Factor This Into Your Portfolio (Part 2 of 2)
Factor investing is an old concept that has taken on renewed interest with investors. Part 1 of this article explained what factors are and how they are measured. This article highlights the benefits of developing a multifactor portfolio and risks therein.
In a nutshell, factors are risks that explain portfolio return. The first factor that business students learn about is market beta. This is the non-diversifiable risk in a market. All portfolios that are composed of constituents in a market have an element of market risk in them. The more market risk, the higher the portfolio’s beta.
There are other risk factors that occur within a market. For example, in the U.S. equity market, small cap stocks have outperformed large cap stocks historically after adjusting for their higher beta; stocks with value characteristics and profitability characteristics have outperformed those that do not have them; and stocks with strong price momentum tend to outperform stocks with low price momentum.
The return premiums created by risk factors do not occur all the time, although they do occur often enough and with enough statistical robustness to dismiss the notion that these are mere random occurrences or market anomalies. The expectation, at least as told by the providers of factor based products, is that investors will be rewarded in the future as these risk premiums persist.
Outsized performance is the obvious goal of building a multifactor portfolio around a model, such as the Fama-French Three Factor model. This model is based on three factors: beta, small company size, and value stock exposure.
I estimate that the expected excess return over the market for an equity portfolio that has a 25% tilt to small-cap value stocks is 0.5% annualized. This is based on my estimate using historic returns that small-cap value stocks are expected to earn 2.0% annually over the broad US stock market. Whether this premium materializes or not cannot be known in advance. Only an expectation can be known.
One disadvantage with any multifactor portfolio is the excess cost to own funds that have more than market risk factors. The extra cost exists even among the cheapest fund providers. For example, the Vanguard Total Stock Market ETF (VTI) has an expense ratio of 0.05% while their Small-Cap Value ETF (VBR) has a 0.10% fee. Granted, the difference isn’t much, but costs matter as John Bogle, founder of the Vanguard Group is fond of saying.
Some multifactor fund providers charge more than others. The DFA US Small Cap Value Portfolio - I (DFSVX) has an expense ratio of 0.52%. Is it worth paying 0.42% more in annual fees over VBR for small value exposure? Perhaps it is. DFSVX has heavier factor exposures than VBR because it holds smaller stocks and digs deeper into value factors. Thus, a pricing model can be created that compares cost per using of factor risk – but that’s getting ahead of things.
A second benefit of multifactor investing is that it combines different risk factors in a portfolio; hence, investors may be able to achieve higher risk-adjusted returns. Size, value, profitability and beta factors are not highly correlated with each other. There is the possibility of engineering lowering portfolio risk without lowering expected return. This results in higher risk-adjusted performance.
The flip side of mixing risks is the possibility that the risk factors will go away for a long period or stop working altogether. There have been long stretches when small-value stocks underperform the rest of the market. The worst spell lasted more than 18 years from 1984 through 2001. I documented some of these periods in Expect Years of Pain before Market Gain.
The question of factor premium persistence is especially important considering the costs of pursuing this strategy are always higher. Many people have observed that the size effect has all but disappeared in the marketplace since its discovery. Columbia Business professor expressed his concern in a recent working paper titled Factor Investing:
“Some factors disappear. The size effect – that small stocks outperform large stocks – was brought to investors’ attention by Banz in 1981 and reached its peak just after that… Since the mid-1980s, however, there has been no size premium after adjusting for market risk. That is, small stocks do have higher returns than large stocks, but not after taking out their exposure to the market factor. The creation of small stock mutual funds allowed the ordinary investor to participate and bear size-related risk. Thus, the risk-bearing capacity of the economy changed after industry created new products to capitalize on the size premium. Those industry developments caused size to disappear.” [pgs. 22-23]
Noted author and economic historian William Bernstein has his own explanation for the disappearance of factor premiums. Bernstein considers the process to be evolutionary, “I tend to think of the various betas as the earth's species diversity (albeit on a much, much smaller scale). A long time ago, there was a wonderful proliferation of them: arbs on convertible securities, currencies, and yield spreads galore. But, along with the dodo and auroch, they're gone, leaving a few shrunken monocultures.”
If Bernstein is right, then the destruction of risk premiums could have serious implications for multifactor investors going forward. My investment motto is: there is risk, there is return, and there are costs, all else is marketing. In the absence of factor premiums, there’s just market beta and higher cost. Net, net, it was a nice experiment that was all for naught.
Factor investing is an intellectually enriching idea and that in itself is the third benefit. The journey to learn about multifactor investing is academically stimulating and rewarding in its own way. This adds a certain irony to the endeavor. I don’t believe it is very important to some investors that they actually earned an excess return from multifactor risk because their reward has been the educational experience.
The lack of concern about earning a return premium from risk factors is evident in the fact that many clients pay 1% or more per year in fees to an advisor to access DFA funds. That’s an unusually high annual sum and that’s nearly impossible to overcome, even if excess returns are earned from multifactor investing. Recall that my estimate for having 25% equity exposure in small-cap value stocks is only expected to add 0.5% per year to equity returns and this is before fund expense.
Finally, tracking error is the name give to a strategy that falls short of a market benchmark. It could mean the downfall for many multifactor investors. How long will an investor stick with a mulitfactor strategy when risk premiums disappear for several years and all that is left are higher fees? Uncommitted investors are sure to abandon ship when things look the worst, literally locking in underperformance for the rest of their lives.
This leaves us with one question, should you multifactor invest? The answer lies within each investor. How deep is your knowledge? Are you going to get help from an advisor? How much are you paying an advisor? How long are you willing to hold on when the strategy isn’t working?
There are the questions one must ask before proceeding with a multifactor strategy. Either way, you should know that facts, and know your limits.