I'd like to propose a useful (and, I believe, superior) way to think about banks -- or, to be more specific, about how to identify a publicly traded bank that is worthy of a long-term investment.
An industry beset by bank swans
The typical approach is to cycle through a host of financial metrics with an eye toward estimating a bank's profitability in the, more or less, ordinary course of business. This is generally done by looking at metrics such as net interest margin, efficiency ratio, loan growth, credit quality, capital levels, and so on.
The idea is that a bank that is firing on all cylinders is likely to generate a return on assets that, assuming responsible allocation of capital (which can't be taken for granted), will generate a respectable return on investment in the quarters and years ahead.
The problem is that this approach ignores a fundamental reality about banks -- namely, that long-term returns in the bank industry are not dictated by what happens in the ordinary course of business. They are dictated instead by cataclysmic events that besiege the industry on an irregular, but not infrequent, basis.
The financial crisis of 2008-2009 serves as a case in point. From the beginning of 1993 through the first half of 2007, the KBW Bank Index, which tracks 24 of the nation's leading lenders, generated a total return of 328%. Unfortunately, it then lost 84% of its value over the next 20 months before bottoming out in the first quarter of 2009 at 33% below its 1993 level.
This was a textbook example of a "black swan" event -- relatively rare, extremely impactful, and retrospectively (though not prospectively) predictable. And it was only the latest in a long stream of cataclysms to obliterate bank earnings and the value of their stocks since the end of the Civil War in 1865.
This is a point Nassim Taleb addressed in his book The Black Swan: The Impact of the Highly Improbable, in which he observed that the same thing happened to banks in 1982 and again in the early 1990s:
[I]t took just one summer to figure out that [banking] was a sucker's business and that all their earnings came from a very risky game. All that while the bankers led everyone, especially themselves, into believing that they were "conservative." They are not conservative; just phenomenally skilled at self-deception by burying the possibility of a large, devastating loss under the rug.
Indeed, the bank industry's vulnerability to black swans stems from a pair of inescapable realities. First, because most banks are leveraged by a margin of 10 to one, it only takes a small drop in asset values to trigger insolvency. Second, because asset values are dictated in no small part by fluctuations of the credit cycle, they are virtually guaranteed to drop on that irregular, but not infrequent, basis that I noted earlier.
For these reasons, prudent and informed investors in bank stocks should care less about how a specific bank is performing at any one particular point in time, and more about how it will weather the next crisis.
The central importance of credit quality
Determining whether a bank is positioned to survive or even thrive in the next crisis is not hard, but it does require that you think differently about the traditional metrics that measure banks' performance. Specifically, every metric should be viewed with respect to its impact on credit quality.
The efficiency ratio is a great place to start. This measures the percentage of a bank's net revenue that is consumed by operating expenses. Suffice it to say that this is an important metric in and of itself, as a bank that spends most of its revenue on things like compensation and rent has little chance of generating a respectable rate of return for investors.
But beyond its direct impact on profitability, the efficiency ratio serves an even more critical, indirect purpose. Namely, by keeping operating costs low, a highly efficient bank can compete aggressively on loan prices -- i.e., interest rates -- without inordinately impairing its profitability. This allows a bank to attract creditworthy borrowers who are less likely to default during the next downturn in the credit cycle.
The evidence supporting the relationship between efficiency and loan losses is compelling if not dispositive. You can see this by plotting a sample of average long-term efficiency ratios at the nation's biggest banks against their nonperforming loan ratios over the same period. As the following chart reveals, there is an obvious fit between the two.
A second metric that should be approached in the same way is the percentage of net revenue a bank generates from noninterest, or fee-based, sources.
Noninterest income first gained prominence in the 1980s when high short-term interest rates rendered obsolete the then-traditional business model at most banks, which focused on interest rate arbitrage. This coincided with a dramatic increase in competition among lenders following the proliferation of money market funds, commercial paper, and the direct participation of institutional investors in the credit markets.
In addition to the fees themselves, good bankers have since gained an appreciation for the role that noninterest income can play in mitigating credit risk. This follows from the fact that a bank with considerable fee-based income does not have to rely exclusively on loan origination volumes to fuel profitability.
Mike Hagedorn, the president and CEO of UMB Bank, which is a subsidiary of UMB Financial, touched on this point in a recent interview with American Banker:
If you're competing for the highest quality credit, by definition it's going to be on the lower end of the pricing spectrum. [...] If that's what's important to you, then how do you supplement lower yields in your loan book? You do that with the diversity that fee businesses bring to your revenue streams.
Finally, it can't be denied that a bank's credit culture, evidenced most acutely by loan performance in previous downturns, shines a revealing light on that bank's inclination to become mired in the next boom and bust cycle. I say this because history is replete with examples such as Bank of America and Citigroup, which have been embroiled in, and often existentially so, multiple past swoons.
The recidivism rate among banks follows from the fact that a prudent credit policy involves both institutional culture and organizational structure, neither of which is easy to change.
It is cultural because prudence is a function of temperament, and temperament reflects the rare willingness on the part of bank executives and shareholders to forgo short-term gains in favor of returns that take longer to materialize but offer a more predictable payoff. And it is structural because it requires a greater allocation of authority to risk managers than it does to the sales force that is charged with originating loans.
The net result is that investors and analysts should assume a bank's performance in the last cycle will weigh heavily on how it will perform in the next one. There's no guarantee that this will be the case in any particular situation, of course, but when you're dealing with an industry prone to black swan events, it's best to act with an abundance of caution.
Coming to terms with black swans
When it comes to investing in banks -- or, for that matter, running them -- I've found that it's easy for people to be overconfident (if not outright cavalier) about their ability to recognize and therefore minimize risk. But in an industry beset by black swans, as the bank industry is, these feelings are inappropriate.
At the end of the day, the only intelligent way for long-term investors to succeed with bank stocks is to anticipate the next downturn by aligning themselves with lenders that have demonstrated histories of prudence and thrift, as wells as healthy streams of noninterest income. This won't guarantee the safety of your capital, but it will give you the best shot at generating returns over the long run that are proportionate to the risk.