In this report, we make the case that Congress should permit the Federal Housing Administration (FHA) to insure zero–down payment mortgages for first-time homebuyers. The down payment is the largest single barrier to homeownership. Many first-time homebuyers rely on family assistance for a down payment, and allowing zero–down payment mortgages would level the playing field for potential homeowners without family wealth. If Congress were to allow the FHA to insure zero–down payment loans, it could result in millions of renter households transitioning to homeownership earlier, giving homeowners more time to build housing wealth. Moreover, FHA zero–down payment loans would also replace the inconsistent patchwork of down payment assistance programs.
What We Found
Buying a home requires a sufficient amount of income, a high enough credit score, and enough savings for a down payment. We quantified that millions of borrowers have the income and credit score to meet underwriting criteria but not the down payment.
Although the down payment is the biggest obstacle to homeownership, we found it is not the largest determinant of the probability of default—credit scores are. Zero–down payment mortgages would raise the probability of default by a marginal amount, but this risk could be more than offset by tightening the criteria for eligible borrowers. We recommend initially limiting the eligibility for borrowers with credit scores of 700 and above (or 660 with either a good rental payment history or sufficient reserves).
We show that moving renters into homeownership a few years earlier would help them accumulate home equity faster.
How We Did It
We used Federal Housing Finance Agency (FHFA) public data from the National Survey of Mortgage Originations linked with the National Mortgage Database. The data have a nationally representative sample of mortgage loans, with borrower survey responses taken soon after origination and including the eventual loan performance. We constructed a single model to forecast mortgage performance based on origination characteristics. Our data consisted of all purchase and refinance transactions (minus jumbo loans) from 2013 to 2021, totaling about 47,000 loans. Performance was measured by whether the loan went 90 days delinquent in the first three years. We limit our data to 2013 through 2021, as performance data are available only until the end of 2024. We weigh all the results according to the nationally representative weights provided in the FHFA’s data.
The model coefficients show that moving from loans with 3.5 percent down payments (current FHA loans) to 0 percent down payments would only incrementally increase default risk. Moreover, such an incremental risk increase can be compensated for by an increase in the credit score threshold. We show that loss given default might suffer, but that can be compensated for by a moderate increase in the mortgage insurance premium for zero–down payment borrowers.
