Deathbed Designations

One of the biggest issues outstanding in financial reform is which institutions the Financial Stability Oversight Committee (FSOC) will deem “systemically important,” and thus subject to all the enhanced Title I regulations. These systemically important financial institutions, known as SIFIs, will presumably be subject to the new resolution authority rather than the bankruptcy code — although for some reason SIFIs aren’t automatically subject to the resolution authority, a problem which I pointed out within hours of the Treasury releasing its initial legislative language back in 2009. In spite of that anomaly, SIFIs will be required to continually submit comprehensive “resolution plans” to the FDIC, so that the FDIC will have all the information they need to plan, design, and execute a successful resolution of a SIFI. As I’ve said before (see here and here), and as Sheila Bair has been emphasizing repeatedly in recent weeks, the resolution plan requirement is hugely important.



Dodd-Frank does, however, leave open the possibility that a failing financial institution could be designated as a SIFI at the last minute, and then immediately handed over to the FDIC to resolve under the new resolution authority. In that situation, the FDIC would be forced to resolve an institution without the benefit of a resolution plan — that is, without the comprehensive information on organizational structure, funding practices, major counterparties, etc., that will be required in resolution plans. Bair calls these “deathbed designations” (which, you have to admit, is a clever name), and has been understandably arguing that this situation “should be avoided at all costs.”



But how do you avoid “deathbed designations”? Bair argues that the FDIC should be allowed to collect information from a broad class of potential SIFIs:

[W]e need to be able to collect detailed information on a limited number of potential SIFIs as part of the designation process. We should provide the industry with some clarity about which firms will be expected to provide the FSOC with this additional information, using simple and transparent metrics such as firm size, similar to the approach used for bank holding companies under the Dodd-Frank Act. This should reduce some of the mystery surrounding the process and should eliminate any market concern about which firms the FSOC has under its review. In addition, no one should jump to the conclusion that by asking for additional information, the FSOC has preordained a firm to be “systemic.” It is likely that, after we gather additional information and learn more about these firms, relatively few of them will be viewed as systemic, especially if the firms can demonstrate their resolvability in bankruptcy at this stage of the process.

Frankly, I don’t see any other option. Bair is right that “deathbed designations” are the worst of all worlds. At the same time, we can’t rely on the FSOC to accurately identify every single SIFI ahead-of-time, in perpetuity.



The question, then, is: how much and what kinds of information should the FDIC collect from potential SIFIs? One thing to keep in mind is that the FSOC, in figuring out which institutions should be designated as SIFIs, will necessarily be collecting some information from potential SIFIs as well. But that information probably won’t be the same information that the FDIC would ideally want to collect — and since the FDIC is the one responsible for planning and executing these difficult resolutions, I think they should have reasonably broad discretion to say what information they need to collect ahead-of-time.



I think the best approach would be to allow the FDIC to collect the important information that isn’t constantly changing — e.g., typical funding practices mapped the institution’s legal structure, descriptions of the institution’s major business lines, potential acquirers of the major business lines, and so on. Detailed balance sheet information at major financial institutions is typically stale in a matter of days, so requiring that kind of information would probably be more trouble than it’s worth (especially if the institution has 10-Qs and 10-Ks available).



I also don’t see why the required information couldn’t be collected from potential SIFIs discreetly. That would eliminate Bair’s concern about people “jump[ing] to the conclusion that by asking for additional information, the FSOC has preordained a firm to be ‘systemic.’” Financial institutions communicate confidentially with regulators all the time, and I don’t see why this should be any different.



In any event, it’ll be interesting to see if Bair can press upon the FSOC to create this separate information-gathering authority.