The banking industry makes me ill.

The more you learn about banks, the less you realize you know about them. After covering the industry for several years, I've gotten used to muttering the phrase, Are you serious? I've realized that I fail to comprehend the complexities of banks' balance sheets -- and so do the people running them. For better or worse, I'm turned off by the industry and won't hear otherwise.

But I realize that's an extreme position. And probably an unhealthy one. So to force myself out of the fingers-in-my-ears-lalala objection to anything positive said about banks, I decided to find three areas where things look like they're headed in the right direction.

1. Asset quality is generally improving
With third-quarter earnings reported for the four big commercial banks -- Bank of America (NYSE: BAC), Citigroup (NYSE: C), JPMorgan Chase (NYSE: JPM), and Wells Fargo (NYSE: WFC) -- here's how broad loan quality is holding up:


 

Source: Capital IQ, a division of Standard & Poor's.

Wells Fargo aside (I'm convinced it's in perpetual misery), the trend in asset quality has been clear for the past year: It's getting better, for several reasons. One, employment has improved, albeit slowly. Two, a lot of the really atrocious loans have already been purged and written off. Three, banks are doing a better job resuscitating loans after they fall into delinquency.

To that last point, here's the Federal Reserve last month: "Transitions from early (30-60 days) into serious (90 days or more) delinquency improved sharply in 2010Q2, falling from 39% to 33%, the lowest rate of deterioration since 2008Q2." That "cure rate" is still high, but it's headed in the right direction -- encouraging, because it had been a major factor in bank losses over the recent past.

2. Capital is thick
Banks need capital to withstand losses. Right now, they have a ton:


 

The black line in this chart denotes the recent Basel III international banking standards minimum Tier 1 capital standard. Every major bank is well above the threshold, often by a factor of two. Banks can stand to lose a ton of money right now. More than they have in years.

3. Valuations look reasonable
As fellow Fool Anand Chokkavelu recently wrote, "I get interested in bank stocks that have price-to-tangible-book values below 1.5. Bank of America sits at 0.9, and JPMorgan sits at 1.3." Additionally, all major banks trade at less than 10 times forward earnings.

Two other factors could make current valuations look cheap. First, banks set aside so much cash for bad loans during the recession that now, with business not as bad as expected, reserves are being released and counted as net income. As The Wall Street Journal wrote last week:

There are 18 commercial banks in the U.S. with at least $50 billion in assets, and together they earned an adjusted $16.8 

billion in the third quarter. Of those profits, nearly half, or 48%, were from drawing down what bankers call loan-loss reserves, according to an analysis by Dow Jones Newswires. A year ago, the same 18 banks earned $6.2 billion in quarterly profits; at that time, they added more than $7.8 billion to the same reserves, a move that reduced their profits.

This is, of course, transitory profit, and I hesitate to even call it that. It's a pro-cyclical accounting maneuver that exacerbates booms and busts. But if we're at the beginning of a new boom -- or just a recovery -- profitability could enjoy this cyclical accounting boost for several more years. Enjoy it while it lasts.

Second, most banks still don't pay significant dividends -- a protective carryover from the 2008 bust. With capital levels high, loan quality improving, and major bank reform codified, that anti-dividend policy could change fairly soon. When it does, investors will bid valuations up accordingly. Rationally or not, there's a premium on yield these days.

Again, I generally believe that most investors should avoid banks' common stocks. They're like cocaine: feels good until it doesn't. Instead, I recently bought some B of A high-yield preferred stock, which is a completely different beast and, as I view it, a much better risk-reward bet. When these banks get into trouble, the U.S. Treasury will respond (as history shows) by injecting cash and diluting common shareholders into oblivion, and then protecting that diluted stake as taxpayers become owners. That leaves a decent buffer around for preferred shareholders and bondholders. Call it a safe wager "on too big to fail." If tempted, I'd suggest this as a more sensible way to invest in banks.

Disagree? Let me know in the comments section below.

Check back every Tuesday and Friday for Morgan Housel's columns on finance and economics.