Speeches & Testimony
Remarks by FDIC Chairman Sheila Bair to the Exchequer Club of Washington D.C.
June 18, 2008
Good afternoon everyone. Thank you, Ron, for that very kind introduction. I am pleased that you have invited me back again this year to speak to you. In fact, with your indulgence, I'd like to pick up where I left off in my speech last year. As you may recall, I talked about why the FDIC is spending time worrying about what then seemed such a low probability event as the failure of a large bank. I ended my speech by noting that–
[S]ome have come to believe that the FDIC should not spend any time worrying about or planning for a large bank failure because these banks have become so well diversified and sophisticated in their risk management.
What a difference a year makes. Since then, the FSA has taken over Northern Rock after announcing it would protect all depositors to prevent a bank run and the Fed has intervened to prevent the bankruptcy of Bear Stearns. The failure of a large bank remains highly unlikely–and I am certainly not predicting one–but for many it is no longer unthinkable.
Update on FDIC Readiness Efforts
Last year I spoke about the FDIC's efforts to prepare for such an event. At the FDIC, we must plan for the worst and work with the industry to try to get the best outcome. Today, I'd like to give you a brief update on what we have been doing this past year before I proceed to my main topic, which will be the implications of Bear Stearns for the way we treat investment banks.
Yesterday, the FDIC's Board adopted a final rule to modernize the claims process. The rule reflects the comments we've received on several proposed rulemakings on the claims process issued over the past few years. It also reflects extensive talks we held with industry representatives.
The rule requires that large banks have the ability, in the event of failure, to do several things. They must be able to place holds on a fraction of large deposit accounts, produce depositor data for the FDIC in a standard format, and automatically debit uninsured deposit accounts so that they will share losses with the FDIC.
This approach should give most depositors uninterrupted access to virtually all their funds, thus diminishing the likelihood that liquidity problems for individuals and businesses will lead to disruption in the financial system. To complement the industry's efforts, we have been extensively modernizing our computer systems and expanding our ability to categorize large numbers of claims in a very short time–one to two days.
I also expect that the Board will soon consider an NPR on qualified financial contracts (or QFCs), which include derivatives and some other financial contracts. When a bank fails, the FDIC has only one business day to decide how to treat the bank's QFCs. In addition, we must decide whether to accept or repudiate all positions held with an individual counterparty. When a bank has a large volume of QFCs, this can be challenging. Banks may not keep their QFC records in a way that provides the information we need quickly. I anticipate that the NPR would specify the information that troubled banks would have to maintain on QFCs and how it would be provided to the FDIC. We will also seek comment on whether all banks should be held to some minimum recordkeeping requirements on their derivatives portfolios.
Over the past year, we have conducted a series of tabletop exercises to test and improve our ability to handle the failure of a large bank if it were ever to occur. These exercises usually target a single, hypothetical bank, but sometimes target several banks.
In each case, we work through the FDIC's preparedness plans and identify areas for improvement. The most recent exercise was held earlier this year and posed the hypothetical failure of a very large commercial bank. We plan to continue these exercises and are hoping to bring other regulators in to participate.
The Evolution of Too Big To Fail
When I was here last year, I described the evolution of "too-big-to-fail" for commercial banks and the two events that served as the book ends for how the U.S. has approached the tradeoff between stability and moral hazard - the resolution of Continental Illinois in 1984 and the passage of the FDIC Improvement Act of 1991, known as FDICIA.
The resolution of Continental Illinois was a turning point for the FDIC, just as Northern Rock and Bear Stearns appear to be for the FSA and the Fed. In Continental Illinois, the FDIC provided open bank assistance and protected more than just insured deposits, but there were some key differences.
At the time, the FDIC had the statutory authority to act as receiver for a failed bank, and the deposit insurance fund-built by industry contributions-was available for open bank assistance. However, the FDIC still maintained a credible threat of closing the bank. Although shareholders had to approve the transaction, their interests were greatly reduced and subject to potential elimination.
After Continental, in response to concerns about disparate treatment of large and small banks, the FDIC used its discretion to resolve failures in ways that protected most, if not all creditors.
In 1987, the FDIC received bridge bank authority. Bridge banks allow the operations of a bank to continue under FDIC management without requiring shareholder approval. The FDIC used this authority to handle the next few larger banks that failed-including First Republic and Bank of New England-in a way that eliminated shareholders' interests and imposed losses on creditors.
In 1991 Congress sought to further reduce costs and the potential for moral hazard by requiring that the FDIC always use the least costly method of resolving a failed bank. The only exception is for systemically important banks, and Congress deliberately made it difficult to declare a bank systemically important.
Northern Rock and Bear Stearns raise new questions about how to strike the right balance between stability and containing systemic risk, on the one hand, and containing moral hazard and protecting the federal safety net, on the other.
The handling of Bear Stearns kept the institution open, preserved some shareholder value, and protected all other creditors. It also extended the federal safety net by providing discount window liquidity support and an express credit guarantee of $29 billion. In the case of Continental, the shareholders were eventually wiped out and the management was removed.
Should we view the extension as a one-time event or as permanent? In my view, it is almost impossible to go back. As Gary Stern has said, "There is no way to put the genie back in the bottle. Even if we were to announce that we're never going to lend to investment banks again, would that be credible given what we've done?"
If this is the case, it makes sense to extend some form of greater prudential regulation to investment banks as well as a process or protocol for dealing with a systemically significant investment bank approaching failure. The government cannot be put in the position of having to simply write a blank check when these institutions get into trouble.
At a minimum, there should be greater parity between commercial banks and investment banks over how they manage risk, liquidity and capital. There should be a Prompt Corrective Action-like mechanism with mandatory triggers for supervisory intervention and, if necessary, closure if capital is not restored. While many cite Bear Stearns and Northern Rock as liquidity failures, they were both over-leveraged. Greater capital improves access to liquidity.
However, this is not meant as criticism of the Fed. There is a playbook for the failure of a commercial bank, even a systemically important one, but there isn't any for the failure of an investment bank. The Fed had to invent one on the fly. The Fed was in essentially the same boat as the FSA, which had no ready mechanism for handling the failure of Northern Rock. Lack of a playbook was part of the reason the UK had to protect all depositors and nationalize the bank.
Receivership Process for Investment Banks
In my opinion, we need predefined rules to handle potential failures. As Larry Summers has noted:
The authorities had no realistic choice but to provide support as Bear Stearns faced bankruptcy. They do have a choice as to whether to put in place a regime where such problems can be managed with no government financial support provided directly or indirectly to shareholders or unsecured creditors. A resolution regime that could apply to any financial institution that became a source of systemic risk should be an urgent priority.
I believe that we need a special receivership process for investment banks that is outside the bankruptcy process, just as it is for commercial banks and thrifts. The reason goes back to the public versus private interest.
The bankruptcy process focuses on protecting creditors. When the public interest is at stake, as it would be here, we need a process to protect it. This process must achieve two central goals. First, it should minimize any public loss and impose losses first on shareholders and general creditors. Second, it must allow continuation of any systemically significant operations.
As I've previously suggested, the FDIC's authority to act as receiver and to set up a bridge bank to maintain key functions and sell assets offers a good model. A temporary bridge bank allows the government to prevent a disorderly collapse by preserving systemically significant functions. It enables losses to be imposed on market players who should be at risk, such as shareholders. It also creates the possibility of multiple bidders for the bank and its assets, which can reduce losses.
The authorities that the FDIC has are a good model, but there are still many open issues.
In an intervention, access to liquidity and a method for bearing losses are necessities. For commercial banks, the deposit insurance fund fronts money for losses, but it recoups them from the bank's assets and from assessments on the banking industry. Should there be a similar fund for investment banks? An alternative might be to provide for special ex-post assessments on all investment banks over a certain size to recoup government losses where support has been provided to a systemically significant investment bank. FDICIA also has an ex-post special assessment for recovering the FDIC's costs when a systemic risk determination has been made.
Another question is whether all investment banks would be subject to the receivership process. Many commentators suggest that only systemically important investment banks are of concern, but determining how and when a decision is to be made on whether a potential failure poses a systemic risk is not simple.
The systemic risk determination process for banks is complex and is made only when failure threatens. It requires a two-thirds majority of both the Boards of the FDIC and the Federal Reserve, as well as the approval of the Secretary of the Treasury, who must first consult with the President.
If a systemic risk determination for investment banks is similarly not made until the bank is in trouble, the rationale for imposing costs and heightened prudential regulation on investment banks that are not systemically important is undercut. On the other hand, if the determination is made well before failure, moral hazard could greatly increase. At a minimum it would seem that investment banks qualifying for access to the discount window should be subject to heightened prudential supervision.
The treatment of creditors and shareholders in receivership also poses issues. For commercial banks, as I mentioned, the least cost test generally imposes losses on classes of creditors and on shareholders, thereby maintaining market discipline and reducing resolution costs.
But in a systemic risk situation, the least cost test can be relaxed. However, my assumption is that even if a systemic risk determination were made for an insured depository institution, shareholders would take a complete loss and general creditors would at least take significant hair cuts. We will need to decide whether the rules for paying claimants in an investment bank receivership should be the same.
As we've worked through issues associated with a potential large bank failure, we have found that the inter-relationships between the institution, its parent and other affiliates within the holding company could potentially complicate an orderly receivership. The same situation would hold true for investment banks. Congress may want to also consider addressing this issue.
There are many other issues, but let me just mention a few more -
who will be the prudential regulator?
who will make the decision to close the investment bank and appoint a receiver?
should cross-guarantees apply to all affiliates?
who will be the receiver in charge of the bridge bank?
I doubt there would be many volunteers to be the receiver. For that job - there would probably be a stampede for the exit. Running a bridge bank, as the FDIC has done on several occasions, is a thankless job. Nevertheless – while I'm agnostic about how this is handled for investment banks – housing all receivership and resolution responsibility in a single federal agency may make sense. It would ensure that expertise is at the ready. Also, large banks and investment banks have many interrelationships with each other, including counterparty exposures. This again argues for putting all resolution authority in a single agency.
At least for the last century or so, every federal agency confronting the choice of allowing a systemically important financial institution to simply fail has chosen to act. Poets may take the road less traveled by and be happy with the choice. When systemic risk threatens, the government cannot allow the institution to simply collapse.
Congress has recognized that processes must be in place before a systematically important commercial bank threatens to fail. The FDIC plans and prepares for the possibility of such a failure–however remote–seriously. We need to do the same for major investment banks, since they can pose systemic risks at least as severe.