The Economics of the Cadillac Health Care Tax, Part II
By David Henderson
A little analysis goes a long way.
Yesterday, I wrote:
I’m sympathetic to what it’s [the Cadillac tax is] trying to achieve, but I’m against it. Moreover, I’m against it for the same reason that I’m sympathetic to what it’s trying to achieve. I want to reduce distortions caused by the tax code. The Cadillac tax in some ways adds to the distortions.
But now that I have done a more careful analysis with some plausible numbers, I am seriously undecided. Congress made a reform in December that got rid of one of my objections and I didn’t notice that until today.
I gave part I, the background about the tax treatment of employers’ contributions to employees’ health insurance premiums, and the distortions that causes.
Now to analysis of the Cadillac tax.
Remember that the stated goal of many of the health economists who support a 40% excise tax on employers’ contributions to employees’ health insurance above a threshold is to reduce the distortion I discussed in Part I.
Let’s see how it would work.
The thresholds are $10,200 for an individual plan and $27,500 for a family. Health insurance expenditures above these thresholds would be taxed at 40%. Whereas the original version would have forbidden employers from deducting this tax as an expense against income, Congress did get rid of this distorting provision in its tax bill last month. (Why would that have distorted? Because the tax system would have treated this expense differently from virtually all other employer expenses.)
Consider the relatively low-income person who is in a 15% federal income tax bracket and, say, a 3% state income tax bracket. That person is in a 7.65% payroll tax bracket (Social Security and Medicare.) Total marginal tax rate: 25.65%. That 25.65%, plus the 7.65% payroll tax paid by the employer, gives a measure of the distorting influence of taxes that I discussed in Part I.
So consider the pre-Cadillac tax world (the world we’re in) where the employee/employer health insurance plan is right at the threshold. For a family plan, that amount is $27,500. The employer is trying to decide whether to make the plan more generous in a way that adds $1,076.50 to it. (Why that number? To make subsequent calculations easier.) Alternatively, the employer could pay the employee $1,000 in pay and pay $76.50 in payroll taxes. The employee would then get net pay of $743.50 ($1,000 minus (0.15 + 0.03 = 0.0765)*$1,000).
So here’s how the 40% Cadillac tax would offset that distortion. If the employer spent $1.076.50 on added health insurance plus tax, the amount spent on added health insurance would be x. x*1.4 = $1,076.50. So x = $768.93. Notice that that is close to $743.50. So the Cadillac tax in this case does a fairly good job of substantially reducing the distortion.
Now consider the really high-income employee who is at the threshold of $27,500. He is in a 39.6% federal tax bracket and is, in most states, probably in a 5% (or more) state tax bracket. He is in a 0 percent Social Security (FICA) bracket, but a 2.35% Medicare (HI) tax bracket. The total marginal tax rate is therefore 46.95%. Imagine that the employer is deciding whether to spend $1,014.50 on the employee’s health insurance. Alternatively, he could pay the employee $1,000 and spend $14.50 on his (employer’s) portion of the payroll tax. If he pays the employee $1,000, the employee nets only $530.50.
Notice how huge the distortion is. For the same $1,014.50 expenditure by the employer, the employee can get only $530.50 in net pay or $1,014.50 in health insurance. Now, imagine that the Cadillac tax is imposed and the employer spends an additional $1,014.50 on health insurance and tax. Let x be the additional amount spent. Then x*1.4 = $1,014.50. Therefore x = $724.64. $724.64 is much higher than $530.50. So there still remains a large distortion. The employer still has too high an incentive to spend additional compensation on health insurance instead of straight employee pay.
Still, the Cadillac tax almost eliminates the distortion for lower-income employees and reduces it substantially for high-income employees.
That raises the question: why was I against the Cadillac tax 2 days ago. The first reason is that I thought that the tax was not deductible against income for the employer. But, as noted above, Congress changed that a month ago. The second reason is that I thought the tax would be too heavy on health insurance for lower-income people. I no longer think that. Of course, you can show that it is too heavy for someone in a 10% federal income tax bracket. But how many people would be both (1) in a 10% federal income tax bracket and (2) getting very expensive health insurance from their employer? Not many.