Monday, November 20, 2017

Factor Investing: Factors Hated and Loved

From Validea's Guru Investor blog:

The Most Hated (And Most Loved) Investing Factor
Factor investing requires a lot of patience. Despite the fact that research shows that many factors can produce outperformance over long periods of time, all of them will struggle at times in the short-term. And those struggles are typically long and difficult enough that most investors will abandon underperforming strategies in favor of what is working now. When that happens, that typically signals a bottom for the factor is near.

At Validea, we track several hundred factors in our guru-based models that run the gamut from value to growth to momentum. Our historical testing shows that mean reversion in factors, particularly value factors, can be a very powerful force. The longer and more a factor is out of favor, the stronger its performance tends to be when things reverse.

So I thought it would be interesting to take a look across the factors we follow to see what is the cheapest right now.

Value stocks have been underperforming for a decade now, so it’s no surprise that nearly all the cheapest factors are value related. The cheapest one, however, also has the distinction of being the most loved factor, and the most unloved factor, all at the same time, which makes the discussion of its future prospects an interesting one.

The Price/Book ratio is probably the most commonly used factor in value. When Fama and French published their three factor model in the early 90s, they found that a low Price/Book ratio was positively correlated with future stock returns. That led to a huge uptick in money following it.

Price/Book has more money following it than any other factor. It is the primary factor Russell uses when it build its value indices, which have a lot of capital invested in them. It is also the value factor used by Dimensional Fund Advisors, which manages over $500 billion, for its funds.
That huge pool of money following Price/Book makes it the most loved value factor in terms of the assets following it.

With that amount of capital following the factor, you would expect its effectiveness to be reduced. And data indicates that may be exactly what has happened. Since January of 1995, the Russell 3000 Value Index has returned 9.99% annually, while the Russell 3000 Index has returned 10.12%. So adopting a value tilt using Price/Book has not produced the outperformance predicted in the academic research for the last 20+ years.

There are a couple of arguments as to why this has happened. First, as previously mentioned, when you put a huge amount of capital behind a factor, and when that capital tends to be permanent (sticking with the factor through ups and downs), that factor should lose some or all of its effectiveness. Second, share buybacks have become more common as time has gone by. When a company buys back shares, it reduces both its market capitalization (by reducing shares outstanding) and its book value (since either cash is subtracted to buy the shares or debt is added). The net result of this is a higher Price/Book ratio.  This can have the effect of making a company look less attractive from a valuation standpoint, even though it is engaging in behavior that is beneficial to shareholders.

This combination of significant permanent capital tied to the Price/Book and the accounting basis for the reduction in its effectiveness has led to a transition. The best and most thoughtful minds in the quantitative investment business now almost universally hate the factor. They tend to prefer more advanced ratios like Enterprise Value to Operating Earnings or even different more common metrics like the Price to Earnings ratio. And those factors have all worked better in recent years. So despite being the most loved value factor in terms of money following it, the Price/Book has become the most hated one in the active management community....MORE

HT: Alpha Ideas, Nov. 17