2002: Will Berkshire maintain and grow its float, at a reasonable cost, in the future?

AUDIENCE MEMBER: I’m John Bailey from Boston, Massachusetts and I hope I’m not asking you to repeat your insurance presentation, but I have a question about the growth of our float.

There’s an increasingly popular piece of analysis out there where people project the growth of float for a large number of years into the future in order to determine the value of our business here.

But I wanted to ask more fundamentally, the existing float that we have runs off annually at a pretty considerable rate.

In order to maintain that, we have to replace it through our operations.

And then going the next step, to achieve the growth, we have more than replace it.

And so I wanted to ask you to address the characteristics of the — maybe the non-GEICO insurance businesses — that should give us the confidence to expect large amounts of replacement and growth float, at reasonable costs, going forward?

WARREN BUFFETT: Yeah, in a sense, float is somewhat similar to being in the oil business. I mean, you know, every day, some goes out as you pay claims, and the question is, did you find more oil than you produced that day? And it’s very relevant.

It’s a good question to, you know, what is the permanence of the float? What is the cost of the float? What’s the likelihood of it growing? Could it actually run off?

As you saw up on the slide, we have $37 billion-plus of float. I think we have more float in our property-casualty business. A little bit of that float is in General Re’s life and health business, but very small.

So, basically you’re looking at property-casualty float when you look at that 37 billion.

I believe that’s more than any company has in the United States and it’s possible — I haven’t checked Swiss Re and Munich — but it’s even possible it’s larger than anybody in the world.

Now, if you go to 30th and Harney Street, here in Omaha, you’ll see National Indemnity building. It’s the same building that was there when we bought the company in 1967 from Jack Ringwalt, when it had, maybe, 12 million of float.

And I had no idea that that 12 million, or whatever the number was, would turn into 37 billion. I mean, sometimes I can’t really quite figure out how it happened.

But in any event, it did, and it’s — we don’t want people focusing on growth in our insurance business. I mean, we want them focusing on intelligent growth when we can do it at a GEICO or whatever it may be.

But I think it’s suicide, from a business standpoint, to tell a bunch of insurance managers to go out and grow a lot.

So, you can say, well, with that lack of push from the home office, you know, how is that 37 1/2 billion going to grow? And I would say, just as I would have said to you for the last, you know, 30-odd years, I don’t know.

But I think that — well, I can tell you this, that our float would have less natural runoff than the float from just about any company in the world.

I mean, we have a longer duration to our float because it arises from these retroactive contracts and from reinsurance, long-tail reinsurance, and that sort of thing.

So, our float has less natural erosion than any — just about any — that I know of in the world, but it erodes. It is a long-lived oil field, but it — we’re pumping it every day.

You know, if I had to bet my life on whether the float would be higher or lower three years from now, or five years from now, I would certainly bet it would be higher.

And it’s turned out, over the decades, that it’s grown at a very significant rate. But I don’t want to push anybody to do it. It grew at $1,800,000,000 the first quarter.

Now, there are a few special transactions in that, but we seem to attract special transactions.

There’s nothing more important to Berkshire than to — to have that float, at least, be maintained, but I would say grow. And it will grow, I think. And to have it be obtained at low cost.

That float did us no good last year at all. That float was — lost us a lot of money in the year 2001, because it cost us, I think, 12.8 percent. And we didn’t have a way to make money with 12.8 percent money.

We will make a lot of money if we can obtain a float at no cost, as we did in the first quarter.

The answer to your question is that without knowing any specifics that — without being able to promise you any specifics — you know, I think the float is more likely to grow than to erode.

I said last year at this meeting that there — you know, that the float of the American property-casualty business was 300-and-some billion, and I thought we were sneaking up on 10 percent of it.

I was corrected later on. Ferguson pointed out to me that — he sent me the figures. The float of the American property-casualty business is well over 400 billion.

But even at that, you know, we are 8 or 9 percent, or some figure like that, of the float of the whole country.

And, obviously, we can’t grow at the same percentage rate starting from that kind of a base as we could when we started back in 1967.

But I still think we can grow it.

Charlie?

CHARLIE MUNGER: Yeah, I think the questioner realizes that growing float at a good clip, with very low costs, is extremely difficult.

It is. It’s almost impossible. But we intend to do it anyway. (Laughter)

WARREN BUFFETT: See, of the two variables, though, that the most important thing to do is to focus at getting it at a very low cost. If we get $37 billion at no cost, or very low cost, you know, then if we don’t do — if we don’t make money with that, shame on us.

I mean, the troops have delivered and then it’s up to Charlie and me to figure out ways to use that money.

So the important thing is the cost of the float and not the size of the float, although, obviously, we would like it to grow and we will do what we can to make sure that happens.