The recent boom-bust episode in home values was accompanied by an equally striking boom-bust pattern in the volume of mortgage-backed security issuances. My recent research focuses on understanding the reasons behind the pre-crisis rise of asset-backed securitization in the United States. It also suggests that the rise of securitization contributed to the loosening of lending standards that characterized the pre-crisis period, which helped fuel the housing boom on the way up and magnified the default crisis on the way down.
In a recent paper (“The rise of securitization: a recursive security design approach,” working paper, 2012), Dean Corbae (Department of Finance, Investment, and Banking at the Wisconsin School of Business) and I formalize and quantify the “Saving Glut” idea closely associated with U.S. Federal Reserve Chairman Ben Bernanke. Chairman Bernanke pointed out that demand for U.S. investment-grade paper has increased dramatically in the 15 years, driven in part by the fast emergence of nations with high saving rates and a strong preference for safe U.S. assets. Since structured finance is inter alia a method to produce investment-grade securities collateralized by risky assets, it should come as little surprise that the scope of securitization widened when the demand for—and the market value of—U.S. safe assets rose.
Dean Corbae and I describe a macroeconomic model where an increase in the willingness to pay for safe assets causes the volume of securitization to rise. Our primary goal is to quantify the importance of this phenomenon for the behavior of key macroeconomic variables. For instance, what are the consequences of increased foreign demand for U.S. safe assets on the volatility of U.S. GDP? At the same time, however, our model suggests that the rise of securitization may have played a key role in the recent crisis.
In a model Corbae and I have developed, financial intermediaries maximize the quantity of safe assets they can extract from the risky assets they have pooled. Therefore, a supply response to an external demand shock must come from the extensive margin: new assets have to be brought into the securitization fold. At the same time, intermediaries always select the safest assets first for securitization purposes. Our model, therefore, predicts that as asset securitization widens, ever riskier projects should enter the securitization fold. This, we argue, could yield a possible answer to one of the most vexing open questions when it comes to the recent crisis: why did mortgages with risky features such as low down payments become so popular a few years ago? Our model points to a simple explanation: the increasing market value of investment-grade assets raised the profitability of funding these risky mortgages.
Other recent work of ours (“Mortgage Innovation and the Foreclosure Boom,” working paper, 2010) argues that, in turn, the availability of mortgages with risky features boosted participation in housing markets during the boom and magnified the foreclosure crisis when home values started falling. Specifically, we lay out a model of housing and mortgage choice that captures key features of U.S. housing markets. In the context of that model, we calculate that popularity of high-leverage loans between 2003 and 2007 can account for half of the subsequent spike in foreclosure rates.
All told, this line of work suggests that the rise of securitization that characterized the pre-crisis period was driven at least in part by independent factors, played a key role in the housing boom, and contributed to the subsequent default crisis.
Join me for my Rays of Research presentation to hear more about my research on this topic. I’ll be presenting from 3:00-4:00 p.m. today (October 31, 2012) in Grainger Hall, room 2339.