Posted on CFO.com by Edward Teach:
On March 27, Omni National Bank of Atlanta was closed by federal regulators. The Office of the Comptroller of the Currency put the $956 million (in assets) bank in receivership because losses had depleted most of Omni National's capital, and there was "no reasonable prospect that the bank [would] become adequately capitalized without federal assistance," according to the Comptroller's office.
Omni National became the 21st U.S. bank to fail during the first three months of 2009. That number approaches the 25 bank failures for all of 2008 and is seven times 2007's total. More are sure to fail. In December, the Federal Deposit Insurance Corp. disclosed that its list of problem banks numbered 252, the most since the middle of 1995, with combined assets of $159 billion. The FDIC won't name the banks, for fear of triggering panics, but it's probable that the list is not comprehensive. Analysts point out that giant Washington Mutual ($307 billion in assets) wasn't on the problem list prior to its demise.
CFOs can take comfort that analysts expect relatively few of the nation's more than 8,200 FDIC-insured banks to go bust. "The vast majority of banks will be OK," says Scott Valentin, a managing director at FBR Capital Markets. Historically, only about 13% of the banks on the FDIC's problem list have failed. FDIC chairman Sheila Bair said in March that 98% of banks were well capitalized by regulatory standards.
Still, loan portfolios are deteriorating rapidly in the downturn, and some banks are going to become just another statistic. The FDIC forecasts that bank failures over the next five years will cost the agency $65 billion, on top of last year's $18 billion tab. Christopher Whalen, managing director and senior vice president of Institutional Risk Analytics (IRA), a bank-rating firm, predicts that some 100 U.S. banks with assets totaling more than $800 billion will fail in 2009. Other observers have been even more pessimistic.
CFOs appear braced for some bad news. According to CFO's Q4 2008 Business Outlook Survey, 72% of finance chiefs said they had "moderate" or "significant" concerns about the condition of the financial institutions they deal with.
Federal regulators are expected to announce some results of their stress tests for the nation's 19 largest banks in May. But companies can perform their own check of a bank's health the same way that analysts do, by reviewing the bank's key capital ratios and loan-performance trends. That information is easily accessible on the Internet, via Securities and Exchange Commission filings and the quarterly call reports that every FDIC-insured bank is required to file (see www.fdic.gov/quicklinks/analysts.html).
Evaluating a bank's health falls into two parts, says Mark J. Flannery, a finance professor at the University of Florida's Warrington College of Business Administration. One, how well capitalized is the bank — how much loss can it stand without failing? Two, what is the quality of its assets — how much loss risk is the bank exposed to?
In the FDIC's eyes, a well-capitalized bank has a ratio of Tier 1 capital to total risk-weighted assets of at least 6% (analysts prefer to see 8%); a ratio of total capital to total risk-weighted assets of at least 10%; and a Tier 1 leverage ratio of at least 5%. (Tier 1 capital includes common stock, some preferred stock, and retained earnings, among other things. The leverage ratio is Tier 1 capital divided by average total consolidated assets.) All three ratios can be found in Schedule RC-R (Regulatory Capital) of a bank's call report.
The trouble is, the risk-based capital ratios "don't work very well," says Frederick Cannon, chief equity strategist at Keefe Bruyette Woods, specialists in financial services. That's because the risk weightings that the government uses are out of date. For example, a mortgage-backed security is weighted at 20%, meaning that it requires one-fifth the capital of whole loans. "But some of those securities have declined in value a lot more than the values of whole loans," says Cannon. The option ARM, which "proved to be an absolutely horrible product in terms of performance," is weighted at 50%; "in hindsight it probably should have been weighted at 200%," he says. As for the leverage ratio, "it doesn't pay any attention to the composition of assets and their risk," says Flannery.
Many investors no longer trust the regulatory ratios. Shareholders, conscious that they will be the first to lose if a bank fails, are turning to the tangible common equity (TCE) ratio as a better measure of solvency. The TCE ratio, which isn't a GAAP metric, is tangible common equity divided by tangible assets; hybrid equity instruments and all intangibles are excluded. "It's a harsh measure," notes Whalen. There's no general rule of thumb for an adequate level of TCE, but many analysts like to see a ratio of at least 4% for large banks and 5% or 6% for regional banks.
An officially well-capitalized bank may have a dangerously thin TCE ratio. Take Citigroup. At the end of December, the $1.9 trillion (in assets) bank holding company had a Tier 1 ratio of 11.9%, a total capital ratio of 15.7%, and a leverage ratio of 6.1%. (Capital ratios for bank holding companies can be found at the National Information Center's Website, www.ffiec.gov/nicpubweb/nicweb/nichome.aspx.) But its TCE ratio was just 1.5%. Acknowledging the importance of TCE to investors, this past February Citigroup announced that it would offer to exchange up to $52.5 billion of its existing preferred stock for common stock, thus raising its TCE ratio to about 4%. "This securities exchange has one goal — to increase our tangible common equity," said CEO Vikram Pandit.
All About the "A"
But the TCE ratio is not infallible. Right before Washington Mutual failed, its TCE ratio was 7.8%. For regulators and analysts, TCE is one more metric in the tool kit. That tool kit is typically based on CAMELS, the supervisory rating system that looks at a bank's capital, asset quality, management, earnings, liquidity, and sensitivity to market risk (hence the acronym). Earnings are always important, but "these days it's more about the 'C' and the 'A,'" says Valentin.
The "A" is a growing source of discomfort as the recession drags on. With growth slowing and unemployment rising, a broad swath of consumer and business loans is beginning to sour. "Most of the banks that have failed to date have had significant early credit-cycle exposure — subprime, option-ARM, residential construction loans," says Cannon. "We're starting to see significant deterioration in midcycle credit: prime mortgages, home-equity loans, some nonresidential construction. And there's increasing concern about late-cycle credit instruments such as commercial real-estate mortgages and commercial loans."
It's prudent, therefore, to keep an eye on a bank's loan-loss reserves and nonperforming assets. "If you put a couple of quarters together and you see a trend increase in the loan-loss reserve and an increase in the amount of delinquencies, you start to get a picture of what is likely to happen to these assets in the near future," says the University of Florida's Flannery. "There are two categories [of delinquent loans] reported. One is 30 to 89 days late, and that's kind of noisy; it includes all the people who mailed their checks late. Ninety days–plus is a problem; you know you're going to have some sort of loss on that."
Banks are taking a beating on write-offs of delinquent loans. In February, credit-card defaults rose to their highest level in 20 years. IRA predicts that net charge-offs will peak in 2009. Total net charge-offs for the industry reached 2% of assets in 1990–91, the tail-end of the savings-and-loan crisis; this time around they will reach 3% or 4%, IRA predicts. "These charge-offs are already baked in," says Whalen. "They are the result of portfolio decisions made during the last three or four years."
How many banks will be done in by the deterioration of their loan books? Whalen agrees that most banks, particularly smaller ones, are sound. Still, while most community banks (typically banks with assets under $1 billion) weren't affected by subprime or Alt-A loans, they don't have a diversified geographic footprint and are "very exposed" to the local economy, points out Valentin. "Banks that have large construction portfolios are probably at risk," he says, particularly in states like California and Florida. Layoffs and unemployment will mean higher vacancy rates in commercial buildings, which will result in more defaults.
When All Else Fails
Despite all of the attention on banks' earnings the past year and a half, a flawless diagnostic tool for bank soundness has yet to be invented. "There's no single perfect measure of solvency," says Cannon. "Accounting statements are not a perfect measure of value. They are all shorthand for solvency. We spend a lot of time looking at balance sheets and we still haven't got it right on some banks, given how quickly things have deteriorated."
Scott Bugie, a managing director at Standard & Poor's, thinks capital ratios are unreliable. "Unfortunately, and much to the consternation of the global regulatory authorities, the correlation between creditworthiness and capital ratios is weak," he says. Regulatory capital ratios and the capital measures used by S&P "have by themselves been mediocre indicators of relative strength of credit," says Bugie. "If one institution has a 12% capital ratio and another has a 6% ratio, that doesn't necessarily mean that the latter is more likely to encounter problems or to default." Bugie says S&P is working on a new measure of bank capital, called the risk-adjusted capital framework, that it believes will correlate with a bank's creditworthiness better than its current ratios do.
Of course, there's another, forward-looking indicator of a bank's strength: what the market is willing to pay for a share of stock. As of April 1, the KBW Bank Index had fallen more than 60% in the past year, while the KBW Regional Banking Index had dropped more than 40%. Shares in Citigroup traded at just $2.68. While that is a big improvement over the stock's all-time low of 97 cents in March, it's a long way from a price that would indicate the hard times are over. For many banks, unfortunately, the worst may lie ahead.