The Clearing House Association, a trade group comprised of the 18 largest banks, spent 10 months and $2 million to plan a simulation of a large bank failure, a bold and massive undertaking.
"Had this not gone well, it would have complicated the debate immensely," says Paul Saltzman, the president of the group.
That may be the understatement of 2012.
Whether Dodd-Frank's Title II ends Too Big to Fail will affect how rules that bring the reform law to life are written. If regulators do not trust Orderly Liquidation Authority — the process regulators will use to resolve a failing megabank — will work, then rules governing capital to corporate governance and everything in between will need to be a whole lot tougher.
"Demonstrating to policymakers that there is an effective resolution framework is a critical part of the regulatory dialogue now," Saltzman says. "It will inform all of the macro-prudential rules and regulations that are being put in place to prevent a future failure."
Industry critics will never be convinced; they view The Clearing House's simulation as self-serving.
But it's hard to deny the industry some credit for trying to test how the government might take down a large bank without causing a full-blown crisis followed by an expensive bailout.
Was the simulation perfect? No, but that's the wrong yardstick. It's enough to say the simulation was constructive. And it provides perhaps the single most visible example of the industry proactively participating in Dodd-Frank's implementation, which is so much more effective than whining, or worse, undermining from the sidelines.
The simulation took place in Hurricane Sandy's wake in early November at the Dorral Arrowwood in Rye Brook, N.Y. The 180 participants, including 150 bankers from 23 institutions, split into 11 teams to play a variety of roles. The simulation focused on the failure of a $1.9 trillion-asset U.S.-based bank, but included two others: another U.S.-based bank with $850 billion of assets and a U.K.-based bank with $2.5 trillion of assets.
The eight other teams covered the roles played by the Federal Deposit Insurance Corp., other U.S. regulators, U.K. regulators, Congress, buy-side investors, market utilities, the media and finally a "command central" that oversaw the simulation.
Each of the 11 groups had a distinct briefing book, with a list of tasks and objectives. Each had its own room, equipped with a TV that provided a media feed, which included a variety of rumors designed to inject some chaos and make the simulation more real. "Reporters" on the media team were instructed to seek out information and participants were allowed to call the media to plant stories and try to influence events.
The Clearing House hired Ernst & Young as the project manager as well as Promontory Financial and a who's who of law firms including Davis Polk and Cleary Gottlieb.
The simulation was divided into four phases and assumed that Dodd-Frank had already been implemented. In other words, any limits proposed but not yet adopted by federal regulators were assumed to apply.
The "crisis" began with the news that Bank A had suffered a $6 billion loss in its U.K. broker-dealer on a Friday morning. That afternoon the company's U.S. commercial bank was hit with a $50 billion loss. Runs followed; confidence faded and Bank A's holding company faced a big debt repayment that it could not meet. It became clear the government had to step in or risk systemic contagion.
Dodd-Frank bars regulators from bailing out any individual bank so Bank A could not simply turn to the Federal Reserve Board for financial help. Under Dodd-Frank, the FDIC must decide that a bank's failure could create a crisis that would threaten the entire financial system and then convince other policymakers to go along.
FDIC Chairman Marty Gruenberg has said the agency's preferred approach to these types of resolutions will be a "single point of entry" where the holding company is seized and operating subsidiaries continue to function, so that's what the Clearing House tested.
In the simulation, the FDIC wiped out shareholders, fired the CEO (played by Greg Baer of JPMorgan Chase) and converted holding company creditors to owners of the new bridge bank.
To cover the bank's losses, the FDIC arranged an $80 billion line of credit that was funded by institutional investors and guaranteed by the Orderly Liquidation Fund, which was mandated in Dodd-Frank. The law requires other large banks to repay this money. (The people playing FDIC officials, led by Mike Krimminger of Cleary Gottlieb, asked for $200 billion but the folks playing Treasury Department officials, led by Michael Helfer of Citigroup, only approved $80 billion; in the end only $40 billion was tapped.)
There's more, but that's the basic set up.
The Clearing House is still compiling its final report, which it plans to take on the road and share with regulators here in the U.S. and abroad early next year. But interviews with various participants yielded a long list of recommendations.
Among the biggest — more cooperation among U.S. and foreign regulators — has already seen significant progress with the FDIC-Bank of England agreement.
The industry also wants the FDIC to work more closely with individual institutions as banks crafts their resolution plans.
Currently, large banks submit "living wills" to the FDIC, showing how they think they could be unwound in the event of a failure. The FDIC writes a separate resolution plan, laying out how it thinks a failure would be resolved. This is not a public document and only bits of the living wills are public so it's impossible to know how far apart these two plans are for any particular institution.
The Clearing House thinks resolutions would go more smoothly if the FDIC worked with individual banks to craft a joint plan.
The FDIC declined my request to interview Jim Wigand, who as director of the agency's office of complex financial institutions, is leading the charge on Orderly Liquidation Authority.
So I turned to Krimminger, who before joining Cleary Gottlieb this year was the FDIC's general counsel and worked at the agency for two decades.
Krimminger expects the FDIC will eventually decide to work with the large banks.
"There is no way, in isolation, you can develop a resolution plan for a huge institution without involving the institution because they are always going to know … where the issues are that will cause you problems," he says. "I think they need to sit down, institution by institution, and say, 'Okay, this is what we are thinking about with you. This is what our strategy would be. What are the problems with it?'
"They haven't gotten to the point where they are ready to do that. I think they will."
The Clearing House also wants the FDIC to spell out in more detail just how it sees one of these resolutions unfolding. The term of art for this is "presumptive path" and it's something the industry is pushing hard on. Ideally, banks want the FDIC to spell out in a policy statement how it plans to use its bridge-bank authority.
The FDIC wants to preserve as much discretion as it can because every resolution will be different. But there is a trade-off between flexibility and clarity. In the simulation, when investors or counterparties were confused, they didn't wait for more information, they dumped assets.
"The more certainty and clarity market participants can get about what the FDIC's plan is, the better," Krimminger said.
To ensure markets that function effectively, the government needs to convince investors that Dodd-Frank really did end Too Big to Fail.
"Regulators … want to maximize flexibility, but if nobody is certain what you are going to do, then you have a problem because the market may react in ways you don't anticipate," Krimminger says.
For example, if all short-term debt holders are left behind in the receivership, as the simulation assumed, then creditors may stop lending to other large banks. However, protecting short-term creditors could increase moral hazard and eliminate market discipline.
The Clearing House didn't reach a firm conclusion on this question, but it's high on its list of policy questions that the FDIC needs to answer.
Bankers also want, and expect, the Fed to issue a rule or guidance setting a minimum amount of long-term unsecured debt at the holding company. It's necessary because, in a resolution, this debt would be converted into equity of the bridge bank.
While the industry wants some guidance, it's also worried the Fed will come out with some sort of one-size-fits-all approach that, well, won't fit some banks. There also may need to be a limit on how much of this debt may be held by other large banks to reduce knock-on effects.
And that brings us to the single counterparty credit limits that the Fed has proposed. These were required by Dodd-Frank as a way to cut down on the interconnections among the large banks.
The simulation assumed these limits were adopted as written and that prevented the two other banks from buying any of the failed bank's assets — they were already too exposed to that bank.
"Many Dodd-Frank reforms intended to be stabilizing actually hindered the resolution process and amplified contagion risk," Saltzman says.
In addition to the counterparty limits, other policies that complicated resolution included Dodd-Frank's derivatives push-out provisions and the regulators' unofficial ban on dealmaking by any of the large institutions.
There are plenty of other issues to consider and it's likely the industry will attempt another simulation, perhaps in conjunction with the FDIC.
"We have briefings for regulators scheduled in early January, and look forward to finding ways to get them more engaged in any future simulations," Saltzman says.
The FDIC has, rather quietly, been doing a series of more narrowly focused tabletop exercises, and did do a broader simulation to learn more about the decision-making process needed to pull off one of these large-bank resolutions.
It, too, was conducted in December and it sounds like the Fed participated but all I could get was a description of the simulation from the FDIC. It "involved evaluating the required steps and possible alternatives when making the decision to potentially implement a Title-II resolution for a failing SIFI (systemically important financial institution)," the FDIC said. "The simulation tested the intra- and inter-agency decision making process leading up to Title-II resolution, identified issues and resolution alternatives, and improved interagency communication and coordination in the context of Title-II."
Sounds interesting. But the more detail regulators and the industry can provide on these and other simulations, the more we may finally begin to believe that Too Big to Fail is over.