The home mortgage interest deduction is the largest federal housing subsidy in the United States. It is also, as the sociologist Matthew Desmond wrote in the New York Times Magazine, “what may very well be the most regressive piece of social policy in America.” Because this policy is an income tax deduction, it is most valuable to people in the highest income brackets—and in particular, to those who borrowed the most money in order to buy the very most expensive homes. That is an unusual way for a government to encourage the provision of housing.

How did we end up with this peculiar social policy? Who benefits from it? And why, given its apparent disadvantages, does it persist?
The Mortgage Tax Reform Working Group met on July 15, 2017 at the Urban Democracy Lab to discuss several new research projects in progress on these questions. Three research presentations guided the discussion. The authors, titles, and abstracts of these presentations were as follows:
Joshua McCabe, “The Road Not Taken: The Politics of Mortgage Tax Relief in the U.S. and U.K.”
Abstract:
Both the U.S. and the U.K. introduced tax deductions for (mortgage) interest paid as part of their original income tax legislation. Whereas the home mortgage interest deduction (HMID) has come to be seen as an untouchable “third rail” in American politics, the British government quietly eliminated mortgage interest relief (MIR) in 2000. This paper traces the divergence in outcomes to the 1970s with an emphasis on the interaction between institutions and policy sequence. The 1974 decision of British policymakers to place a nominal cap on MIR led to substantial erosion in the inflationary decade that followed. This weakened political support, allowing successive governments to actively reduce and eliminate it between 1991 and 2000. The structure of American political institutions prevented policymakers from successfully placing a nominal cap on HMID until 1987, at which point inflation was back under control. As a result, the political cost of directly attacking HMID remains strong until this day.
Monica Prasad, “The Problem of the Wellesley Democrat.”
Abstract:
Contrary to popular perception, it is not impossible to reform the Home Mortgage Interest Deduction (HMID).  Congress has done so on at least two occasions.  The first section of this paper briefly discusses these two episodes in the context of several failed attempts to reform the HMID over the last half-century. One conclusion from this overview is that Democrats have in fact been able to reform the HMID when they have chosen to try. The second section then considers why Democrats may not be particularly interested in HMID reform today, asking whether this is because—as some commentators have speculated–demographic and partisan changes have led to a situation in which HMID reform would affect Democratic constituencies more than Republican ones.  Wellesley, Massachusetts, is an example of the kind of constituency that traditionally supports Democrats, but would be hardest hit by HMID reform. We use polling data and state and county-level analysis of election results to answer this question, but reach conflicting results: there is indeed a strong negative correlation between housing prices and Republican voting, at both state and county level.  But polls do not show income having an effect on respondents’ attitudes to HMID.
Isaac William Martin, “How Inegalitarian is the Home Mortgage Interest Deduction?”
Abstract:
Sociologists have described the HMID as a regressive subsidy for the rich, and have argued that it exacerbates economic inequality, especially inequality between black and white Americans. But just how much does the HMID contribute to inequality in America? This project presents new evidence about the incidence of the HMID from an analysis of the Panel Study of Income Dynamics that takes economic sociology seriously, by modeling alternative tax-and-transfer distributions that might be possible on the assumptions that housing markets are embedded in regulatory institutions, potentially characterized by path-dependent development, and segmented by status. I argue for a sociological approach to incidence analysis that involves comparisons among multiple counterfactual scenarios, judged on the grounds of their sociological tenability. By setting logical bounds on the parameter values considered in these scenarios, it is possible to show that the aggregate distributional effects of the HMID approach the maximally inegalitarian extreme, in the sense that almost any other way of distributing the equivalent tax revenues would reduce the inequality of disposable income. It is also possible to estimate bounds on the inequality-reducing effects of eliminating the HMID. In the simulations presented here, the effect of eliminating the HMID is shown to be, at most, a 4% reduction in selected measures of aggregate income inequality. Eliminating the HMID in favor of some more egalitarian tax and transfer policy of equivalent budgetary magnitude is not the largest egalitarian policy intervention that might be contemplated, but it is also not trivial.
In addition to discussing these three research projects, the Mortgage Tax Reform Working Group discussed the possible implications of these preliminary findings, debated other research priorities, and engaged in preliminary planning for a panel discussion of the comparative historical sociology of the HMID that will take place at the annual meetings of the Social Science History Association in Montreal, November 2-5, 2017.