The United States spends more than four times as much on homeowner subsidies as it does on affordable housing for those most in need.
That stat comes courtesy a detailed NBER working paper by Robert Collinson and Ingrid Gould Ellen, my colleagues at NYU, and Jens Ludwig of the University of Chicago. The U.S. shells out roughly $46 billion a year on affordable housing—$40 billion on means-tested programs and another $6 billion in tax expenditures through the Low Income Housing Tax Credit (LIHTC) program, which supports affordable housing investments for low-income Americans. Compare that to $195 billion in subsidies that flow largely to wealthy and middle class homeowners via tax deductions for mortgage interest. And the study itself notes, the actual subsidy to homeowners may run as high as $600 billion based on the non-taxation of imputed rent, as estimated by Wharton real estate economists Todd Sinai and Joseph Gyourko.
The NBER paper is organized around two key issues. The first examines changes in the government programs that provide housing assistance to the low-income Americans who need it most. Since its inception during the New Deal, the study shows, America’s housing policy has undergone a major shift, from public housing to less direct forms of housing assistance. From the 1930s to the mid-1970s, housing assistance was oriented around the direct provision of public housing. The number of public housing units, which date back to the early 1930s, peaked in 1991, and have shrunk by nearly 23 percent—over 300,000 units— since. Federal policy has shifted, as the chart below shows, away from public housing programs and toward rental assistance programs supported by Section 8 and low-income tax credits.
The second and more profound issue has to do with the effectiveness of America’s housing policy to address the needs of lower income people. Here the researchers contrast providing housing assistance with the alternative of providing cash payments or transfers. Their exhaustive review of dozens upon dozens of earlier studies suggests that cash transfers may actually be the better route. Policymakers’ goals, they write “should hinge at least partially on the assumption that inkind programs will lead to more housing consumption than would cash transfers of equal cost.” And yet, “there is remarkably little evidence available to date on this first-order question.”
This, the researchers contend, is surprising for two reasons. On the one hand, these programs consume significant amounts of government resources each year, and so are important and certainly worthy of study. On the other, the excess demand for these program services (the researchers find that fewer than one in four income-eligible families in the U.S. participates in such a program) would seem to offer numerous opportunities for researchers to carry out studies with truly comparable comparison groups—those who qualify for assistance and get it, and those who qualify but don’t.
Part of the problem, the study points out, is that Americans are more comfortable with the idea of housing programs than they are with cash transfers. The researchers cite a 2003 survey that found 39 percent of Americans supported cash payments to the non-disabled poor, compared to 89 percent who supported low-income housing assistance programs.
Still, the need is growing, as housing burdens faced by low income families have increased significantly. As the chart below shows, the share of rent paid (in blue) compared to income (red) has increased substantially in recent years. Between 1980 and 2007, the two moved mostly in tandem. But since the economic crisis, real incomes have dropped off while rents have recovered from their recession-related dip. As a result, Americans now spend more of their incomes on housing. The median renter spent 25 percent of their income on rent in 1980. Today, that figure has risen to 29 percent. It’s even worse for low-income renters. In 1980, renters in the bottom fifth of earners spent 53 percent of their incomes on rent. Now they spend 63 percent. Though part of this rise may be attributable to an improvement of housing stock, that significant rise can't only be explained by better housing.
Lower income households, which receive the smallest and most limited kinds of assistance, face the biggest housing burdens. By far the biggest subsidy continues to go to wealthy and middle class homeowners.
This makes little economic sense, particularly at a time when many economists believe America is over-investing in housing. As Nobel laureate Edmund Phelps has said: “It used to be said that the business of America was business. Now the business of America is homeownership.”