The Stark Geographic Inequality of the Home Mortgage Interest Deduction
The home mortgage interest deduction is America’s largest housing program. It costs the federal government (in lost tax revenue) more than Washington spends on the entire Department of Housing and Urban Development, more than it spends on veterans benefits and education. The subsidy is simply massive. And it has long been treated as untouchable because, in theory, so many middle-class Americans get their own little piece of that pie.
In reality, the bulk of the benefit goes to high-income households. The more money you have, and the higher the value of your home, the more this subsidy helps you. This chart, from a report last year by the Reason Foundation, estimates the average annual tax savings from the mortgage interest deduction, by income, in 2008.
Unmasking the Mortgage Interest Deduction, by Dean Stansel and Anthony Randazzo (2011).
Families at the bottom of the income ladder (those who might need the most housing assistance) don’t get much from the subsidy at all. Meanwhile, as Matthew O’Brien noted over at TheAtlantic.com earlier this year, 75 percent of the entire benefit goes to the top 20 percent of earners.
This criticism of the deduction – that it primarily subsidizes wealthy people living in expensive homes – has grown louder as Washington has gotten more serious about reforming the tax break. But homeowners everywhere still clinging to the benefit should also take note of another little-noticed pattern in who really wins here: The mortgage interest deduction is dramatically skewed by geography, too.
More specifically, a small number of coastal metros get the bulk of the benefit (hat tip to Brookings’ latest policy paper on the topic for pointing us to this research).
"The winners are the Bostons, the New Yorks, the Washington, D.C. metro area, San Francisco and L.A." says Todd Sinai, an associate professor at Pennsylvania’s Wharton School.
Back in 2003, Sinai and co-author Joseph Gyourko examined the geographic distribution [PDF] of the subsidy among homeowners in 1979, 1989 and 1999. The pattern stayed true for most of that time, and Sinai says he has little reason to believe it’s changed since then, even with the housing crash. If anything, those cities Sinai mentioned are the same ones that have already rebounded from the housing downturn.
"The thing that was surprising was just how concentrated the benefits were, not only in just a handful of states, but in a handful of metropolitan areas within those states," Sinai says. "That’s been a pretty persistent phenomenon."
In 1999, the average subsidy per owner-occupied housing unit in the San Francisco/San Mateo/Redwood City metropolitan area was $26,385. In McAllen/Edinburg, Texas, on the other hand, it was $1,696. This is not the amount of interest these households deducted from their income on average. It’s the full amount they were able to save on their final tax bill (or, viewed from another angle, it’s the money the federal government didn’t receive as a result of subsidizing homeownership).
About a third of this total directly comes from the mortgage interest deduction. We’ll let Sinai explain what he means by this: "I find that the easiest way to think about this is to consider what would happen to my tax bill if I sold my house and rented. (a) I would no longer get a deduction for mortgage interest, so my tax bill would go up. (b) I would invest the proceeds from selling my house (after paying back my mortgage), let’s say in bonds, and pay tax on the interest income. That would make my tax bill go up more."
As of 1999, this is what the top 20 metropolitan beneficiaries looked like. The list would look fairly similar today:
The (Un)changing Geographical Distribution of Housing Tax Benefits: 1980 to 2000, by Todd Sinai and Joseph Gyourko (2003)
Two patterns jump out. A majority of these cities are in California or the Northeast corridor. And the subsidy they receive quickly drops off from the really big winners at the top of the list. In this graph, Sinai and Gyourko plotted the benefit to more than 300 metro areas in the country (with San Francisco sitting at the right and McAllen/Edinburg at the left):
This geographic pattern is directly related to the one we more commonly think about by income bracket. The top cities on the list are essentially those cities with high home values and high incomes to pay for it.
"There’s nothing particularly deep about it once one thinks about it," Sinai says. "It’s simply that we’ve known for a very long time that the high-income households tend to get a larger subsidy per dollar of house."
And high-income people and high-priced houses tend to cluster together, in places like Boston, New York and San Francisco. It’s not necessarily that it costs more to build homes in these places. "They’re on land that’s expensive to buy," Sinai says. "The tax subsidy is a subsidy in large part to owning more expensive land."
In short, a lot of people benefit a little bit from the mortgage interest deduction, while a few people – and a few pockets of the country – benefit from it a lot. And now that policymakers are starting to look at overhauling the benefit, those same big winners suddenly look vulnerable. They stand to lose the most from any major reform (unless, that is, mortgage reform comes bundled with other tax changes that blunt its impact).
If you live in, say, Oklahoma City, this might change how you view the deduction. The massive benefit must be financed somehow – in the broadest sense, it's paid for by all of us – and homeowners in San Francisco are getting a lot more out of it than you are. If you live in San Francisco, Boston, L.A. or New York, well, get ready.
"They’ll be the ones who will be writing larger checks to the IRS," Sinai says. "When you think about the politics of this, that is a constituency that we generally think of as being politically powerful. So it’s going to be interesting to see how the next few months play out."
Top image: A realtor shows a home during an upswing in the housing market, in Riverside, California. (Alex Gallardo/Reuters)