A fair number of analysts are hawking bank stocks as the beneficiaries of a perfect storm of higher interest rates, deregulation and expectations that the Fed — with the departure of Gov. Dan Tarullo — will be more accommodative of shareholder payouts. This narrative was seemingly reinforced last week with the Fed’s announcement that all 34 banks subject to the Dodd-Frank Act Stress Test, or DFAST, would remain “well capitalized” in a severe recession. The DFAST is a prelude to the more important Comprehensive Capital Analysis and Review, or CCAR, which the Fed will announce later this week. The CCAR, run in tandem with DFAST, will determine whether those 34 banks have sufficiently strong capital plans to increase their capital distributions to shareholders.

Amid all the ebullience in the media over the DFAST results, I was somewhat surprised when I actually dug into the numbers. True, the ever unreliable risk-based ratios — which allow banks to shrink their denominators if their assets are deemed “safe” by regulatory standards — looked pretty good, even in the Fed’s most adverse economic scenario. However, the more reliable (in my view) leverage ratios were unimpressive. The leverage ratios do not allow banks to boost their regulatory capital by lowering the risk weights of their assets. They are simpler metrics that compare equity capital to non-risk-adjusted, average total assets. This year, the Fed used a supplementary ratio that also includes some off-balance-sheet risks. Using this more stringent standard, several banks, to wit, Morgan Stanley, Goldman Sachs and State Street, were hovering around 4% — hardly a robust capital cushion during distressed economic times. And these ratios are calculated before any increased capital distributions planned by these institutions. That is, if the Fed approves increases at these banks next week, their leverage ratios in a stressed scenario would be even weaker.

This is not to detract from the progress made by banks and their regulators in improving the capital strength of the banking system. Progress has been impressive. Regulators and the banks they regulate have a right to be proud. However, let’s not get ahead of ourselves. The memories of the financial crisis are still fresh in my mind (even if they are only a chimera to the much younger financial reporters and analysts who cover bank stocks these days). And here is what I remember: Citigroup — the sickest of the sick big banks requiring the biggest bailout — was one of the top dividend payers in 2007, distributing precious capital that taxpayers had to later back fill with Tarp investments. Regulators were saying (truthfully) well into 2008 that all the big banks were well capitalized under regulatory standards — standards that proved to be spectacularly wrong in gauging banks’ true solvency. A market in turmoil had zero confidence in the complex and opaque “risk-based standards” in place at the time, and instead looked to the leverage ratio as a better indicator of a bank’s solvency. Studies performed by the Basel Committee staff and others after the crisis showed that the leverage ratio did a much better job of predicting whether a bank would get into trouble during the crisis than the risk-based standards.

Regulators got it wrong in the run-up to 2008. Though regulatory rigor has improved post-crisis, it would be hubris for the Fed to believe that it has found a magic formula that will predict with accuracy how any of the big banks will truly perform in the next downturn. Yes, the assumptions around the Fed’s severely adverse scenario are transparent and dire — 10% unemployment, 50% drop in the stock market (though the drop in inflation to 0.5% seems a little weak). However, we have less transparency around how the Fed determines bank losses based on that scenario. For instance, in a stock market with a 50% drop, why does the Fed think securities losses will only be $5 billion? Moreover, in those kinds of market conditions, does the Fed really think large financial institutions will maintain access to funding if their leverage ratios show capital margins of only 4%? Markets react quickly, pulling funding at the first hint of potential insolvency. If the purpose of this exercise is to prove big banks have sufficient capital to keep functioning during a crisis without government support, I would say many of them still have some way to go.

Some marginal increase in capital distributions may be warranted. As a group, the banks’ solvency position has improved significantly. But I would advise extreme caution for payouts in excess of earnings. If JPMorgan Chase CEO Jamie Dimon’s definition of a financial crisis is accurate — something that happens every five to 10 years — then it won’t hurt banks, or their shareholders, to retain some portion of their earnings and keep building their capital positions as we approach the decennial of the 2008 debacle.

Sheila Bair

Sheila Bair

Sheila Bair is the former chair of the Federal Deposit Insurance Corp.

BankThink submission guidelines

Want to contribute to BankThink, American Banker's platform for informed opinion about the ideas, trends and events reshaping financial services? View our detailed submission criteria and instructions.