Friday, December 15, 2017

Share Share    Print Print    Email Email

What the Mortgage Interest Deduction can Teach us About Government Failure

Is it hypocritical for a business or a politician to publicly oppose a government program only to turn around and ask for a share in it? Stephen Koff of the Cleveland Plain Dealer posed this question to me a few weeks ago. Likening it to the mortgage interest deduction, I said I didn’t think so.

I oppose the mortgage interest deduction. It pads the pockets of housing-industry special interests. It puts pressure on marginal tax rates to rise in order to make up for the lost revenue. And it likely has little impact on the incidence of home ownership since the value of the deduction is capitalized into the price of homes. (Even if it worked as intended and didn’t end up being capitalized into the value of home prices, it would be a regressive privilege for relatively wealthy home-mortgagers). For all these reasons, I—like most economists—oppose the mortgage interest deduction.

But come April 15, I take the deduction. And though I can’t say for certain, I suspect the same is true of most economists. Why?

The main reason is that not taking the deduction would have approximately zero impact on the problems I mentioned. Those problems only go away if the whole policy goes away. So, I do what I can to expose its faults when I talk to journalists but I don’t forgo it myself.

Whether or not you find this hypocritical, it is an empirical fact that lots of people see things this way. More importantly, this phenomenon is at the heart of the public choice critique of government failure. It explains why bad policy exists.  

Consider pork-barrel spending, as modeled in the prisoner’s dilemma (economists should skip the next two paragraphs). Imagine a community of two constituents. And imagine that each has the choice of two options: take pork or abstain from pork. Taking pork yields a private benefit of $10 for the taker. But because there are deadweight losses associated with taxation, $10 in pork will cost the community $12 in taxes and unrealized economic gain. If this cost is split evenly between the two constituents, and Constituent A is the only one who takes it, then he obtains $4 = $10 – 0.5*$12 (his gain, minus his share of the cost). Constituent B, however, only gets the cost of A’s pork: -$6 = -0.5*$12. The situation is reversed if B takes and A abstains. If both take, then each pays -$2 = $10 – 0.5*24 (the value of the pork minus the cost of paying for two peoples’ shares). Lastly, if both abstain, neither is taxed and neither obtains a benefit. The table below shows these outcomes. The first number indicates Constituent A’s payoff, while the second indicates Constituent B’s.

Player B
Abstain Take Pork
Player A Abstain 0.0 0.0 -6.0 4.0
Take Pork 4.0 -6.0 -2.0 -2.0


Irrespective of what Constituent B does, it always makes sense for Constituent A to take the pork. If B abstains, then A should take it because he gets $4.00 instead of $0. And if B takes, then A should also take because losing $2.00 is better than losing $6.00. Similarly, it is always in B’s interest to take. So the “equilibrium” of the game is for both A and B to take pork and for them both to be worse off than if neither took.

The point of the exercise is to show that the incentives of the system lead people to a socially suboptimal outcome. If they could somehow change the entire system—say by prohibiting taxes that fund special interests instead of general welfare—then they could get to the optimal outcome. But without changing the incentives of all players, it makes little sense for any one person to act against his or her interest.

MSNBC has lately taken to airing commercials that highlight federal funding for parochial projects. The commercials are apparently supposed to convince people that the federal government should fund all sorts of local project. As a one-time resident of Arizona I’m sure I benefited from the electricity generated at Hoover Dam. And whenever pork-barrel projects are considered, you can generally count on the local constituents who benefit from them to support them. But that doesn’t mean that the residents of the 48 states other than Arizona and Nevada should have had to pay for the Dam. In fact, the simple model of the prisoner’s dilemma teaches us that the incentives of such a system can lead to suboptimal outcomes.

I take the mortgage interest deduction. And the problem is that I—like every other homeowner—am incentivized to do so, even though the total costs outweigh the total benefits.