WASHINGTON — Four years after it began charging risk-based premiums, the Federal Deposit Insurance Corp. is set to revamp them again for the largest institutions.
The agency, which says it has learned a lot more about the riskiness of big banks from the financial crisis, wants to add new factors — including the level of concentration of higher-risk assets and a bank's ratio of core to total funding — to the formula to calculate assessments.
The plan would also create greater gradation between possible premium prices for large banks and emphasize as a factor the magnitude of FDIC loss from a bank failure.
"We'd like it so that over the long term the behaviors and practices that tend to create more risk in an institution get priced into the system — so that the factors that point to risk over the long term are better captured," said Arthur Murton, the director of the FDIC's division of insurance and research. "We'd like to do that as accurately as possible."
Although some critics say that the FDIC is tilting at windmills, that it will never be possible to gauge an institution's risk to the system, the agency is intent on trying to perfect its methods.
The FDIC's current formula stems from the 2006 deposit insurance reform law, which gave the agency more assessment authority to charge premiums according to risk. For most healthy institutions, the plan combines an institution's Camels rating with performance ratios, such as bad loans to total assets, to calculate a rate. For large banks the FDIC added other factors, such as debt ratings, to determine their assessment.
Though the agency has largely left the small-bank assessment calculation alone, it is clearly worried its system does not fully capture the risk that the largest banks pose to the Deposit Insurance Fund. Last year it added risk factors for such institutions and expanded its discretion to raise or lower a large institution's premium from its calculated rate.
But the FDIC is set to go even further, suggesting changes for institutions with more than $10 billion of assets at the same time the recently enacted Dodd-Frank law already made alterations that will result in higher premiums for many large banks.
As a result of the proposal, large banks, a universe of a little more than 100 institutions, would be measured by up to nine risk factors. One previous factor, debt ratings, would be retired because of the Dodd-Frank law, which required regulators to stop relying on credit rating agencies given their poor track record in the financial crisis.
The FDIC first unveiled the plan in April, but issued a revised version on Nov. 9 to conform with changes made by the Dodd-Frank law. Agency officials say the new system would reduce procyclicality by affecting a large bank's premium at the moment it assumed risk, rather than increasing prices when it suffered losses down the road.
"During the crisis, it became clear that our large-bank pricing metrics were lagging indicators of financial deterioration, to a greater extent than the metrics we use for smaller institutions," FDIC Chairman Sheila Bair said in a Nov. 9 statement.
Observers said the process has similarities with that of the Basel capital standards, which appear in near constant flux given the changes and updates regulators have suggested over the years.
"Both deposit insurance and Basel need to be works in progress because … we're going to continue to have these blowups from the gap between regulation and products," said Joseph Mason, Louisiana Bankers Association Professor of Finance at Louisiana State University. "In both realms you have to realize that it's going to be an ongoing effort to address dynamic industry risks. The industry is never going to be static."
Many observers praised attempts to keep the pricing model up to date.
"They are adjusting it as they gain experience and as they're responding to different conditions," said Richard Carnell, a former Treasury Department official who is now a Fordham University law professor.
Robert Eisenbeis, a former regulator and now the chief monetary economist at Cumberland Advisors, said that, unless a bank's premium can correspond to its risk appetite in real time, it can escape the higher assessment that it should be paying.
"It's pretty well known that unless you have continual monitoring and constraining of risk-taking positions, institutions can always game the risk-based pricing by taking on more risk after the fact," Eisenbeis said.
But some argued that changing the plan may prove too difficult, and that accumulating more risk factors in the formula does not necessarily mean a premium truly reflective of risk.
John Douglas, a former FDIC general counsel, said a new large-bank pricing plan overlooks the fact that most FDIC losses in the crisis came from small-bank failures.
"I don't think the FDIC has it right, and I am not sure it is necessarily more predictive," said Douglas, a partner at Davis Polk & Wardwell. "It is more sophisticated" than the current formula. "I do anticipate they will continue to refine it, just as Basel keeps trying to refine capital requirements."
Others said once the door is open to adding factors every time a crisis spotlights new risk, it becomes hard to draw the line.
"It's an attempt to micromanage risk, which I think we've learned just doesn't work," said Robert Litan, a senior fellow at the Brookings Institution and research chief at the Kauffman Foundation.
Litan suggested a simpler approach: assess institutions solely on the basis of their size. "It's a slippery slope," he said. "Once you start making distinctions … you won't be able to stop."
Murton said officials "try to be sensitive" to overdoing changes.
"We've proposed a scorecard that has nine elements … to capture the complexities of large financial institutions. That doesn't seem like too many," he said. "And these are things that, I think, people are familiar with. We're not just pulling arbitrary measures out that people don't understand. We'll be judicious about how we make changes to the system.
"I don't think you can say that this scorecard proposal will result in us micromanaging the largest firms."
The FDIC reissued the proposal amid concerns the first version was too complicated and could conflict with provisions of Dodd-Frank, which was still pending at the time and contained numerous measures on deposit insurance.
Under the current premium system, all banks are placed in one of four price "buckets." In the first bucket, where most of the industry resides, institutions now pay between 12 and 16 basis points per domestic deposits. (A separate rulemaking, required by Dodd-Frank, will change the assessment base to assets minus capital.)
Where a bank's price falls within that 12-to-16 range depends on the risk-based formula.
But under the new plan, each large bank would receive an individual scorecard corresponding with their price, and subject to a much more complicated formula. The score would combine elements of the bank's ability to withstand asset- and liquidity-related stresses, along with supervisory ratings and a "loss severity score" — meaning how big a hit the FDIC would take should the bank fail. The formula has more steps for "highly complex" institutions — essentially those with assets of more than $50 billion owned by holding companies with assets of more than $500 billion. Such behemoths would be measured by additional market indicators, such as trading volatility.
The FDIC said the new proposal is in fact simpler than the prior one, but that oversimplifying it would defeat its goal.
"The FDIC recognizes that the scorecard and some risk measures in the scorecard continue to be somewhat complex; however, this complexity simply reflects the complexity of large" institutions, the agency said in the staff proposal. "Further reducing the complexity would lead to considerably less accuracy in predicting risk."
Murton said the new assessment scheme would be similar to internal assessments that banks already conduct.
"There are things that large banks do on a regular basis in terms of their regular business dealings that are far more complex than this," he said.
