Housing Choice Under Borrowing Constraints

Abstract

A dramatic decline in the maximum loan amount eligible for FHA mortgage insurance in the Salt Lake City metropolitan statistical area between 2013 and 2014 provides a natural experiment in the impact of borrowing constraints on housing decisions. Using a differences-in-differences design within a seemingly unrelated regression model, we estimate the impact of this constraint on the size and location of homes purchased by FHA borrowers. We find that borrowers likely constrained by the new loan limits purchased smaller homes with larger downpayments than similar borrowers prior to the loan limit decline. Likely constrained borrowers do not appear to compromise on location, including the quality of local schools. The net effect of the housing choices compelled by the reduction in loan limits does not appear to change the credit risk of borrowers.