Machine Learning for Financial Crises

Below are the slides from my discussion of Helene Rey et al., "Answering the Queen: Machine Learning and Financial Crises", which I gave a few days ago at a fine NBER IFM meeting (program and clickable papers here). I also discussed it in June at the BIS annual research meeting in Zurich. The key development since the earlier mid-summer draft is that they actually implemented a real-time financial crisis prediction analysis for France using vintage data, as opposed to quasi-real-time using final-revised data. Moving to real time of course somewhat degrades the quasi-real-time results, but they largely hold up. Very impressive. Therefore I now offer suggestions for improving evaluation credibility in the remaining cases where vintage datasets are not yet available. On the other hand, I also note how subtle but important look-ahead biases can creep in even when vintage data are available and used. I conclude that the only fully-convincing evaluation involves implementing their approach moving forward, recording the results, and building up a true track record.

Gary Gorton, Harald Uhlig, and the Great Crisis

Gary Gorton has made clear that the financial crisis of 2007 was in essence a traditional banking panic, not unlike those of the ninetheeth century.  A key corollary is that the root cause of the Panic of 2007 can't be something relatively new, like "Too Big to Fail".  (See this.)  Lots of people blame residential mortgage-backed securities (RMBS's), but they're also too new.  Interestingly, in new work Juan Ospina and Harald Uhlig examine RBMS's directly.  Sure enough, and contrary to popular impression, they performed quite well through the crisis.

On the Great Financial Panic of 2007

(a) I've always felt that the "Great Financial Panic of 2007" was a good old-fashioned banking panic, even if the modern version at first looks quite different from those of the nineteenth and early twentieth centuries. Gary Gorton's wonderful book, building on his earlier research, gets it exactly right:

"Holders of short-term liabilities...refused to fund "banks" [that is, various vehicles in the shadow banking system] due to rational fears of loss. ... As with the earlier panics, the problem at root is a lack of information."

(b) Simultaneously, I've always felt that the Great Panic of 2007 was largely driven by "too big to fail" (TBTF) (e.g. see Gary Stern here), which creates incentives that promote excessive risk-taking. (If you win, you make a fortune; if you lose, you get bailed out.)

What didn't occur to me until a dinner with Gorton during the 2015 PIER Workshop is that (a) and (b) are largely incompatible. That is, if the Panic of 2007 really is like those of the nineteenth century, then it can't have been driven by TBTF, which didn't exist back then. And the Panic of 2007 really was like those of the nineteenth century. 

So my view has evolved significantly: TBTF may well have made the Great Panic more likely than it otherwise would have been, and TBTF may well have increased its severity relative to what otherwise would have been, but TBTF simply can't be "responsible." Thanks, Gary for pushing me forward.