the paradox of diversification

I read an entry on Fred Wilson’s blog on The Power Of Diversification. I don’t disagree with anything in his column, or the earlier one he links to where he describes basic portfolio theory. But the concept of diversification has always puzzled me a bit. Taken to it’s conclusion, portfolio theory suggests that optimal investment is one that is extremely diversified across all investments – i.e. the market portfolio. That’s often the reason given to invest in market index funds (though optimally diversification should be across asset types that may not be reflected or inadequately reflected in market indices – e.g. bonds, real estate, precious metals, etc.). This is because diversification reduces the impact of company or investee specific risk. Going further, if one diversifies more broadly, you eliminate industry specific risk, geographic specific risk, and so on. Therefore, the optimal investment strategy to maximize your return for any given level of risk is to invest only in the market portfolio, then leverage or deleverage to meet your personal risk tolerance. I’m probably not explaining it all too well – try Wikipedia for a more detailed and better written explanation.

Diversification can be quite a powerful tool. In fact, in certain (rather limited) circumstances, it can even change two losing prospects into a winner, so long as you alternate between the two. Not really the subject of what I wanted to chat about today – you can read more about Parrondo’s Paradox in this NYT archive or on io9. Or just google it.

Anyway, the mention of paradox is fortuitous. I don’t necessarily know if it’s the right term for what I’m about to describe, or whether it really does precisely fit within the definition of paradox. Nonetheless, to me, it seems that advice on diversification seems somewhat paradoxical. And by that, I mean that if you follow the logical conclusion of the lesson taught by modern portfolio theory on diversification, then it doesn’t really make much sense for anyone to specialize say, for example, in early stage technology companies. The greater the departure from the market or efficient frontier portfolio, the less optimal the risk/return ratio. And yet, despite this, there are many, many specialists that take a narrow focus (despite their diversification amongst investee companies), just like Union Square Ventures, who often do quite well, even though portfolio theory suggests that all specialists are using sub-optimal strategies exposing them to more risk than they need to be exposed to for a given return.

I suppose the same could be said of  entrepreneurs who, very often, put all their eggs into one basket – and, to address a point made by Fred Wilson, this is even after they’ve succeeded and accumulated a great deal of wealth. Elon Musk may be a good example of that. I think (but don’t know with certainty) that most of his wealth is invested in two, and only two, highly risky and focused ventures (being Tesla Motors and SpaceX).

Moreover, it would seem to me that if everyone in the world pursued the optimal diversification strategy as suggested by portfolio theory, including, for example, all narrowly focused or industry specific venture capital funds, then the diversity of assets (and possibly asset classes) would, I think diminish. Sort of like the paradox of efficient markets, I suppose.

But who knows. I don’t pretend to be an expert in modern portfolio theory. Would be interested in hearing from those more knowledgeable in the area.