David R. Henderson  

Is Taxing One Item Because It's in a Subsidized Category a Good Idea?

PRINT
Why I favor taxing subsidized ... Can fiscal transfers solve the...

Co-blogger Scott Sumner advocates keeping a tax on medical devices. His reason:

the medical devices tax was far smaller than the subsidy on medical devices, which is enormous. (Subsidies include Medicare, Medicaid, tax deductibility of insurance, etc.)

But normally when people advocate a tax to offset a subsidy, it's because the subsidy is on the item taxed. As Scott admits in the above, medical devices per se are not subsidized. Instead, medical care for the poor and near-poor is subsidized through Medicaid; medical care for the elderly is subsidized through Medicare; medical care for employees is implicitly subsidized by the tax code.

So a way to offset the subsidy, if the first-best solution of eliminating or reducing the subsidy is, for some reason, off the table, is to tax the thing being subsidized.

The problem is that to single out medical devices does not clearly reduce the deadweight loss that Scott would like to reduce. If, say, 70%, of expenditures on medical care were on medical devices, probably singling them out for a tax would make sense. But I'm guessing that medical devices account for under 10% of expenditures on medical care. Taxing just medical devices will cause the price of them to rise relative to the prices of other inputs: doctors, nurses, hospital rooms, drugs, etc. So the tax would cause a distortion in inputs used: fewer medical devices used and more of the other inputs.

Actually, this post overlaps with a comment made on Scott's post by Jim Glass.


Comments and Sharing






COMMENTS (3 to date)
J Mann writes:

Tax deductibility is a somewhat easier case, but I'm having trouble thinking through whether this moves consumption for Medicare and Medicaid recipients in a direction that reduces deadweight loss more than it increased administrative costs, if at all.

- My rough understanding is that recipients of those programs don't pay premiums, but do pay co-pays or coinsurance, at least up to some maximum out of pocket payment.

- So assuming that the medical device tax increases the coinsurance payment for Medicare and Medicaid users of medical devices that aren't otherwise going to hit their maximum out of pocket for the year, I guess it would reduce deadweight usage of medical devices somewhat.

- On the other hand, if the tax just increases the costs to the Medicaid and Medicare programs, isn't that just an administratively cumbersome way of both increasing the subsidies on one side and then cancelling that increase out on the other?

Scott Sumner writes:

I don't agree with your premise that medical devices are not being subsidized. They are. Subsidies can be applied to either the supply side or the demand side of a market.

David R. Henderson writes:

@Scott Sumner,
We’re agreed that there’s a subsidy to health care on the demand side. That’s not the issue here, and I’m surprised that you’re raising it in this context.
My critique is, as I wrote in the post, of your idea of singling out one relatively small component of health care and taxing it.

Comments for this entry have been closed
Return to top