Fuzzy Thinking on Breaking up the Banks.

Noam Scheiber considers the problem of Too Big To Fail, writing that skeptics of "breaking up the banks" who focus on the fact that size wasn't the major factor that lead to the crisis are merely gaining a "rhetorical advantage." Linking to James Kwak, who admits that the problem is not just size but other factors, Noam concludes:

Why don't we call it "too spooky to fail" and agree that we don't want to live with anything like it.

Now we can get out our spooky-meters and use them to measure the banks, and break up the ones that are too spooky. Sounds like awesome public policy.

Jokes aside, the problem breaking up the banks is that no one -- not Noam, not Kwak, neither Paul Volcker nor Mervyn King, has come up with a standard for when it becomes necessary to break them up. If they merely want regulators to force banks to divest based on their discretionary identification of systemic risk, then they're set -- those powers are already in the House and Senate regulatory reform bills, and they should be. But if they want to institute broad limits on bank size and interconnectedness, which is what the argument appears to be, then they should say what that standard would be and why.

Some people suggest returning to Glass-Steagall. Fine, but separating commercial banking from investment banking wouldn't stop investment banks like Bear Stearns and Lehman from getting into trouble. Prevent commercial banks from running hedge or private equity funds? Do it, but that's not what caused the crisis. Regulating derivatives, stopping sub-prime loans, and all the other parts of regulatory reform are all more important to preventing a repeat of 2008's crash than splitting banks apart.

But we do want to prevent future bank bailouts in the (likely) event that some unexpected financial innovation (ugh) leads banks to fail again. Kwak says it's tricky to measure the overall importance of financial institutions. And it is! Maybe that's a sign that, in order to prevent taxpayers from holding the bag on future finance disasters, we shouldn't be focusing on finding an exact standard of Too Big To Fail, but instead making sure that even important institutions can be dissolved in such a way that they don't impact the overall economy. Sure, that's also a hard thing to do, but at least we know that federal regulators are already really good at dissolving banks -- certainly much better at that than identifying systemic risk in advance.

-- Tim Fernholz