When too big to fail isn’t systemically dangerous

By Felix Salmon
October 20, 2009
Economics of Contempt finds something very odd in the legislation that Tim Geithner has proposed for beefing up the regulation of too-big-to-fail institutions, or "Tier 1 FHCs" as they're known in the jargon:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Economics of Contempt finds something very odd in the legislation that Tim Geithner has proposed for beefing up the regulation of too-big-to-fail institutions, or “Tier 1 FHCs” as they’re known in the jargon:

The administration makes a big mistake by requiring a separate “systemic risk” determination in order to use the proposed resolution authority for Tier 1 FHCs. This introduces needless uncertainty. Remember, a financial company is a Tier 1 FHC, by definition, if “material financial distress at the company could pose a threat to global or United States financial stability or the global or United States economy during times of economic stress.” An institution thus can’t even be a Tier 1 FHC in the first place if it doesn’t pose a systemic risk. Why require an additional, albeit slightly different, determination of “systemic risk” before the new resolution authority can be used? This will leave the market guessing as to which resolution regime — the Bankruptcy Code or the new resolution authority? — will be used to resolve distressed Tier 1 FHC. Creditors, unsure which resolution regime will apply and thus how their claims will be treated, will be less likely extend credit at exactly the time we don’t want creditors to be pulling back from a Tier 1 FHC.

I’m with EoC here: it makes no sense to require two separate determinations of when a bank is too big to fail. I guess the idea is that the government can preserve the fiction that Tier 1 FHCs aren’t necessarily too big to fail, and that therefore investors should be wary of lending to such institutions, instead of brainlessly making the moral-hazard play that all debt of Tier 1 FHCs is implicitly backed by the government.

Still, it’s silly. If you’re going to impose tougher capital and leverage requirements on too-big-to-fail institutions just because they can’t be allowed to fail, then you shouldn’t have to make a second determination that they pose a systemic risk before you intervene in times of crisis. After all, if you’re in a time of crisis, then simply making that determination would probably suffice to force some kind of intervention and conservatorship, so there’s going to be a lot of pressure on Treasury not to make it, unless and until bankruptcy is the only other option. Much better to make the determination ex ante, for all Tier 1 FHCs, and then address the moral-hazard problem directly.

One comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

seems like the problem could be the other way — what if the institution hadn’t been on the fed’s radar and so wasn’t a tier 1 fhc, but turned out to be a problem at a time of crisis?

aigfp?

Posted by q | Report as abusive