Fool's Eye View

[ October 30, 2000 ]

Diversification

By Alan Oscroft (TMFAlan)

A version of this article first appeared in January.

Bournemouth, Dorset -- A number of things have recently got me thinking about diversification. It all started with the attention that Index Tracker funds have been attracting in recent months and the effects that they might be having on share price movements within the tracked indices.

Rob Davies (TMF Essex) wrote a great piece on the subject of Index trackers, the title of which I plagiarised unashamedly in one of my attempts to debunk a subsequent Wise criticism of said funds.

And James Carlisle (TMF JimmyC) pondered the whole issue of diversification from the point of view of minimising risk, and based on individuals' expectations and spending patterns. A fascinating topic, I think, though purely from an intellectual point of view. But where should practice meet theory?

The idea of diversification seems to contain a lot of common sense; if we spread our investments across different companies in different industries, then we are spreading our risk. If our investment in OnlinePants.com turns out to be a dog, then we will hopefully be saved by the money that we have in house builders, or banks, or whatever.

But I've always had a niggling doubt about the whole idea of diversification, and it is based on my deep rooted suspicion of everything that emanates from the mouths of the Wise. For it is those same doyens of pointless academia that have been spouting all that stuff about portfolio theory and perfect markets over the decades, and nobody ever got rich listening to that drivel.

No, successful investment is all about finding great companies with great prospects, and holding them for decades. We all know that, of course, so why do we diversify into companies that we just don't feel so strongly about? Basically, we feel the need to diversify our investment portfolios simply to cover the possibility that we got it wrong. So is it all down to confidence? Yes, it is.

So, here's an unqualified and, no doubt, insulting suggestion...

Diversification is for wimps.

But let me hastily qualify it. To an extent, we are all wimps. Every time any of us has doubts about a chosen investment and feels less than 100% confident in our company's chances of success, we are being wimps. To a degree. So what do we do? We go and stick some of our money into a safer investment too, a less risky one. Just in case. And it makes us feel better.

But isn't that just an illusion? Instead of investing half the amount in a company that we like less than our chosen star, albeit a more boring, less risky company, shouldn't we pay more attention to making the right first choice in the first place? Shouldn't we only ever be investing in companies that we are convinced by, whose fundamentals we understand, and whose risks we are confident we have assessed accurately?

That, then, is my first conclusion. The "correct" degree of diversification for any individual investor depends on that investor's degree of confidence in his or her choice of companies. That's all. The more convinced we are of the long term prospects for our companies, the less we should diversify. And don't forget, ultimately the only thing that diversification guarantees is mediocrity. If you buy more than about 10 companies, particularly if they are diversified across sectors, you will be very hard pushed to beat a simple index tracker. And you will be spending a lot of time failing to beat it too.

My second conclusion is that real Fools, those whose investment horizon is measured in decades, are actually getting diversification into the bargain too. We might only focus on a small number of companies that we have great confidence in, but over time, we will achieve profitable diversification.

My background, before becoming a Fool, was in telecommunications, and I can't help but see a similarity between diversification in investment and multiplexing in telecommunications. Multiplexing is the name given to a number of methods of shoving multiple connections down the same bit of wire, and the two most common approaches are Time Division Multiplexing, in which the different connections get specific time slots (usually measured in milliseconds) during which to do their stuff. The alternative is known as Frequency Division Multiplexing, and each connection gets a different frequency on which to communicate, all of them being sent simultaneously.

Conventional diversification is a bit like Frequency Division Multiplexing really, and I think it is useful to think of it as a kind of Cash Division Multiplexing -- we divide our cash across multiple simultaneous investments.

But if we focus on a smaller number of investments, and our investment horizon is long enough, we will achieve a form of Time Division Multiplexing instead. We may get a poorly performing share from time to time, but we will dump it as soon as the fundamentals change, and will compensate for it in our next investment. We will only ever invest in companies that we feel very confident of, and will therefore not have to make compromise investments "just to be safe".

So those are my thoughts then. If we don't feel that confident about companies we invest in, or we really can't handle short term volatility, then we should perhaps go for the traditional Cash Division Multiplexing approach to diversification. But if we really do feel confident about the companies we buy, perhaps we should forget traditional diversification and instead rely on Time Division Multiplexing. If it achieves nothing else, it will make us think a lot more carefully about our investments.

What say ye, Fools?

Thoughts, please, on the Fool's Eye View message board