CRS Finds Uneven Mortgage Deduction Distribution

An Analysis of the Geographic Distribution of the Mortgage Interest Deduction: Before and After the 2017 Tax Revision (P.L. 115-97)

February 18, 2021

Congressional Research Service
https://crsreports.congress.gov
R46685

Policymakers may be interested in the mortgage interest deduction (MID) because it is associated with homeownership and is one of the largest tax benefits available to homeowners in terms of forgone federal tax revenue. P.L. 115-97, commonly referred to as the Tax Cuts and Jobs Act (TCJA), reduced the maximum mortgage balance eligible for the deduction and restricted the deduction of interest associated with home equity loans starting in the 2018 tax year. The TCJA also increased the standard deduction and limited the deduction for state and local taxes (SALT), which reduced the number of taxpayers who claim itemized deductions generally, including for mortgage interest. Barring congressional action, the changes to the mortgage interest deduction (and the standard and SALT deductions) will revert to pre-TCJA law after 2025.

The mortgage interest deduction is one of hundreds of tax benefits considered a tax expenditure. Tax expenditures can generally be viewed as government spending administered via the tax code, and may also be intended to achieve particular policy objectives. Regardless of the interpretation, tax expenditures like the mortgage interest deduction provide a benefit to qualifying taxpayers by lowering their federal tax liabilities. For this reason, and because some policymakers have expressed interest in increasing equity (fairness) in the tax code, it is important to understand how the mortgage interest deduction's benefits are distributed by income class. Additionally, understanding how the deduction's benefits are distributed across taxpayers in different states may help Members of Congress assess how potential policy changes might affect their constituents.

This report's analysis — which examines the mortgage interest deduction's geographic distribution in the years preceding (2016) and following (2018) TCJA's enactment — indicates that the deduction's benefits are not distributed uniformly across the states according to several measures of benefit. A number of reasons likely explain why the variation exists, including differences in homeownership rates, home prices, state and local tax policies, and area incomes. The data used in this report, however, are not detailed enough to isolate and quantify the individual factors' effects on the variation across states.

The report concludes by providing a number of policy options and considerations for Congress. Analysis of these options suggests that some of them may provide a benefit that is more uniformly distributed across geographic areas. For example, limiting the size of mortgages that qualify for the deduction could reduce some of the variation caused by regional differences in home prices. Replacing the deduction with a credit, or limiting the rate at which interest could be deducted, could reduce variation in benefits caused by differences in area incomes. Still, it is important to understand that any change to the mortgage interest deduction would likely require careful consideration of how to transition to the new policy while minimizing disruptions to the housing market and overall economy.

---

Contents

Introduction

Background

Data Analysis

Policy Options and Considerations

Figures

Figure 1. Mortgage Interest Deduction Tax Expenditures by State, 2018

Figure 2. Percentage Change in Mortgage Interest Deduction Tax Expenditure by State, 2016 to 2018

Figure 3. Percentage of Homeowners Who Claimed the MID by State, 2018

Figure 4. Percentage Change In Share of Homeowners Who Claimed the MID by State, 2016 to 2018

Figure 5. Mortgage Interest Deduction Tax Expenditure Per Claimant by State, 2018

Figure 6. Percentage Change in Mortgage Interest Deduction Tax Expenditure Per Claimant by State, 2016 to 2018

Figure 7. Homeownership Rate by State, 2018

Tables

Table 1. Distribution of Mortgage Interest Deduction Tax Expenditure by Income Class, 2016 and 2020

Appendixes

Appendix. Estimation Methodology

Contacts

Author Information

---

Introduction

This report analyzes the geographic distribution of the mortgage interest deduction (MID) tax expenditure in the years preceding and following enactment of P.L. 115-97, commonly referred to as the Tax Cuts and Jobs Act (TCJA).¹ The TCJA reduced the maximum mortgage balance eligible for the deduction and restricted the deduction of interest associated with home equity loans, starting in the 2018 tax year. The TCJA also increased the standard deduction and limited the deduction for state and local taxes (SALT),² which reduced the number of taxpayers who claim itemized deductions generally, including for mortgage interest. Barring congressional action, the changes to the mortgage interest deduction (and the standard and SALT deductions) will revert to pre-TCJA law after 2025.

The mortgage interest deduction interests policymakers because it is associated with homeownership and is one of the largest tax benefits available to homeowners in terms of forgone federal tax revenue. Tax expenditures can generally be viewed as government spending administered via the tax code. These tax benefits may also be intended to achieve particular policy objectives. Regardless of the interpretation, tax expenditures like the mortgage interest deduction provide a benefit to qualifying taxpayers by lowering their federal tax liabilities. For this reason, and because some policymakers have expressed interest in increasing equity (fairness) in the tax code, it is important to understand how the mortgage interest deduction's benefits are distributed by income class. Additionally, understanding how the deduction's benefits are distributed across taxpayers in different states may help Members of Congress assess how potential policy changes might affect their constituents.³

Background

Homeowners have had the ability to deduct the interest paid on home mortgages since the introduction of the modern federal income tax code in 1913. Congress recognized the importance of allowing for the deduction of expenses incurred in the generation of income, which is consistent with traditional economic theories of income taxation.⁴ At the time the framework for the federal income tax code was being laid, most interest payments were business-related expenses. Compared to the present day, households generally had very little debt on which interest payments were required — credit cards did not yet exist and the mortgage finance industry was in its infancy. As a result, all interest payments were deductible, with no distinction made for business, personal, living, or family expenses.

For more than 70 years, there was no limit on the amount of home mortgage interest that could be deducted. That changed with the Tax Reform Act of 1986 (TRA86; P.L. 99-514), which restricted the amount of mortgage interest that could be deducted and limited the number of homes for which the deduction could be claimed to two. Mortgage interest deductibility was limited to the purchase price of the home, plus any improvements, and to debt secured by the home but used for qualified medical and educational expenses.⁵ Subsequently, the Omnibus Budget Reconciliation Act of 1987 (P.L. 100-203) established the basic deduction limits in effect until enactment of the TCJA.

Overview of the TCJA MID Rules

Under current law (i.e., since the TCJA), a taxpayer may claim an itemized deduction for “qualified residence interest,” which can include interest paid on a mortgage secured by a principal residence and a second residence. The amount of interest that is deductible depends on when the mortgage debt was incurred. For mortgage debt incurred on or before December 15, 2017, the combined mortgage limit is $1 million ($500,000 for married filing separately). For mortgage debt incurred after December 15, 2017, the deduction is limited to the interest incurred on the first $750,000 ($375,000 for married filing separately) of combined mortgage debt.

If a taxpayer has mortgage debt exceeding the applicable mortgage limit ($750,000 or $1 million), he or she may still claim a deduction for a percentage of interest paid equal to the applicable mortgage limit divided by the remaining mortgage balance. For example, a homeowner whose mortgage was originated after December 15, 2017, and who has a balance of $1 million could deduct 75% ($750,000 divided by $1 million) of the interest payments.

Refinanced mortgage debt is treated as having been incurred on the origination date of the original mortgage for purposes of determining which mortgage limit applies ($750,000 or $1 million). The balance of the new loan resulting from the refinance, however, may not exceed the balance of the original loan. This may occur, for example, when a homeowner “cashes out” equity in the home by obtaining a larger loan than is necessary to pay off the current mortgage balance.

For purposes of the deduction, mortgage debt includes home equity loans secured by a principal or second residence that are used to buy, build, or substantially improve a taxpayer's home. Mortgage debt does not include home equity loans when the proceeds are used for purposes unrelated to the property securing the loan. For example, interest associated with a home equity loan that is used to pay off a credit card balance, go on a vacation, or send a child to college does not qualify for the mortgage interest deduction. The restrictions on the use of home equity loans apply irrespective of when the loan was originated.

Comparison to Prior Law

Under prior law (i.e., immediately preceding the TCJA), a homeowner was allowed an itemized deduction for the interest paid on the first $1 million of combined mortgage debt associated with a primary or secondary residence. As with current law, a homeowner could deduct a percentage of interest paid if the mortgage balance exceeded the $1 million limit. Additionally, a homeowner was allowed to deduct the interest on the first $100,000 of home equity debt regardless of whether or not the taxpayer incurred the debt to finance costs associated with the home. For example, under prior law, a homeowner could use a home equity loan to purchase a boat, pay for a child's college, cover medical costs, or any number of other things not involving the property that secured the loan and still deduct the associated interest. In contrast, a home equity loan used for these purposes does not qualify for the mortgage interest deduction under current law.

Data Analysis

The Joint Committee on Taxation (JCT) has estimated that the mortgage interest deduction reduced federal tax revenues by $25.5 billion in FY2020 and will reduce revenues by $23.7 billion in FY2021.⁶

The following analysis describes how the mortgage interest deduction's benefits are distributed across states, and what impact, if any, the TCJA has had on that distribution. For the purposes of this analysis, the benefits of the deduction to claimants are measured in terms of the revenue loss associated with the provision. Because the JCT does not produce tax expenditure estimates on a state-by-state basis, CRS used an approach that accounts for state-level differences in incomes and in amounts of mortgage interest deducted to allocate the JCT's national expenditure estimate to the states. This approach is explained in the Appendix.

The analysis focuses on FY2016 and FY2018. Although the JCT has published more recent estimates, the Internal Revenue Service (IRS) tax data necessary for a state-by-state analysis are only available through 2018. These years also correspond to the years preceding and following enactment of the TCJA. Unless otherwise noted, the JCT estimates used in the analysis below are from the JCT's distributional estimates (by income), which vary slightly from JCT's line item analysis.

The following sections analyze the mortgage interest deduction by state and changes that occurred in each state after the TCJA. An analysis of factors that may have led to these changes is provided in the subsequent section (“Reasons for the Variation in MID Beneficiaries”).

Tax Expenditure by State

Figure 1 displays the estimated mortgage interest deduction tax expenditure for each state in 2018.⁷ The data presented in the figure may be interpreted in one of two ways: (1) the amount of federal spending per state administered through the tax code that is attributable to the mortgage interest deduction; (2) the reduction in total federal tax liability realized collectively by individuals in each state from allowing mortgage interest to be deducted. Nationwide, the average tax expenditure per state in 2018 was $489 million. Naturally, more populous states tended to realize larger benefits from the deduction because they generally had more total homeowners who were eligible to claim it. For example, the tax expenditure attributable to California (most populous) was $6.3 billion, while the expenditure for Wyoming (least populous) was $22.2 million.

Figure 2 displays the percentage change in the estimated mortgage interest deduction tax expenditure by state between 2016 and 2018. Following the TCJA, there was an average 61% decrease in the estimated tax expenditure per state. The largest decreases occurred in four midwestern states — Iowa (-77%), Nebraska (-75%), Ohio (-75%), and Wisconsin (-76%) — plus West Virginia (-75%) and Vermont (-76%). The smallest decreases occurred in California (-52%), Washington, DC (-46%), Hawaii (-55%), and Maryland (-56%).

Share of Homeowners Claiming MID by State

Many homeowners do not claim the mortgage interest deduction. Several factors may explain this, including not having a mortgage, having low mortgage payments (from being toward the end of the mortgage period, from living in a low-cost area, or because of historically low interest rates), or living in a state without an income tax. Following the TCJA's enactment, one of the largest factors driving the relatively low MID claim rates was a significant reduction in overall itemization rates caused by the near doubling of the standard deduction and the $10,000 limit placed on the deduction for state and local income taxes (SALT).⁸

Figure 3 shows that about 18% of all U.S. homeowners claimed the deduction in 2018.⁹ Homeowners in West Virginia had the lowest claim rate at 5%, while homeowners in DC had the highest claim rate at 52%, followed by Maryland at 40%. States in the middle and southern portions of the country tended to have the lowest percentages of homeowners who claimed the deduction. States on the West Coast, in parts of the mid-Atlantic, and in the Northeast had some of the highest claim rates, as did Colorado, Utah, and Georgia.

Figure 4 displays the percentage change between 2016 and 2018 in the share of homeowners who claimed the mortgage interest deduction by state. Following the TCJA, the estimated claim rate among all homeowners decreased by 60% on average. The largest decrease occurred among homeowners in Iowa (-78%), while the smallest decreased occurred among homeowners in DC (-30%), followed by California (-40%). Overall, reduced claim rates appeared to be mostly concentrated in the middle of the country, though several New England states experienced greater-than-average declines.

Tax Expenditure Per MID Claimant by State

Figure 5 displays the geographic distribution of the average mortgage interest deduction tax expenditure per claimant for each state. The data show that U.S. taxpayers claiming the mortgage interest deduction saved approximately $1,772 in taxes on average in 2018. Figure 5 indicates that there was variation among states in the average benefit received by those claiming the deduction. Claimants in Washington, DC, received the largest average benefit ($2,404) from the deduction, followed by claimants in California ($2,272). At the other end of the spectrum, claimants in Mississippi received the smallest average benefit ($1,090), followed by claimants in Iowa ($1,179). Stated differently, on average, claimants in DC had their tax liability reduced by a little more than twice as much as claimants in Mississippi.

Figure 6 displays the average percentage change between 2016 and 2018 in the mortgage interest deduction tax expenditure per claimant by state. Following the TCJA, there was an average 8% ($157) decrease in the estimated tax expenditure per claimant across the country. The largest decreases occurred in DC (-30%) and Maryland (-27%). Claimants in Iowa and Tennessee experienced no change in their benefits, on average. Claimants in six states — Idaho (3%), Indiana (3%), Kentucky (5%), Ohio (1%), South Dakota (6%), and West Virginia (1%) — realized increased benefits on average.

Reasons for the Variation in MID Beneficiaries

A number of factors likely contribute to the state-by-state variation in the per-claimant mortgage interest deduction tax expenditure. Isolating and quantifying each factor's precise effect is complicated by the interaction of the various factors and the use of state-level data. Still, it is useful to highlight general differences among states that are likely contributing to the variation. Understanding the causes of state-by-state variation may be helpful in analyzing potential policy changes.

The TCJA

The TCJA's lower mortgage limits reduced the amount of interest that can be deducted relative to prior law, but that reduction may not be the TCJA's most significant contribution to variation across states. Other TCJA changes, specifically the near doubling of the standard deduction and the $10,000 SALT deduction limit, are estimated to have reduced the overall itemization rate. Shortly after TCJA enactment, the Tax Policy Center estimated the act would reduce the overall itemization rate from 26.4% of taxpayers to 10.9%.¹⁰ Because taxpayers must itemize to claim the mortgage interest deduction, fewer homeowners now benefit from the deduction (see Figure 4). As a result, the TCJA also reduced the overall cost associated with the mortgage interest deduction (see Figure 2).

Table 1. Distribution of Mortgage Interest Deduction Tax Expenditure by Income Class, 2016 and 2020

	2016		2020
Income Class	Share of Claimants	Share of Tax Expenditure	Share of Claimants	Share of Tax Expenditure
Below $30k	1.5%	0.3%	0.7%	0.1%
$30k to $40k	2.0%	0.5%	1.0%	0.2%
$40k to $50k	3.4%	0.9%	1.9%	0.5%
$50k to $75k	13.9%	5.6%	9.0%	2.9%
$75k to $100k	15.0%	8.5%	11.3%	5.8%
$100k to $200k	43.1%	38.3%	38.8%	27.6%
$200k and over	21.2%	45.9%	37.3%	62.8%
Total	100%	100%	100%	100%
Source: CRS calculations using estimates reported in Tables 3 of U.S. Congress, Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2016-2020, 115th Cong., 1st sess., January 30, 2017, JCX-3-17, and U.S. Congress, Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2020-2024, 116th Cong., 2nd sess., November 5, 2020, JCX-23-20. Notes: 2020 data are presented instead of 2018 data, given that 2020 is the most recent tax year completed. However, the 2018 distribution is nearly identical to the 2020 distribution.

The lower itemization rate and the fact that higher-income homeowners have larger mortgage balances on average means that the mortgage interest deduction's benefits disproportionately accrue to taxpayers at the upper end of the income distribution (see Table 1), to a greater degree than under prior law. However, this does not necessarily mean that homeowners who no longer claim the mortgage interest deduction will pay higher taxes since the increased standard deduction and other TCJA changes may more than compensate for the loss of the deduction.

Homeownership Rates

Because the mortgage interest deduction is only available to homeowners, geographic variation in homeownership rates will naturally contribute to geographic variation in the distribution of the deduction's benefits. Figure 7 shows that homeownership rates varied across states, from a low of 42.3% in DC to a high of 72.5% in Minnesota in 2018.¹¹ Homeownership rates appeared to be lowest in several states that have concentrations of their populations in relatively higher cost of living areas, such as New York and California, and highest in less densely populated and lower cost of living areas, such as portions of New England and the Midwest, Delaware, Idaho, South Carolina, Utah, and Wyoming.

Homeownership rates changed very little immediately following the TCJA. The overall homeownership rate in 2016 was 63.1%, compared to 63.9% in 2018. This is perhaps not surprising given the transaction costs associated with becoming a homeowner and the size of the investment involved. It can take a number of years for households to adjust to changes that affect the cost of homeownership. Even with a number of years of data, it can be difficult to isolate the impact of any change because of other variables. For example, the homeownership rate began to slightly increase before the COVID-19 outbreak, then appeared to accelerate as the virus spread. In the third quarter of 2020 (most recent data), the homeownership rate was 67.4%.¹² However, homeownership data may reflect collection issues encountered during the COVID-19 pandemic. On March 20, 2020, the Census Bureau suspended in-person interviews, and it did not fully reintroduce them until September 2020. Although the bureau explained it had reweighted its sample to reflect lower response rates, it also cautioned researchers that homeownership rate estimates could be impacted.¹³ In the last quarter before the pandemic — the fourth quarter of 2019 — the homeownership rate was 65.1%. An increase in homeownership would increase the overall mortgage interest deduction expenditure and could impact the geographic distribution. It remains to be seen how the pandemic will impact the geographic distribution of homeownership and, in turn, the geographic distribution of the mortgage interest deduction.

All else equal, states with higher homeownership rates should expect to see higher MID claims rates because more taxpayers would be eligible for the deduction. It is less clear how well variation in the homeownership rate explains variation in the average amount of interest homeowners deduct or the average tax savings realized. Two states could have different homeownership rates but have similar average home prices and incomes, resulting in homeowners in both states deducting similar amounts of interest on average. Of course, all else is not equal in reality and other factors influencing the claims rate may also interact with the decision to become a homeowner, which in turn will influence how many people benefit from the deduction.

Home Prices

Area home prices contribute to the geographic variation in the mortgage interest deduction data. Homeowners are more likely to claim the deduction in higher-priced areas because higher home prices generally require larger mortgages, and hence larger amounts of deductible interest, leading to larger average benefits from the deduction. Thus, homeowners in two different states whose situations are otherwise identical except for the prices of their homes will realize different benefits from the deduction. Home prices are typically lower in less-populated markets than in densely populated areas and metropolitan markets.¹⁴ Accordingly, higher average home prices along the East and West Coasts likely explain some of the concentration of mortgage interest deduction benefits in these areas.

State and Local Taxes

Variation in state and local taxes, particularly state income and property taxes, likely contributes to variation in the mortgage interest deduction data.¹⁵ Homeowners can only claim the mortgage interest deduction if they itemize their deductions. An individual will only itemize if his or her itemized deductions exceed the standard deduction. As state and local income and property taxes increase, all else equal it becomes more likely that homeowners will itemize and claim the mortgage interest deduction. Nine states currently have no broad-based income tax: Alaska, Florida, Nevada, New Hampshire, South Dakota, Tennessee, Texas, Washington, and Wyoming. These states account for roughly 21% of all homeowners in the United States, with Florida and Texas combined accounting for 14% of all homeowners.

Incomes

Area incomes also influence the decision to claim the deduction. Higher area incomes will support higher home prices, which implies larger mortgages and higher interest payments. But higher incomes also imply that the same dollar of mortgage interest deducted will be more valuable than the same dollar deduction at a lower income level. Thus, all else equal, markets with higher incomes should be expected to have a higher MID claim rate.

Policy Options and Considerations

Congress has a number of options regarding the mortgage interest deduction. It is important to note that any change to the mortgage interest deduction would likely require careful consideration of how to transition to the new policy so as to minimize disruptions to the housing market and overall economy. Depending on its design, a policy modification could result in a more evenly distributed benefit to homeowners, both geographically and across incomes.

Retain the Current Deduction

One option available to Congress is to retain the deduction in its current form. This would require legislative action because the deduction is scheduled to revert to the limits in place prior to TCJA after 2025. In other words, beginning in 2026 — absent any legislative action — interest will be deductible on the first $1 million of combined (first and second home) acquisition debt, plus interest on $100,000 of home equity debt. Retaining the deduction in its current form would prevent the revenue loss associated with allowing reversion to the pre-TCJA rules.

Leaving the mortgage interest deduction unaltered would result in continued differences across states in the deduction's beneficiaries. States with higher homeownership rates, home prices, and average incomes would continue to benefit the most on average. This could be of concern to those who view tax expenditures as government spending administered via the tax code because the spending would continue to be distributed unevenly (in per capita terms). If Congress decides to assist homeowners via the tax code, several alternatives to the mortgage interest deduction may accomplish that objective in a more geographically equitable, and possibly more efficient, manner.¹⁶

Eliminate the Deduction

Alternatively, Congress could eliminate the mortgage interest deduction. This option can be evaluated along several dimensions, starting first with its effect on the tax treatment of taxpayers. The variations in claim rates and benefit values documented in this report suggest that eliminating the deduction could help promote more uniform tax treatment across taxpayers. Eliminating the mortgage interest deduction would result in two homeowners who are equally situated in terms of financial resources but located in different states being treated more equally for tax purposes. For example, two homeowners with similar incomes and mortgages may benefit differently from the current mortgage interest deduction if one lives in a state with an income tax (and claims the SALT deduction) and another does not. Eliminating the deduction would also result in equally positioned homeowners and renters being treated similarly by the tax code.

Eliminating the deduction can also be evaluated by its effect on economic performance or its contribution to improving economic efficiency. Eliminating the deduction could improve the economy's overall performance if the deduction is currently leading labor and capital to be allocated to less productive uses in the owner-occupied housing sector. A number of studies have found that owner-occupied housing is generally taxed favorably compared to other sectors of the economy.¹⁷ Eliminating the deduction would be a step toward creating more uniformity in the tax treatment of various sectors, which could assist in a more efficient allocation of resources across the economy. The increase in federal revenue from eliminating the deduction could also improve the long-term federal budgetary situation, implying less reliance on deficits to finance spending.

Additionally, eliminating the deduction can be analyzed by examining the potential effect on the homeownership rate. Economists have identified the high transaction costs associated with a home purchase — mostly resulting from the down payment requirement — as the primary barrier to homeownership.¹⁸ Because the deduction does not directly address the largest barrier to homeownership and is not well targeted to the group of potential homebuyers most in need of assistance — lower-income households, which includes younger first-time buyers who do not itemize — the effect of eliminating the deduction is likely to be relatively small in the long run.¹⁹

While eliminating the deduction may lead to improved economic efficiency with potentially little effect on the homeownership rate in the long run, careful consideration would still be required to minimize the likelihood of short-run negative consequences. For example, suddenly eliminating the deduction could cause a drop in home ownership demand, leading to a decrease in home prices. The decrease in home prices would impose capital losses on current owners and perhaps produce a lock-in effect — current homeowners could be reluctant to sell at a loss. In addition, a decrease in home prices could lead to a reduction in new home construction, a reduction in homeowner wealth, and the possibility of increased defaults because some homeowners could find themselves underwater on their mortgages (i.e., owing more on their mortgages than their homes are currently worth). These three events could have a negative impact on the broader economy in the short run.

Gradually phasing out the deduction over time could help mitigate the negative consequences for the economy and housing market. Researchers Steven Bourassa and William Grigsby propose eliminating the deduction over a 15-to-20-year period with a fixed date after which the deduction would no longer be available.²⁰ For example, if January 1, 2041, were chosen as the cutoff date, taxpayers who bought a home in 2021 could claim the deduction for 20 years, buyers in 2022 could claim the deduction for 19 years, and so on. Bourassa and Grigsby postulate that there would be no effect on home demand or prices, although they present no modeling to support their proposal. It is possible that gradually eliminating the deduction could simply delay the negative short-term consequences for the economy and housing market. This could happen if households do not anticipate the full effects of the deduction's elimination until closer to the chosen cutoff date.

Limit the Deduction

As a middle option between retaining the deduction unaltered or eliminating it entirely, Congress could choose to retain it but limit its scope. Currently, for mortgage debt incurred on or before December 15, 2017, the combined mortgage limit is $1 million ($500,000 for married filing separately). For mortgage debt incurred after December 15, 2017, the deduction is limited to the interest incurred on the first $750,000 ($375,000 for married filing separately) of combined mortgage debt.

To better target the deduction to those who may need homeownership assistance, Congress could limit it to interest paid on a mortgage amount that more closely resembles that of a first-time homebuyer. In 2009, the Congressional Budget Office (CBO) estimated the revenue effect of gradually reducing the maximum mortgage amount on which interest can be deducted from its then $1.1 million cap to $500,000.²¹ The CBO option would not have taken effect for four years (i.e., 2013 at the time the report was published) and would have decreased the maximum mortgage amount by $100,000 annually until it reached $500,000. CBO estimated this option would raise a total of $41.4 billion between 2013 and 2019.

Another option would be to leave the maximum mortgage amount unchanged, but limit the amount of interest that could be deducted. For example, the amount of interest that a taxpayer may deduct could be limited based on their adjusted gross income (AGI) such that low- and moderate-income homeowners could deduct a greater share of their mortgage interest.

Limiting the deduction in this way would likely help reduce interstate variation. As discussed, a portion of the variation is attributable to differences across states in income levels. States with higher average incomes should, all else equal, expect to benefit more from the deduction because higher-income households are more likely to itemize, higher incomes can support larger mortgages, and higher incomes imply a higher deduction value (i.e., reduced taxes) per dollar deducted. Limiting the amount of interest that could be deducted would be expected to decrease the variation to some degree, although deductions in general will typically display some variation simply because they increase in value as incomes increase.

Replace the Deduction with a Credit

Congress could also choose to replace the mortgage interest deduction with a tax credit. The current deduction tends to provide a proportionally larger benefit to higher-income homeowners because they buy more expensive homes and are subject to higher marginal tax rates. The requirement that homeowners itemize their tax returns also limits the number of owners who receive the tax benefit. A tax credit for mortgage interest could provide a benefit to more homeowners because itemization would no longer be required. A credit, unlike the current deduction, would have the same dollar-for-dollar value to a homeowner regardless of income, creating a more consistent rate of subsidization across homeowners. Making the tax credit refundable would help it reach lower-income homeowners.

Over the years, several mortgage interest tax credit options have been proposed. Five of the more prominent ones are listed below. All five would limit the deduction to a taxpayer's principal residence. Four out of the five would allow a 15% credit rate. Three of the five credit options would be nonrefundable. Two of the options would limit the size of the mortgage eligible for the credit to $500,000, while one would limit eligible mortgages to no greater than $300,000 (with an inflation adjustment). Another option would limit the maximum eligible mortgage to 125% of the area median home price. Finally, one would place no cap on the maximum eligible mortgage, but would limit the maximum tax credit to $25,000.

• CBO, in its 2016 Options for Reducing the Deficit report, presented the option of converting the mortgage interest deduction to a 15% nonrefundable tax credit.²² The credit would be restricted to a taxpayer's primary residence. No credit would be allowed for interest associated with home equity loans. Under this option, the deduction would still be available for five years as the credit is phased in. Simultaneously, the maximum eligible mortgage amount would be reduced to $500,000 during the phase in. After five years, the credit could only be claimed on mortgage amounts up to $500,000. A similar option was presented by CBO in 2009 and 2013.²³

• The American Enterprise Institute's Alan Viard proposed converting the deduction into a 15% refundable tax credit.²⁴ The credit would be limited to the interest on the first $300,000 of mortgage debt (in 2013 dollars) associated with one's primary residence (second homes and home equity debt would be excluded). The qualifying mortgage amount would be adjusted annually for inflation. Homeowners could still claim the deduction but only at 90% of its current value, decreasing by 10% annually. A homeowner could switch to the tax credit regime at any time.

• President Obama's National Commission on Fiscal Responsibility and Reform recommended replacing the mortgage interest deduction with a nonrefundable credit equal to 12% of the interest paid on mortgages of $500,000 or less.²⁵ The credit would be restricted to a taxpayer's primary residence. No credit would be allowed for interest associated with home equity loans.

• The Bipartisan Policy Center's Debt Reduction Taskforce, cochaired by former Senator Pete Domenici and former CBO Director Alice Rivlin, proposed a 15% credit for up to $25,000 of interest paid on a mortgage associated with a principal residence — interest paid on home equity loans and second homes would be ineligible.²⁶ The tax credit would be refundable, which would help lower-income homeowners benefit from it. The proposed credit would be administered via mortgage lenders, who would apply for the credit and transfer it to homeowners by lowering their interest payments in amounts equal to the credit.

• In 2005, President George W. Bush's Advisory Panel on Federal Tax Reform (Tax Reform Panel) also proposed replacing the mortgage interest deduction with a credit.²⁷ Specifically, the Tax Reform Panel proposed a tax credit equal to 15% of mortgage interest paid. Under the proposal, the credit would be restricted to a taxpayer's primary residence. The size of the mortgage for which the interest credit could be claimed would be limited to 125% of the median home price in the taxpayer's region. It appears from the panel's report that the credit would be nonrefundable.

Appendix. Estimation Methodology

The estimates for the geographic distribution of the mortgage interest deduction tax expenditure were produced using an approach developed by economist Martin A. Sullivan.²⁸ Sullivan's method accounts for differences in incomes across states — and therefore, differences in tax rates — and differences in the amount of interest deducted in each state. The two data sets needed to carry out this methodology are:

• Historic Tables 2 in IRS's Statistics of Income, Individual Income Tax State Data, 2016 and 2018 (https://www.irs.gov/statistics/soi-tax-stats-historic-table-2); and

• Tables 3 in U.S. Congress, Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2016-2020, 115^th Cong., 1^st sess., January 30, 2017, JCX-3-17 and in U.S. Congress, Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2018-2022, 115^th Cong., 2^nd sess., October 4, 2018, JCX-81-18 (https://www.jct.gov/).

The first step used in the analysis was to compute national “average marginal” tax rates for various income groups. The tax rates were calculated by first consolidating the income classes used by the JCT in its distributional estimates so that they matched the smaller number of income classes in IRS's Statistics of Income (SOI) data. Next, the JCT expenditure estimate for each income class was divided by the amount of mortgage interest deducted in each income class as reported in the SOI data. This produced an estimate of the national “average marginal” tax rate for each income class.

For each state, these tax rates were then multiplied by the amount of mortgage interest deducted in each respective income class and then summed. This produced an estimate of each state's share of the JCT's mortgage interest deduction tax expenditure estimate.

Author Information

Mark P. Keightley
Specialist in Economics