What Banks Can Learn from Citi's Stress Test Mishap

WASHINGTON — Citigroup Inc.'s failure to see that it would not pass the Federal Reserve Board's stress tests has left the firm in a challenging position — and may serve as a stark lesson to other banks.

It was the second time in two years that a major bank publicly committed to deploying capital only to have the central bank block the move based on the exam results.

Vikram Pandit, like Bank of America Corp.'s Chief Executive Brian Moynihan before him, was forced to play defense this week, raising questions about the Fed's methodology and arguing that the bank had not really "failed" the stress tests.

The episode is likely to make other firms more conservative in the future to ensure they are not overly aggressive in plans to return capital, analyst said.

"Bank of America found out a year ago that they were in the Don't-Return-Capital bucket, and they're still in that bucket," said Fred Cannon, an analyst for Keefe, Bruyette & Woods. "Morgan Stanley would appear to continue to be in that one, and Citi found out they're still in it. The problem is do those companies know yet what can get them out of it?"

That did little to answer questions, however, about why Citi had appeared confident enough of its results that it repeatedly assured shareholders it would deploy capital later this year, and then found itself on the wrong side of the Fed's targets.

"Clearly there was some miscommunication, misunderstanding or miscalculation on the part of Citigroup that led them to miss it," said Brian Monteleone, an analyst for Barclays Capital. "Certainly, I would expect they would have wanted to avoid the types of headlines we've seen."

To be sure, Citi missed the 5% minimum capital target by the smallest of margins. Under the Fed's worst case scenario, it ended up with a low of 4.9% — but only if it moved forward with its relatively aggressive capital plans. If it did not move forward with those plans, Citi's ratio would have been 5.9%, passing the Fed's test.

Presumably the bank would have not pursued such plans if it had believed it would be rejected by the Fed (the central bank will allow Citi to continue its existing 1 cent dividend, but it can not do anything additional). Neither Citi nor the Fed has disclosed Citi's capital plan.

Observers said the issue isn't whether Citi asked for "too much" in its capital plans, but rather one of perspective. The firm saw itself among the top of its peers as an institution that had safely recovered from the crisis.

The Fed, on the other hand, did not agree.

"Citi is still classified by the Federal Reserve in the Bank of America, Morgan Stanley group of very large institutions that have still not truly emerged from the financial crisis and as a result are very hesitant to allow them to deploy capital," said Cannon.

Like Citi, other banks have raised questions about the entire exercise, noting that while the banks provided all the data, the Fed made its own assumptions and changes to the information. The banks can now appeal those decisions, but they were not given time to do so before the stress tests results were released.

"Despite all this analysis at the end of the day these are judgment calls by the Fed," said Cannon. "They were very specific about which banks they wanted that capital to be and Citi happened to be on the wrong side of that point."

Some institutions appeared to benefit from the stress tests, namely Wells Fargo & Co., JPMorgan Chase & Co. and U.S. Bancorp., which beat analysts' predictions by offering larger dividend increases or repuchases.

"The Federal Reserve is permitting the strongest banks to flex their muscles, while the weaker banks stay in the penalty box," Jaret Seiberg, an analyst for Guggenheim Partners' Washington Research Group wrote in a recent research note. "That is why we believe this test could be an inflection point for the strong banks. It creates the perception that strong banks now are free to manage their business in a way that benefits shareholders."

The question remains, however, as to why Citi did not know it was going to fall short. Banks were alerted well in advance to the economic assumptions like 13% unemployment and a sharp decline of gross domestic product to minus -8%, but it's unclear how much they knew about the models the Fed was using in its testing and the specific assumptions behind them.

"The banks then had to make their own assumptions about what assumptions the Fed was making and this may have caused some confusion," said Monteleone.

A spokeswoman for the Fed declined to comment .

Citi, meanwhile, has called on the central bank to disclose its models and the related benchmarks and assumptions. Institutions will receive confidential feedback from the Fed on their stress tests, as well as capital planning, by April 30.

Some analysts questioned whether Citi should have received the benefit of the doubt, given that it was only 10 basis points shy of the Fed's minimum threshold.

"Is it so perfect as to base regulatory policy on a 10 basis point or 20 basis point difference of opinion on one of many measures? I think the answer to me, anyway, is no," said Karen Shaw Petrou, a managing partner at Federal Financial Analytics Inc. "It's a false science to view these very useful, and indeed important, exercises as so precise and as so predictive as to base policy on 10 basis points."

Others are even more concerned about the lack of disclosure on why a particular banks' capital plan was approved or denied. The Fed has said there could be a number of reasons to reject a banks' capital plan, including not keeping pace with projected Basel III requirements.

"There are these other factors the Fed said it was going to consider," said Ernest Patrikis, a partner with White & Case LLP and former general counsel of the New York Federal Reserve. "Theoretically, based on those other factors the Fed can say, 'I do not object to your capital plan.' But we don't know that. We don't know what the Fed said."

The test also favored institutions that played it conservatively. Clearly smarting from last year, Bank of America passed because it did not ask to raise its dividend, not necessarily because it is in a better position than Citi.

Indeed, Bank of America and Morgan Stanley both ranked below Citi when it came to its Tier 1 common capital ratio without any capital deployment. MetLife Inc., Ally Financial Inc. and SunTrust Banks Inc., who also failed the Fed's test, were also below Citi in that particular ranking.

The Fed's rejection, however, may not ultimately be a bad thing for Citi — at least from a creditors' vantage point.

"For a bank like Citi, it's important to look at where they stacked up prior to any capital return because in the near term they won't be returning capital to shareholders," said Monteleone. "They've said they're going to submit a new plan. If we assume they're not allowed to return capital to shareholders when we get to this point next year Citi is going to be in a much stronger relative position."