Federal Reserve Board Economists Present “On the Effect of Student Debt on Access to Homeownership” at a REALTOR® University Speaker Series

The rising level of student debt and the relatively high default rates for student loans have raised concerns about the impact on future homeownership among student borrowers. In a presentation at a REALTOR® University Speaker Series held recently, Federal Reserve Board economists Dr. Daniel Ringo and Dr. Alvaro Mezza presented the results of a paper estimating the impact of an increase in student debt on homeownership.[1]

(To listen to the webinar, please click here.  To access the paper by Alvaro A. Mezza, Daniel R. Ringo, Shane M. Sherlund, and Kamila Sommer “On the Effect of Student Loans on Access to Homeownership”, please click here.)

The authors use a uniquely constructed administrative data set for a nationally representative cohort of 34,891 individuals aged 23 to 31 in 2004 and followed over time, from 1997 to 2010. This administrative dataset includes anonymized information pertaining to credit bureau records, student-level information (e.g., loans and grants availed of, years in school, and degree earned), school-level information (e.g. tuition fee, whether public, private, for-profit or non-for-profit), and economic variables of the state that student was enrolled in[2].

Among the paper’s interesting results are:

  • A 10 percent increase in student loan debt decreases the homeownership rate by one to two percentage points 24 months out of school.
  • In terms of numbers, a 10 percent increase in tuition fee (which is associated with student debt) reduces the number of potential homeowners by 280 individuals per 10,000 college goers two years after exiting school, which is equivalent to 170 individuals per 10,000 individuals (both college and non-college goers). [3]
  • A 10 percent increase in student loan debt causes a 0.6 percentage point increase in the probability that the borrower falls into the subprime category (credit score of 620 or less) and a 0.8 percentage point increase in the probability that a borrower falls into deeply subprime (500 or less).[4]
  • A 10 percent increase in debt is associated with a 0.7 percentage points increase in delinquency rates.
  • The authors did not find conclusive evidence that an increase in student loan leads to a lower mortgage balance.
  • Still, all is not gloom and doom– the authors point out that the homeownership rate is increasing almost linearly over time during the five-year window. In other words, an increase of 10 percent in student debt only delays the home purchase rate of a given cohort, by about three months, based on the authors’ estimates.
borrowersThe authors are careful to point the following caveats:

  • The paper captures the impact of an increase in student debt, not the impact of student’s access to education loans on homeownership. The impact of the access to student loans on homeownership can be positive for as long as the income returns over time exceed the level of debt.
  • Tighter credit underwriting standards after 2005 suggest that the drag of student debt on homeownership may be greater, with lenders more sensitive to debt-to-income and loan-to-value ratios. However, the authors note that the introduction of income-based student loan repayment plans may moderate the link between student loan debt and homeownership.

About the Authors

Alvaro A. Mezza is a senior economist of the Consumer Finance Section. Daniel R. Ringo is an economist of the Real Estate Finance Section. Shane M. Sherlund is assistant director of Program Direction Section. Kamila Sommer is a senior economist of the Real Estate Finance Section. Information on the research areas, expertise, and education of the authors can be accessed from the Federal Reserve Board.

About REALTOR® University

REALTOR® University provides on-line education on real estate and other topics at the MBA and undergraduate levels. The REALTOR® University Speaker Series provides a venue to learn about and stimulate discussion of economic and real estate issues in support of NAR’s mission as the Voice of Real Estate. The Speaker Series presentations can be accessed on this webpage.

[1] The REALTOR® University Speaker Series was held on March 25, 2016. The authors make the disclaimer that the views expressed in the paper and in the presentation do not reflect the views of the Federal Reserve Board.

[2] Anonymized administrative data are from TransUnion LLC, National Student Clearinghouse, National Student Loan Data System, Integrated Postsecondary Educations Data System, and the College Board.

[3] The authors use the in-state tuition rate at public 4-year colleges in the student’s home state as an instrumental variable for student debt, and the authors conduct a number of validation exercises to ensure that the tuition rate is a valid instrumental variable. In econometric theory, omitted variables, whose effect is picked up by the error term, lead to biased estimates of the effect of the observed explanatory variables on the dependent variable being explained. The authors use Two-Stage Least Squares and maximum likelihood methods to estimate the coefficients of the likelihood of owning a home.

[4] The credit scores are based on TransUnion LLC credit score system.