Kevin Drum is right that the bad distributional consequences of the home-mortgage-interest tax deduction could be partly fixed by lowering the cap on the size of the the mortgages covered. But of course that would also harm residents of high-housing-price areas, who are already over-taxed compared to residents of lower-housing-price area, since to maintain a given standard of living in an expensive area means earning a higher income than would be required to maintain that same standard of living in a lower-housing-price area. In general, the federal system redistributes income toward the (mostly Republican) voters who live outside the big cities. It wouldn’t be technically hard to do a cost-of-living adjustment to income taxes, as we now do to federal employee salaries, but it would be politically impossible. In the meantime, I’m wary of further screwing over New York for the benefit of Kansas.
Footnote Limiting mortgage-interest deductibility also benefits those rich enough to pay cash for their housing. The technically right way to handle problem is to leave the interest deductible but treat what would be the value of owner-occupied housing as income to the owner. That’s a complete non-starter, for administrative as well as political reasons, but that’s the right baseline for comparison to actual proposals, in that it would equalize treatment between renters and owners and between owners-for-cash and owners-with-a-mortgage.
Part of the difficulty is that renters benefit from deductions available to their landlords. These include not just interest, but other items, such as maintenance costs, not deductible to homeowners. It is not easy to see how the tax system can be made to treat owners and renters equally.
While I would like to believe it, I'll admit that I have always been a little uncomfortable with the assumption that cities subsidize less populated areas. I don't know the direction of the subsidy for sure, but I do know that it is a little more complicated than conventional wisdom would suggest for several reasons. (1) The same person is paid more in a big city than a small one for the same job. This is due to agglomeration externalities or locational returns to scale. (2) People are compensated for loss of housing, space, high prices, etc. in the form of access and amenities in high rent areas. i.e., given the same house for the same price, I'd rather live in Santa Monica or Manhattan, than Buffalo or Tulsa. (3) In the long run people are mobile in the U.S., and these price/salary differences have been around a long time.
The existence of externalities suggest subsidies might be in order, but which direction the subsidies run currently, and which way they should run is a hard problem to figure out. It depends on the marginal relationship between agglomeration, congestion, housing supply, and demand for housing and amenities.
I can't fit the footnote with the main part of the post. If we adopt the perspective of the footnote, we don't end up with the conclusion that residents of high-house-price areas are over-taxed.
I think Bernard Yomtov means to say "In a perfectly competitive rental market with an unlimited supply of rental units, renters would benefit from deductions available to their landlords."
Let's say the cap is $500,000 and the house you buy is $700,000 — you still benefit, and you are still pretty well off if you can spend that much for a house. Even in SF or NY, where people typically spend a lot of money on housing, my guess is that the "average" homeowner isn't spending anywhere near the cap. The point is, how much of a taxpayer subsidy do you really want to give to people who can afford to spend that much on a house? There is a lot of research that shows the mortgage interest deduction contributes to price escalation and, if not quite a net zero in terms of contributing to the affordability of home ownership, is certainly not the glowing torch of economic populism the real estate industrial complex tries to make it.
P.S. — The truly rich pay cash for their houses anyway because it is almost always advantageous not to pay interest, whether you can deduct it or not.
You cannot even if you tried level every relative advantage.
Broadly, I can see mortgage subsidy as a good thing to the extent that homeownership is a good thing. Okay, mostly - homeownership is really god-awful for people with houses chained around their necks in Camden and Flint and East St Louis and all the jobs have gone to Tulsa. But generally, all the social science research says homeowners get more involved in their communities and vote more and feel more empowered to call the cops when things go awry. So I favor getting people into houses, and some subsidy is okay. I don't see any reason I care to make it easier and more favorable for people to get into a marble-counters house instead of living their lives with Formica. We are grotesquely over-invested in luxury housing. I think my fave is a tax credit of 50% of what you spent on mortgage interest up to your having spent $25000. So the maximum subsidy is $12500. That gets a moderate number of postal delivery people and ironworkers and seamstresses into houses, who otherwise maybe couldn't. Podiatrists in Piedmont who want a million-five house versus having to make do with three quarters of a million? Cry me a river.
@Barbara:
It's advantageous to pay interest if you can get a rate low enough to put the money into other investments (or avoid liquidating current investments and taking a tax hit). Right now the rich can get mortgages with an after-tax interest rate of 1.5-2%, and there are a lot of plausibly safe investments yielding more than that.
I should also point out that in countries (Canada leaps to mind) where mortgage interest is not tax-deductible, mortgage interest rates are significantly lower than in the US. Also, their housing prices tend to be lower. The tax deduction (from which my wife and I benefit handsomely, thank you very much) distorts the market and should be phased out.