any set of regulations that attempts to assign risk to all securities in the same way for all banks would necessarily create some regulatorily-favored group of securities because the objective algorithm assigned too low an amount of risk capital to those securities purely as a result of statistical flukes and selection bias. Were each bank to do the analysis itself, their errors would be more r andom and less likely. Unfortunately, with a comprehensive system of regulations, identical errors pervade all banks.
Tyler Cowen points to this article today, although it was written several months ago. David Henderson mentioned it back then.
READER COMMENTS
thruth
Aug 10 2011 at 11:17am
Agree that capital regs are flawed. But is capital regulation coupled with deposit insurance worse than no regulation at all? His simple model doesn’t allow the possibility of runs, so I think it is basically silent on the critical issues.
Chris Koresko
Aug 10 2011 at 1:07pm
This is way outside my area of expertise, but it looks like Maymin makes some really good points, especially that regulation reduces the natural diversity that might enable the financial ecosystem to survive distress.
He also makes the critical observations that “volatility” estimates based on asset price variance are not a good measure of real risk, because the variance estimates based on the small available number of data are necessarily noisy, and because the real variations are not Gaussian.
thruth: But is capital regulation coupled with deposit insurance worse than no regulation at all?
It’s really hard to be sure, but my guess, based on the history of banking problems caused by regulation, and what I’ve heard about the success of “free” banking in Scotland, would be yes.
His simple model doesn’t allow the possibility of runs…
Maymin’s article explicitly mentions bank runs, not as a problem but as a natural disciplinary mechanism that helps keep banks from taking on too much risk.
Lord
Aug 10 2011 at 1:09pm
Sadly, this is probably not true; that in the absence of regulation we would probably have both more closely aligned errors and larger ones. Regulations probably just slow the process.
mark
Aug 10 2011 at 4:05pm
This is a very important point. To which I would add that the risk capital guidelines from Basel II in the mid 80’s not only steered financial institutions toward certain kinds of assets, particularly agency MBS and sovereign debt, which have been at the heart of the ongoing crisis, but also created an open invitation to financial engineers to securitize and repackage assets that would have independently been given higher risk weights into packages that had AAA ratings and thus were given low risk weights, although the very essence of that packaging is to increase risk correlation.
Finally, I believe, although there is no theoretical proof, as it certainly doesn’t have to be this way, that allowing banks to merge and grow large also increases risk correlation. When this happens there is pressure on other banks to follow the leader’s strategy and that causes further correlation.
steve
Aug 10 2011 at 9:38pm
Banks can hide their losses for a long time. The bad money would drive out the good. In order to compete, everyone would need to use the capital standards set by the banks with the lowest requirements. This would probably work really well with something like a, hmmm, maybe a real estate bubble? As long as real estate prices keep going up, the banks with the lowest requirements win. The others have to follow, and they will.
Steve
Comments are closed.