If you were trying to create a system for controlling investment risk in equity investing, how would you do it?  What I would do is look at the factors that are the least positively correlated in terms of return generation, and focus on them.

But what do investment managements consultants do?  They divide the world up into managers that look at two factors: large/mid/small capitalization, and value/core/growth.  This has been popularized by the Morningstar “Style Box.”

Looking over the last 15 years, the style box is very correlated with itself.  The lowest correlation is 75%, between largecap value and smallcap growth.  That is not a reason to categorize managers; the difference between the average largecap value and growth manger is teensy. It is even true between largecap value and smallcap growth.  And in more recent years, the correlations have been tightening to nearly 90% at worst.

So, consider country allocations.  Over the last 15 years, the correlations in developed markets have been 45% at worst, with the average being near 70%.  Looking at the last few years, both figures are higher.  My opinion: the advent of naive quantitative investing has pushed all correlations higher.

But now consider correlations across economic sectors.  Over the past 14 years, the correlations have been 32% at worst.  Across industries, which are more diverse than sectors, some of the correlations are negative, perhaps affording true diversification.

My point here is that those that look at capitalization size and value/growth are missing the boat.  If you classify managers based on that, you are focusing on minor concerns that do not aid much in diversification.  Better to focus on the industries that a manager invests in, and/or the countries that those companies are located in — there is a real oportunity to limit risks through either of those two methods.

Now, as for me, when I pick stocks, I start with the industry.  I ignore the factors in the style box.  I look for industries that are near the bottom of their pricing cycle, and buy the highest quality companies there.  For industries that are doing well, but are undervalued, I buy companies with undervalued growth prospects, with good quality balance sheets.

I strongly believe that the investment consultant community has shortchanged its clients by focusing on the “style box.”  Very little of the risks of the market result from factors in the style box, while much resluts from industry selection, which is a richer model.

So, as for me, if I have to be squeezed into the style box, call me midcap value with some style drift, buying companies larger and smaller, and outside the US, as conditions dictate.  I’m looking for the best value over the next three years, and I don’t like non-economic factors distracting me.  Why should that be such a crime, that the ignorant gatekeepers screen me out?

The risk model for the investment consultants is broken.  Let them find one that better reflects the way that the market works.

This entry was posted on Tuesday, November 4th, 2008 at 1:46 am and is filed under Portfolio Management, Quantitative Methods, Speculation, Stocks, Value Investing. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.