Tax Analysts®Tax Analysts®

My Subscriptions:

Featured News

April 10, 2012
The Fake Third Rail of Tax Reform

Full Text Published by Tax Analysts®

by Dennis J. Ventry Jr.

Dennis J. Ventry Jr. is a professor at the University of California Davis School of Law.

This report condemns the mortgage interest deduction for its inequities, inefficiencies, ineffectiveness, and cost. It also exposes the deduction as the fake third rail of tax reform and provides policymakers with ample justification for putting the tax code's sacred cow out to pasture and replacing it with a tax credit for homeownership.

Copyright 2012 Dennis J. Ventry Jr.
All rights reserved.

* * * * *

Table of Contents

I. The Inequities of the MID

II. The Inefficiencies of the MID

III. The Costs and Benefits(?) of the MID

    A. The MID and the Rate of Homeownership

    B. The MID and Social Capital

    C. The MID and Housing as a Good Investment

IV. Reforming and Replacing the MID

    A. Slowly Phase Out the MID

    B. Capping the MID for High-Income Taxpayers

    C. Replacing the MID With a Tax Credit

V. Conclusion

Dennis J. Ventry, Jr. The mortgage interest deduction (MID) has long been considered a "symbol"1 of the American Dream, both a facilitator and protector of homeownership. It has been described as the "third rail of tax reform,"2 the "most sacred tax break in the code,"3 a member of the "Holy Trinity of U.S. social programs,"4 and "an American birthright"5 so "sacrosanct"6 that the "mere thought of tampering with it was unpatriotic."7

In fact, touching the MID is not just treasonous but ruinous. According to the subsidy's most resolute defenders, namely the housing lobby led by the National Association of Realtors8 and the National Association of Home Builders,9 disturbing the MID for any reason would induce housing prices to plummet by 15 percent and result in "trillions of dollars in wealth destruction."10 Attacking the MID, moreover, would be tantamount to "a de facto tax increase on the middle class."11 The subsidy has "lowered the cost of ownership," its supporters insist, boosted the nation's homeownership rate,12 and been "vital to the stability of the American housing market and economy."13 The taxpaying public seems to have embraced these sweeping positive connections between the MID and the housing market, with nearly three-quarters of respondents to opinion polls expressing support for the deduction.14 (It is worth noting that the housing lobby sponsors many of these polls, and surveys conducted by independent organizations reveal considerably less public enthusiasm for the MID.15)

But what if it turned out that the beneficial effects of the MID on homeownership were spurious? That is, what if the subsidy failed to accomplish the stated goal of promoting and protecting widespread homeownership and, in fact, lowered rates of homeownership? Or what if the subsidy had the effect of destabilizing rather than stabilizing the national economy, and of producing less rather than more national wealth? In other words, what if it turned out there was no valid reason to consider the purported third rail of tax reform untouchable?

This report argues that notwithstanding the assertions of its powerful and well-financed supporters, the MID is in fact the most inequitable, inefficient, and ineffective tax expenditure provision. It is also the second most expensive, costing $100 billion each year.16 It is time for American taxpayers and policymakers to understand that the fake third rail of tax reform helps only one-quarter of all households, distorts the allocation of capital in our economy and thereby reduces economic growth and national wealth, and may even lower the rate of homeownership. By the same token, it is time for Congress to put the tax code's sacred cow out to pasture and replace it with a policy alternative -- a tax credit for homeownership -- that accomplishes everything the MID purports to do but fails.

I. The Inequities of the MID

If we assume that promoting homeownership is the desideratum of national housing policy (rather than, say, dignified and affordable housing), the MID is precisely the wrong vehicle (it would also be the wrong vehicle for promoting dignified and affordable housing). It is the classic upside-down subsidy with its primary beneficiaries being taxpayers who would own homes even in the absence of the subsidy rather than taxpayers residing on the margin between owning and renting. According to the Joint Committee on Taxation, taxpayers reporting incomes exceeding $200,000 reflect just 2.7 percent of all tax filers17 but receive 35 percent of the MID's total tax benefits (that is, the actual amount of taxes saved for taxpayers and the actual amount of revenue lost for the government).18 Moreover, taxpayers reporting incomes exceeding $100,000 (12.4 percent of filers) capture 78 percent of the benefit, while those reporting incomes greater than $75,000 (20.6 percent of filers) receive 89 percent.19 That leaves very little for everyone else, including taxpayers reporting incomes less than $50,000 (a group representing two-thirds of all tax filers), who take home just 2.7 percent of the MID's annual tax benefits.20

Additional empirical work by economists underscores the upside-down effects of the MID. According to James Poterba and Todd Sinai, the deduction for mortgage interest delivers 10 times the tax savings for households with incomes exceeding $250,000 as for households with incomes between $40,000 and $75,000.21 Moreover, the regressive features of the MID have worsened over the last 20 years. In 1987 households earning less than $50,000 took home 48 percent of the tax savings provided by the MID.22 By 2010 and after adjusting for inflation, households earning less than $100,000 (roughly the equivalent of $50,000 in 1987) received just 21.7 percent of the MID's tax benefits.23

Several factors explain the inequitable distribution of the MID's largesse.

First, the subsidy takes the form of an itemized deduction. To reap the tax benefits of itemization, the sum of a taxpayer's total itemized deductions (such as the MID, property taxes paid, state and local taxes paid, and qualifying charitable contributions) must exceed the dollar value of the standard deduction. For 2011, that amount equaled $11,600 for a married couple and $5,800 for a single person.24 In any given year, only one-third of all taxpayers itemize while the remaining two-thirds claim the standard deduction.25 Thus, members of the two-thirds majority -- homeowner, renter, or squatter -- are denied tax savings from the MID.

Second, higher-income households claim a disproportionate share of itemized deductions compared to lower- and middle-income households. Only 15 percent of tax units with incomes less than $50,000 (a cohort that includes more than two-thirds of all taxpayers and a figure that approximates the median family income for 201026) report itemized deductions.27 Meanwhile, 76 percent of tax units with incomes between $75,000 and $200,000 itemize, and 96 percent of returns reflecting incomes more than $200,000 report itemized deductions.28

Third, the value of the deduction for mortgage interest, or of any deduction for that matter, depends on the taxpayer's marginal tax rate (that is, the rate imposed on the last dollar earned). A taxpayer's marginal rate, in turn, depends on the size of a taxpayer's income, with increasingly higher rates of tax levied on increasing increments of income. Thus the MID, like other deductions, delivers greater dollar-for-dollar benefits to high-income households than to low- and middle-income households (unlike a tax credit, which delivers the same dollar-for-dollar benefit to all qualifying claimants).

For example, consider three married households earning three different levels of income: $50,000 (again, below which two-thirds of taxpayers reside and the national median family income); $75,000 (an income level that we will consider representative of the "middle class," even though four-fifths of all taxpayers report income below that level); and $225,000 (an income that falls into one of the top two brackets and part of the political discussion over capping the value of deductions for high-income households29). Under our current tax rate schedules,30 the marginal rate imposed on the three households is, respectively, 15, 25, and 33 percent. Consequently, the value of a dollar of itemized deduction varies considerably, from 15 cents for the $50,000 household to 25 cents for the $75,000 household and 33 cents for the $225,000 household.

The incidence of the MID reflects these distributional inequities, which are inherent in delivering tax benefits in the form of a deduction. According to the most recent tax return data, only 26 percent of tax units in 2009 claimed the deduction for mortgage interest paid.31 An even lower percentage of filers (21 percent) benefited from the MID after accounting for the number of nontaxable returns claiming the deduction.32 In addition, as with itemized deductions generally, higher-income households claimed the MID at significantly elevated rates compared to middle- and lower-income households. Nearly 78 percent of taxpayers with incomes exceeding $200,000 claimed the MID, while only 15 percent of taxpayers below $75,000 and 10 percent below $50,000 claimed the deduction.33

Higher-income households capture a disproportionate share of the MID not just as a percentage of taxpayers claiming the deduction, but also in terms of total dollar amounts claimed. Households with income above $200,000 represent 2.7 percent of all taxpayers, but they claim nearly 16 percent of the total amount of MIDs reported.34 Moreover, households with income exceeding $100,000 snag 47 percent of the total but represent just 12.4 percent of all taxpayers, while returns reflecting income greater than $75,000, or the top 20.6 percent of all taxpayers, claim nearly 64 percent of mortgage interest paid.35 At the same time, taxpayers with income below $50,000 report less than 20 percent of the value of all MIDs claimed.36

Wealthier households also nab greater benefits per individual return from the MID due to their generally larger homes and bigger loans. Taxpayers with more than $200,000 in adjusted gross income claim, on average, nearly $22,000 in mortgage interest paid, while taxpayers reporting up to $75,000 claim, on average, $9,000.37 Higher incomes enjoy still greater per-return benefits with households reporting income exceeding $1 million (or 0.17 percent of all returns) claiming an average of nearly $33,000 in mortgage interest paid.38

Add to the discrepancy in absolute benefits the additional discrepancy associated with delivering the tax subsidy in the form of a deduction, and the inequity widens. Take a household earning $75,000 and paying $9,000 in mortgage interest. The MID reduces this household's tax liability by $2,250 ($9,000 x 0.25) or 3 percent of total income. Comparatively, a household earning $225,000 and paying $22,000 in mortgage interest could save $7,260 in taxes thanks to the MID ($22,000 x 0.33) or 3.2 percent of total income. As between these two households, the MID worsens both the absolute and relative income disparity.

Defenders of the MID routinely ignore the effect of marginal tax rates on the distribution of the deduction's benefits.39 They would say of the two taxpayers in the preceding paragraph, for example, that the $75,000 earner with $9,000 in mortgage interest received a larger benefit from the MID than the $225,000 earner with $22,000 in mortgage interest, because the MID claimed by the $75,000 earner equals 12 percent of total income versus 9.8 percent of total income for the $225,000 earner. But that calculation ignores the real-world effect of marginal tax rates on the value of deductions (which, as we just witnessed, increased the relative income disparity between the $75,000 and $225,000 households), and it certainly is not the way the IRS calculates the deduction. It also ignores the effect of the standard deduction on the "net benefit" of the MID. While the amount of mortgage interest paid might push lower- or middle-income taxpayers past the tax-free threshold of the standard deduction, which would allow them to start itemizing, it might not push them very far past the threshold. For these taxpayers, the tax benefit of the MID is negligible or nonexistent. Let me explain.

Measuring the net benefit of deductions like the MID provides a more accurate distributional analysis of the relative tax savings across the income spectrum. The calculation is straightforward and requires us to (i) take the dollar amount of mortgage interest paid by the taxpayer; (ii) determine by how much (if at all) that dollar amount, when added to the dollar amount of the taxpayer's other itemized deductions, exceeds the standard deduction; and (iii) multiply that amount by the taxpayer's marginal tax rate.

For consistency, we will use our two households from above, one with $75,000 in annual income and $9,000 in mortgage interest, and the other with $225,000 in annual income and $22,000 in mortgage interest. In addition to claiming the MID, the $75,000 household claims $3,000 in property taxes, $4,000 in state and local taxes, and $1,000 in charitable contributions. This household would itemize its deductions rather than claim the standard deduction of $11,600 because its itemized deductions of $17,000 exceed the standard deduction by $5,400. The value of the MID to this taxpayer, however, is not the full $9,000 of interest paid nor even that figure multiplied by the taxpayer's marginal tax rate. This household would have been able to shield income up to the value of the standard deduction with or without the MID or itemization. To ascertain the net benefit of the MID for this household, we need to measure by how much its total itemized deductions (including its mortgage interest of $9,000) exceed the standard deduction, or $5,400, and multiply that amount by the household's marginal tax rate of 25 percent for a net benefit of $1,350 ($5,400 x 0.25) or 1.8 percent of total income.

As for the $225,000 household, in addition to mortgage interest of $22,000, it claims $6,000 in property taxes, $9,000 in state and local taxes, and $2,000 in charitable contributions for total itemized deductions of $39,000, well above the $11,600 standard deduction threshold. The net benefit of the MID to this household equals the full $22,000 of mortgage interest paid (because the sum of its other itemized deductions exceeds the standard deduction), multiplied by the household's marginal tax rate of 33 percent, or $7,260 ($22,000 x 0.33), which equals 3.2 percent of total income.

Thus, the net benefit of the MID as between the middle-income and high-income households in our example is skewed decidedly against the middle-income taxpayer. In other words, the MID worsened both the absolute and relative income disparities between these two households once we properly accounted for marginal tax rates and the excess of itemized deductions over the standard deduction. The high-income taxpayer saved $7,260 in taxes from the MID versus $1,350 for the middle-income taxpayer, which represented 3.2 percent of total income for the high-income taxpayer versus 1.8 percent for the middle-income taxpayer.

In the end, the net benefits of the MID are so skewed toward higher-income households that it is effectively worthless to lower-and middle-income taxpayers.40 Almost any other policy aimed at subsidizing owner-occupied housing would be more effective and would increase the progressivity of the income tax.41 Currently, the MID offers no help to three-quarters of all taxpayers42; no help to two-thirds of taxpayers who claim the standard deduction43; no help to more than half of all homeowners44; no help to more than 20 percent of mortgaged homeowners45; no help to renters; and very little help to the elderly who no longer service mortgages or who have too little taxable income to enjoy any savings from the deduction.46

The inequities associated with the MID extend beyond differences between income cohorts, owners versus renters, and elderly versus young households. They also include disparate treatment across different regions of the country and among different races.

Regarding regional disparity, the tax benefits of the MID accrue disproportionately to wealthy cities and states. The average per capita tax benefit for the 20,000 residents of Beverly Hills, Calif., for instance, amounts to $1,873, while the average benefit for the 20,000 residents of Clarksdale, Miss., totals a mere $45.47 Similarly, residents of Atherton, Calif., a wealthy suburb of San Francisco, enjoy per capita MID benefits of $2,400 compared to $7.61 for residents of Oklahoma City (as well as $7.45 in Erie, Pa.; $7.94 in Milwaukee; $12.01 in El Paso, Texas; $12.50 in Rochester, N.Y.; $13.25 in Memphis, Tenn.48) You get the point. So much so that you would not be surprised to learn that three metropolitan areas receive more than 75 percent of MID benefits: (i) New York City-Northern New Jersey, (ii) Los Angeles-Riverside-Orange County, and (iii) San Francisco-Oakland-San Jose.49

For the same reasons that tax savings associated with the MID skew heavily toward wealthy cities (higher incomes, larger loans, more mortgage interest paid, higher marginal tax rates), they disproportionately advantage wealthy states. The distribution of the MID among the states, observers have demonstrated, also benefits residents of blue states considerably more than residents of red states. Maryland ($499 MID per capita), California ($464), Connecticut ($446), Virginia ($438), New Jersey ($436), and Massachusetts ($393) all enjoy significantly higher MID benefits than Oklahoma ($131), Arkansas ($119), South Dakota ($118), North Dakota ($110), Mississippi ($108), and West Virginia ($102).50 In the 2008 U.S. presidential race, Democratic candidate Barack Obama won the popular vote in 14 states with the highest per capita MID benefit, as well as 19 out of the top 20 states.51 Meanwhile, his Republican opponent, John McCain, won the vote in the six states with the lowest per capita MID benefit, as well as nine out of the bottom 10 and 14 out of the bottom 16.52

As to racial disparity, researchers have shown that the MID discriminates against minority taxpayers.53 These households generally report lower incomes, lower rates of homeownership, and lower home values than non-minority households. As a result, they are more likely to take the standard deduction than to itemize deductions, less likely to claim the MID, and less likely to receive meaningful net benefits from the MID. In addition, research indicates that in certain housing markets, homeowners act like "local cartels" and restrict entry into the market based on various factors, including race.54

II. The Inefficiencies of the MID

While the inequitable effects of the MID are both far-reaching and overwhelming, the subsidy's inefficiencies might very well outweigh its inequities. Researchers have long condemned the MID for distorting the housing market by artificially propping up home prices and creating a false baseline for the cost of housing, encouraging taxpayers to pay for homeownership with debt rather than cash or financial assets, causing wasteful and unproductive misallocation of physical capital, and dragging down the American economy. Economists of every ilk consider the MID "a huge subsidy that causes massive, efficiency-draining distortions in the economy," which creates "less business capital, lower productivity, lower real wages, and a lower standard of living."55 The macroeconomic effects of the MID are so destructive that every economist (excluding only those employed by the housing industry) believes "the most sure-fire way to improve the competitiveness of the American economy is to repeal the mortgage interest deduction."56 This section inventories the many inefficiencies of the MID.

The MID distorts the cost of owner-occupied housing relative to other investments,57 and thereby contributes to overinvestment in the asset class and economy-wide misallocations of capital stock.58 According to economist Kevin Hassett, the MID tells taxpayers, "Don't build a factory, build a mansion."59 The distortion caused by housing tax policies (of which the MID is by far the most prominent) is so strong that they might account for 50 percent of all distortions from misallocated capital in the economy,60 and a 30 percent larger housing stock than an optimal allocation would produce.61 (It is worth noting that these estimates, although disturbingly large, preceded the boom in housing stock created by the caffeinated housing policies of the 1990s and 2000s.) The overinvestment in housing has diverted capital away from productive assets to such an extent that it shrinks U.S. GDP by as much as 10 percent.62

Thanks to the MID and other housing tax subsidies (primarily the exclusion of net imputed rental income, the deduction for property taxes on owner-occupied homes, and the exclusion of capital gains on the sale of primary residences), housing enjoys a significant tax advantage over other capital investments. In fact, the economy-wide tax rate on housing investment is close to zero, while the tax rate on corporate investment exceeds 30 percent.63 At certain points over the last 50 years, moreover, housing has enjoyed a negative rate of tax.64

The MID distorts not only where to invest but also how to invest. Specifically, it alters the decision to pay for homeownership with debt rather than with financial assets or cash. Thus, the MID distorts the allocation of financial capital in the same way that it distorts the allocation of physical capital. Research shows that if homeowners used other means to finance home purchases, including financial paper assets such as savings accounts and brokerage accounts, they could reduce household mortgage debt by almost 30 percent.65 If they deployed all non-housing assets in shifting away from mortgage debt, they could reduce such obligations by 70 percent.66 By subsidizing mortgage debt, the MID encourages home buyers to take on massive debt and precariously high loan-to-value ratios.67 Thanks in large part to the MID, mortgage indebtedness soared in the decade preceding the recent collapse of the housing market, with mortgage debt as a percentage of GDP rising from 47 percent in 1995 to 81 percent by 2007.68 That excessive leverage "played a prominent role in the credit meltdown."69

By creating artificial demand for owner-occupied housing through subsidized mortgage debt, the MID also encourages home buyers to consume larger and more expensive homes than they would purchase in the absence of the subsidy.70 It thereby encourages suburbanization and decentralization of metropolitan areas as housing markets respond to the demand for bigger and costlier homes.71 In fact, econometric research has found that without the MID and other housing tax subsidies, American taxpayers would purchase homes that were 9 to 17 percent less costly.72 Researchers have also found that the MID raises the cost of housing by as much as 10 percent,73 although the figure is probably closer to 3 to 6 percent as upper bound estimates assume an inelastic supply of housing.74 Even a 5 percent effect would mean that the MID is grossly overcapitalized into the cost of housing. In a well-functioning market, one dollar of subsidy would result in a price increase that approximated one dollar as the market adjusted to the subsidy and reached a new equilibrium. But in the housing market, historically a poorly functioning market, one dollar of MID subsidy seems to count for more than 100 cents such that the net effect of the MID is to raise rather than lower the cost of housing.75

Artificially bidding up the cost of housing helps no one (except perhaps the housing industry and then only in the short term). Higher prices prevent millions of potential home buyers from entering the housing market. And even though current homeowners may express a preference for hyper-inflated prices (either for wealth effects or to maximize gain upon sale), the perceived benefit to sellers is illusory as sellers become buyers in the same overheated market. The only reason to desire higher home prices in and of itself would be if homes were viewed as viable investments. But as we will see in Section III.C, while a home may serve as a decent savings account, homeownership amounts to an irrational long-term investment, particularly if wealth creation is the goal. Suffice it to say that in the same way subsidized mortgage debt distorts the allocation of capital economywide, it also distorts the allocation of individual households' investment decisions.

Two other distortions caused by the MID deserve mention, both of which involve the subsidy's destabilizing effect on U.S. labor markets. First, the MID restricts labor mobility and contributes to higher rates of unemployment.76 Recent findings indicate that labor immobility caused by the MID and homeownership increased unemployment nationwide by as much as 2 percent during the last recession.77 In addition to creating an inflexible labor force, the MID further erodes economic stability due to "dramatic swings in employment associated with the construction industry and the volatility of housing investment."78 It is no wonder that economists -- who recoil against unnatural price supports, artificially depressed debt yields, misallocated capital, wasted resources, aberrant behavioral distortions, and lower rates of economic growth -- condemn the MID.79

III. The Costs and Benefits(?) of the MID

The inequities and inefficiencies of the MID cost the American economy trillions of dollars every year by rewarding select households, distorting individual choices, and allocating capital away from more productive uses. In addition to these economywide losses, the MID costs the federal government critical tax revenues, creating a shortfall that must be replenished by higher taxes on all taxpayers. In 2011, more than 170 tax expenditures in the Internal Revenue Code reduced federal tax receipts by nearly $1.3 trillion,80 or as much as the entire federal deficit.81 The MID ranks as the second most expensive tax expenditure item, costing $100 billion.82 Over the next five years, the MID will drain $600 billion from the federal Treasury.83 And while the tax expenditure budget contains additional items that subsidize owner-occupied housing, the MID is by far the most expensive.84 All told, the four largest tax expenditures related to housing reduce federal tax revenues by almost $200 billion every year, and by $1.26 trillion over the next five years.85

Notwithstanding the MID's various and substantial costs (inequities, inefficiencies, and lost tax revenue), its defenders extol a list of purported benefits in an effort to justify its existence. Admittedly, if the MID accomplished its stated goal of raising the homeownership rate, and if government intervention to promote homeownership was deemed a worthwhile national policy, then we might be inclined to overlook the MID's considerable costs. Similarly, we might look the other way if the MID produced the kind of social benefits associated with homeownership celebrated by its supporters (such as higher rates of community and political activism, lower rates of crime and juvenile delinquency, and elevated student achievement). Finally, we might view the many costs of the MID in a different light if the subsidy provided Americans a road toward wealth creation on par with other investment opportunities. Below we consider the relationship between the MID and these ostensible benefits.

A. The MID and the Rate of Homeownership

Substantial empirical research over the last 30 years shows that the MID has "almost no effect on the homeownership rate."86 Tax policies seeking to promote homeownership "should emphasize the purchase decision, not the quantity decision."87 To further influence the purchase decision and to reduce unnecessary distortions, moreover, the policies "should be only the minimum amount necessary to switch people from renting to ownership," and "should not be available for anyone who would buy a house anyway."88 Even if we assume that increasing the rate of homeownership is a desirable policy goal, the MID "is not a cost effective tool," because "its main beneficiaries are not individuals on the margin between renting and owning."89

Both the historical and cross-country comparative data underscore the ineffectiveness of the MID in promoting homeownership. At the end of 2011, the homeownership rate in the United States stood at 66 percent, down from a high of 69.2 percent at the end of 2004 but still high by historical standards.90 In 1940, only 43.6 percent of Americans owned homes.91 Over the next 20 years, the number of owner-occupied households experienced extraordinary growth thanks to direct government intervention in the housing and mortgage markets by New Deal agencies such as the Federal Housing Administration and postwar agencies such as the Veterans Administration and its mortgage insurance program.92 The rate of homeownership rose steadily and rapidly to almost 62 percent by 1960, and then basically flat-lined for the next 30 years.93

Recent research suggests that the dramatic postwar increase in homeownership between 1940 and 1960 can be explained primarily by a change in the demographics of homeownership. Federal lending policies shifted the age profile of homeownership by decreasing the age of entry into owner-occupied housing. In other words, Americans became homeowners at a younger age and not necessarily at a higher rate.94 The rapid increase in homeownership during this period was accompanied by an increase in the number of taxpayers paying and deducting mortgage interest. But neither policymakers nor taxpayers yet considered the deduction for mortgage interest as integral to national housing policy or homeownership.95

The relatively stagnant rate of homeownership in the 30 years following the remarkable postwar period of growth began to improve only after the successive two-term administrations of Presidents Clinton and Bush made raising the rate of homeownership an explicit policy priority.96 The resulting overaggressive government intervention in the housing and mortgage markets (which included relaxing lending standards for Fannie Mae and Freddie Mac, encouraging securitization of risky mortgages, endorsing subprime lending, and authorizing down payments close to zero) created the housing bubble as well as its subsequent burst, which brought the rate of homeownership back to historical levels.97

Over this latter period, the value and cost of the MID grew considerably, but the rate of homeownership changed little. In 1986, the percentage of homeowners stood at 63.9 percent (negligibly higher than the rate in 1960),98 and the MID provided tax savings to select homeowners of $27 billion.99 Between 1986 and 1996, the subsidy doubled in cost,100 while the rate of homeownership ticked upward by only 1.5 percentage points to 65.4 percent.101 The cost of the MID doubled again over the next 15 years with its price tag topping $108 billion in 2010,102 the same year the rate of homeownership equaled 66.5 percent.103 Thus, between 1986 and 2010, the MID quadrupled in cost but homeownership grew a meager 2.6 percentage points.

Regarding cross-country comparisons of homeownership, one would expect the United States to boast higher percentages than other countries given the extent to which the U.S. government subsidizes owner-occupied housing (not only through tax expenditures but also with the government-sponsored enterprises, Fannie and Freddie, whose operations unnaturally depress mortgage prices104). The numbers reveal that the United States is somewhat of a laggard when it comes to owner-occupied housing, however. There is great variation across countries regarding homeownership, with only 37.5 percent of Swiss citizens owning their homes and over 90 percent of citizens in several former socialist republics achieving the "American" dream.105 For its part, the United States hovers near the middle of the pack at 66 percent.106 Among the 27 member countries of the European Union, moreover, the average rate of homeownership equals 73.5 percent, a rate never achieved in the United States.107

Nor is there any cross-country evidence of a correlation between a deduction for mortgage interest and rates of homeownership (much less a causal connection). For example, Switzerland permits taxpayers to deduct mortgage interest but it reflects a low rate of homeownership (37.5 percent), while Australia reports a relatively high rate of homeownership (68.3 percent) but does not allow taxpayers to deduct mortgage interest.108 Meanwhile, Canada (68.4 percent in 2008) reports a rate of homeownership similar to the United States (67.5 percent in 2008), but its tax system does not provide a mortgage interest deduction.109 Additional comparative research reinforces the conclusion that a tax deduction for mortgage interest does not correlate with increases in homeownership.110

In the end, based on both empirical and observational evidence, the best we can say about the influence of the MID on rates of homeownership is that it is "not . . . particularly effective in altering the choice between renting and owning."111 In fact, according to some researchers, it might even reduce the supply of housing to the extent it provides homeowners an incentive to restrict supply in order to raise prices.112

B. The MID and Social Capital

If the MID has "almost no effect" on rates of homeownership, then it cannot be said to generate the social benefits that allegedly flow from owner-occupied housing.113 However, it is worth considering these purported benefits as part of a more fundamental inquiry into whether the government should subsidize homeownership in the first place. As commentators have observed, the "main argument for subsidizing homeownership is that ownership may provide positive spillover effects to individuals other than the owner."114 The housing industry touts these ostensible benefits in celebrating the virtues of homeownership.115 If they exist, such widespread public benefits would supply an argument for preserving and even extending current housing subsidies despite the many costs described in this report associated with subsidizing owner-occupied housing (that is, the inequities, inefficiencies, ineffectiveness, and foregone revenue).

In both theory and practice, homeownership may produce positive social effects that create public benefits accruing to society at large rather than merely accruing to individual homeowners. In other words, homeownership may create social capital. Owning rather than renting might provide independent incentives to improve the quality of one's neighborhood and community, participate in local politics, and establish an environment inhospitable to crime. In fact, researchers have found correlations between these kind of social benefits and homeownership. Studies have shown that homeownership (compared to renting) correlates with higher rates of participation in community and political activism,116 as well as positive effects on children's well-being and behavior.117 Studies have also found a negative correlation between homeownership and the incidence of crime.118 Still other research suggests that homeowners take better care of their homes regarding internal and external improvements.119

None of the studies, however, identify a causal connection between homeownership and generating social capital.120 Children of homeowners might be found to exhibit lower rates of juvenile delinquency or truancy than children of renters, but that does not mean that renting will land your child in juvenile hall or that owning will get your child into Harvard. It may simply mean that homeowners as a group are wealthier than renters and can afford to supplement their children's upbringing with educational environments that provide optimal structure and opportunities. Such a scenario would correlate homeownership with staying in school and student achievement. But it does not mean that homeownership causes those things. The same can be said of the observed correlation between homeownership and lower rates of crime. It turns out there are fewer single-family houses and more rental units in densely populated urban areas where land values are generally higher than less densely populated areas. It also turns out that urban centers and cities suffer from higher rates of crime. But that observation merely correlates renting with crime. It does not mean that renting is a cause of crime or that homeowning magically deters crime.

The challenge for researchers attempting to isolate effects of homeownership has been to account for the methodological biases of correlations that encourage homeownership versus those that generate social capital. Several studies have recognized that failing to account for these correlations produces spurious connections between homeownership and spillover effects. These studies have attempted to account for these "endogenous" variables (that is, those variables whose value is dependent on other variables in a model) and to identify and study the "exogenous" variables (that is, those variables whose value is independent of other variables in the model) to isolate causal influences of homeownership.121 To varying degrees, these studies have located exogenous instruments free of observed and unobserved endogeneity.

The researchers that have come closest to employing exogenous instruments in designing their studies have found that the alleged social benefits of homeownership largely disappear and in some instances become negative. A recent study using an exogenous instrument for homeownership found "no evidence" that homeowners participated more actively in politics than renters or raised more money or volunteered for more community organizations; in fact, becoming a homeowner made people less likely to participate in these activities.122 The study also found mixed evidence that homeowners take better care of their homes: Homeowners were more likely to perform internal maintenance but not external maintenance, the former of which confers only private benefits and has no effect on social capital.

C. The MID and Housing as a Good Investment

Thus far, the weight of authority has demonstrated that the purported benefits of the MID are dubious and empirically unsubstantiated. The overwhelming evidence reveals a negligible (and perhaps negative) correlation between the MID and rates of homeownership, as well as a weak correlation (and certainly no causal connection) between homeownership and positive social benefits. Nonetheless, if homeownership could be said to offer Americans a solid investment opportunity and a path toward financial security, and if the MID actually helped families invest in homeownership, the costs of the subsidy, on net, could be diminished.

Viewing homeownership as a wise financial investment that generates solid gains and financial security permeates the American psyche. The connection between homeownership and wealth creation formed the basis of President Clinton's National Homeownership Strategy and President Bush's Ownership Society. It is also reflected in national public opinion polls even in the aftermath of the most recent collapse in the housing and mortgage markets. The housing lobby widely publicizes these opinion polls (many of which the housing lobby also funds) to convince Congress to preserve government subsidies for housing, including the MID. According to one such poll sponsored by the National Association of Home Builders, 74 percent of respondents stated that "owning a home is the best long-term investment they can make," while 73 percent opposed eliminating the mortgage interest deduction.123 The "clear message," the pollsters warned, and one that "candidates running for the White House and Congress would be wise to heed," is that American voters express "broad support for government policies that encourage homeownership and oppose efforts to make it more difficult to get a home loan and to tamper with the mortgage interest deduction."124 Another such survey touted by the National Association of Realtors (although conducted independent of the housing industry) found that 81 percent of respondents agreed that owning a home amounts to "the best long-term investment a person can make."125

Unfortunately, the investment returns on homeownership are not nearly as robust as Americans believe nor as the housing lobby champions. Adjusted for inflation, housing prices were flat throughout most of the postwar period until prices temporarily deviated from their historical pattern beginning in the 1990s as an overheated housing market bubbled, burst, and led to global financial meltdown. According to economist and housing guru Robert Shiller, from 1950 to 1995 (or until immediately before the bubble) real home prices grew a paltry 1.67 percent.126 From 1950 to 2000, moreover, the average annual investment return to housing amounted to less than 0.5 percent.127 Historical real returns to housing were so unimpressive over the postwar period that $100 invested in a home in 1950 only grew to $104 by 1997.128 Over the longer term, housing posted equally unimpressive returns, with inflation-adjusted prices growing only 0.4 percent per year from 1890 to 2004.129

In addition to struggling to produce positive real returns for more than a century,130 owner-occupied housing has grossly underperformed compared to other investment opportunities. Real stock prices jumped 1,176 percent between 1950 and 2000,131 for instance, while the Dow Jones stock index grew more than 2,700 percent.132 In addition, compounded annual returns between 1926 and 2009 for small stocks (11.9 percent), large stocks (9.8 percent), long-term U.S. government bonds (5.4 percent), and U.S. (30-day) Treasury bills (3.7 percent) produced strong and reliable gains that far outpaced housing.133 In fact, $100 invested in 1928 in, respectively, stocks, Treasury bonds, and Treasury bills would have been worth $166,787, $1,970, and $6,726 at the end of 2011.134

Owning a home is not the path to prosperity. Nor is it an alternative to investing actively for retirement with a sufficiently diversified portfolio that, among other things, does not contain any single asset representing 90 percent or more of total wealth (like a house, for instance). At best, homeownership amounts to a form of forced savings and a decent savings account (and that assumes an amortizing mortgage). Or, as economic commentator Anthony Randazzo has written, "A house is not a stock to be wielded as an investment, but rather it is a savings account that maintains its value with inflation."135

Even as a savings account, homeownership has become less effective over the years. Due to the MID and other policies that encourage would-be and current homeowners to finance homes with debt rather than with financial assets or cash, the percentage of equity that U.S. households hold in their homes dropped from 80 percent in the 1950s to 38.5 percent by the end of 2010.136 "Without the homeowner putting equity into their home," Randazzo observed, "there is no actual wealth building. And if the government juices prices, then there is no investment gain either."137 So, not only does the MID fail in any meaningful way to influence the rate of homeownership as we saw in Section III.A, it also discourages homeowners from actually owning their homes. So much for the promise of the Ownership Society.

IV. Reforming and Replacing the MID

If the housing industry cared at all about homeownership, it would call on Congress to abolish the MID immediately. Instead, it ignites and then stokes fears that touching the MID would result in a precipitous drop in housing prices and "trigger yet another crisis in home values."138 But the housing industry's dire predictions of a collapse in the housing market in the event Congress reconsidered the MID are dramatically "overstated."139

In fact, eliminating the MID (and not even replacing it with any number of policy alternatives better suited to promote homeownership) might actually raise rates of homeownership, particularly in high-priced areas. According to researchers, positive effects on homeownership rates from lower home prices could more than offset any negative effects on homeownership from loss of the deduction. This is particularly true of younger, urban buyers living in high-priced, space-constricted markets such as San Francisco where the price effect would provide an opportunity for them to jump into the market.140 Over time, the natural increase in demand would raise home prices -- but not in an artificial or distortionary way -- as demand began to outpace supply. Other studies have found that to the extent prices dipped in the event Congress reformed or replaced the MID, the downturn would be temporary and would affect bigger, more expensive homes.141 If Congress was still concerned about preserving artificially inflated home values for sellers of bloated homes, any change to the MID could be phased in over several years, allowing those homeowners to capture their phony appreciation. Either way, reforming or replacing the MID would encourage the buildup of home equity (that is, real ownership rather than leveraged ownership), increase the national saving rate,142 help households absorb future income shocks,143 encourage "fewer financial eggs in a single basket,"144 and create "less risk of financial catastrophe."145

Assuming that national policymakers and the American public still consider homeownership a goal worthy of subsidizing, repealing the MID would achieve that objective. It would also free the U.S. economy from the wealth-destroying and growth-retarding distortions caused by the MID. And depending on which policy alternative Congress endorsed, abolishing the deduction for mortgage interest would immediately generate much-needed tax revenues, provide all Americans a more equal chance of achieving homeownership, and encourage the growth of equity rather than debt in one's home. Below we consider the three most prominent replacements to the MID.

A. Slowly Phase Out the MID

Congress could gradually eliminate the MID over several years, phasing out the deduction rather than eliminating it immediately, to minimize adverse impacts on the housing and mortgage markets. Under a sample proposal outlined by the Congressional Budget Office, the maximum mortgage amount eligible on which interest could be deducted might decrease in annual decrements of $100,000 from the current level of $1.1 million until zeroing out 11 years later.146 Revenues would rise modestly in the beginning years of the phaseout ($215 billion over the first 10 years), but then increase substantially thereafter (averaging over $100 billion annually) and then grow relative to the size of the economy.147

If Congress desired eliminating the MID immediately rather than slowly, it could generate more than $100 billion under current law (which assumes expiration of the 2001 and 2003 tax cuts) in the first year after repeal and $1.26 trillion over 10 years.148 These gains decline slightly when evaluated under current policy rather than current law (which assumes the 2001 and 2003 tax cuts are extended) to $88 billion in the first year after repeal and $1.03 trillion over 10 years.149 The estimated revenue gains also adjust downward if we assume that some taxpayers with mortgages respond to repeal by reallocating portfolios to pay down mortgage debt.150 After accounting for that portfolio modification, repealing the MID would yield $91.4 billion (current law) or $74.8 billion (current policy) in the year after repeal, while the 10-year estimate would equal $1.07 trillion (current law) and $879 billion (current policy).151

The revenue gains from repealing the MID -- whether totaling slightly more or less than $100 billion per year -- could be used to achieve other policy goals. Among those purposes, Congress could decide to apply the funds to lowering marginal tax rates across the board (by some estimates, the average effective income tax rate could be cut almost 8 percent from 18.2 to 16.8 percent152); providing targeted refundable tax credits and grants (either to promote homeownership or for other goals); spending more on existing programs; or reducing the federal budget deficit. Regarding closing the deficit window, if repealing the MID recouped, say, $100 billion in forgone tax revenue, the projected 2012 federal budget deficit would shrink by 7.5 percent while the projected 2013 deficit would decline by 11.1 percent.153 None of this discussion, by the way, accounts for the additional revenue and economic prosperity gained from eliminating the MID, which experts agree would "boost the amount of capital available to other sectors of the economy and increase total economic output."154

Regarding the distributional effects of eliminating the MID, micro-simulation models reveal precisely what we would expect: High-income homeowners would experience the largest reductions in after-tax income from repeal, because they are more likely to own homes, itemize deductions, and face high marginal tax rates (the latter of which optimize the value of deductions). Research shows that taxpayers in the 80th to 99th percentiles of the income distribution lose the most from eliminating the MID, while the very highest income households (that is, the top 1 percent) are relatively unaffected, because mortgage interest payments, although large in absolute terms, comprise a fraction of total income.155

B. Capping the MID for High-Income Taxpayers

In each of his administration's four budget proposals, President Obama has included a recommendation to cap itemized deductions at 28 percent for high-income taxpayers in the 33 and 35 percent marginal tax brackets. The effect of the proposal would limit the value of the MID to 28 cents on the dollar rather than as much as 35 cents on the dollar. For tax year 2012, the limitation would kick in for households with taxable income exceeding $178,650 (for singles) and $217,450 (for married couples).156 The President's Economic Recovery Advisory Board, also known as the Volcker Commission for its chair, former head of the Federal Reserve, Paul Volcker, included a similar recommendation in its 2010 report on tax reform. Specifically, the commission proposed limiting itemized deductions to 75 percent of certain expenses and to use the revenue generated to raise the standard deduction as a way to simplify the tax system, improve fairness, broaden the tax base, and smooth effective tax rates.157 Economists have praised these plans to limit itemized deductions on both equity and efficiency grounds. "It seems unfair to subsidize higher income households' activities at higher rates," Diane Lim Rogers has written, particularly because "the deductions may merely reward behavior that would have taken place anyway."158

The revenue gains from limiting the value of the MID for high-income taxpayers would not be nearly as large as eliminating the MID. Without accounting for behavioral responses to the reform, the president's proposal to cap itemized deductions at 28 percent would raise $40 billion over 10 years (compared to more than $1 trillion for repealing the MID).159 In addition, the distributional consequences of the proposal would be concentrated exclusively at the upper end of the income spectrum with 40 percent of tax units in the 95th to 99th percentiles experiencing tax increases and with 55 percent of the top percentile paying higher taxes.160

C. Replacing the MID With a Tax Credit

The third and most attractive option for replacing the MID -- a tax credit for mortgage interest paid -- would succeed in every way that the MID fails. In particular, it would be more equitable, efficient, and effective in promoting homeownership.

A tax credit would apply more equitably than the MID in that it would be available to all taxpayers regardless of income or whether they itemize or take the standard deduction. Using a credit rather than a deduction to deliver homeownership benefits through the tax system would break the connection between the value of the subsidy and a taxpayer's income or tax rate, and instead provide the same dollar-for-dollar benefit to all qualifying claimants. A $5,000 subsidy delivered in the form of a credit would be worth an equal $5,000 to taxpayers in the 15 and 35 percent brackets (assuming each paid at least $5,000 in mortgage interest) rather than the unequal amounts of $750 ($5,000 x 0.15) and $1,750 ($5,000 x 0.35) as under the MID. Equitable treatment would be the rule irrespective of whether a taxpayer claimed the standard deduction or itemized as well as whether a taxpayer's itemized deductions exceeded the standard deduction by $1 or by $100,000.161 In this way, a tax credit would help the three-quarters of all taxpayers that the MID ignores, and it would make national housing policy more inclusive as well as more progressive.162

In addition, a tax credit for mortgage interest could address the inequities of the MID pertaining to disparate treatment across different regions of the country and among different races.163 A home credit could be capped and geographically indexed to prevent households in high-priced, space-constrained areas from receiving disproportionately larger subsidies.164 A credit rather than a deduction, moreover, would equalize treatment for minority households to the extent that those households report lower incomes, lower rates of homeownership, and lower home values than non-minority households. Unlike a deduction, a credit would offer homeownership assistance independent of income level, ability to itemize, or marginal tax rates.

A tax credit would also be more efficient than the MID, because it would inject less distortion into the housing and mortgage markets. Of course, any intervention aimed at lowering the cost of mortgage debt distorts choices. But subsidizing that debt with a tax credit would distort the allocation of financial capital to a lesser degree than the MID. High-income homeowners possess the greatest ability to pay for housing with cash and financial assets rather than debt, and the MID, by providing incentives to take on bigger loans to purchase bigger homes, encourages those taxpayers to skew portfolios heavily toward debt financing. Those same taxpayers would receive smaller benefits under a tax credit and thus be less inclined to distort allocations of capital regarding housing. A tax credit would further reduce the incentive to take on more debt by eliminating the deduction for home equity indebtedness, the "house-sized credit card" that helped fuel homeowners' excessive leveraging in the 1990s and 2000s.165 At the same time, the biggest beneficiaries from substituting a credit for the MID would be low- and middle-income households that do not have the luxury of swapping in cash or financial assets to purchase a home.

Finally, regarding effectiveness, a tax credit would allow considerably more Americans the opportunity to purchase a home. In many respects, this analysis is merely a numbers game. Currently, the MID helps only one-quarter of taxpayers with the decision to own or rent, while ignoring the other three-quarters of taxpayers. A tax credit, on the other hand, offers the same dollar-for-dollar benefit to all qualifying taxpayers, to high incomes as well as low incomes, to itemizers as well as non-itemizers, and to owners of big and small homes as well as big and small loans. In addition, a tax credit would more effectively promote homeownership than the MID by targeting taxpayers on the margin between owning and renting -- that is, low- and middle-income households -- rather than those taxpayers for whom the decision is not about whether to buy a house but how big and expensive a house to buy.

In other words, a tax credit would work where the MID has failed. In fact, researchers have found that replacing the MID with a tax credit for mortgage interest could raise the national homeownership rate anywhere from 3 to 5 percentage points in the aggregate and 8 percentage points for the lowest income households.166 Generally, a credit that provided a fixed dollar amount rather than a fixed percentage of mortgage interest paid would boost rates of homeownership to a larger degree, because such a subsidy would provide bigger benefits to households on the margin of homeownership and renting. A tax credit that targeted first-time home buyers could raise the rate by 5 percentage points,167 as could a credit that replaced both the MID and the deduction for property taxes paid.168

Given these findings, it is no wonder that calls for replacing the MID with a tax credit for homeownership have grown louder from both sides of the political spectrum. In 2005, President Bush's blue-ribbon tax reform panel issued a report recommending that tax benefits for mortgage interest "be retained, but shared more evenly" through a credit rather than a deduction to "encourage home ownership, not big homes."169 In particular, the panel recommended replacing the MID with a credit equal to 15 percent of mortgage interest paid.170 It also limited the principal mortgage amount to an average regional price of housing as determined by the Federal Housing Administration (FHA) (indexed and ranging from $227,000 to $412,000, down from the current mortgage cap of $1.1 million), a move explicitly designed to "encourage homeownership without subsidizing overinvestment in housing."171 In the same spirit, the panel further proposed repealing the interest subsidy on second homes and for home equity indebtedness. In 2009, the Congressional Budget Office recommended a similar tax credit for mortgage interest as part of its annual survey of spending and revenue options to address the federal budget deficit.172 And in 2010, President Obama's National Commission on Fiscal Responsibility and Reform called on Congress to substitute a 12 percent nonrefundable tax credit for the MID with mortgage principal capped at $500,000, no interest deduction on second homes, and no home equity indebtedness.173

To mitigate potential concerns over the perceived unfairness of repealing the MID for homeowners who relied on the deduction during the purchase decision, as well as to lessen any possible adverse market responses to the change in policy, a tax credit for homeownership could be phased in over several years for preexisting mortgages. The Bush tax reform panel, for its part, recommended that its tax credit proposal phase in over five years.174

The potential revenue gained from substituting a tax credit for the MID would be wholly dependent on how policymakers designed the credit (for example, using a fixed dollar amount versus a fixed percentage of mortgage interest or a refundable versus a nonrefundable credit) as well as which taxpayers were targeted (for example, lower-income taxpayers versus shared benefits throughout the income spectrum). In 2009, the CBO estimated that a 15 percent nonrefundable credit for mortgage interest and a gradually reduced cap on mortgage principal of $500,000 would generate $13 billion in the first year and $388 billion over seven years. The net gains in revenue reflected the fact that the credit would benefit some taxpayers (primarily low- and middle-income households) while disadvantaging others (high-income households).175

If policymakers wanted to provide an even bigger boost to homeownership, they could consider a revenue-neutral credit replacement for the MID. Of course, a revenue-neutral plan would prevent policymakers from selling the proposal as a debt reduction measure. But it would offer the opportunity to mobilize all $100 billion in tax subsidy from the MID to a policy alternative that could achieve widespread and shared gains in homeownership for all levels of income. From a structural and design standpoint, a revenue-neutral credit could provide a subsidy level equal to 20 percent of mortgage interest paid rather than 12 to 15 percent of mortgage interest paid as under the Bush, CBO, and Obama plans.

Researchers have examined the distributional effects of several different versions of a revenue-neutral tax credit for homeownership. The Tax Policy Center has simulated the effects of (i) a nonrefundable 20 percent credit for mortgage interest; (ii) a refundable 17 percent credit; (iii) a nonrefundable 100 percent credit on the first $2,030 of mortgage interest paid; and (iv) a refundable 100 percent credit on the first $1,490 of mortgage interest paid.176 While none of the options specifically caps the amount of qualifying mortgage principal, two of the plans have the effect of restricting the mortgage amount by limiting the credit to a certain dollar value of mortgage interest paid.

All four credit alternatives examined by the Tax Policy Center benefit taxpayers in the bottom four-fifths of the income distribution with a larger subsidy for homeownership. At the same time, all four plans disadvantage the top one-fifth of income earners compared to current treatment under the MID due to lower subsidy rates (17 and 20 percent) than the cohort's statutory marginal tax rates (25, 28, 33, and 35 percent). Among the four alternatives, the 100 percent refundable credit benefits the bottom 60 percent of taxpayers the most, while the 100 percent nonrefundable credit provides taxpayers in the fourth quintile (that is, the 60 to 80 percent cohort) the largest tax savings. Finally, taxpayers in the top quintile -- subject to higher taxes under all four plans -- lose the most under the 100 percent refundable credit and the least under the 100 percent nonrefundable credit.

V. Conclusion

Substituting a tax credit for the failed MID would redirect national housing subsidies to taxpayers on the margin between owning and renting. In this way, a tax credit would signal a welcome shift from the debased housing policies of the past and effectively promote widespread homeownership for all taxpayers. The only question is whether politicians can overcome their historical addiction to housing subsidies and enact a policy that actually works.

Indeed, in the immediate aftermath of the most recent collapse in the housing and financial markets, Congress began pursuing precisely the same ruinous policies of over-subsidizing housing as it had over the entire postwar period.177 In 2007, Congress bailed out homeowners for bad investments in residential housing by forgiving taxes on discharge of indebtedness income associated with forgiven mortgage debts.178 Months later, it created a new and poorly targeted tax subsidy for first-time home buyers.179 Research indicated that as many as 85 percent of the program's recipients would have purchased a home without the subsidy.180 Undeterred by the program's ineffectiveness, politicians raised the maximum subsidy amount and expanded its reach the following year,181 before extending the subsidy yet again and allowing certain repeat homebuyers to receive benefits under the program.182 With its appetite for propping up housing still not satiated, Congress doubled the maximum principal amount for FHA loans in 2009 to $730,000 and permitted the agency to underwrite mortgages with loan-to-value ratios as low as 3.5 percent, resulting in a quadrupling of these loans at a time when the FHA's default reserves neared zero.183 Also in 2009, Congress enacted the Worker, Homeownership, and Business Assistance Act, which allowed companies to carry back losses incurred in 2008 and 2009 to taxable profits earned as far back as five years earlier, an extension of the existing two-year carryback provision that resulted in sizable tax refunds for home builders, even those with large amounts of cash on hand.184 Faced with indisputable evidence that an overheated housing market contributed directly to financial Armageddon, Congress responded the only way it knew how: subsidize homeownership with artificial supports to jump-start and fuel another boom-to-bust cycle.

Fortunately, there are signs that politicians are finally getting the message that the MID has failed and needs to go. As we saw in Section IV.C, blue-ribbon tax reform panels commissioned by both Republican and Democratic presidents have recommended abolishing the MID and substituting a tax credit for homeownership. We saw the same recommendation from Obama's bipartisan debt commission. Moreover, tax and economic experts across the political spectrum are unanimous in their opposition to the MID, from the Brookings Institution185 and Tax Policy Center186 to the American Enterprise Institute187 and Tax Foundation.188 In addition, politicians are reaching across the aisle to reform and replace the MID.189 We already know that Obama has proposed limiting the subsidy's benefit to high-income households.190 And even current Republican presidential hopefuls are prepared to challenge the historical third rail of tax reform.191

It is time to repeal the MID and replace it with a tax credit for homeownership. By severing the connection between the nation's largest housing tax subsidy and a recipient's income or marginal tax rate, a tax credit would provide equal dollar-for-dollar benefits to qualifying claimants across the income distribution and thereby have an immediate and positive effect on homeownership. As important, a credit rather than a deduction would break the relationship between the value of the subsidy and the size of a taxpayer's mortgage debt or household square footage. In other words, and unlike the MID, a tax credit for homeownership would imbue national housing policy with equity, efficiency, and effectiveness.


1 Lou Cannon, "Reagan to Keep Home Mortgage Tax Deduction," The Washington Post, May 11, 1984, at F1 (quoting President Reagan assuring the National Association of Realtors (NAR) that he had instructed the Treasury Department to "preserve that part of the American Dream which the home mortgage interest deduction symbolizes").

2 Bruce Bartlett, "Tax Reform's 'Third Rail': Mortgage Interest," 139 NCPR Pol'y Backgrounder 1 (1996).

3 Jeffrey H. Birnbaum and Alan S. Murray, Showdown at Gucci Gulch: Lawmakers, Lobbyists, and the Unlikely Triumph of Tax Reform 246 (1987).

4 Christopher Howard, The Hidden Welfare State: Tax Expenditures and Social Policy in the United States 93 (1997) (naming Social Security and Medicare as the other two members of the "trinity").

5 Id.

6 Pat Jones, "Housing Industry Lobbying Campaign Rankles Some Taxwriters," Tax Notes, Jan. 23, 1989, p. 409.

7 Birnbaum and Murray, supra note 3, at 246.

8 According to the Center for Responsive Politics, the housing industry spent $65.4 million lobbying Congress and federal agencies in 2011. See Of that amount, NAR spent $22.5 million, ranking it the third most free-spending organization across all industries, behind only the Chamber of Commerce ($66.4 million) and General Electric ($26.3 million). See

9 The National Association of Home Builders (NAHB) offers a website,, that warns visitors about threats to the MID from Congress and economic experts. According to the NAHB, its site "is dedicated to preserving the mortgage interest deduction and protecting homeownership." See

10 Lawrence Yun (chief economist, NAR), "Why the MID Deserves to Stay," Realtor Mag (Sept. 2010), available at

11 Ron Phipps (president, NAR), "Should the Mortgage Interest Deduction Be Phased Out?" The Costco Connection, at 19 (Nov. 2011), available at See also Tami Luhby, "Slash Mortgage Deductions for the Rich? Fat Chance," (Feb. 15, 2011) (quoting Jerry Howard, CEO of NAHB, opining that an attack on the MID amounts to "an attack on the middle class"), available at

12 Ben Steverman, "A Taxing Debate: The Mortgage Interest Deduction," Bloomberg (Oct. 18, 2011) (quoting Yun), available at See also Robert D. Dietz (NAHB), "The Tax Benefits of Homeownership," (Mar. 27, 2009), available at

13 Phipps (president, NAR), "Realtors® Say Mortgage Interest Deduction Vital to Home Ownership, Economy" (Dec. 1, 2010), available at; Phipps, supra note 11.

14 See e.g., NAHB (73 percent of voters polled oppose eliminating the MID), available at; NAR (74 percent of homeowners and 62 percent of renters consider the MID "extremely" or "very" important to homeownership), available at; and; American Strategies Inc., "Annual PULSE Survey Shows Opposition to Down Payment Requirements and Elimination of Mortgage Interest Deduction," prepared for NAR (June 22, 2011) (two-thirds of respondents oppose eliminating the MID to reduce the deficit), available at

15 The Editors, "Tax Break for Mortgage Debt Is Ready for the Wrecking Ball: View," Bloomberg (June 23, 2011) (reporting that 49 percent of respondents to a Bloomberg National Poll indicated they would eliminate the MID if it coincided with across-the-board tax cuts), available at

16 Office of Management and Budget, "Analytical Perspectives: Budget of the United States Government, Fiscal Year 2013" 261 (2012), Doc 2012-2944, 2012 TNT 30-43 (estimating cost of the MID at $100.91 billion).

17 All citations to tax filer characteristics, unless otherwise noted, were calculated using Justin Bryan (IRS), "Individual Income Tax Returns, 2009," Statistics of Income Bull. 23 (Fall 2011) (hereinafter cited as "IRS").

18 JCT, "Estimates of Federal Tax Expenditures for Fiscal Years 2011-2015," JCS-1-12, at 53 (Jan. 17, 2012), Doc 2012-894, 2012 TNT 11-21.

19 Id.

20 Id.

21 James Poterba and Todd Sinai, "Tax Expenditures for Owner Occupied Housing: Deductions for Property Taxes and Mortgage Interest and the Exclusion of Imputed Rental Income," 98 Am. Econ. Rev. 84, 85 (2008). See also Anthony Randazzo and Dean Stansel, "Unmasking the Mortgage Interest Deduction: Who Benefits and By How Much?" Reason Foundation, at 30 (July 2011) (showing the distribution of MID benefits across income levels and over time without adjusting for inflation).

22 JCT, "Estimates of Federal Tax Expenditures for Fiscal Years 1987-1991," at 20 (Mar. 1, 1986).

23 JCT, supra note 18, at 53.

24 Rev. Proc. 2011-12, 2011-2 IRB 297, Doc 2010-27298, 2010 TNT 247-8.

25 IRS, supra note 17, at 42. If we only count taxable returns among the itemizers, the population shrinks still further to 26.3 percent of tax units.

26 Carmen DeNavas-Walt et al., "Income, Poverty, and Health Insurance Coverage in the United States: 2010," U.S. Census Bureau, Current Population Reports P60-239, at 5 (Sept. 2011) (reporting median family income of $49,445).

27 IRS, supra note 17, at 42.

28 Id.

29 See infra Section IV.B.

30 There are six statutory (nonzero) tax rates under the individual federal income tax: 10, 15, 25, 28, 33, and 35 percent. Rev. Proc. 2011-12, supra note 24.

31 IRS, supra note 17, at 46.

32 Id.

33 Id. at 23 and 46.

34 Id. at 46.

35 Id.

36 Id.

37 Id.

38 Id.

39 For examples of flawed analysis ignoring the effect of marginal tax rates on the absolute and relative distribution of MID benefits, see Kevin Drum, "Why Everyone Loves the Mortgage Interest Deduction," Mother Jones, July 13, 2011, available at; Dietz, "Housing Tax Incentives Are Particularly Helpful to Younger Households," (May 3, 2010), available at; Dietz, "The Mortgage Interest and Real Estate Tax Deductions," (July 1, 2008), available at

40 Eric Toder et al., "Distributional Effects of Tax Expenditures," Schwartz Center for Economic Policy Analysis, the New School of Research, and Urban-Brookings Tax Policy Center (2009), available at; Adam Carasso et al., Making Tax Incentives for Homeownership More Equitable and Efficient (2005), at Appendix Table 2 (reporting that only 3 percent of benefits from the MID and the deduction for property taxes accrue to taxpayers in the bottom three quintiles); Charles A. Capone Jr., "Taxation and Housing Tenure Choice: The Case for Moderate Income Homeownership," 4 J. Housing Econ. 328, 345-347 (1994); Richard Voith, "Does the Federal Tax Treatment of Housing Affect the Pattern of Metropolitan Development?" Fed. Reserve Bank of Phil. Bus. Rev., at 8 (Mar.-Apr. 1999).

41 See, e.g., John E. Anderson and Atrayee Ghosh Roy, "Eliminating Housing Tax Preferences: A Distributional Analysis," 10 J. Housing Econ. 41, 55-56 (2001). See also Joseph Gyourko and Todd Sinai, "Spatial Distribution of Mortgage Deduction Benefits Across and Within Metropolitan Areas in the United States," 137, in Using Tax Policy to Increase Homeownership among Low- and Moderate-Income Households (2001) (finding the MID more regressive than the tax system is progressive).

42 See supra text accompanying note 31.

43 See supra text accompanying note 25.

44 In 2009, there were 74,843,000 owner-occupied housing units (Census Bureau, 2012 Statistical Abstract (2011), at Table 996, available at, but only 36,541,819 of those homeowners (or 48.8 percent) claimed the MID on their tax returns (IRS, supra note 17, at 46).

45 Of the 46,703,000 homeowners with mortgages on their property, only 36,541,819 (or 78.2 percent) claimed the MID. U.S. Census Bureau, supra note 44, at Table 998.

46 JCT, "Present Law, Data, and Analysis Relating to Tax Incentives for Homeownership," JCX-50-11, at 27-28 (Sept. 30, 2011), Doc 2011-20797, 2011 TNT 191-33; Randazzo and Stansel, supra note 21, at 6-7; Toder et al., "Reforming the Mortgage Interest Deduction," Tax Policy Center, at 10-11 (Apr. 2010); Poterba and Sinai, supra note 21.

47 Martin A. Sullivan, "The Rich Get 100 Times More Mortgage Subsidy Than the Poor," Tax Notes, Mar. 7, 2011, p. 1110, Doc 2011-4466, 2011 TNT 44-2. See also Ike Brannon et al., "The Geographic (and Political) Distribution of Mortgage Interest Deduction Benefits," Andrew Young School of Policy Studies Research Paper No. 1932845 (Sept. 23, 2011), available at and; Toder et al., supra note 46, at 14-15; Joseph Gyourko and Todd Sinai, "The Spatial Distribution of Housing-Related Ordinary Income Tax Benefits," 31 Real Estate Econ. 527, 557 (2003); Peter Brady et al., "Regional Differences in the Utilization of the Mortgage Interest Deduction," 31 Pub. Fin. Rev. 327, 360-361 (2003).

48 Sullivan, supra note 47, at 1112.

49 Gyourko and Sinai, supra note 41, at 137.

50 Sullivan, "Mortgage Deduction Heavily Favors Blue States," Tax Notes, Jan. 24, 2011, p. 364, Doc 2011-1232, 2011 TNT 15-3 .

51 Id. at 367.

52 Id.

53 See, e.g., Toder et al., supra note 46, at 13-14; Dorothy A. Brown, "Shades of the American Dream," 87 Wash. U. L. Rev. 329 (2010); Beverly I. Moran and William Whitford, "A Black Critique of the Internal Revenue Code," 1996 Wis. L. Rev. 751, 775-779 (1996).

54 Edward L. Glaeser and Jesse M. Shapiro, "The Benefits of the Home Mortgage Interest Deduction," National Bureau of Economic Research, Working Paper No. 9284, at 6 (2002).

55 Sullivan, "The Economics of the American Dream," Tax Notes, Jan. 24, 2005, p. 407, Doc 2005-1275, 2005 TNT 15-9.

56 Id. See also Gerald Prante, "Some Facts About the Mortgage Interest Deduction," Tax Foundation Tax Policy Blog (Nov. 5, 2006) ("Ask any economist that does not speak for the homebuilding or real estate industry and he or she will tell you that the home mortgage interest deduction has little economic justification"), available at

57 John E. Anderson et al., "Capping the Mortgage Interest Deduction," 60 Nat'l Tax J. 769 (2007) (finding that the MID distorts "the user cost of owner-occupied housing for taxpayers"). See also James Poterba, "Tax Subsidies to Owner-Occupied Housing: An Asset Market Approach," 99 Q. J. Econ. 729, 748-749 (1984); Patric Hendershott and Joel Slemrod, "Taxes and the User Cost of Capital for Owner-Occupied Housing," 10 AREUEA J. 375, 376 (1983).

58 Martin Gervais, "Housing Taxation and Capital Accumulation," 49 J. Monetary Econ. 1461, 1482 (2002); Lori Taylor, "Does the United States Still Overinvest in Housing?" Fed. Reserve Bank Dallas Econ. Rev. 10, 16 (1998).

59 Roger Lowenstein, "Who Needs the Home Mortgage-Interest Deduction?" N.Y. Times Magazine, at 79 (Mar. 5, 2006).

60 Dale Jorgenson and Kun-Young Yun, "Tax Reform and U.S. Economic Growth," 98 J. Pol. Econ. 151, 190 (1988).

61 Edwin S. Mills, "Has the United States Overinvested in Housing?" 15 J. Am. Real. Est. & Urb. Econ. Ass'n. 601 (1987).

62 Viral V. Acharya et al., Guaranteed to Fail: Fannie, Mae, Freddie Mac, and the Debacle in Mortgage Finance 168-169 (2011), citing Edwin S. Mills, "Dividing Up the Investment Pie: Have We Overinvested in Housing?" Fed. Reserve Bank Phil. Bus. Rev. 13 (Mar.-Apr. 1987).

63 The President's Advisory Panel on Federal Tax Reform, "Simple, Fair, and Pro-Growth: Proposals to Fix America's Tax System," 71 (Nov. 1, 2005), Doc 2005-22112, 2005 TNT 211-14. See also Jane G. Gravelle, "The Corporate Tax: Where Has It Been and Where Is It Going?" 57 Nat'l Tax J. 903 (Dec. 2004) (estimating the effective tax rate on owner-occupied housing in 2003 at 2 percent compared to 18 percent on noncorporate investment and 32 percent on corporate investment).

64 Gravelle, The Economic Effects of Taxing Capital Income 204 (1994).

65 Poterba and Sinai, "Revenue Costs and Incentive Effects of the Mortgage Interest Deduction for Owner-Occupied Housing," 64 Nat'l Tax J. 531, 546 (June 2011).

66 Id.

67 William G. Gale et al., "Encouraging Homeownership Through the Tax Code," Tax Notes, June 18, 2007, p. 1171, Doc 2007-13495 (concluding that "theoretical considerations and empirical evidence suggest" that one of the main effects of the MID is to increase loan-to-value ratios).

68 Sullivan, "Deleveraging the Tax Code," Tax Notes, Sept. 29, 2008, p. 1241, Doc 2008-20522, 2008 TNT 190-9.

69 Id.

70 Anthony Randazzo, "How Would Getting Rid of the Mortgage Interest Deduction Impact Housing Prices" (July 12, 2011), available at; Anderson et al., supra note 57, at 769 (The MID "provides an incentive for people to increase their consumption of housing [because] larger subsidies are provided to those purchasing more expensive homes."); Gale et al., supra note 67, at 1171 (stating that the MID "serves mainly to raise the price of housing and land"); Richard K. Green et al., "Metropolitan-Specific Estimates of the Price Elasticity of Supply of Housing and Their Sources," 95 Am. Econ. Rev. 334, 335 (2005) (finding significant price premiums associated with the MID); Harvey S. Rosen, "Housing Subsidies: Effects on Housing Decisions, Efficiency, and Equity," at 395, in Handbook of Public Economics, eds. Alan J. Auerbach and Martin Feldstein (1985). Price increases due to the MID are greatest in areas with high home prices, high-value loans, high tax rates, and relatively fixed housing stock (either due to geographic limitations or community efforts to restrict supply or land-use restrictions).

71 See Joseph Gyourko and Richard Voith, "Capitalization of Federal Taxes, the Relative Price of Housing, and Urban Form: Density and Sorting Effects," 32 Reg'l Science & Urban Econ. 673, 685 (2002).

72 Harvey S. Rosen, "Housing Decisions and the U.S. Income Tax: An Econometric Analysis," 11 J. Pub. Econ. 1, 21-22 (1979).

73 Dennis R. Capozza et al., "Taxes, Mortgage Borrowing, and Residential Land Prices," in Economic Effects of Federal Tax Reform 171, eds. Henry J. Aaron and William G. Gale (1996). See also Donald Bruce and Douglas Holtz-Eakin, "Will a Consumption Tax Kill the Housing Market?" at 96, in Transition Costs of Fundamental Tax Reform, eds. Kevin A. Hassett and Glenn R. Hubbard (2001) (assuming an infinitely elastic supply of housing).

74 For 3 to 6 percent, see Poterba and Sinai, supra note 65. Researchers have also attempted to estimate the effect of the MID on homeownership and the price of housing stock by examining the subsidy's effect on mortgage affordability; that is, the extent to which the MID actually makes homes more affordable for potential buyers. See Randazzo and Stansel, supra note 21, at 14 (finding that the MID increases mortgage affordability by less than 1 percent across all income levels).

75 As of January 2012, the value of U.S. housing stock equaled $16.1 trillion ($6.2 trillion in home equity plus $9.9 trillion in mortgage debt). See Freddie Mac Investor Presentation, "Freddie Mac Update" (Jan. 2012), available at If we assume that the MID raises the cost of housing by 5 percent, it should be capitalized into the value of housing stock to the tune of $805 billion ($16.1 trillion x 0.05). But for tax year 2009 (the most recent year for which we have microdata), taxpayers reported $420.8 billion in mortgage interest paid on their tax returns. IRS, supra note 17, at 46. Moreover, the actual tax savings from the MID amounts to roughly $100 billion per year, far less than $805 billion. OMB, supra note 16, at 261.

76 George McCarthy et al., "The Economic Benefits and Costs of Homeownership: A Critical Assessment of the Research," Research Inst. Hous. Am., Working Paper No. 01-02, at 34 (2001) (stating that "increased mobility costs" to homeownership "make the national economy inflexible").

77 Hernan Winkler, "The Effect of Homeownership on Geographic Mobility and Labor Market Outcomes" (Sept. 10, 2011), available at

78 McCarthy et al., supra note 76, at 34. See also Andrew Caplin et al., "Collateral Damage: Refinancing Constraints and Regional Recessions," 29 J. Money Credit & Banking 496, 512-514 (1997).

79 See supra text accompanying notes 55 and 56.

80 Donald B. Marron, "How Large Are Tax Expenditures?" Tax Notes, Mar. 28, 2011, p. 1597, Doc 2011-6124, 2011 TNT 62-50.

81 Center on Budget and Policy Priorities, "Where Do Our Federal Tax Dollars Go?" (Apr. 15, 2011), available at; Charles Riley, "Obama Unveils $3.8 Trillion Budget," CNNMoney (Feb. 13, 2012), available at

82 OMB, supra note 16, at 261. The MID has ranked second on the government's list of costliest tax expenditure items since 2003, and it has not ranked lower than third since 1996. See OMB, "Analytical Perspectives: Budget of the United States Government for Fiscal Years 1996-2012" (for 1996, id. at 64; 1997, at 86; 1998, at 79; 1999, at 99; 2000, at 114; 2001, at 117; 2002, at 63; 2003, at 107; 2004, at 110; 2005, at 294; 2006, at 324; 2007, at 296; 2008, at 296; 2009, at 298; 2010, at 308, 2011, at 300, 2012, at 252).

83 OMB, supra note 16, at 261 (estimating cost of MID at $606.42 billion between 2013-2017).

84 Id. The four costliest tax expenditures related to housing included the MID ($100.91 billion), the exclusion for net imputed rental income ($50.08 billion), the exclusion for capital gains on the sale of a primary residence ($23.44 billion), and the deduction for state and local property taxes on owner-occupied housing ($22.32 billion).

85 Id.

86 Glaeser and Shapiro, supra note 54, at 3. See also Gale et al., supra note 67, at 1179 (concluding that the MID "has little if any positive effect on homeownership").

87 Alan L. Feld, "Redeployment of Tax Expenditures for Housing," Tax Notes, June 25, 1984, p. 1441.

88 Calvin H. Johnson, "Was It Lost?: Personal Deductions Under Tax Reform," 59 SMU L. Rev. 689, 717 (2006).

89 Toder et al., supra note 46, at 3.

90 Census Bureau, "Housing Vacancies and Homeownership," Table 14, "Homeownership Rates for the U.S. and Regions: 1965 to Present," available at

91 Census Bureau, "Historical Statistics of the United States, Colonial Times to 1970," at 646 (1975), available at

92 For a history of the MID and other housing tax policies in the United States, see Dennis J. Ventry Jr., "The Accidental Deduction: A History and Critique of the Tax Subsidy for Mortgage Interest," 73 Law & Contemp. Probs. 233 (2010).

93 Census Bureau, "Statistical Abstract of the United States: 1999," at 727 (1999).

94 See Daniel K. Fetter, "How Do Mortgage Subsidies Affect Home Ownership?: Evidence from the Mid-Century GI Bills," NBER Working Paper No. 17166 (June 2011).

95 It was not until the mid-1960s that policymakers and the housing industry began to consider the connection between the potential tax savings for deducting mortgage interest and the housing market. See Ventry, supra note 92, at 252-277.

96 Clinton's pro-homeownership policy was embodied in his National Homeownership Strategy (see, e.g., William J. Clinton, "Remarks on the National Homeownership Strategy" (June 5, 1995), available at, while Bush's pro-homeownership efforts became a core component of his "Ownership Society" (see, e.g., Joe Becker et al., "White House Philosophy Stoked Mortgage Bonfire," The New York Times, Dec. 31 2008, at A1, available at

97 Census Bureau, supra note 90. See also JCT, supra note 46, at 15-16.

98 Census Bureau, supra note 90.

99 JCT, "Estimates of Federal Tax Expenditures for Fiscal Years 1986-1990," at 13 (1985).

100 OMB (for fiscal 1996), supra note 82, at 64 (reporting cost of $54 billion).

101 Census Bureau, supra note 90.

102 OMB (for fiscal 2010), supra note 82, at 308.

103 Census Bureau, supra note 90.

104 For a discussion of distortions in the mortgage market caused by both tax expenditures and the GSEs, see Randazzo, "Rethinking Homeownership: A Framework for 21st Century Housing Finance Reform," Reason Foundation Policy Brief 93 (Oct. 2010).

105 JCT, supra note 46, at 17-18.

106 Id.

107 Id.

108 Id. at 18.

109 Id.

110 See Gale et al., supra note 67, at 1180-1182 (finding deduction for mortgage interest not correlated with increases in homeownership); Roberta F. Mann, "The (Not So) Little House on the Prairie: The Hidden Costs of the Home Mortgage Interest Deduction," 32 Ariz. St. L. J. 1361 (2000) (looking at 10 countries and finding no relationship between interest deduction and rates of homeownership); William G. Gale, "What Can Americans Learn from the British Tax System?" 50 Nat'l Tax J. 753 (1997) (finding increases or decreases in deduction for mortgage interest not correlated with higher rates of homeownership).

111 Anderson et al., supra note 57, at 769-770. See also Glaeser and Shapiro, supra text accompanying note 54; Rosen, supra note 70, at 395-402.

112 See Glaeser and Shapiro, supra note 54, at 33.

113 Id. at 3.

114 Toder et al., supra note 46, at 2.

115 See, e.g., Yun, supra note 10 ("Academic studies have demonstrated positive social benefits, including lower juvenile delinquency rates and higher student achievement among children of home owners").

116 Denise DiPasquale and Edward L. Glaeser, "Incentives and Social Capital: Are Homeowners Better Citizens?" 43 J. Urban Econ. 354 (1999); William M. Rohe and Michael A. Stegman, "The Impact of Homeownership on the Social and Political Involvement of Low-Income People," 30 Urban Affairs Rev. 152 (1994).

117 Joseph Harkness and Sandra J. Newman, "Differential Effects of Homeownership on Children from Higher-and Lower-Income Families," 14 J. Housing Research 1 (2003); Donald R. Haurin et al., "Does Homeownership Affect Children's Outcomes?" 30 Real Estate Econ. 635 (2002); Thomas P. Boehm and Alan M. Schlottman, "Does Homeownership by Parents Have an Economic Impact on Their Children?" 8 J. Housing Econ. 217 (1999); Richard K. Green and Michelle J. White, "Measuring the Benefits of Homeowning: Effects on Children," 41 J. Urban Econ. 441 (1997).

118 Richard D. Alba et al., "Living with Crime: The Implications of Racial/Ethnic Differences in Suburban Location," 73 Social Forces 395 (1994); Edward L. Glaeser and Bruce Sacerdote, "Why Is There More Crime in Cities?" 107 J. Pol. Econ. 225 (1994).

119 George Galster, "Empirical Evidence on Cross-Tenure Differences in House Maintenance and Conditions," 59 Land Econ. 107 (1983); Neil S. Mayer, "Rehabilitation Decisions in Rental Housing: An Empirical Analysis," 10 J. Urban Econ. 76 (1981).

120 Glaeser and Shapiro, supra note 54, at 30 (characterizing these claims as based on observed "correlations without any strong evidence for causality"); McCarthy et al., supra note 76, at 43 ("Evidence regarding the societal economic benefits of homeownership is highly conjectural"); Peter Rossi and Eleanor Weber, "The Social Benefits of Homeownership: Empirical Estimates from the National Surveys," 7 Housing Pol'y Debate 37 (1996) (concluding "the claims for some social and individual benefits from homeownership are supported but only weakly").

121 See, e.g., Harkness and Newman, supra note 117; Haurin et al., supra note 117; Daniel Aaronson, "A Note on the Benefits of Homeownership," 47 J. Urban Econ. 356 (2000); DiPasquale and Glaeser, supra note 116; Green and White, supra note 117.

122 Gary V. Engelhardt et al., "What Are the Social Benefits of Homeownership?: Experimental Evidence from Low-Income Households," 67 J. Urban Econ. 249 (2010).

123 NAHB, "Voters Place High Value on Homeownership, Oppose Policies That Make It More Difficult to Own a Home" (Jan. 11, 2012), available at

124 Id.

125 NAR, "Realtor Nationwide Open House Weekend Opens Doors to Home Ownership" (June 2, 2011) (referring to a PEW Research Center national poll, "Home Sweet Home. Still." (Apr. 12, 2011), available at, available at

126 According to Shiller's historical data sets, the real home price index rose from 105.8948 in 1950 to 107.666 in 1995. See Online Data Robert Shiller, available at

127 Dennis Cauchon, "Why Home Values May Take Decades to Recover," USA Today, Dec. 12, 2008, available at From 1940 to 2000, home prices rose at an annual real rate of just 0.7 percent (these slightly higher returns compared to the 1950-2000 period reflect lower prices at the beginning of the time series). James Hagerty, "Outlook for Home Prices Clouded by Spat Over Historical Trends," (Apr. 24, 2009) (quoting Shiller), available at

128 Id.

129 See Online Data Robert Shiller, supra note 126. Data also reflected in Figure 2.1 in Shiller, Irrational Exuberance, 13 (2005).

130 Morgan Housel, "Robert Shiller on Why Home Prices Could Fall for Several Decades," The Motley Fool (Dec. 23, 2011) (interview with Shiller), available at

131 See Online Data Robert Shiller, supra note 126.

132 Randazzo, "Debunking Mortgage Interest Deduction Benefit Myths," Reason (July 11, 2011), available at

133 Morningstar and Shooter Financial, "Ibbotson SBBI: Stocks, Bonds, Bills, and Inflation 1926-2009" in Long-Term Investment Performance (Mar. 1, 2010).

134 Damodaran Online, "Annual Returns on Stocks, Treasury Bonds, and Treasury Bills: 1928 to Current," available at This downloadable dataset is one among many generated by Aswath Damodaran, professor of finance at New York University's Stern School of Business.

135 Randazzo, "The Myth of Homeownership Wealth Creation," Reason (Apr. 26, 2011), available at

136 Id. See also Calculated Risk, "Q4 Flow of Funds: Household Real Estate Assets Off $6.3 Trillion From Peak" (Mar. 10, 2011), available at

137 Randazzo, supra note 135.

138 Letter from Charles McMillan, (president, NAR) to President Obama (Feb. 26, 2009) (on file with the author). See also supra text accompanying notes 8-13.

139 Donald Bruce and Douglas Holtz-Eakin, "Apocalypse Now?: Fundamental Tax Reform and Residential Housing Values," NBER Working Paper No. 6282 (1997).

140 See, e.g., Steven C. Bourassa and Ming Yin, "Tax Deductions, Tax Credits and the Homeownership Rate of Young Urban Adults in the United States," 45 Urban Studies 1141 (2007).

141 See Voith, supra note 40, at 7 (Eliminating the MID "would lower the demand for housing, especially for large houses, which would result in a short-run oversupply of these homes. The excess supply of large houses would result in declining values for these properties until natural growth in demand restored the balance between supply and demand."). See also Steven C. Bourassa and William G. Grigsby, "Income Tax Concessions for Owner-Occupied Housing," 11 Housing Pol'y Debate 520, 533-537 (2000); Capozza et al., supra note 73, at 173.

142 Capozza et al., supra note 73, at 173.

143 McCarthy et al., supra note 76, at 21-22.

144 Theodore P. Seto, "The Problem with Single-Mindedness," Cal. Daily J., at 6 (Apr. 22, 2009).

145 Id.

146 CBO, "Reducing the Deficit: Spending and Revenue Options" 146-147 (Mar. 2011), Doc 2011-5145, 2011 TNT 48-18.

147 Id. at 146.

148 Toder et al., supra note 46, at 8 ($108 billion for 2012). The MID was estimated to cost $98.5 billion in fiscal 2012 (OMB, supra note 82, at 252) and $100.9 billion in fiscal 2013 (OMB, supra note 16, at 261).

149 Toder et al., supra note 46, at 18.

150 Static revenue estimation (that is, equating revenue gained from repealing or altering the MID with the pre-reform estimated cost of the MID) overstates net savings due to these behavioral responses. In particular, static estimates fail to account for the response of higher-income taxpayers who would likely pay down mortgage debt by reallocating portfolios and shifting leverage to other assets. See, e.g., John L. Buckley, "Tax Expenditure Reform: Some Common Misperceptions," Tax Notes, July 18, 2011, p. 255, Doc 2011-13056, or 2011 TNT 138-6. Researchers offer wide-ranging estimates about how much federal tax revenues would increase after accounting for adjustments in borrowing patterns. See Poterba and Sinai, supra note 65, at 552 (85 percent increase after accounting for portfolio reallocation); Gale et al., supra note 67, at 1178 (84 percent increase); Martin Gervais and Manish Pandey, "Who Cares About Mortgage Interest Deductibility?" 34 Can. Pub. Pol'y 1 (Mar. 2000) (58 percent increase); James R. Follain and Lisa Sturman Melamed, "The False Messiah of Tax Policy: What Elimination of the Home Mortgage Interest Deduction Promises and a Careful Look at What It Delivers," 9 J. Housing Res. 179 (1998) (25 percent increase).

151 Toder et al., supra note 46, at 18.

152 Randazzo and Stansel, supra note 21, at 17.

153 OMB, "2013 Budget Overview," available at

154 CBO, supra note 146, at 146.

155 See Toder et al., supra note 46, at 8-11. For the same reasons that static estimates can overstate revenue gains from eliminating the MID (see supra note 150), they can also skew distributional impacts to the extent that high-income taxpayers respond to repeal by paying down mortgages. See Buckley, supra note 150, at 264-267.

156 See 2012 Form 1040-ES (depicting 2012 tax rate schedules), available at

157 The President's Economic Recovery Advisory Board, "The Report on Tax Reform Options: Simplification, Compliance, Corporate Taxation" 44-46 (2010), Doc 2010-19068, 2010 TNT 167-50.

158 See Diane Lim Rogers, "Limiting Itemized Deductions Still Makes Sense," Tax Notes, Jan. 23, 2012, p. 465, Doc 2012-910, or 2012 TNT 14-11.

159 Toder et al., supra note 46, at 8. Revenue estimates for capping all itemized deductions (and not just the MID) are more robust. The CBO estimates that the president's proposal to cap itemized deductions at 28 percent could raise $293 billion over 10 years (CBO, "An Analysis of the President's Budgetary Proposals for Fiscal Year 2012," 6 (Apr. 15, 2011), Doc 2011-8368, 2011 TNT 75-16), while limiting deductions to 15 percent (rather than 28 percent) could generate $460 billion over five years and $1.2 trillion over 10 years (CBO, supra note 146, at 151-152).

160 Toder et al., supra note 46, at 9.

161 For a fuller explanation of this phenomenon, see Section I. See also Buckley, supra note 150, at 263-264 (noting existence of "wasted" housing deductions).

162 See Gale et al., supra note 67, at 1184; Peter Dreier, "The New Policies of Housing: How to Rebuild the Constituency for a Progressive Federal Housing Policy," 63 J. Am. Planning Assoc. 5, 22 (1997).

163 See supra text accompanying notes 47-54.

164 For such a recommendation, see the President's Advisory Panel on Federal Tax Reform, supra note 63, at 73-74 and 237-238.

165 Robert J. Wells, "It's Time to Revisit the Interest Deduction Rules," Tax Notes, Aug. 2, 1993, p. 649.

166 Adam Carasso et al., "How to Better Encourage Homeownership," Urban-Brookings Tax Policy Center (June 2005); Richard K. Green and Kerry D. Vandell, "Giving Households Credit: How Changes in Tax Policy Could Affect the Homeownership Rate," 29 Regional Sci. & Urban Econ. 419 (1999) (3 to 5 percentage point increase in the overall homeownership rate and 8 percentage point increase for lowest income households); Andrew Reschovsky and Richard K. Green, Tax Credits and Tenure Choices 401, Proceedings of the National Tax Association Ninety-First Annual Conference (1998) (3 percentage point increase for nonrefundable fixed mortgage interest credit and 5.3 percentage increase for fixed refundable credit); Gale et al., supra note 67, at 1185-1186. For foundational work on the subject of homeownership rates and the design of housing tax subsidies, see Harvey S. Rosen and Kenneth T. Rosen, "Federal Taxes and Homeownership: Evidence from Time Series," 88 J. Pol. Econ. 59 (1980); Harvey S. Rosen, "Owner-Occupied Housing and the Federal Income Tax: Estimates and Simulations," 6 J. Urban Econ. 247 (1979); Harvey S. Rosen, "Housing Decisions and the U.S. Income Tax: An Econometric Analysis," 11 J. Pub. Econ. 1 (1979).

167 Gale et al., supra note 67, at 1183.

168 Green and Vandell, supra note 166. For a survey of the literature on replacing the MID with tax credits for homeownership, see Adam Carasso et al., "Making Tax Incentives for Homeownership More Equitable and Efficient," Urban-Brookings Tax Policy Center (June 2005).

169 The President's Advisory Panel on Federal Tax Reform, supra note 63, at 73.

170 Id. at 68-75 and 237-238.

171 Id. at 73. The panel estimated that between 85 and 90 percent of mortgages would have been unaffected by the lower mortgage cap.

172 CBO, "Budget Options" (vol. 2), 187-189 (Aug. 2009), Doc 2009-17869 , 2009 TNT 150-22.

173 National Commission on Fiscal Responsibility and Reform, "The Moment of Truth: Report of the National Commission on Fiscal Responsibility and Reform," 31 (Dec. 1, 2010), Doc 2010-25486, 2010 TNT 231-35.

174 The President's Advisory Panel on Federal Tax Reform, supra note 63, at 74 and 238.

175 CBO, supra note 172, at 187.

176 Toder et al., supra note 46, at 9-10 and 16.

177 See Ventry, supra note 92.

178 See Mortgage Forgiveness Debt Relief Act of 2007, P.L. 110-142. In 2008 Congress extended the program another four years to the end of 2012. See Emergency Economic Stabilization Act of 2008, P.L. 110-343.

179 See Housing and Economic Recovery Act, P.L. 110-289.

180 Ted Gayer, "Should Congress Extend the First-Time Homebuyer Credit?" Brookings Institution (Sept. 24, 2009), available at See also Howard Gleckman, "The Homebuyer Tax Credit: When Will They Ever Learn?" Tax Policy Center TaxVox Blog (Oct. 22, 2009), available at

181 See American Recovery and Reinvestment Act of 2009, P.L. 111-5.

182 See Worker, Homeownership, and Business Assistance Act of 2009, P.L. 111-92. In 2010 Congress extended the closing deadline of the homebuyer tax credit by five months. Homebuyer Assistance and Improvement Act of 2010, P.L. 111-198.

183 David Streitfeld, "With FHA Help, Easy Loans in Expensive Areas," The New York Times, Nov. 21, 2009, at B1.

184 See supra note 182. Beneficiaries of the five-year carryback included Pulte Homes ($450 million refund, $1.5 billion cash on hand), Hovnanian Homes ($250 million refund, $550 million cash), Standard Pacific ($80 million refund, $525 million cash), and Beazer Homes ($50 million refund, $557 million cash). Gretchen Morgenson, "Home Builders (You Heard That Right) Get a Gift," The New York Times, Nov. 14, 2009, available at

185 See Robert C. Pozen, "The Mortgage Interest Deduction Needs to Be Slashed," Brookings Institution (Mar. 28, 2011), available at; Ted Gayer, "The Mortgage Interest Deduction and Negative Equity," Brookings Institution (Dec. 15, 2010), available at

186 See; Howard Gleckman, "What Would Happen If Congress Rewrote the Mortgage Interest Deduction?" Tax Policy Center TaxVox Blog (Mar. 1, 2011), available at Gleckman, "Should We Dump the Home Mortgage Interest Deduction?" Tax Policy Center TaxVox Blog (May 27, 2010), available at

187 Alex Brill, "A Pro-Growth, Progressive, and Practical Proposal to Cut Business Tax Rates," American Enterprise Institute (Jan. 19, 2012), available at

188 See

189 See House Committee on the Budget, "The Path to Prosperity: Restoring America's Promise, Fiscal Year 2012 Budget Resolution" (2012); Stephen Ohlemacher, "Bipartisan Tax Plan Trims Mortgage Deduction," Yahoo News (July 20, 2011) (discussing the "Gang of Six" debt-reduction proposals), available at; Gleckman, "Tom Coburn: Tax Subsidies Are Socialism," Tax Policy Center TaxVox Blog (July 19, 2011), available at

190 See supra Section IV.B.

191 See N. Gregory Mankiw, "A Better Tax System (Assembly Instructions Included)," The New York Times, Jan. 21, 2012 (Mitt Romney's economic adviser calling for abolishing the MID), available at; Paresh Dave, "GOP Economy: Jon Huntsman's Economic Plan," Marketplace, American Public Media (Nov. 28, 2011) (discussing Jon Huntsman's plan for limiting the MID and other housing subsidies), available at


About Tax Analysts

Tax Analysts is an influential provider of tax news and analysis for the global community. Over 150,000 tax professionals in law and accounting firms, corporations, and government agencies rely on Tax Analysts' federal, state, and international content daily. Key products include Tax Notes, Tax Notes Today, State Tax Notes, State Tax Today, Tax Notes International, and Worldwide Tax Daily. Founded in 1970 as a nonprofit organization, Tax Analysts has the industry's largest tax-dedicated correspondent staff, with more than 250 domestic and international correspondents. For more information, visit our home page.

For reprint permission or other information, contact