Why Tax Deductions Are Stupid (and Why We Love Them)

Human beings like to hang on to the things they've already got.

It's a difficult relationship.

Photographer: Andrew Harrer/Bloomberg

Here is the basic argument against using tax deductions to encourage certain behavior -- in this case, buying houses. It's from the late, great Harvard Law School tax scholar Stanley Surrey, as quoted by journalist T.R. Reid in his fun new tax reform book, "A Fine Mess": 1

The federal government wants to help some Americans pay their mortgage. Here’s how it works: For a couple with $200,000 of income, and a mortgage interest payment of $1,000 per month, the government will pay the bank $700 and the homeowners will have to pay only $300. For a couple with $20,000 of income and a mortgage interest payment of $1,000 per month, the government will pay $190, leaving the couple to pay $810. For a couple making less than $10,000, the government will pay zero -- so the low-income couple has to pay the full $1000 of mortgage interest.

That is ... perverse. It's what inevitably happens, though, with itemized deductions in an income tax system with progressive rates. The exact numbers in the above quote no longer hold up, because it's from the 1960s or 1970s, when the top tax rate was 70 percent (it's 39.6 percent now). But the basic idea does. Those with higher incomes get more out of deductions because they face higher marginal tax rates. A $1 reduction in taxable income -- which is what a tax deduction is -- is worth 39.6 cents to somebody in the top tax bracket and 28 cents to one in the 28 percent bracket (for married couples filing jointly, those with incomes from $153,100 to $233,350). And it's worth nothing at all to those whose mortgage payments, charitable donations and the like add up to less than the standard deduction ($12,700 for married couples).

There's nothing especially unfair about this. Higher-income taxpayers get a bigger bang out of deductions because they pay a bigger share of their income in taxes. And those who take the standard deduction are in some sense getting a windfall by deducting more from their taxable income than they would otherwise be allowed to do. When we talk about deductions that exist for reasons other than behavior modification, such as the state and local tax deduction that was included in the original income tax legislation in 1913 as a federalist compromise with the states -- and is now under threat from Republican tax writers in Washington -- none of what I've written above is really an argument against them (there are other arguments, however).

If the idea is to encourage people to buy houses or save for retirement, though, then the perversity outlined above matters. It means that tax deductions can be among the most expensive possible ways for lawmakers to accomplish such things, and nearly useless in affecting the behavior of the low- and middle-income taxpayers most in need of an assist. With the mortgage deduction, almost all available evidence indicates that it causes very few people to buy houses who wouldn't otherwise buy them, 2 instead mainly causing those who would be buying a house in any case to buy a more expensive house, thus helping to drive up housing prices. With the deduction for retirement savings, the (less copious) evidence indicates that while it may motivate some high-income taxpayers to save slightly more than they otherwise would -- which doesn't seem like a bad thing, really -- it does little to increase saving among those in lower tax brackets. In both cases, tax credits (which directly reduce taxes rather than taxable income) or outright grants targeted at middle- and lower-income taxpayers would deliver much more bang for the buck. 3

Among economists and many other technocratty types, these are not even remotely controversial assertions. Conservative wonks want to shrink deductions to pay for cuts in tax rates. Liberal wonks want to shrink them to pay for government programs. All sorts of wonks want to shrink them to pay for expanded child tax credits.

But whenever such arguments are laid out before the general public, as I have done a couple of times over the past few months and as congressional Republicans may try to do soon, they are greeted with something like apoplexy. Some of this may just be straightforward economic interest talking -- it's not unreasonable to oppose having money taken from you to give to somebody else. But the emails, comments and tweets that I and others have been getting about this tend to focus on the injustice of it. Some are silly, but most are earnest complaints from seemingly reasonable people. What are they so worked up about? I have some theories:

1. The endowment effect. That's 2017 economics Nobelist Richard Thaler's term for the human preference for holding on to what we've already got. As he described it in 1980:

Mr. R bought a case of good wine in the late 50's for about $5 a bottle. A few years later his wine merchant offered to buy the wine back for $100 a bottle. He refused, although he has never paid more than $35 for a bottle of wine.

If you already get a mortgage tax deduction, then, it might take a more-than-commensurate amount of tax-rate cuts to make you happy about giving it up.

2. The HENRY effect. The prime beneficiaries of the mortgage and 401(k) deductions are not the rich -- the maximum mortgage amount one can deduct interest on is $1 million, and the maximum pre-tax contribution to a 401(k) is $24,000 a year 4 -- but those trying to get there. "High earners, not rich yet," my former Fortune magazine colleague Shawn Tully dubbed them in 2003: people with serious incomes ($250,000 to $500,000 a year was the income range Tully chose in 2008) but also serious mortgage payments, property-tax payments, graduate school debts, private-school bills and such, and no trust fund to tap into. I've done my part to popularize the HENRY acronym, with limited success, but the key thing to remember is that these are people who fork over much more of their income in taxes than those below them on the income distribution. This seems fair to me, but it can understandably feel burdensome. And to make it feel all the more burdensome, there's this twist: Those in the HENRY class also fork over more of their income in taxes than those well above them on the income distribution. As Reid reports in his book, a family of four making around $200,000 faced an average federal income tax rate of 19 percent in 2011, while those making between $500,000 and $10 million a year paid an average of 24.5 percent. So that's still progressive taxation. But those with incomes above $10 million paid 20.4 percent, while those making more than $100 million paid just 18 percent. There's a reason for this regressivity: When you make that much money in a year it's usually because you sold something, and capital gains are for various non-silly reasons taxed at a lower rate than other income. But it still doesn't seem quite fair.

3. Common sense. When econowonks call for eliminating tax deductions, it's usually in the context of "all else being held equal, this deduction isn't a great way to accomplish X." But all else isn't being held equal. Policy decisions in Washington these days are often driven by billionaire donors with specific interests, and two of the specific interests evident in this year's tax plans involve (1) avoiding estate taxes and (2) reducing taxes on pass-through businesses. The great majority of the benefits from eliminating the estate tax and creating a special, lower tax rate for owners of pass-throughs such as S corporations, partnerships and limited liability companies would seem to flow to the very, very rich. Cutting corporate taxes could bring more widely shared gains, but those are indirect and uncertain. If tax deductions are going to be squeezed in order to cut taxes for billionaires and corporations, which is not an unreasonable assumption given who currently controls Congress and what we've learned about the tax plan so far, then harping on the inefficiency and distributional unfairness of tax deductions -- as I just spent the first half of this column doing! -- can seem like missing the point. So OK, I'll stop. For a while, at least.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
  1. The subtitle is "A Global Quest for a Simpler, Fairer, and More Efficient Tax System." It came out in April, but that seems pretty new to me. Also, I searched Google for prior uses of the phrase "fun new tax reform book," and there weren't any.

  2. I'm leaving aside the question of whether it's a good idea for government to encourage people to buy houses.

  3. Refundable tax credits such as the earned income tax credit, premium tax credit for health insurance and additional child tax credit are hybrids of tax credit and grant.

  4. And you have to be 50 or older to do that.

To contact the author of this story:
Justin Fox at

To contact the editor responsible for this story:
Brooke Sample at

Before it's here, it's on the Bloomberg Terminal.