You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

Securitization Is to Blame, Just Not the Way You Think

Felix Salmon calls me out for arguing that securitization might not be to blame for the decline in lending standards.

If lending standards dropped at the same time as the securitization rate soared, I’d say there’s a strong correlation between the two, and a pretty good prima facie case for a causal relationship too.

I actually agree with Felix here. As I'll explain, it's hard to argue that lending standards didn't decline because of the introduction of securitization. But perhaps where we differ, and what I should've been more clear about in the first post, is in on my view on the mechanism behind the decline.

On that point, I'm with Hyun Song Shin, who wrote in March that securitization allowed new sources of savings from money market funds, pension funds, foreign central banks, etc. to flow into the mortgage market and expand the level of available credit. The fact that much of this credit was extended to subprime borrowers has to mean that lending standards declined since the average creditworthiness of the entire pool of mortgage holders fell. But it's important to note that this explanation has nothing to do with unscrupulous lenders trying to hoodwink borrowers and banks, as the typical news story or politician is likely to say. Of course, at the time many believed that securitization prevented this average creditworthiness level from dropping, but as we now know that didn't turn out to be the case.

Further evidence on the points above comes from the new paper I wrote up in the last post (which was just released by the Boston Fed). The authors find that securitizers were apparently aware of potential moral hazard issues with securitization and responded by buying and packaging loans that had lower default probabilities.

Also little-known is that the percentage of loan denials actually increased 75% between 2002 and 2007. Perhaps 750% would've been better, but this is not a sign that standards declined across the board. The importance of these details is that they should inform how financial regulation gets done as they strongly suggest that stronger "skin in the game" provisions would have done little to prevent the crisis.

Finally, I'd like to clear up some potentially confusing wording in this post: I gave the impression that this paper by Amit Seru et al argues that a decline in lending standards -- made possible by the growth of securitization -- was the primary cause of the crisis. But that's not the case. The researchers found evidence that securitization reduced incentives to screen, but they say nothing about the role of securitization in causing the crisis. (Seru, who has done some of the best research I've read into the forces surrounding the crisis, told me he plans a response to the Boston Fed study, which critiques the paper.)