Townhouses by Shutterstock.com
- The federal tax treatment of owner-occupied housing cries out for reform.
- Current tax policy offers unwarranted subsidies for the purchase of expensive homes by high-income taxpayers.
- Current tax policy does little to promote homeownership by those of more modest means.
In this policy proposal - part of The Hamilton Project's 15 Ways to Rethink the Federal Budget - Alan Viard proposes to replace the mortgage interest deduction with a refundable credit as a way to reduce the artificial incentive for the construction of high-end homes by better targeting the tax breaks for housing. The article has been revised from the published version to remove the unrealistic $300 billion 10-year revenue gain estimated in the published version. The author thanks Sheila Crowley for detecting the flaw in the original revenue estimate.
Deficit Reduction (10-year): Uncertain
Broader Benefits: Reduces the artificial incentive for the construction of high-end homes by reducing and better targeting the tax breaks for housing.
The federal tax treatment of owner-occupied housing cries out for reform. Current tax policy offers unwarranted subsidies for the purchase of expensive homes by high-income taxpayers, but does little to promote homeownership by those of more modest means. To address these problems, I propose to replace the mortgage interest deduction with a refundable credit and to reduce the size of the mortgages eligible for the credit while providing transition relief. Although this proposal is not ideal in every respect, it offers an effective way to scale back and better target the tax system's housing tax breaks while raising revenue in a progressive manner.
THE CURRENT TAX PREFERENCE
An owner-occupied home provides a return in the form of housing services, the value of which can be measured as the cost of obtaining the same services from a rental property. To maintain neutrality with respect to the current taxation of business capital, the tax system would need to tax homeowners on this return, often called imputed rent, while allowing a deduction for the associated costs, including mortgage interest payments.
The current income tax system does not do this. Instead, it gives homeowners the best of both worlds, sparing them from tax on imputed rent, yet allowing many of them to deduct their mortgage interest payments. Although taxpayers who claim the standard deduction may not deduct mortgage interest, itemizers may deduct the interest paid on up to $1 million of mortgage debt plus up to $100,000 of home equity loans. The dollar limits are not adjusted for general inflation or for home price fluctuations. Mortgage interest on a second home may be deducted, provided that the total interest deduction remains within the dollar limits. Essentially the same rules apply under the alternative minimum tax, except that home equity loan interest cannot be deducted.
The tax advantage for owner-occupied housing is not the mortgage interest deduction, which would be allowed under a neutral tax system, but rather the tax exemption for imputed rent. It is convenient, however, to break down the tax advantage into two components, one of which is linked to mortgage interest. Suppose that a taxpayer who itemizes deductions and is in the top 39.6 percent bracket (rounded to 40 percent for simplicity) owns a home worth $1.5 million with a $1 million mortgage. If the home provides a 5 percent rate of return in terms of housing services and the mortgage rate is also 5 percent, then the taxpayer receives $75,000 of imputed rent and pays $50,000 of mortgage interest. Under a neutral tax system, the homeowner would pay $10,000 of tax on imputed rent minus mortgage interest; under the current tax system, the homeowner actually receives a $20,000 tax saving from deducting the mortgage interest. The $30,000 total tax advantage provided by the current tax system, which is equal to 40 percent of the imputed rent, can be broken down into a $20,000 benefit from the mortgage deduction and a $10,000 benefit from the failure to tax imputed rent minus mortgage interest. Table 8-1 summarizes these calculations.
Applying this breakdown to national data, the Department of the Treasury listed the mortgage deduction as a $111 billion tax expenditure, and the failure to tax imputed rent minus mortgage interest as a $59 billion tax expenditure for fiscal 2014 (Office of Management and Budget [OMB] 2012, 250).1
Because the basic flaws of the current tax treatment are well known, I cover this ground only briefly. There may be good economic grounds, and there is certainly strong political support, for promoting homeownership, but there is no case for subsidizing bigger or more-expensive homes. Yet, the current tax treatment is more geared toward the latter objective, offering the largest benefits to taxpayers in the highest brackets and providing more-generous treatment to taxpayers who itemize than to those who claim the standard deductions. Indeed, the current tax policy may actually impede homeownership for taxpayers of more modest means because the preferences for high-bracket itemizers drive up the demand for homes and boost home prices.
The tax advantage is likely to have a powerful effect on the demand for owner-occupied housing, particularly for high-income people. James Poterba and Todd Sinai conclude that, relative to a neutral system that taxes imputed rent, the current system reduces the cost of investing in owner-occupied housing by about 20 percent on average and by almost 40 percent for the highest-income households (Poterba and Sinai 2011, 559-561).
The tax system should be changed to curtail this artificial incentive that inefficiently diverts resources away from business capital and toward the construction of high-end homes.
SUMMARY AND COMPARISON TO OTHER PLANS
Starting in 2015, the mortgage interest deduction is converted to a 15 percent refundable tax credit available to all homeowners, including those who claim the standard deduction and those with no income tax liability. (The 15 percent value is illustrative; as discussed below, a lower value may be necessary to ensure a revenue gain, relative to current law.) The credit is limited to interest on $300,000 of mortgage debt (in 2013 dollars), with no tax relief for mortgages on second homes or on home-equity loans. The dollar limit is indexed to the consumer price index (CPI) in the same manner as the bracket endpoints and other dollar values in the tax code. Taxpayers with existing debt are allowed to claim 90 percent of the current-law deduction in 2014 on that debt, declining 10 percent per year thereafter, with the option to switch to the credit at any time.
By replacing the deduction with a credit, the proposal follows an approach that has been embraced by many economists and that has appeared in several recent reform plans.2 The President's Advisory Panel on Federal Tax Reform's 2005 plan also featured a 15 percent refundable credit, on mortgages up to 125 percent of the median home price in the taxpayer's county, and no relief for second homes and home-equity loans. Taxpayers were allowed to choose between the deduction and the credit for five years, with the dollar limits phased in over four years and the second-home and home-equity provisions effective immediately (President's Advisory Panel on Federal Tax Reform 2005, 73-74, 237-238). The Bipartisan Policy Center's (BPC's) November 2010 plan called for a 15 percent refundable credit, administered as a matching grant, on up to $25,000 of interest payments, with no tax relief for second homes. The limit would remain fixed in nominal terms and no transition relief was mentioned (BPC Debt Reduction Task Force 2010, 35-36, 126). The Simpson-Bowles December 2010 illustrative tax reform plan featured a 12 percent credit with a $500,000 limit and no tax relief for second homes and home-equity loans, with "appropriate transition relief" (National Commission on Fiscal Responsibility and Reform 2010, 26-27). A proposal in President Obama's budget would limit high-income households' federal income tax savings to 28 percent of deductible mortgage interest payments, with no transition relief. For the affected taxpayers, the deduction would effectively be replaced by a 28 percent tax credit, but there would be no tightening of the $1 million limit (U.S. Department of the Treasury 2012, 73-74).
This proposal seeks to promote sound economic policy while being sensitive to political realities. Like the other recent reform plans, this proposal does not end the tax preference for homeownership, but merely scales it back and retargets it toward less-expensive homes and taxpayers of more modest means. The economic merits of a homeownership preference depend on whether homeownership generates spillover benefits for society as a whole, perhaps by promoting social stability or by encouraging residents' neighborhood involvement. Rather than wading into this contentious debate, however, this proposal accepts the political reality that complete removal of the tax preference, or even of the mortgage deduction, is impossible, and instead seeks to target the tax preference in a more rational manner. Opinion polls suggest that many Americans who are unwilling to eliminate the mortgage deduction are willing to restrict it.3
Political realities also shape another feature of the proposal. Although it would be preferable to directly eliminate the tax advantage for expensive homes by taxing imputed rent on such homes, imputed rent taxation is politically impossible and administratively difficult. Like the other reform plans, the proposal allows imputed rent to remain untaxed and instead limits the mortgage deduction. As discussed further below, this approach regrettably leaves fully intact the current tax advantage for the equity that homeowners have in their homes and limits the tax advantage only on the mortgaged portion of home value.These concerns should not overshadow the fundamental advantages of the proposal. For the mortgaged portion of home purchases, everyone receives the same 15 percent marginal incentive on modestly priced homes and no one receives any additional incentive for expensive homes. The proposal substantially limits the tax preference for expensive homes while increasing homeownership assistance for taxpayers who are less well off.
The proposal sets a uniform national limit on the size of a mortgage that can receive tax relief, which is the approach taken by the current $1 million limit and the BPC and Simpson-Bowles plans. Arguments can be made for the alternative approach of having the limits vary with local home prices, as in the Tax Reform Panel's plan. Linking the limit to local home prices might help ensure that tax relief applies to modestly priced homes everywhere in the country by accounting for variations in the price of modest homes. It might also ease political opposition in high-cost areas.4 But there are countervailing considerations. Although it might be desirable to link the limit to a measure of the local cost of buying a home of fixed quality, the median home price may be a poor proxy for that unavailable measure. In areas with higher median home prices, homeowners may be living in homes of higher quality and enjoying better community amenities. They should not receive additional tax relief to accommodate those choices, particularly if they are affluent. In addition, setting higher limits for higher-cost areas might increase political resistance to the proposal in low-cost areas. Moreover, if the limit were linked to each area's home prices, then it also would presumably rise and fall over time with swings in home prices. But, there is no reason for the creditable portion of a homeowner's mortgage payments to change year to year over the life of the mortgage in response to home price movements. A uniform nationwide limit that is indexed to the CPI avoids those problems and is also simpler.
There is no easy way to select the right level of the limit. Previous reform plans have made a variety of choices. The BPC's $25,000 interest limit is consistent with a $312,500 mortgage at an 8 percent mortgage rate or an $833,333 mortgage for a borrower with a 3 percent rate. To avoid this sensitivity to interest-rate fluctuations, this proposal follows current law and the Panel and Simpson-Bowles plans by applying the limit to the mortgage value rather than to the interest payments. This proposal adopts a $300,000 limit, significantly more restrictive than the $500,000 Simpson-Bowles limit. Census Bureau data show that the median sales price for a new home was $248,900 and the mean price was $304,000 in December 2012; the nominal values of these series peaked in March 2007, with the median at $262,600 and the mean at $329,400 (U.S. Census Bureau 2013). The proposal's $300,000 limit therefore accommodates a mortgage roughly equal to the mean new-home sales price.
Although the $300,000 limit may seem stringent, it provides even expensive homes with a substantial tax advantage, though not to the same extravagant extent as the current tax system. Recall the previous example of the owner of a $1.5 million home who received a $30,000 tax advantage under current law by avoiding $10,000 of tax on imputed rent minus mortgage interest and reaping a $20,000 tax saving from the mortgage deduction. Under this proposal, that owner would keep the $10,000 tax advantage from the exemption of imputed rent minus mortgage interest and also would receive $2,250 from the 15 percent mortgage credit. The proposal seems stringent only when compared to the unrestrained tax breaks in place today.
The proposal offers significant transition relief, a policy that is desirable in its own right as well as being politically necessary. Because housing is a large investment and taxpayers have relied on a longstanding policy, they should receive some protection from unexpected changes. It certainly is far better to address concerns about market disruption by providing transition relief than by scaling back the underlying reform.
There is ample room to alter the proposal's parameters and design features. Disagreements about details should not be allowed to impede the adoption of a reform that better targets housing tax preferences.
The proposal seeks to direct economic resources away from expensive homes, which have been artificially advantaged by the tax system, and toward other sectors of the economy. Like any proposal that limits the mortgage deduction rather than taxing imputed rent, however, its effectiveness may be diminished by undesired changes in assets and debts. Consider yet again the taxpayer with the $1.5 million home and the $1 million mortgage. If the taxpayer responds to the proposal by selling off $1 million of other assets and paying off the mortgage, then the proposal does not diminish the housing tax advantage and raises no revenue. The tax savings previously obtained from deducting interest on a $1 million mortgage are replaced by the tax savings from no longer paying tax on the income from $1 million of other assets, as the taxpayer continues to fully enjoy the benefits of tax-free imputed rent. The same results occur if the taxpayer pays off the mortgage with $1 million borrowed against other assets and deducts the interest on the new debt as investment interest. Limits on the mortgage deduction can be thwarted because they withdraw the tax advantage only for home purchases financed by mortgages, sparing home purchases financed by other borrowing or by drawing down other assets.
In practice, though, the homeowner may not have $1 million of other assets, or may be unwilling or unable to sell or borrow against other assets. Limits on mortgage tax relief can remain effective if homeowners have little ability to change their portfolios. Fortunately, the evidence suggests that this is generally the case. Poterba and Sinai (2011) survey the extensive literature on this topic and provide new estimates of the scope of potential portfolio changes based on an analysis of the 2004 Survey of Consumer Finances. Examining a proposal to lower the mortgage deduction cap to $250,000, they conclude that, even under relatively generous assumptions about households' ability to liquidate other assets, the portfolio changes will undo less than one-quarter of the proposal's potential revenue gain. They find that changes to the mortgage deduction curtail the tax advantage of housing by almost as much as if there were no portfolio changes at all (Poterba and Sinai 2011, 555-556, 559-560).
BUDGET AND DISTRIBUTIONAL CONSEQUENCES
The Urban-Brookings Tax Policy Center recently provided budget and distributional estimates for a reform option that would phase in a 15 percent credit and a $500,000 cap over a five-year period. The Center's option differs from the proposal here; the option sets a higher cap, but offers less transition relief and is nonrefundable. Relative to an August 2011 current-policy baseline, the Center estimated a ten-year revenue gain of $324 billion from the option (Urban-Brookings Tax Policy Center 2011a).5 The Center estimated that in 2015, 81 percent of the tax increase would fall on tax units with cash income above $200,000 in 2011 dollars, with 18 percent falling on those above $500,000 and 6 percent on those above $1 million (Urban-Brookings Tax Policy Center 2011b).6
The proposal presented here would not have the same revenue gain as the option examined by the Tax Policy Center. Due to the refundability of the credit in this proposal, there might be no revenue gain relative to current law. A revenue gain could be ensured, however, by lowering the credit rate sufficiently below the proposed 15 percent value.
Any retrenchment of the mortgage deduction is likely to reduce the value of existing homes, compounding the recent declines in home values. The proposal offers transition relief to cushion the blow to current homeowners. Moreover, the price impact is likely to be more modest than some observers have suggested.
In general, a reduction in housing tax benefits has price and quantity effects, reducing both the value of existing housing and the quantity of new housing. As Jane Gravelle, John Diamond, George Zodrow, and others have explained, the relative sizes of the two effects depend on the flexibility of the housing supply. If the housing supply is completely fixed, the quantity effect disappears and the price effect is very strong, with the value of existing homes falling dollar for dollar with the present value of the lost tax benefits. But if the quantity of housing falls quickly in response to the tax change, the price change is dampened as the scarcity of housing bolsters home prices. Gravelle (1996) and Diamond and Zodrow (2008) point out that previous predictions of large declines of home prices from tax reforms were based on the assumption that the housing supply is completely fixed (which is clearly untrue) or that it is very slow to adjust. They find that, under more realistic assumptions about the responsiveness of housing supply, even tax reforms that are far more sweeping than this proposal have modest price effects. For example, under an assumption of moderate flexibility in housing supply, Diamond and Zodrow (2008) estimate only a 4.2 percent decline in home equity values from a flat-tax reform that completely eliminates the tax advantage for housing. Gravelle (1996) also notes that the historical record does not support large home price impacts of tax changes.
If there is no change in the standard deduction, then the availability of the credit to taxpayers claiming the standard deduction will reduce the number of taxpayers choosing to itemize, diminishing incentives to engage in other tax-deductible activity such as charitable giving. Adam Cole, Geoffrey Gee, and Nicholas Turner estimated that a similar credit proposal would reduce the number of itemizing returns by 21 million in 2021 (Cole, Gee, and Turner 2011, 993). If that result is not desired, it can be counteracted by lowering the standard deduction while increasing the personal exemption and other provisions to prevent a tax increase on low-income households.
Reducing the deficit will require action on many fronts. Replacing the mortgage interest deduction with a refundable credit and reducing the size of the mortgage eligible for tax relief can be an efficient and progressive part of the solution. This approach would preserve the tax incentive for homeownership while targeting it in a more effective and equitable manner.
Alan D. Viard
Resident Scholar, American Enterprise Institute
Alan Viard is a resident scholar at the American Enterprise Institute (AEI). Before joining AEI, he was a senior economist at the Federal Reserve Bank of Dallas and an assistant professor of economics at Ohio State University. Viard has worked for the U.S. Department of the Treasury's Office of Tax Analysis, the White House's Council of Economic Advisers, and the Joint Committee on Taxation of the U.S. Congress. He is a frequent contributor to AEI's "On the Margin" column in Tax Notes. Viard is the co-author of Progressive Consumption Taxation: The X Tax Revisited published in May 2012.
1. The Joint Committee on Taxation lists only the mortgage deduction as a tax expenditure, pegging its value at $100 billion for fiscal 2014 (Joint Committee on Taxation 2012, 36).
2. Economists' support for a credit is discussed by Shawn Zeller (Zeller 2012, 2329-2330).
3. For example, in a Quinnipiac University poll conducted from November 28 to December 3, 2012, respondents rejected abolition of the mortgagededuction 67 percent to 23 percent, but supported limiting the deduction to the interest on a $500,000 mortgage 62 percent to 28 percent and supported eliminating the deduction for second homes 56 percent to 35 percent (Quinnipiac University 2012).
4. It is sometimes suggested that geographic variation in the limit would violate the requirement in Article 1, section 8, clause 1 of the U.S. Constitution that taxes be "uniform throughout the United States." The objection has little force. In its 1983 decision in United States v. Ptasynski, the U.S. Supreme Court stated that the clause permits geographic variation if it is based on neutral factors and unanimously upheld preferential treatment for oil produced in a region with higher production costs. Setting a higher deduction limit for areas with higher housing costs is clearly analogous. The tax system also includes numerous geographically targeted provisions, such as empowerment zones, that are based on areas' economic characteristics. In any event, the proposal's use of a national limit avoids any potential constitutional problem.
5. The option referred to in the text is listed as option 3 in the table. The revenue estimate allowed for some mortgage pay-down and portfolio changes.
6. Under the option's phase-in provision, there is a 19 percent maximum savings and a $600,000 cap in 2015.
Bipartisan Policy Center (BPC) Debt Reduction Task Force. 2010. "Restoring America's Future: Reviving the Economy, Cutting Spending and Debt, and Creating a Simple, Pro-Growth Tax System." November. http://bipartisanpolicy.org/library/report/restoring-americas-future
Cole, Adam J., Geoffrey Gee, and Nicholas Turner. 2011. "The Distributional and Revenue Consequences of Reforming the Mortgage Interest Deduction." National Tax Journal 64 (4, December): 977-1000.
Diamond, John W., and George R. Zodrow. 2008. "Consumption Tax Reform: Changes in Business Equity and Housing Prices." In Fundamental Tax Reform: Issues, Choices, and Implications, edited by John W. Diamond and George R. Zodrow, 227-260. Cambridge, MA: MIT Press.
Gravelle, Jane G. 1996. "Effects of Flat Taxes and Other Proposals on Housing: An Overview." Congressional Research Service Report 96-552, June 17. Congressional Research Service, Washington, DC.
Joint Committee on Taxation. 2012. "Estimates of Federal Tax Expenditures for Fiscal Years 2011-2015." January. JCS-1-12. https://www.jct.gov/publications.html?func=startdown&id=4386
National Commission on Fiscal Responsibility and Reform. 2010. The Moment of Truth. December. http://www.fiscalcommission.gov/sites/fiscalcommission.gov/files/documents/TheMomentofTruth12_1_2010.pdf
Office of Management and Budget (OMB). 2012. "Fiscal Year 2013: Analytical Perspectives: Budget of the U.S. Government." February. http://www.whitehouse.gov/sites/default/files/omb/budget/fy2013/assets/spec.pdf
Poterba, James M., and Todd Sinai. 2011. "Revenue Costs and Incentive Effects of the Mortgage Interest Deduction for Owner-Occupied Housing." National Tax Journal 64 (2, pt. 2, June): 531-564.
President's Advisory Panel on Federal Tax Reform. 2005. "Simple, Fair, and Pro-Growth: Proposals to Fix America's Tax System." 2005. November. http://www.treasury.gov/resource-center/tax-policy/Documents/Simple-Fair-and-Pro-Growth-Proposals-to-Fix-America's-Tax-System-11-2005.pdf
Quinnipiac University. 2012. "U.S. Voters Trust Obama, Dems to Avoid Fiscal Cliff, Quinnipiac University National Poll Finds; Tax The Rich, But Don't Touch Medicare, Voters Say." http://www.quinnipiac.edu/institutes-centers/polling-institute/national/release-detail/?ReleaseID=1821
United States v. Ptasynski. 1983. 462 U.S. 74.
Urban-Brookings Tax Policy Center. 2011a. "Table 11-0271." August 18, Option 3. http://www.taxpolicycenter.org/numbers/Content/PDF/T11-0271.pdf
Urban-Brookings Tax Policy Center. 2011b. "Table 11-0281." August 18. http://www.taxpolicycenter.org/numbers/Content/PDF/T11-0281.pdf
U.S. Census Bureau. 2013. "Median and Average Sales Prices of New Homes Sold in United States." http://www.census.gov/construction/nrs/pdf/uspricemon.pdf (visited January 31, 2013)
U.S. Department of the Treasury. 2012. "General Explanations of the Administration's Fiscal Year 2013 Revenue Proposals." February. http://www.treasury.gov/resource-center/tax-policy/Documents/General-Explanations-FY2013.pdf
Zeller, Shawn. 2012. "Tweaking the Sacred Mortgage Tax Break: Economists May be Gaining Ground with Their Call to Convert the Current Interest Deduction into a Credit." Congressional Quarterly (November 19).