Losing mortgage interest deduction would have modest effects on economy

Eliminating or curtailing the mortgage interest deduction (MID) would initially result in declines in housing prices and investment but would have only modest aggregate macroeconomic effects in the long run, according to a new paper from Rice University’s Baker Institute for Public Policy.

The MID is the second-largest individual income tax expenditure, according to the congressional Joint Committee on Taxation. Given the severity of the fiscal problems currently faced by the U.S., many recent tax reform proposals have included measures that would curtail or eliminate the home MID.

“The MID is, of course, an extremely popular and thus highly politically sensitive provision,” said John Diamond, the Edward A. and Hermena Hancock Kelly Fellow in Public Finance at the Baker Institute. “It was one of the few provisions that was deemed to be untouchable during the deliberations preceding enactment of the landmark Tax Reform Act of 1986, a highly successful effort at fundamental tax reform that is widely believed to be the most sweeping reform of the income tax since its enactment.” Diamond co-authored the paper with George Zodrow, the university’s Allyn R. and Gladys M. Cline Chair of Economics and a Baker Institute Rice Scholar.

The study uses a computer model developed by Diamond and Zodrow to simulate both the short-run and long-run dynamic aggregate macroeconomic effects of several tax reform plans that propose eliminating or curtailing the MID. The model is a dynamic, overlapping-generations, computable general equilibrium model of the U.S. economy.

The most dramatic reform the authors analyze is complete elimination of the MID. In this case, they find that GDP decreases slightly in the short run, due to the adjustment costs incurred in reallocating the capital stock, and increases slightly by 0.1 percent in the long run. Overall investment increases by less than 1 percent and reflects the expected reform-induced increases in investment in the nonhousing sectors and the rental housing sector, coupled with a decrease in investment in the owner-occupied housing sector, of about 6 percent initially and 3 percent in the long run. Asset values increase in the nonhousing sectors by less than 2 percent and by 3.5 percent in the rental housing sector and are coupled with a decline in the value of owner-occupied housing of roughly 4 percent.

The effects of the other two reforms analyzed — replacing the MID with a 12 percent nonrefundable credit subject to a $25,000 interest cap and limiting the MID to primary residences — are qualitatively similar but significantly smaller. For example, for the capped credit, housing investment in the owner-occupied sector declines by 2.6 percent initially and by 1.5 percent in the long run, and the value of owner-occupied housing declines by roughly 2 percent. By comparison, the effects of the far more modest reform of limiting the MID to principal residences are quite small, with investment in owner-occupied housing falling by 0.7 percent initially and by 0.4 percent in the long run, and the value of owner-occupied housing falling by only 0.5 percent.