BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Will The Obamacare Cadillac Tax Increase Worker Wages?

Following
This article is more than 8 years old.

Asking whether the Obamacare Cadillac tax will increase worker wages might seem nonsensical on its face. But wait until you hear the answer: yes AND no. Don't worry: Schrodinger's cat hasn't wandered into the thicket of health policy (nor am I channeling my inner two-handed economist).

It's a trick question: the answer depends on whether employers respond to the Cadillac tax by trimming health benefits to avoid the tax, or instead simply keep their health benefits as is and pass the tax onto their employees. For employers who trim health benefits I have a high degree of confidence that on average, worker wages will rise . For employers who absorb the Cadillac tax increase, average worker wages will fall .

Surprisingly, Sarah Kliff over at Vox claims that "the evidence is thin" that worker wages will rise in companies that trim their health benefits. This has me scratching my head in light of the available scientific evidence:

  • I've counted 14 solid scientific studies that have demonstrated a (statistically significant) trade-off between the cost of employer health benefits and worker wages.
  • An additional 8 studies confirm that not only does the offset exist, it is individualized. Thus, workers with higher expected health costs--e.g., those who are older, obese, pregnant--experience a higher offset that is proportional to the added costs their medical bills place on the company health plan.
  • Admittedly, there are 12 studies with either null (statistically insignificant) or mixed results (some analyses showing health benefits are associated with lower wages, while other models using the same data show a positive association between wages and health benefits.
  • And there are 5 studies showing that health benefits are associated with higher rather than lower wages.

Where's the Catch?

To be clear: in some studies, there's nearly a dollar-for-dollar offset, i.e., fully 100% of the added cost of health benefits is passed onto workers in the form of lower cash compensation. In other cases, the offset may be as low as 15 cents on the dollar, meaning that either the employer has implicitly absorbed some of the cost or (more likely, in my view) it has been passed onto the employee in a different form such as a reduction in other non-health fringe benefits or a higher employee-paid premium for health coverage.

Thus, when I say that wages will rise, I'm not claiming there will be a dollar-for-dollar offset, since the weight of scientific evidence doesn't support such a strong claim. And there's good reason for this. After all, there's no reason to suppose that every individual worker's valuation of health benefits precisely corresponds to the employer's cost to provide it. If the worker only values the benefit at 50 cents on the dollar, then in competitive labor markets, the employer would be able to pass along only half the cost of providing that benefit. Why? Such workers would self-select into firms where half or less of health benefits costs are passed onto workers (implying higher cash wages) and would avoid firms where, say 75% or 100% of employer health benefits costs come out of cash wages.

Am I concerned about the studies that have contradicted the wage-benefits offset hypothesis? I am not, for the very same reasons that most economists Ms. Kliff interviewed were unconcerned. The reality is, there are plenty of explanations for the various counterintuitive and non-significant findings. Workers differ in how much they value health benefits and their productivity, for example. We know for sure, for example, that the demand for health insurance steadily rises with income. So that alone would create a positive correlation between wages and health benefits costs. The empirical challenge is to control for all the various factors that might interfere with getting a clear picture of how wages vary in response to changes in employer health costs.

As a rough approximation, I would argue that the most well-designed studies have been the ones that successfully detected the wage-benefits offset that economic theory says ought to be there. Studies that have reached the opposite conclusion have tended to have methodological limitations that place a caveat on their findings. This may sound self-serving, i.e., that I'm prepared to determine which studies are well-designed based on the results they produce.

But don't take my word for it. Austin Frakt at The Incidental Economist has promised to do a thorough investigation of this matter. He's made clear he doesn't have time to do so quickly, but whenever he completes this review, I will add an update to this post. I'm very confident he'll do a very even-handed assessment of this literature that will help illuminate its disparate findings.

Evidence Regarding Reductions in Health Benefits and Wages

To be fair to Ms. Kliff, her "thin" claim was based on the following:

But all this research runs in the wrong direction. The argument around the Cadillac tax isn't about whether high health premiums depress wages. It's about the opposite question: whether lowering health insurance costs will boost wages for Americans — or if companies will just pocket the difference.

And, worryingly, there are no studies that of that dynamic. Zero.

So in the absence of studies, she interviewed a half dozen health economists, most of whom gave her pretty enthusiastic predictions that lower health insurance costs will indeed boost wages. All things considered, "thin" evidence.

Unfortunately, Ms. Kliff is wrong in her claim of zero studies examining the impact of lower health insurance. Two such studies exist.

  • One study by Jonathan Gruber (yes, that Prof. Gruber) found that following the introduction of national health insurance in Canada, there was a 1.4-4.2% increase in average weekly wages; moreover, the impact was higher in industries having a low rate of private coverage provided by employers. This is rather inconclusive evidence in that a) wages actually did rise in industries able to drop private coverage following the introduction of NHI; but b) wages rose less in those industries compared to industries with much less private health insurance; if the wage-benefits trade-off hypothesis were correct, the opposite should have been observed.
  • More relevant is a study of New York state examining what happened after community rating rules were imposed. If small firms were adjusting compensation packages prior to reform to offset higher health care costs of older workers, then community rating should have lead to greater relative wages for older workers post reform and not necessarily induce adverse selection that results in changes in who is insured. Indeed, the study found that relative wages of older workers in small firms increased in comparison with other states and with large firms within New York following reform.

In short, the one U.S. study that has examined this question found indeed that when health benefits costs declined for certain workers, that was passed along to the workers in the form of higher wages.

Yes, Sarah Kliff is right that there's many caveats to this general conclusion. The adjustment isn't likely to be instantaneous (the speed depending upon the competitiveness of local labor markets) and neither will it be the same for all workers (it will be much smaller for workers who don't place a high value on the marginal health benefits that are being squeezed out by the Cadillac tax).

Moreover, remember that the 40% Cadillac tax is only imposed on the difference between the cost of a health plan and the maximum upper limit (e.g., $27,500 for family coverage). Thus, an employer with a $28,000 health plan only needs to trim $500 to escape the tax. For an average U.S. worker earning $47,200 annually in 2014, $500 is barely more than a 1% increase in pay. Most employers aren't likely to translate that into a one-time increase in pay all at once. Instead, whenever annual pay raises are awarded, this added bit of available cash compensation will be thrown into the mix. Thus workers may not even realize that a modest fraction of some future pay increase is attributable to the Cadillac tax.

It remains to be seen how long the Cadillac tax lasts, but remember it does not even take effect until 2018. I look forward to Austin Frakt's analysis of this question well in advance of that date.


READ CHRIS’ BOOK   The American Health Economy Illustrated    (AEI Press, 2012), available at Amazon and other major retailers. With generous support from the National Research    Initiative at the    American Enterprise Institute   , an on-line    version complete with downloadable Powerpoint slides and companion spreadsheets has been made available through the Medical Industry    Institute’s Open Education    Hub at the University of Minnesota.

Follow @ConoverChris  on Twitter TWTR -0.67%, and The Apothecary on Facebook. Or, sign up to receive a weekly e-mail digest of articles from The Apothecary.