Warren Buffett on Wells Fargo

Fortune: How is Wells Fargo unique?

Warren Buffett: It's sort of hard to imagine a business that large being unique. You'd think they'd need to be like any other bank by the time they got to that size. Those guys have gone their own way. That doesn't mean that everything they've done has been right. But they've never felt compelled to do anything because other banks were doing it, and that's how banks get in trouble, when they say, "Everybody else is doing it, why shouldn't I?"

What about all the smart analysts who think no big bank can survive in its present form, including Wells Fargo?

Almost 20 years ago they were saying the same thing. In the end banking is a very good business unless you do dumb things. You get your money extraordinarily cheap and you don't have to do dumb things. But periodically banks do it, and they do it as a flock, like international loans in the 80s. You don't have to be a rocket scientist when your raw material cost is less than 1-1/2%. So I know that you can have a model that works fine and Wells has come closer to doing that right than any other big bank by some margin. They get their money cheaper than anybody else. We're the low-cost producer at Geico in auto insurance among big companies. And when you're the low-cost producer - whether it's copper, or in banking - it's huge.

Then on top of that, they're smart on the asset size. They stayed out of most of the big trouble areas. Now, even if you're getting 20% down payments on houses, if the other guy did enough dumb things, the house prices can fall to where you get hurt some. But they were not out there doing option ARMs and all these crazy things. They're going to have plenty of credit losses. But they will have, after a couple of quarters of getting Wachovia the way want it, $40 billion of pre-provision income.

And they do not have all kinds of time bombs around. Wells will lose some money. There's no question about that. And they'll lose more than the normal amount of money. Now, if they were getting their money at a percentage point higher, that would be $10 billion of difference there. But they've got the secret to both growth, low-cost deposits and a lot of ancillary income coming in from their customer base.

So what is your metric for valuing a bank?

It's earnings on assets, as long as they're being achieved in a conservative way. But you can't say earnings on assets, because you'll get some guy who's taking all kinds of risks and will look terrific for a while. And you can have off-balance sheet stuff that contributes to earnings but doesn't show up in the assets denominator. So it has to be an intelligent view of the quality of the earnings on assets as well as the quantity of the earnings on assets. But if you're doing it in a sound way, that's what I look at.

Link to the Article