[FatOtt]

[ QUOTE ]

Great quote from Charlie Munger at the 2003 BRK AGM:

Black-Scholes is a know-nothing system. If you know nothing about value -- only price -- then Black-Scholes is a pretty good guess at what a 90-day option might be worth. But the minute you get into longer periods of time, it's crazy to get into Black-Scholes. For example, at Costco [where he is a board member] we issued stock options with strike prices of $30 and $60, and Black-Scholes valued the $60 ones higher. This is insane.

[/ QUOTE ]

I hate this quote and I was in the audience when he said it. With the Costco example, he's acting as if there was no change in the information environment between the time when those options were issued. If the strike=$30 and the strike=$60 options were issued on the same date, Black-Scholes would obviously value the $30 strike option higher (as would all option pricing formulas). The problem is that things aren't the same.

A trivial example:
On Jan. 1, COST issues options with a strike price equal to market price equal to $60.
On Jan. 2, after a 2-1 stock split, COST issues options with a strike price equal to market price equal to $30.

Most people (including Munger) would say that you'd have to receive twice as many options at the $30 strike to equal the value of the $60 options.

Maybe that's an overly trivial response, but his point is still stupid. He'd make the same argument for stock prices:
"I received a share of stock from COST when it was trading at $60 on Jan 15 and I received another share of stock on May 15 when it was trading at $30. Obviously those two stocks are worth the same amount - they represent the same degree of ownership in COST and the same right to future cash flows. Yet the accounting system would say they actually cost different amounts of money! How stupid." But is that really stupid? Shouldn't you value a particular security based on the information environment at the date of the transaction?

Another example:
You own a security whose payoff is dependent on the roll of two dice. You receive $1,000*the total points rolled. You receive one share of this security before a die is rolled - what's the value of that security at that point in time? It's $7,000. Then the first die is rolled - a 4 comes up. What's the value of the security? It's now $4,000+$3,500 = $7,500. But it's the same security - what kind of pricing system would value the exact same security at $7,000 and $7,500 simply because some time has passed?

Munger's obviously a smart guy, but he often takes shortcuts with his comments that do not reflect his understanding of a topic. That's fine for him - I trust that he understands the details of what he makes offhanded reference to. I feel bad for the people, though, that listen to his quips and use that as the sole basis for economic proclamations. (Not referring to you, Buffett, I've read your posts before and don't think you're guilty of this.)