A key mechanism driving this result is that employees are modeled as being more cost-conscious when they make their own health insurance choices, as opposed to when their employers make the most important of these choices on the employees’ behalves. As HR explain (on pages 7-8): “Thus, when employees are given the opportunity to purchase their own plans, more will purchase less expensive plans than under the current system. The insurance market, in turn, will make less expensive plans more readily available. . . . If employees purchase less expensive plans, then less income is excluded from taxation and more goes towards wages; thus employees enjoy more take home income and the government receives more tax revenue.”
This modeling assumption is supported by a robust body of research, as HR elaborate (on pages 8-10). Moreover, it makes intuitive sense that employees are less cost conscious when they do not directly benefit from the resources they would save from exercising greater cost consciousness.
Indeed, this is related to a broader critique of the tax exclusion for employer provided health insurance and of the U.S. system for health care finance more generally. As a society, we spend almost twenty percent of our economic resources on health care. Yet almost no one seems to want to spend almost a fifth of their own economic resources on health care. The U.S. health care system thus conceals much of the costs of funding health care from the beneficiaries of this spending.
Perhaps, then, we should go further than HR propose and simply repeal the exclusion for employer provided health insurance? One argument against doing so is that employer provided health insurance might partially unravel were the exclusion to be repealed. The reason is adverse selection. When people are exposed to the full cost of purchasing health insurance, then people who are sicker generally seek to purchase more robust health insurance policies, whereas people who are healthier often seek to save money by purchasing skimpier and cheaper health insurance policies. Consequently, unless insurers can charge more or deny coverage to sicker people, the market for more less skimpy health insurance policies faces the threat of adverse selection death spirals—as attracting a sicker pool of insureds forces insurers to raise prices, which drives away healthier insureds and makes the pool even sicker and more costly, which forces further price hikes, and so on.
Note that I am discussing here a form of adverse selection death spiral whereby the market might collapse into only offering skimpier health insurance plans. I am not discussing adverse selection death spirals whereby the market would completely collapse into not offering any insurance. At least so long as the Affordable Care Act (ACA) remains in place, it is unlikely that the market would collapse into not offering any health insurance (in most geographic areas). Yet facilitating healthier people purchasing skimpier insurance offerings would leave less skimpy offerings with a sicker pool of insureds, which—absent countervailing mechanisms—would threaten the continued existence of those less skimpy health insurance offerings.
The threat of this form of adverse selection death spiral is well known and well understood. Within the ACA’s Exchanges, this threat is countered through a risk adjustment program that charges individual market insurance plans that end up with a healthier pool of insureds so as to subsidize plans that end up with sicker pools of insureds.
So how have employers dealt with this threat with respect to employer provided insurance offerings? Traditionally, because the exclusion for employer provided health insurance has made employer-provided policies so substantially tax favored, major employers have mostly just directed substantial portions of employee compensation toward offsetting the cost of health insurance for all employees. In this way, by preventing employees from enjoying most of the cost savings from opting for cheaper health insurance offerings, these employers have negated the incentives that might otherwise result in adverse selection death spirals.
Herein lies a critical tension in HR’s argument. HR want employees to have the choice to purchase cheaper health insurance offerings. But the danger is that healthy employees would exercise this choice, and that sicker employees would not. In other words, the danger is that facilitating this choice would give rise to adverse selection.
HR are aware of this danger. Accordingly, to combat this danger, HR propose that employers provide larger sums to sicker employees than to healthier employees. In their words (on page 11): “The simulation accounts for the concentration of health expenditures by transferring greater sums to the sick. Of course in actuality, both healthy and sick individuals purchase similiarly priced insurance, and redistribution of resources from the healthy to the sick takes place within insurance risk pools. We model this transfer within the transfer from employers to employees to account explicitly for the necessary funds to provide care for the sick and to ensure that these social costs are paid for even while others are able to economize and retain more take home pay.”
This generates my primary concern both about HR’s proposal and about their modeling results. Implementing effective risk adjustment mechanisms of this sort is not simple. Would employers be able to successfully achieve the risk adjustment that HR propose and that they assume in their model?
I am somewhat skeptical, but also rather intrigued. Although risk adjustment is not simple, it is not impossible. Indeed, some employers currently use risk adjustment mechanisms (although, to my knowledge, none to the extent proposed by HR). The recent experience with the ACA’s Exchanges provides further information that could guide the development of better risk adjustment policies.
If the risk adjustment problem can be solved sufficiently, then HR’s analysis and proposal are very persuasive and some version of their proposal should definitely be adopted. I think that further work is needed on how well risk adjustment could work within HR’s policy context, and how robust their results might be if risk adjustment works only partially. Nevertheless, at the very least, HR’s new draft article raises important questions and illuminates a possible path toward future reform. Anyone interested in the intersection of tax and health policy should consider the implications of HR’s work and should look forward to the final version of their article.