Of Cadillacs, Coinsurance, and the People that Love Them
[cross-posted with the Field Clinic]
There has recently been considerable criticism of the ACA’s “Cadillac tax,” which imposes a 40% levy on the excess premiums of high-cost group health plans. These are expensive plans with low cost sharing, access to the most expensive providers, and few limits on access to costly new technologies. I signed on to the statement of dozens of economists supporting the tax—not as ideal but as better than staying with the status quo. My reasons are somewhat different from those generally expressed in the debate.
Let me start with some personal observations on the “Cadillac tax” metaphor. In 1955, when I was 13, my father bought a nugget-gold hardtop convertible Chrysler New Yorker. At almost the same time, his brother Art acquired a pearl-white Cadillac Coupe Deville—for both of them the first decent car they ever had. We boy cousins argued endlessly over which was better—my money was on the totally cool hemi-power Chrysler as opposed to the chrome-plated, bullet bumpered Caddy. But none of us bothered to note that either car got at best about 14 miles to the gallon in city driving – gas-guzzlers if ever we had them.
Applying this to health insurance, the debate over the ACA tax is usually framed as a contest between those who think expensive policies really are luxuries that people might be discouraged from buying and those who think they are just high quality health insurance. That framing ignores two important issues. One is the use of resources under expensive policies, similar to the extra gas needed to power the luxury cars of the 1950s. The other relates to the tax break for group health insurance. My Dad and Uncle Archie both paid the full cost of the cars themselves, which was much more than the Fords and Studebakers they were used to buying. Employees who buy a Cadillac health insurance plan can use pre-tax dollars, which amounts to a huge subsidy from the government.
The more expensive health plans have been shown, beyond a shadow of econometric doubt, to cause higher levels of medical care spending by average risk middle class workers who take them. We also have sterling evidence that this higher spending—to the tune of about 30 to 50% more than under high deductible health plans—increases access to care with almost no perceptible effect on health outcomes (with the possible exception of low income high risk people, who generally would not be the ones with these high-premium plans).
I would not want to claim that expensive policies have no beneficial effect whatsoever—cost sharing under cheaper policies is a bother and may cause people to avoid care that would have reassured them, speeded-up alleviation of discomfort, or even increased the chance of a significantly bad health outcome by a tiny amount. The “guzzling” does not mean the extra care is worthless (just as the Chrysler’s ability to do zero to 60 in 10 seconds was worthwhile as bragging rights). It is just that it is not worth nearly as much as what the saved resources could buy if used in a different way.
Workers who buy expensive plans may be acting rationally in choosing them. Even though our most costly plan at the University of Pennsylvania where I work would not hit the tax limit (yet), I choose it and feel guilty. I select it because by paying the extra $5,000 or so in family premium, I am able to reduce my taxes—by nearly half of the cost of the premium, which is close to the 40% ACA excise tax.
I cannot think of a single reason why it is good public policy to offer generous tax subsidies to lucky full professors at the Wharton School to help them buy health insurance that gives them access to high-cost health care of minor effectiveness. If someone who opposes the Cadillac tax could provide a reason, I would be very grateful because my conscience is bothering me.
Of course, the last-minute political dealing when the ACA was passed was the reason why we got the clunky Cadillac tax rather than an alternative, such as a cap on the amount of health benefits that can be excluded from taxation. There is a great deal of room for improvement in the way the tax is structured, and it is not beyond the limits of human intelligence to waive or reduce the tax for low-income or high-risk people.
Whether it is beyond the limits of our current-day political process remains to be seen—but at least the principle of preserving a policy that strikes at an inequitable and inefficient tax break, which the Cadillac tax makes a start at doing, should be respected.