Paul L. Caron

Wednesday, June 15, 2011

Turnier: Mulligan Should Take a Mulligan on the Mortgage Interest Deduction

New York Times, The Misunderstood Mortgage Interest Deduction, by Casey B. Mulligan (University of Chicago, Department of Economics):

The home-mortgage interest deduction does not by itself significantly distort housing markets. Too much owner-occupied housing has been built because housing is excluded from sales and other taxes owed by businesses.

The housing boom and bust of the last decade, and government revenue shortfalls, have brought back the topic of whether the government excessively encourages home building. Those discussions invariably mention the elimination of the home-mortgage interest deduction. ...

At first glance, the home-mortgage interest deduction would seem to cause homeowners to borrow excessively for home ownership, and thereby deplete government treasuries. But that ignores the fact that one person’s mortgage interest payment produces interest income for another person or a business. The lender may well owe taxes on the interest income.

More home-mortgage borrowing means more home-mortgage lending, and the latter means more interest income that can be taxed. In theory, home-mortgage borrowing could even add revenue to the Treasury if the lender is in a higher tax bracket than the borrower (or if the borrower is not itemizing her tax deductions).

Tax Prof William J. Turnier (North Carolina) disagrees:

It is hard to believe that Professor Mulligan holds a position at the University of Chicago and is apparently unaware of the Haig-Simons definition of income which is in part the work of Henry Simons, a University of Chicago professor of economics. Simons' definition put most simply posits that income is consumption (money spent to further personal living) plus savings (with a negative adjustment made for disgorgement of savings). This definition is very widely accepted by tax policy analysts. The result of applying it to our individual lives is that although one can justify allowing a deduction for business expenses, one cannot do so for personal outlays such as groceries, haircuts, rent, interest, etc, unless such expenditures are in furtherance of profit seeking activities. Just because a party on the other side of a transaction take a payment into income does not justify allowing a deduction for the outlay. For example, although Professor Mulligan's barber reports the sum and tip the professor pays him for a haircut, the professor is not thereby going to be allowed to claim a deduction for the cost of the haircut. The right question to ask when examining the home interest deduction is whether it is an expenditure in furtherance of personal consumption or profit seeking. Quite clearly it is an expenditure in furtherance of personal consumption and as such fails the test of being allowed as a deduction under the Haig-Simons definition of income. Henry Simons' book is a slender offering of only about 100 and is recommended as the first step in reading fir all interested in tax policy analysis.

Update: Dan Shaviro (NYU), There Is No Joy in Mudville - Casey Mulligan Has Struck Out:

Thanks again to a link in the Tax Prof Blog, I see that Casey Mulligan of the University of Chicago Econ Department is saying that the home mortgage interest deduction is no big deal, hence no particular reason to repeal it. He focuses mainly on the point that "one person’s mortgage interest payment produces interest income for another person or a business. The lender may well owe taxes on the interest income."

My god, but Mulligan has missed the boat on this one. William Turnier pretty much makes the point, back in the Tax Prof blog link, by referencing the Haig-Simons definition of income and calling the home mortgage interest payment consumption-related.

But I would make it a bit clearer still. The problem isn't really the home mortgage interest deduction as such, although it may encourage the sort of excessive home leverage that helped contribute to the 2008 financial crisis. The real problem is the fact that homeowners get to exclude imputed rental income because it isn't derived from an observable market transaction. The home mortgage interest deduction merely builds off that. Its allowance would be uncontroversial if imputed rental income were included.

One can't seriously discuss the home mortgage interest deduction without understanding its relationship to the underlying imputed rent exclusion. This is really basic stuff. Mulligan ought to know, not just before he blogs about the issue in the New York Times, but as a supposedly informed professional in the field of public economics. ...

The University of Chicago Law School has a pretty good tax faculty, whose members I am sure would have been happy to bring Mulligan up to speed. But if you don't know that you don't know, I guess it's hard to find out.

Tax | Permalink

TrackBack URL for this entry:

Listed below are links to weblogs that reference Turnier: Mulligan Should Take a Mulligan on the Mortgage Interest Deduction:


The Haig-Simons definition is useful to envision a structure for an isolated income tax system that is independent of other factors. Tax laws affect behavior, and if that effect were ignored, it could have an adverse impact on economic or societal goals. From its inception, the American tax laws have included numerous provisions and basic doctrines that do not fit the Haig-Simons construct. For example, current law provides deductions for alimony, charitable contributions, and medical expenses. An employee's receipt of dependent care assistance from his employer is excluded from income. Credits are allowed for purchasing anti-pollution devices and for dependent care expenses. Certain fringe benefits are excluded from an employee's income. Those are merely a few examples of provisions that do not comply with the Haig-Simons definition The basic doctrine of realization and the numerous nonrecognition provisions conflict with Haig-Simons. The determination of a conflict with Haig-Simons may be a starting point for analysis of the merits of a tax provision, but it is far from concluding the analysis.

Posted by: Douglas Kahn | Jun 19, 2011 11:48:38 AM

The Haig-Simons definition or equation is useful for envisioning a broad structure for the income tax system, but the Internal Revenue Code (IRC) has never adhered to it and should not be required to do so. The tax laws affect behavior and those effects should not be ignored when enacting tax provisions. From its inception, the income tax has had numerous provisions that do not fit within the Haig-Simons formula. Personal deductions are examples. The tax laws operate in a context of dealing with economic, financial and societal goals. The alimony deduction, the medical expense deduction, the charitable deduction, the exclusion from an employee's income of dependent care assistance provided by his employer, the credits for purchasing anti-pollution devices are illustrative of provisions that do not adhere to the Haig-Simons definition. Some basic principles of taxation are inconsistent with that definition. The doctrine of realization and the numerous nonrecognition provisions are striking examples.

Posted by: Douglas Kahn | Jun 19, 2011 11:08:21 AM

you did not say it. but from what you wrote i thought you did not. so i will give your thoughts more weight than i would to someone who didn't understand imputed rent. and i stand by my statement that someone who does not understand the concept of imputed rent should get the value it that it deserves. It is like not understanding basis. Are you going to value the opinion of someone who does not understand basis?

you still have what definition of income our income tax system is based on if it is not the Haig-Simons definition of income. i think that this is not mere criticism, but valid criticism that goes to one of your central theses and criticisms of the Profs.

Posted by: tax guy | Jun 19, 2011 11:01:46 AM

Where did I say that I didn't understand the concept of imputed rent? I do understand it. It's just that I think it's a ludicrous concept.

I think the underlying reasoning in your statement is "he doesn't agree with imputed rent ... therefore he must not understand it". I can certainly understand how you would reason that way if you're basing your opinions on the articles above by Turnier and Shaviro, which are highly dismissive of Mulligan's comments. As you say yourself, it's so hard to criticize constructively. It seems to be beyond Turnier & Shaviro anyway.

Posted by: Nonce | Jun 19, 2011 7:32:33 AM

I have not read everything that Prof. Shaviro has written, but I have seen him engage in discussions about tax scholarship. I am surprised that someone would insult his scholarship ability, let alone his intellect. I cannot speak for Prof. Turnier as I have not read his scholarship nor have I seen him speak.

I am pretty sure that the Haig-Simons definition of income has everything to do with out current tax system, and if it does not please direct me to the definition of income that does relate to our income tax system.

If you do not understand the concept of imputed rent, then your criticism of anyone's tax scholarship demonstrates the value that it deserves. Scholarship is about theory, not about whether if A sells B a widget and gets $x in return how much income does A have and can B deduct the cost of the widget.

It is so easy to destroy and so hard to create. It is so easy to criticize and so hard to criticize constructively.

Posted by: tax guy | Jun 18, 2011 9:50:53 PM

I think that the criticisms of Professor Mulligan may be refuting something that he did not say. I read his major point as responding to the suggestion that because the home interest deduction causes people to borrow more, the larger resulting interest deduction reduces the revenue that the government collects from the income tax. His point was that the revenue lost from allowing the interest deduction is matched by the tax revenue gained from the lender. He was repudiating the suggestion that eliminating the home interest deduction will increase the government's revenue. He did not say that the interest deduction is a good thing or that it did not violate horizontal equity. Rather, he made a modest point that seems correct.

Posted by: Douglas Kahn | Jun 18, 2011 6:32:24 PM

What next, an argument that if I do my own taxes I should be taxed on the imputed income I received as a tax consultant to myself?

That's a logical conclusion from the concept. You also need to add all the personal and even sexual services that occur within a household. Don't forget self-service, which is what this exercise resembles.

Posted by: AMTbuff | Jun 16, 2011 3:23:31 PM

Mulligan's right. Turnier and Shaviro illustrate the amazingly low standards of tax scholarship.

Since the Haig-Simons definition of income has nothing to do with our current tax system, it's irrelevant whether Mulligan is familiar with it or not. How bizarre to insist on an academic model over reality.

'Imputed rent' is too bizarre for words. It isn't income at all, and it really doesn't matter what tax theorists say, it won't (and shouldn't) become part of the system. What next, an argument that if I do my own taxes I should be taxed on the imputed income I received as a tax consultant to myself?

Posted by: Nonce | Jun 16, 2011 7:31:09 AM

I love the phrase "schemes to reform the tax code". As if Simon Legree is twisting his mustache in a scheme where the IRC is "reformed."

And of course "tax reform" is another great phrase (or euphemism). Like a wayward child in need of discipline. Naughty naughty IRC. Maybe some time at Dickens' School for Wayward Codes will straighten you out.

Posted by: tax guy | Jun 15, 2011 6:14:08 PM

To properly understand the histroic reasons for the mortgage interest dedcution, you have to go back to the old Civil War income tax, our first. At that time there was awareness, based on the 1799 British income tax (the first real modern such tax) that the imputed value of owner occupied residences was income. They made it Item 1 on Schedule A of the British tax. The Covil War tax chose to ignore imputed income but to correct for that exclusion by allowing a deduction allowing renters a dedcution for their rent paid on residences. Now only those who were not outright owners of their houses were left out in the cold. Thye were given parity by allowing a dedcution for all interest and for state and local taxes. Basically the Civil War tax was designed to impose no tax advantage on outright owners over renters or those who were financing a purchase of a hoe. That tax was repealed during Reconstruction and as the tax was reinstituted at various times, parts of that full construct were left out.

I find our interest with imputed income fascinating. I know that when I frist started teaching, I felt that I had stumbled on a great truth that few knew. About five years later when I was working on an article on personal deductions, I learned that it was the frist thing that the Brits thought of when they chose to tax income. Thye eventually gave up on taxing it in 1964 as just too complicated.

None of this is to dismiss Dan Shaviro's point. It just shows how difficult all of this can be when we wish to place all on an even plain when it comes to home ownership. It seems to be an impossible task. It gets even worse when we try to apply the even plain principle to other aspects of tax law. For example, should we allow those who send their children to private schools a deduction because we do not tax the benefit bestowed on parents who send their children to public school? And what about Professor Mulligan's haircut? If I have mine cut by my wife or a friend shoulc we attempt to create parity by allowing him a dedcution for his in a barbershop? Clearly not. We jsut have to be willing to accept some imperfections in our tax system as it struggels on to do its job in an even more imperfect world.

Posted by: Bill Turnier | Jun 15, 2011 11:50:07 AM

The problem with schemes to reform the tax code along the lines of reducing Schedule A items to pay for a rate cut for at the top is a political problem. As a practioner I know the kind of noise that is going to result when I tell my clients that they can no long right off their interest and taxes and are subsequently going to pay a lot more in tax...but it is OK. Someone who was paying at the 35% rate is not going to get a 30% tax cut for their sacrifice. I would not want to be the congressman who votes for that idea.

Posted by: George W | Jun 15, 2011 10:57:55 AM

Professor Turnier's objection isn't the biggest one-- it seems to be definitional.

Here's the problem with Professor Mulligan's argument. Suppose John buys a house for $100,000 and pays the bank $5,000/year in interest on a mortgage. John gets to deduct the $5,000, reducing taxes, but, Mulligan notes, the bank has to add the $5,000 to its income, raising taxes. So far, so good.

But we have neglected something. John also pays back some of the principal--- say, $3,000/year. That has no tax implication for anybody. He also gets some implicit benefit from living in the house rent-free. That benefit is $8000/year (3000+5000, to be explained later. That, too, has no tax implication.

But we have to compare all of the above to what happens if John rents instead. The rent on that house would be $8000/year, because the landlord has to get the value of the rental benefit in a competitive market. The landlord will pay taxes on that full $8000. John gets no deduction.

Thus, when John buys the house, the $5000 deduction he gets is balanced by the $5000 income, a net income of 0, but when he rents, the net income is $8000.

I may have gotten some of this wrong, but those are the sorts of things that must be considered.

Posted by: Eric Rasmusen | Jun 15, 2011 10:17:31 AM