Mortgage Interest Deduction, part 3

Now that the softball arguments are out of the way, it’s time to tackle what I consider the one real argument for the MID that might have a wee bit of merit.  As our good old friend Lawrence Yun puts it:

One thing that is indisputable is that eliminating the mortgage interest deduction will lower the home ownership rate in the U.S.,” he said. “While we must ensure that the conditions that led to the artificially inflated home ownership rate of the bubble years do not resurface, we also need to create the conditions for sustainable home ownership, which has been shown to provide myriad social benefits for families and communities.

Hold on to your hats, folks, because I’m about to do the impossible, and dispute the indisputable!  In fact, I disagree with his premise, both in a more theoretical sense and from a practical, empirical perspective.  For now, I’ll just look at whether the MID actually does increase the home ownership rate, and later on in the series I’ll come back to whether the government should care about the home ownership rate at all.

Let’s conduct an admittedly grossly oversimplified thought exercise:

Allen, Bob, and Charles are all interested in buying a widget.  Unfortunately for them, there aren’t very many widgets around in their imaginary world, so the three potential buyers are all very excited to learn that a citizen of their town has grown tired of playing with his widget, and will sell it at auction to the highest bidder!

When they arrive at the auction, they each carefully inspect the widget.  Allen, after considering the quality of the widget, looking over his budget, and thinking about how much he wants a widget, decides that he’d be willing to pay up to $1,000 for it.  Bob, going through the same process, decides he’d pay $1,050.  Finally, Charles, being a bit better off than his fellow buyers, or maybe just wanting the widget a little bit more, settles on $1,100 as his maximum bid.

Before the bidding process officially begins, though, the mayor of the town unexpectedly makes an appearance, and proclaims:  “Citizens, as mayor, last year I commissioned an extensive study on the effects of widget ownership on our town.  The results have just come in, and we have found that widget owners are more likely to finish school, earn higher incomes, and live happier, healthier lives.  Therefore, it is my pleasure to announce that, in order to promote widget ownership, the town will offer a reward of $100 to whoever purchases the widget up for sale tonight!”

Allen takes a moment to process this new development, and thinks about how it should affect his bidding strategy.  “Well,” he figures, “I’m still willing to throw in $1,000 of my own money.  But I saw those miscreants Bob and Charles casting their lustful gaze on my precious widget; I don’t want either of those losers to have it!  I’d better just go ahead and bid $1,100.  If I win, I’ll just add the government bonus to my $1,000, so it’s no skin off my nose, and bidding higher gives me a better chance to win!”

Bob and Charles, miscreants and losers though they may be, have a very similar thought process, and decide to increase their bids to $1,150 and $1,200, respectively.  And, sure enough, after the bidding is over, Charles wins.  He pays the owner $1,100 out of his own pocket, and signs over the $100 check from the mayor to the owner as well, for a total of $1,200.

What was the point of this story, other than to demonstrate that I should not quit my day job to write fiction?  Well, think about the finances of this transaction from Charles’s point of view.  Had the mayor not shown up, he’d have bid $1,100, won the auction, and walked away with one widget and $1,100 less in his pocket.  With the mayor’s little bribe subsidy “tax incentive” in play, though, he walked away with one widget and… $1,100 less in his pocket. Curiously, even though the mayor wrote the $100 check to Charles, as far as Charles is concerned, that $100 might as well have not existed!

Hmm, what’s going on here?  Well, it was the nature of the auction that forced Charles to increase his bid.  The reward from the mayor was available to any bidder, and so all the bidders, just by looking after their own self-interests, increased their bid to increase their chances of winning.  And sure enough, had Charles decided to stand by his original bid of $1,100, insisting that he should be able to enjoy the mayor’s reward for himself, Bob’s revised bid of $1,150 would have beaten his.

Now, I’ll leave it as a hopefully easy exercise to translate the elements of this story to the case of real estate transactions.  And yes, I’m fully aware that real estate transactions and taxes are much more complicated than this, but I think the basic point stands.  In fact, it’s not really even controversial.  Remember Mr. Yun’s argument from part 2Home values could fall 15%, says Yun, as buyers discount the value of the mortgage interest deduction in their purchase offers.  Or, if I may re-phrase, keeping the MID makes homes more expensive, since buyers currently do include the value of the MID in their purchase offers.  Of course, he meant that as an argument that repealing the MID would hurt current homeowners, which is true, but the other side of that coin is that keeping the MID doesn’t help buyers, or at least, doesn’t help them as much as it would seem.  Whatever benefit there is to the MID has to be split between the home buyers (keeping their tax savings for themselves) and the home sellers (buyers being willing to include the tax savings in their bids, making prices higher); reality can of course be in between those two extremes, but both sides can’t take all the benefit for themselves.

So, if we’re to base our thinking just on my silly story, the benefit accrues all or mostly to the sellers, simply resulting in higher home prices.  If there’s no or little benefit to buyers, then you can’t really expect the home ownership rate to increase due to the MID.  Why would more people buy widgets due to the mayor’s reward if, like in Charles’s case, the real cost of a widget stays exactly the same with or without the reward?

Now, we can turn our attention to the empirical side of the question.  Does the MID, in practice, lead to an increased home ownership rate?  Or in other words, is my story too much of an oversimplification, and somehow there is a substantial benefit accrued to buyers (in particular, to buyers who are on the fence between buying and renting)?  I think the best I can do here is to report what other, better-qualified researchers have found:  not really.  There’s an excellent paper from Urban Institute which addresses this subject.  It’s long, but most of the paper is about estimating the effects of changing the MID to other tax incentive structures, which I’m not as interested in for reasons I’ve gone over already.  The introductory section, however, is only 5 pages long, and is a great and balanced overview of research that’s been done on the practical effects of the MID.  I highly recommend reading at least that intro, but I’ll quote the most relevant parts here in case you’re already tired of reading due to my general long-windedness on this subject.

Empirical research has found little evidence that the MID increases homeownership. Glaeser and Shapiro (2003) note that the value of the deduction has risen and fallen by tenfold in the past 50 years while homeownership rates have remained nearly unchanged between 63 and 68 percent. Their formal analysis of time-series data finds no effect of the MID on homeownership rates. Cross-national comparisons in Gale, Gruber, and Stephens-Davidowitz (2007) also conclude that the mortgage interest deduction is not correlated with higher homeownership rates. Gale (1997) finds that, similar to the U.S. experience, changes in the value of the MID in the United Kingdom are not associated with changes in the homeownership rate. Mann (2000) compares homeownership rates and mortgage interest tax treatment across ten countries and also finds no consistent relationship between homeownership and the interest deduction.
The MID also has the potential to affect homeownership through its impact on home prices. The subsidy makes individuals willing to pay a higher price for the same home, but the resulting effect on home prices depends on the elasticity of the housing supply. The effect of the deduction on house prices is likely to vary across and within regions due to differences in tax rates and the availability of undeveloped land. The deduction will raise prices more in densely populated areas with low housing-supply elasticities and high tax rates. At the extreme, Capozza, Green, and Hendershott (1996) estimate an upper-bound estimate of the effect on prices at 10 percent, assuming that the housing stock is totally inelastic.

The indirect effect of the MID on homeownership through higher home prices could fully offset any direct effect from the increased demand to own homes. For example, Bourassa and Yin (2007), in an analysis of urban adults between 25 and 34 years old, estimate that eliminating the mortgage interest deduction would increase, instead of decrease, their homeownership rate from 41.5 to 42.5 percent.6 This occurs because the positive effect on ownership by young adults through lower home prices more than offsets the negative effect from the loss of the deduction. This study also illustrates the regional variation in the effect of the deduction. Cities with relatively high-priced, space-constrained markets, such as San Francisco, experience larger increases in homeownership rates when the deduction is eliminated than cities like Birmingham and Tampa, where the price effect is minimal. The overall estimated change in the homeownership rate from eliminating the MID, however, is fairly small.

2 quick asides:

  1. I found that at least a couple of the papers they refer to are freely available.  In particular, my most patient and determined readers can find Glaeser and Shapiro’s paper here.
  2. Spelling nerd moment:  much of the literature seems to use “homeownership” as if it’s one word, but wordpress doesn’t recognize it as a word, and I don’t like seeing those red squiggles everywhere, so I separate it into 2.

I think there are lots of interesting things to discuss here, but I’ll focus on one new idea I particularly like of cross-national comparisons of home ownership rates.  Housing-related policies, as you might expect, vary quite widely by country, so there are plenty of developed countries that have tax policies fairly similar, in this respect at least, to ours, and also quite a few which have no housing subsidies of this kind.  The fact that there is not a clear relationship between home ownership rate and housing tax incentives is, I think, quite telling.  I have to be careful here, though:  People often mindlessly spout off the old truism “Correlation does not imply causation,” but less well-known is the fact that a lack of correlation does not imply lack of causation, either.  But I think trying to argue that there still is a strong causal effect, despite the lack of a clear correlation, would back you into a pretty tricky corner; you’d end up needing to argue that there’s some other factor affecting home ownership rates which just happens to affect countries with and without housing subsidies very differently.  Possible, but I sure haven’t seen a convincing suggestion.  Going into too much detail here would probably involve writing an entire separate post on correlation/causation, and this has already gotten long enough.

Whew!  I doubt anyone is still reading at this point, but to summarize:  Both from the perspective of simplistic economic reasoning, and through quite a bit of empirical research, it doesn’t look like the MID is very effective at increasing home ownership rates much.  Might there be some other tax incentive structure which would be more effective at doing so?  Sure, but as I’ll get to in part 4 (which, I’m starting to think might actually be 2 parts), I’m not really convinced that promoting home ownership is a worthwhile goal for the government to pursue, at least not by using direct tax incentives.

2 thoughts on “Mortgage Interest Deduction, part 3

  1. I like thought exercises and think they’re underutilized. But I also think we often don’t, in a thought exercise, go beyond the immediate affects, and think about how things shift after that.

    In your example – buyer out of pocket is same, seller gets more money by the amount of the government incentive. But what happens after that? The higher price for sellers, supported by the incentive should bring more inventory onto the market.

    I’m really rusty on this stuff, but I think in a simple supply/demand graph, the demand line shifts right. Prices go up, but so does quantity. There is a new equilibrium: higher home ownership.

    Now you can go further: if there is excess rental capacity, rents may go down, undercutting the impact of the home buying incentive.

    Still, it does seem to me theoretically possible that MID leads to an increase in home ownership over what it would be otherwise. I’m not saying that’s a good thing, but it seems like it could be an outcome. But I also think it does depend on supply elasticity, etc. Long story short – perhaps the thought experiment doesn’t help out here given that MID has been around a long time.

    • Oh, definitely. This story is way too simplistic to capture all the economic effects of the MID. Pinning down the price elasticity of supply of housing, as well as the weird relationship between rents and home prices is a little too tricky for me to grasp completely (though I might have more to say on it later).

      In the end, I rely more on empirical research comparing home ownership rates between countries with and without subsidies to determine the total effect of policies like the MID, though every such comparison is apples to oranges in some way. I just consider this thought exercise to be a way to make sense of the results of those comparisons, at least in my mind.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s