Bailout

Do information asymmetries explain the housing bubble?

By Felix Salmon

Adam Levitin and Susan Wachter have a new paper out which reckons it can explain the entire housing bubble by looking at the supply of private-label mortgage-backed securities in the market, and the information asymmetries embedded in them.

They do have a point: since the banks putting together these private-lable securities, or PLS, knew much better than the buyers (and, for that matter, the ratings agencies) what was going into them, there was an opportunity — grasped with both fists — to take advantage of those asymmetries:

PLS are unusually complex, heterogeneous products. Any particular securitization is supported by a unique pool of collateral and has its own set of credit enhancements and payment structure. Complexity and heterogeneity shrouded the risks inherent in PLS. As a result, investors failed to properly price for risk, as they did not perceive the full extent of the risk involved. The structure of PLS (including the underlying mortgages) allowed investors to underestimate the risks involved and therefore underprice the PLS by demanding insufficiently large yield spreads. The housing bubble was fueled by mispriced mortgage finance, and the mispricing occurred because of information failures. Thus, at the core of the housing bubble was an information failure. Investors lacked adequate information about the risks involved with PLS.

I like the way that Levitin and Wachter look at the commercial real estate market as well as the residential market. And I also like the way that they say that the bubble was actually pretty short-lived, starting in 2004 (or possibly 2003) and bursting in 2006. But if that’s true, then the housing bubble started right when the homeownership rate peaked, and the homeownership rate actually declined for the entire duration of the bubble. Which seems a bit weird.

What’s more, while the paper nods to the odd fact that at one point structured mortgage-backed securities started trading through triple-A-rated corporate bonds, it doesn’t really explain why that happened. And it’s far too quick to dismiss other theories about what caused the housing bubble by simply saying that those theories don’t explain the low spreads on mortgage-backed bonds. Which might be right, but surely a bubble can have more than one cause.

And while there are occasional references to housing bubbles elsewhere in the world, there’s no attempt to see whether they can be explained in similar terms. I suspect that doing so would be hard, in which case it’s not that silly to assume that the causes of the US housing bubble overlapped with the causes of the housing bubbles in countries like the UK, Ireland, Spain, Australia, and even South Africa.

In fact, you don’t even need to look abroad. Private-label mortgages were subprime and alt-A, yet we had a perfectly healthy bubble right here in Manhattan, where there was no subprime mortgage bubble at all.

Most importantly, the people with the upper hand, when it came to the information asymmetries, ended up being the people taking the biggest losses when the bubble burst. Information asymmetry might explain why small Norwegian municipalities ended up losing millions investing in structured notes, but they don’t explain the losses at Merrill Lynch and Citigroup, or, for that matter, at Fannie and Freddie.

So while this paper constitutes an important contribution to the literature, I don’t think the authors have managed to explain the housing bubble with a single concept. In fact, I very much doubt that such a thing would even be possible. Modern economics and finance is far too complicated for that.

Full post as published by Macroprudential on September 02, 2010 (boomark / email).