Initial Comments on the Draft House Bill on Systemic Risk and “Too Big to Fail”

Douglas J. Elliott
Douglas J. Elliott Former Brookings Expert, Partner - Oliver Wyman

October 28, 2009

The House Financial Services Committee just released a 253-page draft bill proposing changes to financial regulation intended to address systemic risk and the problem of financial institutions that are deemed to be “Too Big to Fail.” The bill also lays out an “enhanced resolution authority” to allow regulators to tackle certain failing financial institutions in a similar manner to the way the FDIC can already take over a troubled bank. The Obama administration immediately endorsed the draft bill, which significantly increases the odds that most of these provisions will make their way into law. Here are some initial comments, based on a quick reading. Please see my previous reports, particularly “Reviewing the Administration’s Financial Reform Proposals,” for more about the administration’s proposals and my take on them.

The bill is broadly in line with previous administration proposals in these areas. In previous papers I have strongly favored the enhanced resolution authority proposal, which I consider essential, and the general approach to systemically important financial institutions that requires them to hew to significantly tougher capital and other standards. I have some specific concerns about the proposed legislative language, but remain supportive of the general approaches in both areas.

On the other hand, I remain ambivalent about the proposed steps to monitor and respond to systemic risk more broadly. It is worth trying something along these lines, since the potential gains are quite high and the costs quite low. However, I believe the proposed governance structure would make an already very difficult task even harder. Politics and public attitudes will always make it extremely tough to act against a bubble, since bubbles are very popular. Vesting the power in a council of bureaucrats, rather than a single agency, will make it nearly impossible to successfully act when next we find ourselves in a major bubble.

“Too Big to Fail” financial institutions

Much of the bill focuses on the identification and regulation of “identified financial holding companies.” These are the large banks, and a few large financial institutions that are not banks, that are so important that the government is likely to feel required to rescue them should they approach failure. (The slang term for this is “Too Big to Fail.”) I know of no policy analyst who is comfortable with having banks that are too big to fail, but most analysts, including me, do not see an economically efficient way of completely eliminating this problem. Until we do, I agree with the administration and the Financial Services Committee that such institutions need to be regulated more rigorously than other financial institutions in order to reduce the odds of their failure and the cost to the taxpayer if they do fall apart.

The draft bill deals comprehensively with this tougher regulation and generally in a manner that seems sensible. The Federal Reserve Board (“Fed”) would have wide powers to increase capital requirements, require higher levels of liquidity, and insist on other steps that reduce risk from identified institutions.  (Congress is not eager to prescribe the exact rules, nor should they be in this very technical area.) These requirements set by the Fed would be enforced by the primary federal regulators of the relevant institutions, but the Fed would have a back-up power to enforce them directly if the primary regulators fail to do so.

However, on first reading, there are some points in the bill that concern me. First, the bill specifically forbids the government from publicly identifying the affected institutions. This secrecy is presumably a response to the concern expressed by many that the markets and the public will perceive an implicit federal guarantee of these institutions, which would give these banks an unfair competitive advantage and also encourage them to take on excessive risks, in the expectation of being rescued if things do not work out. These are valid concerns, but I believe they are not pressing enough for us to have a secret set of regulations. The public and the market need and deserve to know which institutions are to be subject to these harsher requirements. I accept that there are serious disadvantages to having institutions that are perceived as too big to fail, but I think we are already in that situation and will remain there regardless. If we polled the markets to find out which 20 institutions they believe are too big to fail, I am confident that there would be near-perfect agreement and that the list would very largely overlap that of the regulators. Given this, there is little advantage to keeping the choice secret. In fact, the ambiguity could create its own problems in a crisis, since an institution on the margin might be perceived as having no implicit government backing when it actually does – a serious predicament, since taxpayers would be on the hook should it fail, but there would be no advance benefit in terms of calming the markets.

Second, the Fed’s powers would be extremely broad and would be allowed to vary between different institutions, without significant constraint on the variations. This makes me uneasy. I accept that there may be reasons to be relatively tougher on the largest of these institutions and less tough on the smaller ones and that we may need to reflect other measures of relative risk. However, we should be able to do this effectively by setting specific standards, rather than making ad hoc differentiations.

Third, one of the powers would be the ability to impose size limits on a financial institution. I would not favor this except perhaps under certain extreme circumstances. It almost always makes more sense to directly regulate the source of the risk, whether it is excessively low capital or concentrations in certain risk categories or bad management, than to cap the size of a functioning organization. (This excludes anti-trust reasons, which are not addressed in this bill.)

Emergency financial stabilization

The law would provide specific authority to rescue financial institutions under emergency circumstances, if a two-thirds majority of the Fed and FDIC Boards agree with the Secretary of Treasury, in consultation with the President. This would presumably only occur with an “identified financial holding company,” but circumstances could arise where an institution that was not previously so identified turns out to be systemically important. The government would be authorized to provide loans or guarantees, but not equity, to solvent financial institutions as “necessary to prevent financial instability during times of severe economic distress.” This is more limited than the TARP restrictions, where the support was primarily provided in the form of preferred stock, which is a kind of equity. (It is, of course, possible that a loan would temporarily be provided while Congressional authorization was sought for an equity investment.) Note that a rescue could not be mounted under this section of the law for an institution which is already known to be insolvent. Any losses to the FDIC or government, net of repayments, would be recouped through a levy imposed after the fact over a period of years on all financial institutions with more than $10 billion in assets.

There are clearly disadvantages to providing a generalized rescue authority such as this, but I believe that we do need some such authority for the administration and regulators to act in emergency situations where Congress could not reasonably be expected to act in a sufficiently timely manner. I also support the general principle that the financial industry should bear the cost of rescues of systemically important financial institutions, although I have not focused on the particular mechanism to do so.

Systemic risks

In addition to tackling risks to the system represented by the most important financial institutions, there is also a need to look for risk factors that pervade the system, such as in the recent housing and other bubbles. Most analysts agree on the desirability of such overall systemic oversight in broad principle. The draft bill proposes to have the Financial Services Oversight Council (FSOC) act as the overseer. The council would be made up of all the major federal financial regulators and chaired by the Secretary of the Treasury. The Fed would provide considerable analytical and ministerial support, but the key decisions would lie with the council as a whole. The council would have very broad authority to limit certain activities by financial institutions or to require higher capital to be held against those activities or to impose higher loan-to-value ratios on mortgages.

I have two major concerns. The primary one is that the FSOC will in practice find it almost impossible to act against the next major bubble. Bubbles, as noted earlier, are invariably quite popular. They arise because a large portion of the population accepts some tenet, such as the inevitability of rising house prices. Further, large paper profits are produced for many people as prices rise unsustainably. Thus, any action would run up against strong lobbying from the wide range of people and institutions who believe they are profiting from the bubble. Our governmental system is designed to make action difficult when there is not a consensus. Vesting the systemic risk powers in a council is an invitation for at least some of the regulators to respond to lobbying and Congressional and public concern by watering down or delaying the necessary responses. This is a hard enough task for a single agency. It will be an almost impossible one for a committee.

On the flip side, the powers of the council are so broad that there is the theoretical potential for a response that does more harm than good. I would prefer to see a more limited range of potential responses, including higher capital requirements on activities that appear to be creating systemic risk. It just does not seem like good practice to rely quite so heavily on regulatory fiat. That said, my fear remains greater about inactivity by the council than about hyper-activity.

Enhanced resolution authority

Regulators currently have the ability to intervene strongly when a bank goes off the rails, but they have much lesser authority in the case of non-bank financial institutions, including the bank holding companies that almost always own the banks. For large banks, these holding companies are so integrated with the banks that I agree with the administration and the committee that it is unworkable to have different regimes to resolve these institutions when they become troubled, as we do now.

The bill would give the regulators powers to intervene in the case of troubled “identified financial holding companies” that are very similar to the powers regulators already have in regard to banks. First, they would be allowed to force actions by these institutions to restore adequate capital levels and sound operations. Failure to take these actions would empower the regulators to step in and take over the institution even if it is still solvent. This contrasts sharply with current bankruptcy law for non-banks, which allows action only against insolvent institutions and even then may not allow an outside party to intervene until the firm actually misses a cash payment. Second, once the regulators take over, the FDIC would have the power to handle the disposition of the financial institution that would supersede any potential bankruptcy court action. Thus, it would be up to the FDIC to equitably divide the assets of the financial institution in event of liquidation, rather than relying on a bankruptcy court.

There are many more details in the bill related to the enhanced resolution authority than I can reasonably address in this short piece. In broad terms, it puts regulators in the same position in regard to non-bank financial institutions as they are now in regard to banks.

I strongly support enhanced resolution authority for bank holding companies and other affiliates of banks, since I believe it creates insuperable problems to resolve related institutions through such disparate mechanisms as those currently in place for bank resolution and bankruptcy law. I also support extending the resolution authority to systemically important insurance companies and other financial institutions, even if they are not affiliated with a bank, since it would provide more flexible ways of dealing with, or avoiding, a taxpayer-financed rescue. However, I feel less strongly about this than I do about covering bank affiliates.

Other parts of the bill

There are many more issues addressed in the bill than I have time to discuss now, including limitations on activity by non-bank financial institutions. These will have to wait for another paper.


I believe that the bill represents a distinct improvement over our current system, but I would ideally like to see a number of changes to reduce or eliminate the concerns expressed above.