WASHINGTON — The Federal Deposit Insurance Corp.'s plan to require banks to prepay three years' worth of premiums appears designed to ensure the agency does not repeat the fallout from the last time the government took dramatic steps to rebuild federal reserves.

Twenty years after the savings and loan crisis, the FDIC finds itself in a situation similar to that of the early '90s, with a negative fund balance and a need for more liquidity to handle expected failures.

Back then the FDIC charged banks as much as 25 basis points — only to find out it had reserved too aggressively. A rapid economic improvement starting in 1992 meant fewer failures than projected, causing a boomerang effect for federal reserves. The banking industry, which had argued the high premiums were coming at the worst time, was left wondering why it had paid so much.

The current FDIC plan is designed to avoid the same situation, letting the agency receive enough to handle expected failures, but ensure that if the situation does improve, banks will not have overpaid.

"It does avoid the somewhat awkward situation of their raising fees and then turning around and lowering them right afterwards," said Robert Litan, a senior fellow at the Brookings Institution and the head of research at the Ewing Marion Kauffman Foundation.

The FDIC, on its way to handling well over 100 failures this year, said its Deposit Insurance Fund at the end of the third quarter was in the red, mostly because of money set aside for future failures, which are expected to cost $100 billion through 2013.

The conditions are very similar to more than a decade earlier.

In the early '90s, the continuing failure toll from the S&L crisis and projections of even more failures led to assumptions the agency was bankrupt. Litan and two other economists completed a study commissioned by lawmakers that said the FDIC was "economically insolvent."

But like today, the agency still had plenty of cash. In 1991, the FDIC, under pressure from a General Accounting Office audit, set aside $16 billion to deal with expected failures for the then-Bank Insurance Fund (which was combined with the thrift insurance fund in 2006). As a result, the funds were technically $7 billion in debt.

Yet conditions improved seemingly overnight as the economy turned around. Failures, which had totaled 127 in 1991, fell by five the next year, and dropped to 42 banks in 1993. By that year the balance of the two funds had shot up to $14 billion.

According to some estimates, about 81% of the loss reserve was returned to the fund balance.

"They never actually turned out to be insolvent," Litan said. "They skated through."

But the agency had acted as if it were bankrupt, and under the strict statutes of the time, charged extremely high premiums. By 1996, federal reserves were so high that a statutory requirement kicked in that resulted in most banks paying nothing in premiums — a situation that lasted until 2006.

"It was a very dramatic shift, and it was controversial," said John Bovenzi, formerly the FDIC's chief operating officer and now a partner at the Oliver Wyman Group subsidiary of Marsh & McLennan Cos. "All of a sudden, going from 1992 into 1993, the economy improved dramatically. The bank failures just fell off a cliff and stopped happening, and the losses never materialized and the fund recovered rapidly."

Bovenzi, who was then a senior official at the agency, recalled "the industry was a little suspicious of the reserving methodology."

"The reserving probably turned out to be too aggressive, but there's no way of really knowing," he said. "The problem is that predicting the number of bank failures and the cost of bank failures is very difficult to do. Those projections are always changing. You may look back and say, 'It never materialized.' It's hard to know that at the time."

Many doubt that a similar rebound will happen this time around, but they agreed the FDIC's plan could avoid the premium volatility of the '90s if events do play out the same way.

Under the FDIC's plan, the agency would receive $45 billion of liquidity up front, but the fees would not hit institutions' income statements until the quarter they would have otherwise paid. Premiums for most institutions are between 12 and 16 basis points, but will rise by 3 basis points starting in 2011.

"If the experience is better than expected, they can return the cash and the expense is never charged against capital by the banks," said William Longbrake, an executive in residence at the University of Maryland, who was a chief financial officer for the FDIC in the mid-'90s. "All things equal, if the economy recovers there is a chance that some of these loss reserves could end up being reversed in the future."

However, he and others argued that the FDIC's current projections may well be more on target than in the '90s, and that the agency is correct to reserve aggressively.

"The initial methodology for determining what those reserves should be was unsophisticated," but today "it's much more realistic in terms of actual results," Longbrake said. "On balance, there's every incentive to be on the conservative side of the range of estimation. This is an art, not a perfect science."

Others agreed that while an instant recovery is conceivable, it would be imprudent to expect a repeat of 1992.

"It's always possible, but of course … recovery rates are considerably less than they were at that time and the macroeconomic outlook as at the very least uncertain," said Karen Shaw Petrou, managing partner of Federal Financial Analytics Inc. "It's important for the FDIC, as one would hope banks would do, to err on the prudent side with their reserving policy."

She added that the FDIC was also helped during the savings and loan crisis by a greater potential for asset recoveries. In the current crisis, most of the assets available to the FDIC are troubled mortgage-backed securities.

"Assets are far more complicated than they were then and therefore inherently riskier," she said. "A lot of the assets, if not virtually all in the late '80s or early '90s, were loans or physical assets with ready rights to collateral. That is clearly not the case in the market" today.

James Chessen, the chief economist for the American Bankers Association, noted the agency's historically high projections for failure losses as a percentage of a failed bank's assets — averaging about 25% — which play a big role in its reserving.

Overall, he said, the agency has improved its process since the '90s. The prepayment also ensures that, if reserves do turn out to be too aggressive, the industry does not bear the burden of it.

FDIC officials "were there in that period," he said. "They've gone through the reserving methodology. The staff is keenly aware of the experience of the '90s. It's obvious they don't want to repeat that."

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.