relations from the federal government to the state. The bill’s sponsor, Tina McKinnor, justified the measure as a response to the Trump administration’s “attack” on workers’ rights. McKinnor accused the administration of taking a “wrecking ball” to the federal government’s administrative apparatus, including the agency responsible for overseeing national labor policy, the National Labor Relations Board. Among other things, she said, the administration had fired one of the Board’s three remaining members, leaving the agency without a quorum. Without a functioning Board, workers had no one to protect their statutory labor rights. And if the federal government wouldn’t protect their rights, California would.

The only problem with that argument was that McKinnor’s bill was probably illegal. Among the other witnesses lawmakers heard that day was a representative from the California Chamber of Commerce, who reminded the committee that federal labor law is generally exclusive. The National Labor Relations Act (NLRA) governs most private-sector labor relations. And when the NLRA applies, it displaces all state policy. That is, the NLRA makes labor relations an exclusively federal concern. California could no more adopt its own labor-relations regime than it could send its own peace delegation to Ukraine.

That detail, however, didn’t stop the committee. The committee voted unanimously to move the bill along. In fact, to date, the bill has drawn no opposition in the legislature and has even collected bipartisan support. It seems well on its way to passage. And if signed, it will join a swelling number of state laws that have chipped away at the once exclusive federal labor regime.

The proponents of these new laws, almost all from pro-union blue states, think they can get around the NLRA by taking advantage of the Board’s lack of a quorum. But they should be careful what they wish for. They may inadvertently trigger a new era of competitive labor relations. And if history is any guide, they aren’t likely to be happy with the results.

One Labor Policy to Rule Them All

For a generation, it has been understood that the NLRA occupies the labor-relations field. In 1959’s San Diego Building Trades Council v. Garmon, the U.S. Supreme Court held that the NLRA created a uniform system of labor relations for the whole country. That system was managed by a single federal agency, the Board. And states had no right to set up alternative systems for themselves. Any alternative system would frustrate the law’s uniformity by balkanizing labor policy. So if an issue was even “arguably” regulated by the NLRA, it was generally outside states’ power to regulate.

Though well established, Garmon’s holding has come under increasing pressure. As union membership has steadily dwindled, labor unions have cast about for someone to blame. One culprit has been the NLRA itself, which unions say systematically discourages union membership. It requires unions to organize workplace by workplace, engaging in metaphorical block warfare against individual employers. Employers pull out all the stops to prevent organizing, which drives up unions’ organizing costs. And even when unions succeed in organizing an employer, that employer is bogged down in the market by more expensive union wages. The employer inevitably goes out of business, and the cycle starts all over again. The result is a national union density in the private sector lower than when the NLRA was passed.

The Laboratories of Labor Policy

Unions have therefore pushed state lawmakers to test the NLRA’s limits. For example, they’ve lobbied blue states to require “labor peace” agreements in select industries, like the cannabis industry. They have also lobbied states to ban “captive audience” meetings—a particularly effective tool employers use to get their message out during union campaigns. And most recently, with California’s AB 288, they have pushed lawmakers to transfer much administrative authority to state officials. If passed, AB 288 would give a California agency authority to conduct union elections, decide unfair labor practices, and police collective bargaining. It would, in effect, transform labor policy into a state-level concern.

Under the traditional view, all of these laws would be preempted. They all deal with matters regulated by the NLRA and overseen by the Board. They involve union organizing, the treatment of union supporters, and bargaining with union representatives. They are not just “arguably” under the Board’s jurisdiction; they sit at the core of Board’s administrative responsibility.

Yet states have begun to argue that the traditional view—the one set out in Garmon—is perhaps wrong. For one, the NLRA itself doesn’t mention preemption. So maybe we’ve been reading the statute wrong all this time. Besides, even if we have been reading it correctly, Garmon presupposed a functioning Board. But today, the Board cannot function. It currently lacks a quorum, which leaves it legally inert. And in fact, even when the Board had a quorum, it had increasingly shown itself too slow and too weak to protect workers’ rights. So Garmon’s premises have been eroded on the ground, leaving a space for the states to step in.

A Race to the Bottom—and to the Top

That view is almost certainly wrong. To start, it treats preemption doctrine like a light switch, toggling on or off depending on how many seats are filled at the Board. But preemption doesn’t work like that; it displaces state policy regardless of the operative status of one federal agency at one moment in time. Nor does preemption depend on whether a state agrees with federal law. States cannot opt out of a federal law just because they think the law doesn’t go far enough. If they could, preemption would be effectively meaningless: states could adopt their own policies whenever they disagreed with the federal solution.

Assume, however, that these states have it right. Let’s say that a state could opt out of the NLRA whenever it thought the Board was failing to protect worker rights. Even if true, that principle would not solve organized labor’s woes. To the contrary, it would probably make them worse.

The reason is competitive federalism. Competitive federalism means that when states have leeway to set their own policies, they can compete with one another for business. A state can experiment with taxes, regulations, and subsidies to attract new residents and investments. The state that finds the best policy mix becomes a magnet for people and companies. The state with the worst mix becomes a desert.

Today, competitive federalism is often tempered by federal law, which sets minimum standards in some policy arenas. States can’t fully compete in those arenas because the most important rules are set at the national level. One such arena used to be labor policy. But if California is right that states can adopt their own labor rules, then the doors to competition are now open. And they are open not only for pro-labor blue states, but also for more conservative states with dimmer views of organized labor.

It’s not difficult to predict how the resulting competition would play out. We already have decades of evidence under “right to work” laws. Whether to adopt right-to-work laws is one of the few choices left to states under the NLRA. The NLRA explicitly allows states to decide whether to let unions to bargain for “agency” fees. Agency fees are the fees that non-member employees pay to a union for the union’s bargaining services. Those fees are controversial, and today, more than half the states have banned them.

The results of those bans are unmistakable. Multiple studies have shown that right-to-work laws not only boost overall employment, but also increase real wages. By one count, employment growth in right-to-work states has been more than twice as fast as in their non-right-to-work counterparts. Real wages have been higher too—as much as $2,900 per person in 2023. And the share of manufacturing employment has been significantly higher—by some estimates, nearly 30%.

More broadly, heavily unionized states have performed poorly in the national competition for people. In 2024, the states with the highest union densities were Hawaii (26.5%), New York (20.6%), Alaska (19.5%), and California (16.3%). But the same year, those states lost more residents than any others. Among the biggest gainers of residents were the states with the lowest union densities—North Carolina (2.4%) and South Carolina (2.7%). In fact, the five states with the biggest population increases from domestic migration all had below-average unionization rates. And all but Delaware had a right-to-work law.

Those numbers are daunting for the supporters of independent state labor policies. These supporters tend to be pro-labor advocates who want to strengthen union rights. But the numbers say that pro-union laws chase jobs into other jurisdictions. That leakage is already happening; businesses have been fleeing New York and California for years. And those states are unlikely to stem the tide by making their laws even less friendly to businesses.

Indeed, other states will no doubt seize the same opportunity to make their own laws even less union friendly. If businesses were already eager to decamp for Texas, one can only imagine their excitement when Texas bans card checks and dues checkoffs. The lesson for pro-union states is clear: be careful what you wish for.