2003: If Black-Sholes is incorrect, how would Buffett price options?

AUDIENCE MEMBER: Good morning, gentleman. My name is Hugh Stephenson (PH). I’m a shareholder from Atlanta.

You had indicated in the past that you did not think that the volatility base to Black-Scholes models for options pricing was correct.

Would you share with us how you would evaluate those options as you use them in the business or see them in the marketplace?

And also if you would update us on your thoughts on the asbestos tort situation, given the recent development of national settlement trusts, et cetera?

WARREN BUFFETT: Yeah, we — Charlie and I have thought about options all our life. I mean, my guess is Charlie was thinking about that in grade school.

And — (laughter) — you know, and I — you have to understand— you don’t have to understand Black-Scholes at all — but you have to understand the utility and, in a general sense, the value of options. And you have to understand the cost of issuing options, which is very unpopular subject in certain quarters.

Any option has value. I mean, I bought a house in 1958 for $31,500. And let’s assume the seller of that house had said to me, “I’d like an option on it, good in perpetuity, at $200,000.” Well, that wouldn’t have seemed like it’d cost me much if I’d give it to him, but an option has value.

Any option has value, and that’s why some people who are, you know, kind of slick in business matters sometimes get options for very little or for nothing. I’m not talking about stock options. I’m just talking about an option to purchase anything.

They get options for far less than, really, a market value would be. Black-Scholes is an attempt to measure the market value of options, and it cranks in certain variables.

But the most important variable it cranks in that might be subject — well, might be a case where if you had differing views you could make some money — but it’s based upon the past volatility of the asset involved. And past volatilities are not the best judge of value.

I mean, if you had looked at a five-year option at — on Berkshire stock — at various times Berkshire stock’s had a fairly low beta, as they call it. Beta is a measure that — people in academia always like to give Greek names to things that are fairly simple, and so that they have sort of a priesthood. (Laughter)

You know, it’s — so it’s like priests talking in Latin or something. I mean, it kind of cows the laity.

But they — beta is a measure of past volatility. Berkshire’s had a low volatility, but that didn’t mean that the option value of it, to anybody that really understood the business, was lower than a stock with a higher beta.

And I think Charlie — what Charlie said is that — last year, is that for over — that for longer-term options in particular, Black-Scholes can give some silly results.

I mean, it misprices things, but it’s a mechanical system. And any mechanical system in securities markets is going to misprice things from time to time, and that’s —

We made one — as I mentioned last year — we made one large commitment that basically was — had somebody on the other side of it using Black-Scholes and using market prices — took the other side of it and we made $120 million last year.

And we love the idea of other people using mechanistic formulas to price things, because they may be right 99 times out of 100 but we don’t have to play those 99 times. We just play the one time when we have a differing view.

Charlie, do you want to comment on —?

CHARLIE MUNGER: Yeah, Black-Scholes is a — what I would call a know-nothing value system.

If you don’t know anything at all about value compared with price — in other words, if price is teaching you all that can be known — then Black-Scholes, on a very short-term basis, is a pretty good guess, you know, for what a 90-day option may be worth in some stock or another.

The minute you get into longer-term options, or you don’t have the know-nothing factor so extreme, it’s crazy to use Black-Scholes. People use it just because they want some kind of a mechanical system.

But at Costco, for instance, within a fairly short period, we issued stock options at 30, and we also issued stock options at 60. And Black-Scholes valued the options we issued at 60 as the strike price way higher than the options we issued at 30. Well, this is insane.

WARREN BUFFETT: But we like a certain amount of insanity. (Laughter)

CHARLIE MUNGER: Yeah, well, it’s good for Warren who picked up this extra $120 million. But —

WARREN BUFFETT: I mean —

CHARLIE MUNGER: — so he’s fonder of this kind of insanity than I am. (Laughter)

WARREN BUFFETT: No, we will pay you real money if you will deliver to our offices at Kiewit Plaza somebody who wants to use the Black-Scholes model and is willing to price 100 options for three years, willing to — using the Black-Scholes model — and letting us pick and choose among those.

Because, as Charlie says, it’s a know-nothing affair. And we are know-nothing guys, in respect to an awful lot of things, but every now and then we find something where we think we know something, and anybody that’s using a mechanistic formula is going to get in trouble in that situation.

But options have value. I mean, we issued options, in a sense, last year when we when we sold those — the 400 million of bonds. And we know what we’re giving up when we sell those bonds.

I mean, we may have gotten, what — a negative coupon of sorts, but that’s because we gave up option value. And it, you know, it wasn’t — it isn’t truly a negative cost instrument at all, because options have value.