2000: What is the best way to calculate option costs?
AUDIENCE MEMBER: My name is Steve Check. I’m from Costa Mesa, California.
My question is regarding stock options. I’ve taken your suggestion and have been attempting to subtract stock option compensation from reported income when evaluating companies. When I read annual reports, I usually find companies estimating option costs using the Black-Scholes model.
However, the assumptions going into the Black-Scholes model seem quite different from company to company. These assumptions, of course, are what is used for risk-free interest rates — quote unquote, “risk-free” interest rates, expected option lives — even though options have stated lives, and expected volatility.
Help me out a little bit. What is the best way to calculate option costs? Do you think Black-Scholes is appropriate? If so, how should we normalize the assumptions?
And just one short follow-up: how can we possibly estimate future earnings for companies, when companies, such as even Microsoft last week, in response to a lower stock price, simply reissue a bunch of new options?
WARREN BUFFETT: Yeah, the — I can tell you, from some personal experience, that companies attempt to use the lowest figure they can, even though it doesn’t hit the income account.
So they like to make fairly short assumptions as to the life of the options, even though they’re granted on a ten-year basis. Because they’ll make certain assumptions about exercise date or forfeiture and so on.
I think the most appropriate way, when you’ve got a pattern, which you have at many companies, of what they do on options, is simply to make an educated guess as to the average option issuance that they’re going to incur, or they’re going to elect to do over time.
And, generally, what you really want to — if you were to be precise — you would try to figure out what they could’ve sold those options for in the open market. Because that’s the opportunity cost of giving them to the employees instead of selling the same option in the market.
I think you’ll find, generally, that if you take a value of about a third, for a ten-year option, if you take a value of about a third — obviously, it depends on dividend rate and volatility and a whole bunch of things — but about a third of the market value, strike price, at the time they’re issued, that’s the expectable cost.
We believe in using the expectable cost versus the actual cost. I mean, that is how we would look at it.
If we were issuing options at Berkshire, and we issued options on $100 million worth of stock a year, we would figure it was costing us, probably in our case, with no dividend, at least $35 million a year to issue those options.
And we would figure that if we gave people $35 million in some other form of result-oriented compensation, that it would be a wash. And that is not the way most managements, of course, figure. At least that’s my experience.
And we would figure we could use that 35 million in a more shareholder-oriented way and one where the employee (who) was productive would be sure of getting results, as opposed to having it be at the whims of the market.
And I think you’ll see a lot of option repricing. Everybody says they won’t reprice their options, until they do it. And, you’ll see that with a lot of schemes.
It would be interesting to see whether CONSICO is willing to bankrupt all the executives who made loans to buy the stock and had those loans guaranteed by the company.
And the company initially said they would enforce those loans. And we’ll see whether they do it. I would say, in many cases, they won’t. I don’t know what CONSICO will do.
But, a lot of things that are said in connection with executive option schemes and that sort of thing are what they’ll do if it works in their favor. And then they’ll do something else if it doesn’t work in their favor. And that’s not spelled out in the initial approval that’s granted.
Charlie, you have anything to add?
CHARLIE MUNGER: Well, Warren’s somewhat critical attitude is very understated compared to mine. (Laughter)
WARREN BUFFETT: We’re going to leave raisins out of this particular — (laughter) — analysis. Let’s go to area 2.
We do believe, incidentally, if a company is going to end up giving out 10 percent of the, company over a 10-year period or 15 percent on options, that is like buying an apartment house and letting the seller keep a 10 or 15 percent interest in the upside.
Or it’s like buying an oil field and giving somebody a 10 or 15 percent interest-free override. It changes the value of the property. Make no mistake about it.
It is a — it has a huge economic impact on the value of a property. And just go out and try and sell your house and say, “I want to keep 15 percent of the appreciation in it,” and ask the buyer whether he’s going to pay the same price for the house.
Options subtract value the moment they are granted. And, like I say, unless companies — some companies follow a practice of making a mega-grant every three or four or five years. A lot of them just issue a fairly constant amount annually. And you can figure out the cost.
And, you know, they don’t want to tell the shareholders there’s a cost. And that’s why they fought through Congress and everything else in order to prevent it from being the truth. But, you know, Galileo had that problem many years ago and finally won out. So maybe we will, too. (Laughs)