Getting Tax Burden Wrong

Assessing tax burdens is an inexact science, filled with variables such as a person’s deductions, property ownership, sales purchases and other factors. (“Fact check: Newsom says working-class taxes are higher in Texas, Florida than California. Is he right?,” Sacramento Bee, 12/15/23)¹

That sort of information, if one had enough of it, would tell us who handed over the money but not who bore the burden of the taxes. To see the difference, consider the taxes that fund Social Security, half on the employer, half on the employee. Of each dollar the employer pays ten cents goes to the government, ninety to the employee, who pays ten cents to the government (rate invented to simplify the argument). It follows that of each dollar the employer pays the employee gets eighty cents, the government twenty.

What matters to the employer, in terms of his own welfare and his willingness to employ people, is what he pays. What matters to the employee, in terms of his own welfare and his willingness to take the job, is how much he gets. Whether the government takes the money before or after the transfer from employer to employee or, as in the present system, half before and half after, does not matter. The result would be the same if the tax was wholly on the employer, who would pay twenty cents of each dollar to the government and eighty cents (untaxed) to the employee. It would be the same if the tax was wholly on the employee, who would receive the full dollar paid by the employer, pay twenty cents to the government, end up with eighty cents of each dollar.

The same applies to other taxes. If there is a ten percent tax on home sales it does not matter if the money is handed over by the buyer or the seller. Sometimes one way of doing it is easier than another, collecting sales taxes from sellers instead of buyers since there are fewer sellers, but that is the only difference.

What determines who really pays a tax is not who hands over the money but what effect the tax has on the terms of the transaction. Suppose the government imposes a tax of fifty thousand dollars a year on physicians. That makes it less attractive to be a physician so over time the number of physicians goes down. Since there are now fewer physicians they find that they can charge more, so their income goes up by (say) twenty-five thousand. All of the tax money is being paid by the physicians but half of it is coming from their patients.

There is no reason it has to be half. Perhaps most physicians are people who really like healing, will go into the profession even if it pays fifty thousand less. If so the number of physicians decreases only a little, the price goes up only a little, forty-five thousand is paid by the physicians, five thousand by their patients. Perhaps it is the other way around, perhaps most people who become physicians now would choose not to if the after-tax pay was a little less. When supply finishes adjusting, wages have gone up by most of fifty thousand, the after tax income of physicians is only a little lower and most of the tax is being paid by their customers.

If supply is very elastic, if the number of physicians is very sensitive to the wage, most of the tax ends up on the patients, if it is very inelastic on the doctors. The same logic applies to demand. If when the price of medical services goes up enough to increase physicians’ income by a few thousand dollars lots of people decide to see the doctor less often, most of the burden ends up on the doctors. If, on the other hand, the demand for medical services is very inelastic, quantity demanded almost independent of price, the doctors end up almost entirely compensated for the tax by the increase in their wages, putting most of the burden on their customers.

A drop in the after-tax income of physicians will not persuade many people who have already gone through med school and the apprenticeship hell of interning to quit the profession, although some might retire a little earlier; supply is typically more elastic in the long run that the short. But it will make people less willing to bear the costs of becoming doctors, so over time the number will fall. Hence we can expect the burden of the tax to fall initially on the physicians, over time to increasingly shift to the patients.

Excess Burden

So far I have looked at the burden of taxes as a transfer of wealth from individuals to the government. That is only part of the story. “Excess burden” is the cost individuals pay that nobody receives.

Consider a tax on physicians’ services so high that nobody becomes a physician. Nobody pays the tax so the government collects no money but there is still a cost, both to those who would have been patients and now cannot be and to those who would have been physicians and have now shifted to some less attractive career. The point is not limited to such extreme cases. If the tax results in a thousand fewer people becoming physicians that is a cost to a thousand people who are now shifted to their second choice career and the patients they would have seen.

A more detailed explanation of all of this would carry us well beyond the limits of a substack post. The essential point is that one cannot tell who bears the burden of a tax by merely seeing who hands over the money. This is a point that is missed in most public discussion of taxation, but it is not new.

Adam Smith starts his discussion of taxation with a series of maxims, of which the first is:

The subjects of every state ought to contribute towards the support of government, as nearly as possible, in proportion to their respective abilities; that is, in proportion to the revenue which they respectively enjoy under the protection of the state.²

That sounds as though he is proposing a flat rate income tax but he isn’t. As Smith’s discussion continues he rejects all forms of income tax with the possible exception of a tax on the income of government employees. What he is proposing is not an income tax but a structure of taxes whose incidence, the tax burden, is proportional to income. To figure out what that might look like he attempts, for each possible tax, to figure out who actually ends up paying it. For example:

A tax upon house-rent, payable by the tenant and proportioned to the whole rent of each house, could not, for any considerable time at least, affect the building-rent. If the builder did not get his reasonable profit, he would be obliged to quit the trade; which, by raising the demand for building, would in a short time bring back his profit to its proper level with that of other trades.

1. The conclusion of the news story is that workers who make less than $55,000/year pay lower taxes in California, workers who make more than $55,000/year pay higher taxes. The article refers to “working class” but Newsome’s claim was “They talk about how they are for the working folks and middle class — they are not. Look at who they tax” and “middle class” should get well above median income. Looking online I find inconsistent figures for the median individual income, but according to the Census Bureau the median household income in California is more than ninety thousand dollars.

2. Adam Smith, The Wealth of Nations, Book V Chapter II, Article IV.