Factor Investing
Unless you like to open the occasional dusty academic tome, chances are you're not intimately familiar with factor investing. It's really not as esoteric as it sounds. You've heard of style investing--small cap versus large cap, or value versus growth. If you've ever tilted to a particular style, you've engaged in factor investing. Style investing is a kind of factor investing, dealing with only two factors: size (large-small) and value (value-growth).
A working definition of a factor is an attribute of an asset that both explains and produces excess returns. Factor investing can be thought of as buying these return-generating attributes rather than buying asset classes or picking stocks.
None of this is new.
The original factor theory, dating back to the 1960s, is the capital asset pricing model, or CAPM, which predicts that the only determinant of an asset's expected return is how strongly its returns move (or, in technical terms, covary) with the market's. The strength of the relationship is summarized in a variable called beta. A beta of 1 indicates that for each percentage point the market moves, an asset's price moves in the same direction by a percentage point. CAPM predicts asset returns are linearly related to market beta. However, since the 1970s, academics have known that stock returns don't seem to be related to beta. This finding spurred many fruitless or convoluted attempts to explain how market efficiency could be squared with a world in which CAPM didn't work.
Eugene Fama and Kenneth French "fixed" the CAPM, at least for stocks, by adding two factors: size and value. They observed that smaller stocks outperformed larger stocks and stocks with high book/market outperformed stocks with low book/market. More importantly, the relationships were smooth; the smaller or more value-laden the stock, the higher its return. Fama and French interpret the smoothness of the relationship as indicating the market is rationally "pricing" these attributes, which implies that size and value strategies enjoy higher expected returns for being riskier.
Further research has uncovered more stock factors, including momentum, quality, and low volatility, in nearly every equity market studied. They also display the same smooth relationship: The stronger the factor attribute, the higher the excess returns. The interpretation of these factors depends on whether you believe the market is efficient. In an efficient market, they must be connected to risk. However, if the market is not perfectly rational, some may represent quantitative strategies that exploit mispricings to produce excess returns.
I don't believe value, quality, momentum, and low-volatility strategies work because they are riskier. The strategies were exploited by investors before academics triumphantly published them in journals as "discoveries." It's also hard to reconcile them all as representing risk because if you lump them all together, you get an eerily smooth return stream.
This does not mean that all factors earn profits by identifying mispricings. Some attributes, such as illiquidity, are associated with higher returns because they obviously represent risk.
So factor investing encompasses two different approaches:
1. Rational factor theory, which deals with the rewards that accrue to different types of risk and how the market prices them. Factor investing in this context is about finding the optimal portfolio of factor risks.
2. Factor investing as commonly understood by practitioners, which is the identification of simple quantitative strategies associated with excess returns.
Though it's been around for decades, factor investing has only in the past decade gained adherents. Recent converts include the Government Pension Fund of Norway, the biggest pension fund in Europe, and CalPERS, the biggest public pension fund in the United States. They've seen the light after realizing that the active managers they were richly compensating were simply offering factor risks and factor-based strategies under the guise of skill.
Redefining Alpha
An implication of factor-based investing is that what was once legitimately deemed "alpha"--excess returns attributable to skill--has morphed into "beta" (or a factor) once researchers identify a simple strategy that replicates the alpha. For instance, certain hedge fund managers in the 1980s and 1990s pursued then-exotic strategies such as merger arbitrage that produced excellent returns uncorrelated to the market. However, once researchers identified how the arbitrage strategies worked and created mechanical replications, the managers' alpha became beta.
A consequence of this process is that the hurdle for being declared a truly skilled manager has risen over time. In the 1980s, it was good enough to beat your benchmarks. These days, studies looking for evidence of skill in equity mutual funds control for exposure to size, value, and momentum factors. In other words, if your excess returns come during times that value, smaller-cap, or momentum stocks outperform, the procedure will adjust your "excess" return to zero and declare you unskilled.
If you believe the excess returns of value and momentum strategies reflect risk, then it's a reasonable adjustment. If you believe value and momentum produce excess returns because of market inefficiency, then it's not--what you've done is redefine outperformance. I'm of the latter view. From my perspective, the mountains of studies purporting to show that active equity managers can't beat the market are really showing that much of their excess returns can be replicated by a handful of factor strategies.
