All of the metrics that analysts like to cite about the health of the bank industry seem to suggest that banks are getting less risky with each passing day. Yet, it's this very logic that lulls bankers and investors into complacency and, by doing so, sows the seeds for future losses.

When it comes to banks, as with many other sectors, investors have a tendency to invert the concept of risk. When the economy is soaring and bank valuations are racing higher, investing in the sector is presumed to carry little risk. But once a panic sets in, causing valuations to plummet, investors race to the exits as if a previously latent risk had suddenly been unearthed.

The implication is that risk and the duration of smooth sailing, so to speak, are inversely related. That is, the longer the period of smooth sailing, the lower the risk. And vice versa.

But the problem with this mind-set is that it confuses the true relationship between risk and periods of prosperity. Indeed, far from reducing riskiness, the longer an economy carries on without signs of trouble, the higher stock valuations climb. And the higher stock valuations climb, the greater the chance that they will soon revert to more sustainable levels.

Carl Richards, author of The Behavior Gap: Simple Ways to Stop Doing Dumb Things with Money, refers to this as "risk creep," recently illustrating the point with an analogy to skiing (emphasis added):

The day after Christmas, a skier was caught in an avalanche near my home in Park City, Utah. Part of a foursome, the skier slid with the snow for about 500 feet before ending up buried to his neck. Luckily, his party found him, dug him out, and discovered he hadn't been injured.

The group knew the area well. Based on the conditions and their experience, they had planned to avoid the steeper, more dangerous section of the slope. Then, things changed.

After skiing another section of the slope four times, they slowly moved into more dangerous terrain. Skiing so close to potential danger without any issues gave them a false sense of security. I've been in this position before, too.

You tell yourself a story. Everything has been fine so far. Things must be safe. These long periods of safety sow the seeds of future failure as risk creeps in incrementally. It's so subtle we rarely notice it.

If you follow the dialogue on banking, it's hard to deny that Richards' point applies with equal force to lending. Namely, as the financial crisis recedes into the background, the collective opinion of bank analysts and financial commentators is increasingly coalescing around the idea that banks have cleaned up their acts.

In their defense, it can't be denied that the data lends itself to this conclusion. Capital levels are at their highest point in many decades. Loan losses are down. And profitability is starting to normalize -- that is, if there is such a thing as normal profits in the bank industry. The implication is that banks no longer expose investors to the same degree of risk that they did during and immediately following the crisis.

Richard Davis, the chairman and CEO of U.S. Bancorp (NYSE:USB), even went so far as to explicitly say that credit quality isn't on most bank analysts' radar right now -- though, to be clear, you can rest assured that it's on Davis'. Here he is at the Barclays 2014 Global Financial Services Conference last September:

Credit quality, whatever. Right? I mean, credit quality is not going to be interesting to you guys for five years. It's good to continue to be steady to where it is. In our case, credit quality is very steady. Each quarter has been about the same. This quarter it will be a little bit better than the last quarter, but it can be pretty much flat, and for all intents and purposes, not a lot is changing.

What's important to appreciate, however, is that the data which fuels apathy toward credit risk when times are good is largely, if not entirely, irrelevant. In fact, because of its tendency to lull analysts and investors into complacency, one could even argue that it's worse than irrelevant -- that it's injurious.

This follows from the fact that long-term profitability in the bank industry isn't a function of how much money a bank makes during the good times. It's instead a function of how much money it doesn't lose when the credit cycle contracts. It's here that the cycle exacts its revenge on investors and institutions who fooled themselves into thinking that current credit metrics mattered.

As Nassim Taleb intimated in his book, The Black Swan: The Impact of the Highly Improbable, this is a defining trait of the American bank industry:

If [bankers] look conservative, it is because their loans go bust on rare, very rare, occasions. There is no way to gauge the effectiveness of their lending activity by observing it over a day, a week, a month, or... even a century! In the summer of 1982, large American banks lost close to all their past earnings (cumulatively), about everything they ever made in the history of American banking -- everything. They had been lending to South and Central American countries that all defaulted at the same time -- 'an event of an exceptional nature.' So it took just one summer to figure out that this was a sucker's business and that all their earnings came from a very risky game.

And for the record, the very same thing happened in the financial crisis of 2008-2009. Just take one glance at the apocalyptic performance of the KBW Bank Index during the past two decades. From the beginning of 1995 to the second quarter of 2007, it increased more than fourfold, going from 25 all the way up to 117. However, it then proceeded to plummet, falling to 24 by the first quarter of 2009.

My point is this: Somewhere along the way, we've inverted the concept of risk. This has led us to think that banks are not risky when they really are, and that banks are risky when they really are not. It's a textbook example of Richards' concept of risk creep, and it's one that enterprising investors would be wise to always keep at the forefront of their minds.