Housing Returns and Speculation
Saturday, Jan. 7, 2017 8:00 AM – 10:00 AM
Sheraton Grand Chicago, Sheraton Ballroom II
- Chair: Chester Spatt, Carnegie Mellon University
Economic Consequences of Housing Speculation
AbstractBy exploiting variation in capital gains taxation across U.S. states as an instrument, we provide novel evidence that housing speculation during the boom period of 2004 to 2006, measured by the fraction of non-owner-occupied home purchases, helps to explain not only the severity of the housing price bust in 2007 to 2009, but also the depth of the subsequent economic recession. Housing speculation, anchored, in part, on extrapolation of past housing price changes, was more pronounced in zip codes with low capital gains taxation. Zip codes that had greater speculation, in turn, experienced more new housing construction during the boom, and more severe declines in employment, per capita income, real payroll, and new business establishment growth during the bust. Our analysis also indicates that supply overhang and local household demand are two key channels for transmitting the adverse effects of housing speculation.
Mortgage Supply and the US Housing Boom: The Role of the Community Reinvestment Act
AbstractThis paper studies the role of the Community Reinvestment Act (CRA) in the recent US housing boom-bust cycle. Using a difference-in-differences matching estimation, I find that the enhancement of CRA enforcement in 1998 caused a 7.7 percentage points increase in annual growth rate of mortgage lending by CRA-regulated banks to CRA-eligible census tracts relative to a group of similar-income CRA-ineligible census tracts within the same state. Financial institutions which are not subject to the CRA, however, do not show any change in their mortgage supply between these two types of census tracts after 1998. I take advantage of this exogenous shift in mortgage supply within an instrumental variable framework to identify the causal effect of mortgage supply on housing prices. I find that every 1 percentage point higher annual growth rate of mortgage supply leads to 0.3 percentage points higher annual growth rate of housing prices. Reduced form regressions show that CRA-eligible neighborhoods experienced higher house price growth during the boom and sharper decline during the bust period. I use placebo tests to confirm that this effect is in fact channeled through the shift in mortgage supply by CRA-regulated banks and not by unobserved demand factors. Furthermore, my results indicate that CRA-induced mortgages went to borrowers with lower FICO scores, carried higher interest rates, and encountered more frequent delinquencies.
Non-recourse Mortgage Law and Housing Speculation
AbstractIn a state with non-recourse mortgage law, borrowers have limited liability on their mortgage loan. Examining price discontinuities at state borders, we show that non-recourse law causes larger swings in housing prices by encouraging speculative investments when housing markets are in a boom cycle. We find that mortgage lending pricing does not fully reflect the higher risk in non-recourse state because the emergence of the originate-to-distribute (OTD) model in the housing markets enables lenders to effectively shift the risks to other investors. The amplified housing cycle has a real effect leading to higher households' consumption during a boom period.
University of California-Berkeley
University of California-Los Angeles
University of Colorado-Boulder
- G1 - Asset Markets and Pricing