In the mortgage interest deduction paper I wrote with Dean Stansel this summer, we focused primarily on what the benefits of the MID were and who got them. We stopped the scope of our study at the where those benefits were concentrated, leaving that for other scholars to assess. And a new paper by American Action Forum's Ike Brannon and Georgia State's Andrew Hanson and Zachary Hawley does just that.
At the congressional district level, the percent of taxpayers claiming the deduction ranges from a low of about 3 percent (in New York’s 16th District) to a high of over 48 percent (in Colorado’s 6th District). The dollars claimed per taxpayer taking the deduction range from over $27,000 (in California’s 30th District) to less than $6,000 (in New York’s 28th District). Our analysis at the ZIP code level reinforces the unequal benefits bestowed by the mortgage interest deduction, using locally weighted polynomial regression we show that taxpayers claiming the mortgage interest deduction are concentrated in the suburbs of major metropolitan areas.
The question of "where" the mortgage interest deduction benefits are concentrated has been looked at before. In our Reason Foundation paper, Dean and I noted the following two studies and their results:
Looking at 1999 data for all housing tax subsidies provided by the federal government (not just the MID), Gyourko and Sinai found that the biggest benefits went to homeowners (in descending order) in Washington, D.C., Hawaii, California, New York, Massachusetts, Connecticut and New Jersey, with benefits per owner occupied unit exceeding $8,000 in each of those high-income, high-tax states. Only 16 states were above the national average of $6,024.
Looking at metropolitan areas, a 2001 study found that the benefits were similarly skewed toward large areas with high incomes, taxes, and housing prices, primarily in California and the Northeast corridor. In fact, just three large metro areas—New York-Northern New Jersey, Los Angeles Riverside-Orange County, and San Francisco-Oakland-San Jose—received over 75 percent of the net positive benefits of the MID.
The Brannon et al paper looks at data that is much more recent, IRS ZIP code level data from individual tax returns for 2007. Here are a few snap shots from their findings:
California deducts the most mortgage interest, over $96 billion or 20 percent of all interest deducted, despite paying only 13.5 percent of all federal income taxes. Florida ($31 billion), New York ($26 billion), Illinois ($20 billion), and Texas ($20 billion) round out the top five. Despite claiming a large dollar amount of mortgage interest, New York and Texas each claim a much lower share of the MID than their share of income taxes paid. For New York, this is likely due to the low rate of home ownership– 55.5% in the first quarter of 2007, 2nd lowest in the country.12 For Texas this is likely due to the low cost of housing– average home values were $82,500 in 2000, 8th lowest in the country. [...]
The top 20 percent of [congressional] districts deduct 40 percent of all mortgage interest and the top 50 percent of districts deduct 75 percent of all mortgage interest. [...]
The distribution of the average amount of MID claimed, shown in Table 6, is skewed toward California districts, with the top 20 average MID claims all coming from California. California districts also make up 34 of the top 37, and 41 of the top 50 average MID claims per tax filer claiming the deduction. [...]
The highest average MID claims are concentrated in California, primarily near San Francisco and Los Angeles... [with] relatively higher average claims in districts near Miami, Washington, D.C. and Seattle. Across the country, the average claims are substantially higher in western congressional districts, those along the Washington, D.C.–Boston corridor, and in Florida.
This paper's findings help provide a sharper edge to the critique that the mortgage interest deduction is a tax benefit highly concentrated among a few number of Americans benefit a wide number of members of the housing industry, but not really benefiting home ownership for the majority of Americans. But that does leave us with the daunting political question Brannon et al pose at the end of their introduction:
Our research suggests that in many communities across the country the MID benefits a relatively small proportion of the population at a cost to all taxpayers. The relevant question from the standpoint of political economy is whether the smaller group that benefits most from the MID would provide a larger political opposition than the larger, more dispersed group that would benefit from policy change.
See the whole paper here.