Facebook Twitter Tumblr Close Skip to main content
A Project of The Annenberg Public Policy Center

Cadillac Plans and Unions: Who Benefits?


When the White House and congressional Democrats agreed last month to scale back a Senate-passed tax on high-value health plans, it was widely portrayed as a giveaway to labor unions. For example, the New York Post reported that it was a "sweetheart deal" that would save union members $60 billion, and on its editorial page called it a "bribe" and a "big, fat wet kiss for labor unions," a view quickly echoed by Republican leaders.

Now a new analysis challenges that view with some pertinent facts. It turns out, the Senate-passed tax would affect mainly nonunion workers, and by a wide margin. The study found that under the compromise version of the tax, 17 percent of those affected in 2019 would be union workers and 83 percent would not. It also projects that 71 percent of the dollar savings from the supposed "wet kiss for labor" would actually go to nonunion workers. So the chief beneficiaries of the "sweetheart deal" are not union members at all.

Actually, this should come as no surprise, because union workers now make up such a small portion of the entire workforce. According to the most recent figures from the Bureau of Labor Statistics, only 12.3 percent of rank-and-file wage earners belong to a union. The "sweetheart deal" would have to be tilted to union workers in a truly massive way for them to make up most of the beneficiaries or get most of the dollar benefit. And it isn't.

The Washington Post's story on the new study was headlined " 'Cadillac tax' on health plans would hit union and nonunion jobs equally." But that's not quite right, either. In the body of the story, the reporter — choosing his words more carefully than did the headline writer — said the tax would fall equally on union and nonunion "plans" (not "workers"). He quoted the study as saying the impact of the tax "is roughly in line with the overall breakdown of nonunion vs. union workers with employer-provided plans." (Our emphasis.) That may be true — but union workers are more likely than nonunion workers to have coverage provided by employers. So the only way it can be argued that the tax falls on both kinds of workers "equally" is by leaving out the uninsured and those who buy their own insurance.

The fact is, if this study is right, union workers are somewhat more likely than nonunion workers to be hit by the tax, or to benefit from its reduction — even though in either case many more nonunion than union workers would be affected by it. So it's no wonder union leaders fought to have it reduced. But because union workers are such a small fraction of the workforce these days, they'll still be a small minority of the workers who are affected either by the tax or by any changes in it. So it's also wrong to argue that unionized workers are the sole or chief beneficiaries of a reduction.

Should we believe the study? It comes from the Center for Labor Research and Education at the University of California, Berkeley, and was funded by the liberal think tank Institute for America's Future and by the California Endowment, a foundation that makes grants to provide health care to more people in that state. But one of its two authors is William Dow, who was senior economist for health with George W. Bush’s Council of Economic Advisers, and now is an associate professor for health economics. The authors explain their methodology and sources openly, and — again — the findings are roughly in line with what we'd expect given how few workers belong to unions.

Footnote: The tax's impact would be even less under a new version supported by President Obama. As passed by the Senate, it would have slapped a 40 percent tax on the value of any health care plan exceeding $23,000 for families and $8,500 for individuals. The compromise increased those thresholds to $8,900 and $24,000. But Obama proposed a "reformed" version on Feb. 22 in a plan posted on the White House Web site. The president increased the thresholds further and would also delay implementing the levy until the year 2018. (As we've explained before, the tax would fall on insurers, but would be passed along to workers one way or another.) Obama's proposal would further reduce the number of workers who would be affected, but one of the study's authors, Ken Jacobs, told us that the union/nonunion split of affected workers should be "about the same" as the breakdown in the compromise.