I thought I’d sketch out some of the basic ideas underpinning the factor-based approach, and Columbia’s Andrew Ang offers an eloquent starting point:
“Factors are to assets what nutrients are to food…Factors are the nutrients of the financial world. Factor theory is based on the principle that factor risk must be compensated and factors are the driving force behind risk premiums…This is the modern theory of asset pricing: assets have returns, but these returns reflect the underlying factors behind those assets.”
The reason factors are so important is that traditional asset classes will generally have exposures to the same underlying risk factor. In other words, an asset allocation based on equities, bonds, and alternatives may ultimately be providing very little diversification in terms of the underlying factors that drive returns. So, by focusing on the factors, an investor can better grasp what asset classes will provide the desired risk exposures. As Ang notes, “Factors allow a more holistic view of the investment and business activities of a fund.”
During the credit crunch, the “asset-oriented” approach didn’t provide the amount of diversification that the funds had expected/hoped. All the assets in their portfolios seemed to be moving in the same direction: down. For example, take CalPERS, which is one of the most diversified investors in the world. It lost $100 billion in roughly 18 months — the fund was worth $260 billion in October 2007 and touched $160 billion in March 2009.
As you can imagine, then, CalPERS is one of the funds leading the charge into factors. As I see it, this new approach simply provides the funds with more information about their risk exposures, and that has to be a very good thing.
As for CalPERS, its new “Alternative Asset Classification” will take effect in July 2011 and have the following components: Income: 15.9%; Growth: 63.1%; Real: 13.0%; Inflation-Linked: 4.0%; and Liquidity: 4.0%. And what does all that mean in “traditional” terms? Here’s a translation (click to enlarge):
This post originally appeared on the Oxford SWF Project website.