My last post discussed how many shares one should hold in a relatively abstract way. Another angle is to ask how many shares it is practical to follow.
There is clearly some sort of trade-off between number of shares held and your quality of knowledge. Many experienced private investors are surprised (and often derisory) about the fact that that I hold around 50 shares. Surely I must know far less about many of them than if I held only 10 or 15 shares?
Well, perhaps. But this is diamonds versus flower bulbs again. If I held 10 or 15 shares, I would want them all to be diamonds. I find diamonds very hard to recognise, and most of the time, I can’t find 10 or 15 of them. A lot of the time, I can’t find any of them.
Flower bulbs much easier to find. At most times I can find some shares on P/E ratios around 5, shares trading below net cash, shares where there’s been a failed bid but the bidder looks likely to return, or shares with some other idiosyncratic cheapness. The quality of these businesses is often low (they aren’t diamonds). But there are plentiful, and easy to recognise.
My portfolio of 50 flower bulbs comprises businesses of lower average quality than if I held 10 diamonds. But the certainty of my insight about the 50 flower bulbs is relatively high. This is because my insights about flower bulbs tend to be (a) simple and (b) short-term. For example: “it’s trading 25% below net cash, and something is likely to happen to make it trade at net cash.”
Insights about diamonds tend to be more complex, long-term and nebulous. “It has long-term comparative advantage and it’s going to be a bigger business in 5, 10 and 20 years’ time.” So you say. But how do you know? Generally, I don’t.
Or to put it another way: to identify diamonds, you need long-term foresight about their durability – market position, susceptibility to technological changes, etc etc. I’ve always found this very difficult. But you don’t need long-term foresight to identify flower bulbs.
A final point in favour of diversification is that even if a very few shares is the optimal policy in theory, many highly concentrated investors eventually come a cropper in practice.