Black swan hunting

If housing bubbles are like extreme weather events, Jim Follain has been chasing tornadoes.
     Follain, a senior fellow at the Nelson A. Rockefeller Institute for Local Government, has been conducting research on the run-up of housing prices and ensuing crashes, with 2008 being the most notable perfect storm. He's been fascinated with the idea that on the whole, "it's almost like we couldn't see this coming."  Like meteorologists and extreme weather events, economists have not been able to predict crashes with great precision -- although they are very good at analyzing what happened after the fact. The Black Swan author Nassim Nicholas Taleb describes events such as Google's success or 9-11 in the same way: unpredictable, carrying a massive impact, but where subsequent explanations suggest less random and more predictable characteristics than was previously evident.
     So what are the warning signs to work with? Private mortgage insurance used to be a canary in the coal mine, Follain says, but that has fallen away. There are macro factors --income levels, employment, monetary policy and interest rates -- combined with the reality that housing markets are intensely local and regional. Subprime lending was prevelant in areas that had the greatest declines, and it's possible, Follain says, to look at the early indications of a "feedback loop" in real estate markets in distress.
     Using advanced econometric techniques applied to a large panel data set of annual data for up to 350 cities allowed Follain to look across cities at a point in time, and to look at individual cities across time. Panel data enables analysts to identify the effectsof city characteristics separately from the effects of temporal changes such as economic cycles. His simulations predicted declines in Florida, Arizona, Nevada, and California, he says,"but what actually happened was much worse. The model was good at predicting the worst-hit (Metropolitan Statistical Areas) ... but not the magnititude of the decline."
     Follain co-authored the Lincoln Institute working paper,  A Look at US House Price Bubbles from 1980-2010 and the Role of Local Market Conditions, with Seth H. Giertz, assistant professor of economics at the University of Nebraska-Lincoln, and plans to include new work in another paper to be posted soon. He presented some of his findings in November as part of the fall lecture series, which can be viewed in its entirety at our Lectures & Video page.

Comments

[David Abraham]

Nice informative blog, thanks for sharing.