Monday, September 10, 2012

How to Diversify Your Portfolio


Very few professional investors would advocate a portfolio of equities without insisting diversification. It is certainly a worthy aim, but what is puzzling is that so few seem to understand it or at least make the effort to think in non-conventional terms about. In my humble opinion the conventional and consensus opinion on the subject is plain wrong and investors with even a rudimentary understanding of the underlying issues can better generate diversified portfolios.

In this article I will briefly focus on two ideas.  The first is the concept of beta (derived from the modern portfolio theory beloved of unintentional index hugging fund managers) and the second relates to mechanical based investment systems. In turn, I’ll share a few thoughts on what investors might do to better achieve diversification.


Beta in Investing

The concept of beta is wonderful in principle. Theoretically all an investor has to do is put together a portfolio of stocks whose average beta works out to one and he/she has a portfolio thaqt approximates market risk.  Bingo he is diversified! Furthermore, when he wants more risk, he just buys more stocks with a beta more than one. When he wants less, he just buys stocks with the opposite property.

Unfortunately, it is not that easy. Beta is- by definition- based on historical data. It tells you what was a high or low beta across previous market conditions. I’m not saying that this isn’t useful. It is. If market conditions are the same going forward then the concept of beta is highly applicable. Alas, markets are not that compliant.

In the last 20 years or so, we have been through a technology boom, a housing market boom, a banking boom, a commodities boom, a banking bust, a housing market bust, a boom in emerging market bonds, boom in Gold, a China housing boom and so it goes on. The fact is that market conditions constantly change and deluding yourself that you can obviate the necessity to think about how macro-conditions are evolving by just relying on some backward looking data like beta, is a recipe for trouble.

Consider, a stock like Intel (NASDAQ: INTC) and then look at Caterpillar (NYSE: CAT). Yes, during the dotcom boom the former was likely to have been a high beta stock as the market priced in every ‘Blade Runner’ technological fantasy whilst Caterpillar would have been a lot less beta. After all, they just provide machinery to the boring housing, mining and construction industries. Fast forward a few years into the dotcom bust/low interest rate era and suddenly housing, mining and anything China are (literally) hot property and no one wants tech. Guess who is high beta now?

This sort of example illustrates that beta changes with market conditions and cannot be relied upon. In a brief aside, I will also note that fans of Soros’ reflexivity would also point out that self reinforcing feedback loops also create investment bubbles (i.e. high beta stocks or sectors) and they can come from origins such as sheer sentiment or regulatory changes. How anyone can think that these factors will be expressed in the historical beta of a stock is beyond me.

Ultimately, if the beta moves around with conditions then structuring a portfolio based on this approach will not lead to diversification because the portfolio will do the same.



Mechanical Based Strategies Lead to Unintentional Style or Sector Biases

Another instrument of self-delusion is the seductive idea that investors can achieve diversification by investing based on a set of mechanical metrics which attempt to capture some sort of attribute or other. Again, I am not completely decrying this idea because I use such metrics to quantify stocks and I think they are useful. What I am saying is that if they are applied without consideration of the macro-conditions or overall portfolio direction they will lead to an unintentional style or sector bias.

In plain English –I can use it in occasion, I promise- if a portfolio is constructed purely using a metric such as , say dividend yield or PE ratio or Price/Assets etc, it will end up manifesting an unintentional market view. To give you an example, consider that in 2008 the highest yielding stocks tended to be banks, insurance companies and house builders. Guess what happened next?  The dividend did not exactly provide a floor to the share price, especially when it had a tendency to disappear as earnings collapsed. Investors can look at the stocks like Citigroup (NYSE: C) as a classic example of this trap.

Another example of a pitfall with this approach can be seen if investors focus on only investing in say an earnings or cash-flow basis. This sort of approach has a tendency to immediately disqualify certain sectors like biotech or oil & gas exploration. No one invests in Vertex Pharmaceuticals (NASDAQ: VRTX) or other such stocks because their growth prospects are all about future cash flows and earnings. If you wait for them, you will be too late to capture the upside.

Putting these two examples together begs a key question. How can a portfolio be truly diversified if it overweights one sector and/or it ignores whole swathes of the market?



An Alternative Approach to Risk Management

I’m going to conclude this article by suggesting another approach. Namely, to try and actively diversify the profit drivers in a portfolio and/or select stocks that have upside drivers which are relatively non-correlated with the macro economy. An example of the former approach would be to say balance an oil services company (which you think is undervalued) with say a company that has a high proportion of its fixed costs in oil or energy. Another example would be to buy a fertilizer company and balance it out with a food producer.

In other words, what you are trying to do is pseudo-hedge away macro risks and create a diversified portfolio that isn’t over exposed to anyone sector of the economy. Of course such an approach requires a lot of forethought into stock selection, but hey, no one said investing is easy!

As for the non-aligned approach, I think special situations investors tend to cross over a lot into this camp. It is all about finding stocks with almost hidden upside potential.  I think an example of this now is Johnson & Johnson (NYSE: JNJ). It isn’t the sexiest stock out there but its growth prospects are mainly about execution. If it sorts out its production difficulties within consumer products, integrates the Synthes acquisition and successfully develops sales of its new drug products then the stock price can appreciate irrespective of the economy.

Investing is more of an art than a science and there is no reason for fund managers to try and bamboozle private investors with the idea that they have the secret formula for diversification. They don’t. In truth it is much more about understanding why stocks move in the way they do and trying to balance the overlying risk. There is no reason why private investors can’t do this just as well as institutional investors.

No comments:

Post a Comment