Discount rates and opportunity cost are two huge important factors in investing. A while ago I shared my notes on what Buffett said at the 1993 Berkshire annual meeting about using government bonds as an appropriate discount rate. As I review more Berkshire meeting notes and DJCO meeting notes, I’ve come across a few other good references on the subject.
First, let’s review what Buffett said in 1993’s shareholder meeting (notes from June 1993 Outstanding Investors Digest):
"Buffett: And once you’ve estimated future cash inflows and outflows, what interest rate do you use to discount that number back to arrive at a present value? My own feeling is that the long-term government rate is probably the most appropriate figure for most assets. And when Charlie and I felt subjectively that interest rates were on the low side – we’d probably be less inclined to be willing to sign up for that long-term government rate. We might add a point or two just generally. But the logic would drive you to use the long-term government rate. If you do that, there is no difference in economic reality between a stock and a bond. The difference is that the bond may tell you what the future cash flows are going to be in the future – whereas with a stock, you have to estimate it. That’s a harder job, but it’s potentially a much more rewarding job. Logically, if you leave out psychic income,that should be the way you evaluate a farm, an apartment house or whatever. And in a general way, Charlie and I do that."
During the 1997 shareholder meeting, a shareholder followed up on the question of discount rate (notes from the August 1997 OID issue):
"Question: When you’re projecting cash flows at a company which is a prospective investment, why do you use the current interest rate of risk-free Treasury bills? Why don’t you use the sort of opportunity cost discount rate Charlie was referring to – maybe the 12% return on equity you noted corporation have averaged historically, maybe your 15% goal, maybe Coca-Cola’s return on equity – as a comparison?
Buffett: We use the risk-free rate merely to equate one item to another. In other words, we’re looking for whatever is the most attractive. In order to estimate the present value of anything, we’re going to use a number. And obviously, we can always buy government bonds. Therefore, that becomes the yardstick rate.
It doesn’t mean we want to buy government bonds. It doesn’t mean we want to buy government bonds if the best thing we can find only has a present value that works out to half a percent a year better than the government bond. But it’s the appropriate yardstick to use, in our view, to simply compare all kinds of investment opportunities across the spectrum: oil wells, farms, whatever it may be.
Now it gets to the degree of certainty, too. But it’s the yardstick rate … It serves as a constant throughout the valuation process.
Munger: Yeah. If you look at a corporate stock, it’s obvious you can buy any maturity of government bond you want. So one opportunity cost of buying a stock is to compare it to the bond. But you may find that in the case of half of the stocks in America, you either know so little about them or you’re so fearful or think so poorly of them that you’d rather have the government bond. So, on an opportunity cost basis, they’re taken out by that filter.
Now you start finding corporations where you like the stocks way better than government bonds. So you’ve got to compare them one against the other. And when you find the one that you regard as the best opportunity – now you’ve got one to buy. It’s a very simple idea. It uses nothing but the most elementary ideas of economics and game theory. And it’s a child’s play as a mental process. Granted, it’s hard to make these business appraisals. But the mental process is a cinch."
Another shareholder followed up on the question: Following up on that question – if you don’t adjust for risk by using higher discount rate, how do you adjust for risk – or do you?
"Buffett: Well, we adjust by simply trying to buy it at a big discount from the present value calculated using the risk-free interest rate. So if interest rates are 7% and we discount it back at 7%, which Charlie says I never do anyway – and he’s correct, then we’d require a substantial discount from that present value figure in order to warrant buying it."
During the 2016 DJCO shareholder meeting, a persistent groupie asked Munger three questions on the subject of discount rate:
"Question: How do you use the discount rate to calculate intrinsic value?
Munger: Obviously, it’s relevant what the return you get on your bonds is, that affects the value of other assets in the general climate. Obviously, your opportunity costs should cover your own investment decision making. If you happen to have a rich uncle who will sell you his business for 10 percent of what it’s worth, you don’t want to think about some other investment.
If the opportunity cost is so great, considering everything else, you should forget about it. And most people don’t pay enough attention to opportunity cost. Bridge players know about opportunity cost. Poker players know about opportunity cost. But in an MBA faculty members and other important people, they hardly know their ass from a plate of hot squash.
Question: When you try to arrive at a valuation number using the discount rate, does it mean that between the two rates (you’ll choose one over the other)?
Munger: We don’t use numeric formulas that way. We take into account a whole lot of factors. It’s a multi-factor thing. And there are tradeoffs between factors. It’s just like a bridge hand. You have to think of a lot of different things at once. There’s never going to be a formula that will make you rich just by going through some horrible process. If that were true, every mathematical nerd who gets A’s in algebra would be rich.
So you have to be comfortable thinking about a lot of things at once, and correctly thinking about a lot of things at once. And we don’t have a formula that will help you. And all that stuff is relevant. Opportunity cost of course is crucial. And of course the risk-free rate is a factor.
Questioner: Do you use the same rate for different types of businesses?
Charlie Munger (Trades, Portfolio): No, of course not. Different businesses get different treatments. They all are viewed in terms of value, and they’re weighed one against another. But a person will pay more for a good business than for a lousy one. We really don’t want any lousy businesses anymore. We used to make money betting on reinventing lousy businesses and kind of wringing money out of them, but that is a really painful, difficult way to make money, especially if you’re already rich. We don’t do much of it anymore. Sometimes we do it by accident, because one of our businesses turns lousy, and in that case it’s like dealing with your relatives you can’t get rid of. We deal with those as best we can, but we’re out looking for new ones."
After the meeting, this groupie asked Munger again about the discount rate. I was lucky enough to listen to their conversation from nearby. Below are the notes I jotted down (not the exact conversation):
Shareholder: I wanted to follow up on the subject of discount rate. So you use different discount rates for different businesses.
Munger: Of course. Each business has different characteristics and different risk profiles. You can't use the same discount rate for Microsoft as Coca-Cola. You have to consider all sorts of factors such as the predictability of the company, the risk profiles and etc. Obviously you'll use a higher discount rates for Microsoft. You are trying to get a mathematical formula. But there's no formula for it. It's a framework. Opportunity cost is a mental framework.
