Factor Exposure and Fundamental Equity Management

A new report by S&P Dow Jones holds some figures that are thoroughly disturbing but perhaps not altogether surprising¹. The study suggests in the ten year period ending December 2015 82% of large-cap managers, 87% of mid-cap managers and 88% of small-cap managers have underperformed their respective benchmarks. While these findings represent a ringing indictment of active equity management, we can’t say we weren’t warned.

It is well known that most active managers destroy value with papers by Sharpe (1991) and Malkiel (2005), to name just a few, giving us sufficient advanced notice of the impending decade-long dry spell. However, why these managers underperformed is not well understood or, at the very least, not well appreciated. Carhart (1997) was among the first to suggest active returns are derived from two sources: exposure to systematic factors like size, value and momentum and from exposure to idiosyncratic risk, often termed ‘stock picking’. His study examines outperforming mutual funds over a 20 year period and concludes all of the excess returns of these funds were the result of factor exposure while returns to ‘stock picking’ were, on average, strongly negative. In other words, the only truly reliable and persistent form of excess return was factor exposure. While this conclusion has been well supported in subsequent studies such as Grinblatt, et. al. (2016), it is not the whole story.

There seems to be a pervasive notion that fundamental ‘stock picking’ and factor exposure are somehow independent - that one can exist without the other. This notion was, no doubt, facilitated by the way in which Carhart and other researchers conducted their studies. To these authors ‘stock picking’ was anything that was left over after factors were accounted. The independence of ‘stock picking’ and factors was a by-product of their method of analysis, or, more appropriately, a misunderstanding of their analysis, and not on fact.

In truth, must fundamental ‘stock pickers’ take very significant exposure to factors. The problem is we would consider much of this exposure disadvantageous or, in other words, pointing in the wrong direction. For example, factors like low volatility, high quality and high dividend yield, have all been shown to outperform historically. To their detriment, fundamental ‘stock pickers’ tend to have factor exposure with the wrong orientation, often tilting towards stocks with higher volatility, lower quality and lower dividend yield.

These disadvantageous tilts are not necessarily intentional but are the result of a search for idiosyncratic risk, the raw material of ‘stock picking’. The table below shows the average per-stock idiosyncratic risk for factors in the Russell 3000 universe. In their search for high idiosyncratic risk ‘stock pickers’ would naturally gravitate towards high volatility, low quality, low dividend yielding names.