Yesterday the House Financial Services Committee held a hearing on the too big to fail problem. Ordinarily a hearing includes a couple of witnesses chosen by the majority who say one thing and one or two witnesses chosen by the minority who say exactly the opposite thing. In this, however, it was hard to tell who was chosen by which side (although I can guess), given the extent of the agreement among former Fed bank president Thomas Hoenig, current Fed bank presidentsRichard Fisher and Jeffrey Lacker, and former FDIC chair Sheila Bair.
All agreed that the too big to fail problem still exists, five years after the financial crisis, and that it continues to distort the market for financial services. For example, Lacker, who is perhaps the most sanguine about Dodd-Frank (he likes the living will provisions of Title I), said, “Given widespread expectations of support for financially distressed institutions in orderly liquidations, regulators will likely feel forced to provide support simply to avoid the turbulence of disappointing expectations. We appear to have replicated the two mutually reinforcing expectations that define ‘too big to fail.’”
There was disagreement over whether Dodd-Frank, and the Orderly Liquidation Authority regime that it created, would continue the practice of bailing out failing financial institutions, with Bair arguing that Dodd-Frank “abolished” bailouts. Of course, everyone is against bailouts, but at the same time everyone is in favor of protecting the financial system and the economy against disaster.
The issue comes down to what I think is a simple ambiguity about the meaning of “bailout.” In the short term, when a megabank is about to collapse, bailout can mean two different things. First, it can mean that creditors and counterparties are made whole to the extent necessary to protect the financial system from domino or contagion effects. Second, it can mean that executives keep their jobs and shareholders continue to hold an equity stake. (These could be divided into two different things, but it doesn’t matter here.) Everyone agrees that Dodd-Frank gives the FDIC, using money borrowed from Treasury if necessary, the power to do the first kind of bailout. The statute says that the FDIC can’t do the second, and that’s what Dodd and Frank pointed to when they said that their bill abolished bailouts.
When push comes to shove in the heat of the moment, it’s the creditors and counterparties that matter, though, and they will be made whole—at least to the extent necessary to protect the financial system. At a moment of great panic and uncertainty, that probably means one hundred cents on the dollar. And that is what really skews the incentives ex ante—not whether managers and shareholders will be bailed out.* As Fisher said, this is why SIFI (systemically important financial institution) really stands for “Save If Failure Impending.”
Bair realizes this, and one of her key recommendations is requiring large banks to have a thick layer of unsecured, long-term debt, so that the FDIC can impose losses on those long-term debt holders. These creditors couldn’t pull their cash out overnight in a crisis, which would give banks a more stable funding structure. I’m not confident this will work, though. As long as banks have any significant amount of short-term debt, creditors will behave as if that debt enjoys a government guarantee, and the FDIC would have to bail them out.
Lacker thinks that the system should have no government guarantees at all—which means that if a crisis erupts, the government commits to sitting by and letting the world end. He thinks we can get there via living wills, which will enable regulators to ensure that banks can be allowed to fail without causing damage to the financial system. We can get there, he argues, because regulators can reject a bank’s living will until they are confident it could be allowed to fail, ordering it to restrict its activities if necessary. But this brings us back to the question of whether regulators can and will make these complex, politically fraught decisions without being duped by bankers.
As before, I still think that simple, structural limits—making banks “too small to save,” again using Fisher’s term—are the best way to go. Hoenig recommended restricting banks to banking, securities underwriting, and asset management. Fisher recommended restricting the megabanks to “an appropriate size, complexity, and geographic footprint.” I’m tired of people like Lacker saying, “I am open to the notion that such restrictions may ultimately be necessary to achieve a more stable financial system.” They’ve had five years to try it their way, and everyone agrees TBTF is alive and well.
* As I mentioned in a previous post, Joshua Mitts has a paper arguing that this dynamic will lead managers to try to position their firms to be bailed out for precisely this reason.
This piece is cross-posted from Baseline Scenario with permission.