post by Paul Kelleher
Most health economists seem to agree that the tax exclusion for employer-provided insurance is bad policy. Last week Aaron Carroll and Austin Frakt added to that critique by illustrating the exclusion's distributive implications and arguing (in part) that the exclusion is unfair. I agreed with them, and simply noted that their fairness-base argument against this policy could not, on its own, justify ACA's tax-financed subsidies. Even if the right policy is to tax the wage-base of each earner, that says nothing about how high the tax should be or what those taxes should be used to pay for. Once again, substantive argument--including moral argument--is needed.
There is one argumentative strategy that I've seen in both conservative and left-leaning discussions of insurance reform that I want to argue is a non-starter. Consider the following two writers, A and B (their identities are revealed at the end of this post; bold text is my emphasis).
In today’s Wall Street Journal, Holman Jenkins offers an idea for improving the [Affordable Care Act]:
[L]et’s permit insurers to design their policies free of ObamaCare’s mandated benefit levels and free of state regulation [...]
What’s the first thing the new nationally-chartered insurers would do? Rush out cheap, high-deductible policies, allaying some of the resentment that the mandate provokes among the young, healthy and footloose affluent. [...]
[T]hese folks could buy the minimalist coverage that (for various reasons) actually makes sense for them. They wouldn’t be forced to buy gold-plated coverage they don’t need so the money can subsidize the old and sick (the hidden tax logic of ObamaCare).
This goes the wrong way in two respects…[...]
Second, what is it that people think they’re buying when they purchase insurance? If it’s not a vehicle for the healthy to subsidize the sick, then what’s the point?
One often-criticized feature of insurance, especially individual insurance, is a preexisting-condition exclusion....The rationale for exclusion is that the purpose of insurance is to pay for uncertain events, not those that are already known to exist...The point to be made here is that you can reduce frustration and argument by paying more...It is interesting that in unregulated markets, waiting periods are rare and brief..., but exclusions are common, though usually limited to two years.
You'd be right to say the second author is less supportive of ACA than the first. And yet each suggests that facts about the "point" or "purpose" of insurance can help determine the correct shape of health policy. I do not agree.
Author A is correct that health insurance essentially involves transferring money from the healthy to the sick. But an inference from what a practice involves to its "point" is usually specious. Consider an analogy I have used before. My buying a television essentially involves transferring a portion of my paycheck into the paycheck of the manufacturer's CEO---TVs would not be available to me unless a company made them. But of course the point of my buying a television isn't to pay a CEO's salary. Similarly, while insurance would not be available to me unless others (those in my risk pool) bought and received protection also, the point of my purchase is not to protect them; it is to protect me. My money being used to protect others is just an outgrowth of how insurance markets work, just as my contributing to a CEOs salary is simply an outgrowth of how markets for televisions work. Quite obviously, nothing about the redistributive features of television purchases justifies corporate welfare. Likewise, nothing about the redistributive features of individual insurance can justify publicly subsidized health insurance. The "point" of my individual insurance purchase is to protect myself against risk. Absolutely nothing bearing on social obligation follows from this fact alone.
Author B's mistake is simpler to diagnose. B can at most be making a definitional claim, and it is a claim he himself does not accept. After all, if the true "purpose" of insurance were to protect individuals from uncertain risks rather than already-existing conditions, then individual insurance that covers pre-existing conditions would be conceptually impossible. But B notes that such insurance is in fact available, so long as one is willing to pay enough or wait long enough. So B himself must concede that the "purpose" of insurance is simply to protect individuals against the harms associated with illness. Different insurance arrangements do that differently, and this leaves it an open question whether policies embodying social insurance principles are justified.
I conclude that claims about the point or purpose of insurance provide no guidance whatsoever in a ethical debates about social insurance policies.