The Effect of Housing Wealth on College Choice

Featured in print Digest

In the lowest income category ... a $10,000 increase in value increased the relative probability of attending a flagship university by 8.3 percent.

In The Effect of Housing Wealth on College Choice: Evidence from the Housing Boom (NBER Working Paper No. 18075), Michael Lovenheim and Lockwood Reynolds find that a $10,000 increase in a family's housing wealth in the four years before they send a child off to college increases the likelihood that that child attends a public flagship school by 2 percent. Public flagship universities are the elite public universities in each state and are characterized by having higher student body SAT scores, higher faculty-student ratios, and more total and instructional spending per student than other public institutions. They usually cost more than other public schools as well. Given "growing evidence of the high labor market and educational attainment returns to college quality," the authors conclude that the housing bust could change attendance decisions in ways that have "long-run effects on the supply of high-skilled labor and on income inequality."

The data for this study come from the 1997 National Longitudinal Survey of Youth. The authors measure college attendance, residence, home prices, and family characteristics for 2,801 students who were under age 18 in 1997, attended college within two years of high school graduation, and whose parents were homeowners living in a Metropolitan Statistical Area (MSA). They combine the self-reported home price in the 1997 NLSY with the MSA-level Conventional Mortgage Housing Price Index (CMHPI) created by Fannie-Mae and Freddie-Mac to measure the average housing price change in each MSA in each year. For the entire United States, the CMHPI increased by 121 percent between 1993 and 2003, the period covered by the sample. However, there were large regional variations: home prices in Syracuse, New York, increased by 19 percent, while those in New York City increased by 90 percent. The average homeowner in this sample experienced a four-year home price increase of $53,310 (in 2007 dollars).

The students in this study attended one of four mutually exclusive higher education sectors: flagship public universities; non-flagship public four year schools; private four-year institutions; and community colleges. Because there was no strong relationship between home prices and the likelihood of being accepted to schools in a particular sector, conditional on applying, the authors conclude that any changes in enrollment were a result of changes in student application behavior.

Family income was divided into three groups: low-income households were defined as those with incomes of $75,000 or less with a sample mean of $50,023; middle-income households had earnings between $75,000 and $125,000 with a sample mean of $97,060; and high-income households reported incomes above $125,000 with a sample mean of $190,340.

The authors find that the effect of home price increases was strongest in the lowest income category, where a $10,000 increase in value increased the relative probability of attending a flagship university by 8.3 percent and decreased the relative probability of attending a community college by 3.8 percent. Lower-income students worked less at outside jobs as housing wealth increased, and the likelihood of their earning a BA increased by 1.8 percent. Changes in housing wealth increased the probability that those in the middle-income category attended a state flagship school by about 3.9 percent. There was no measurable effect of housing prices on attendance among families with incomes above $125,000.

--Linda Gorman