Facing the full weight of expectation

In the first instalment of this series we explained that value investing is about buying things for less than they’re worth. And in the second we saw that stocks and shares are worth the value, in terms of today’s dollars, of the cash that they will generate in future. But the trouble with shares is that we can’t possibly know exactly how much cash they’ll produce in future, and that puts something of a spanner in the works.

In many cases, in fact, it’ll destroy the works completely; if we decide we can’t make a decent stab at a valuation, then we’ll simply declare the company beyond our circle of competence and move on to the next potential opportunity. And there’s absolutely no shame in that (indeed it’s good for the soul if you can take a smug sort of pride from it—we’ll look more closely at the pig-headed contrarianism that passes for the ‘psychology’ of value investing later in the series).

If you decide, however, that you can have a go at estimating a company’s value, then the next step is to do just that. And, rather than painstakingly discounting and then adding together every one of a company’s expected future cash flows, there are a variety of short cuts to help you: such as asset value, dividend yield and PER. We’ll examine these in more detail in coming issues, but for the moment we’ll deal with one crucial, but often overlooked, factor: what is it that we actually mean by our best estimate?

Don’t underestimate the Don

The two main ways that people look at this are to think of the middle point of a range of possible outcomes (the ‘median’ or ‘central case’), or the average of the possible outcomes (the ‘mean’, ‘expectation’ or ‘expected value’). For many situations, the central case is very close to the average (because the range of possible outcomes is evenly spread about the central case). If you line up a group of people by height, for example, the person in the middle is likely to be very close to the average height. But thinking like this can get you into trouble in some situations.

Take Don Bradman’s test batting record, for example. His central-case score (between dismissals, so that not outs are accounted for) was 67. But Douglas Jardine would have been wrong to bank on him making only 67 before being out—because when the Don scored big, he very often scored BIG, and that put his average up to the famous 99.9. (Jardine’s development of the infamous ‘bodyline’ tactic would suggest that he didn’t, in fact, make this mistake.)

So the expected value of a Bradman knock would be 99.9, even though that’s some way above the most likely outcome. And you get the same thing with companies. We backed Forest Enterprises Australia in issue 98/Mar 02 (Speculative Buy—$0.12) for this very reason, noting that ‘there’s a chance this company could fall over. But if it doesn’t, there should be some very big gains to be made.’ We’re pleased to say that it didn’t go broke and we were able to bank a nice profit in issue 150/Apr 04 (Sell—$0.35), at much closer to our expected value at the time.

Margin of safety

So just because you might expect 99.9 runs from the Don, that doesn’t mean you’d put your house on it. After all, he might get an unplayable delivery from Harold Larwood aimed at his head. So you’d want a decent margin of safety. And it’s very much a case of the more the merrier: you’d feel pretty good about backing Bradman to make 50, but you’d feel better about backing him to make 20.

This was the thinking behind our recommendation of SecureNet in issue 101/Apr 02 (Buy—$0.85) . We didn’t know much about the company’s technology, but we did know that its shares were trading at only about two-thirds of their net cash backing.

Even with a fat margin of safety, though, you wouldn’t want to go making too big a bet on a single opportunity. Even the great Don Bradman made seven test match ducks. So you need to spread your bets over a number of different opportunities. With shares this is known as diversification and we go into more detail on the subject in Too much of a good thing .

Of course diversification and margin of safety interplay with one another, because the more you’ve got of one, the less you might be prepared to live with of the other. But there’s a crucial difference. As you increase diversification, you reduce the quality of your selections and thereby reduce your expected return (that is, the difference between what you pay for a stock and its expected value). But as you increase your margin of safety, you actually increase your expected return, because your expected value is that much higher than the price you’re paying for the shares.

Right or wrong

The flip side of this is that margin of safety relies on you making correct assessments of value. Diversification, on the other hand, will tend to take you towards an average return whether you’re tending to get it right or wrong.

So if you’re very confident in your assessment of value, you might focus on finding stocks where you see a huge margin of safety and not worry so much if you end up holding only a few of them. But if you’re less sure about assessing the value correctly, then you’d want to focus more on achieving a decent diversification, with the inevitable reduction in margin of safety from your additional selections.

So although shares are difficult to value accurately, we can deal with that situation by sticking to our circle of competence, making a decent estimate of their expected value, allowing a margin of safety and hedging our bets. Over the next few issues we’ll look at some of the tools we can use to help us estimate value.