Room B: FMG 1
Mortgage Prepayment, Race, and Monetary Policy
1Federal Reserve Bank of Atlanta, United States of America; 2Federal Reserve Bank of Boston, United States of America; 3Harvard Business School, United States of America
Black and Hispanic homeowners pay significantly higher mortgage interest rates than Non-Hispanic White homeowners. We decompose the racial rate gap into (1) higher rates charged to Black borrowers at origination and (2) racial disparities in the timing of originations. We show that the latter explains most of the gap and is driven by racial differences in mortgage prepayment behavior. Non-Hispanic White borrowers are more mobile and are more likely to exploit falling interest rates by refinancing their mortgages. Observable borrower and mortgage characteristics like income, credit scores, and loan-to-value ratios explain most of the differences in prepayment behavior by race. By driving down mortgage rates, monetary expansions such as the first quantitative easing program (QE1) exacerbates the racial rate gap by disproportionately benefiting Non-Hispanic White borrowers. Alternative mortgage contract designs to the standard fixed-rate mortgage with rate originated above mortgage-backed securities yields may have desirable distributional implications across racial groups.
Inferring Expectations from Observables: Evidence from the Housing Market
1Fisher College of Business, The Ohio State University, United States of America; 2National Bureau of Economic Research, United States of America; 3Swiss National Bank, Switzerland; 4International Monetary Fund, United States of America
We propose a new method to identify shifts in price expectations in the housing market through the accumulation of excess capacity. Expectations of future price increases (due to anticipated future demand) cause current supply to increase, creating a temporary vacancy. We implement this intuition in a structural vector autoregression with sign restrictions and explore the effects of price expectations in the U.S. housing market. We find that price expectation shocks were a prime factor explaining the 1996-2006 boom, particularly in the Sand States. Expectation shocks at the peak of boom reflected implausible growth expectations and reversed during the bust.