Goldman, Blackrock, Powershares, and other ETF providers have started to roll out smart beta ETFs for the masses. These smart beta ETFs are designed to replicate the factor investing strategies developed by financial economists over the last couple of decades. Overall the strategy is a sensible one – factor investing can offer investors better returns at a comparable level of risk.
For evidence of this, one need only look to the performance of DFA funds over the last three decades.
While DFA is not as well-known as other large asset managers like Vanguard or Prudential, the company has a track record that many asset managers likely see as an object of envy. Since 1981, DFA’s flagship fund has returned an average of 11.8% annually versus 10.3% for the Russell 2000. The company’s funds are low cost and offer an attractive option positioned between the warring camps of active and passive. But this is not a commercial for DFA.
Indeed, DFA has its own issues – for one thing, its returns would likely be higher if it were more willing to explicitly consider momentum factor investing. For another, DFA is notorious for making it difficult for financial advisors to sell its funds to clients. That is one reason why the company is not particularly well-known.
Still DFA is capitalizing on an investing approach that has undeniable credibility and efficacy. The chart below shows returns to an aggressive smart beta/factor-based portfolio versus a basket of actively managed 40 Act funds. And that credibility is why ETF providers have begun rolling out smart beta products en masse. This could end badly for investors in those ETFs if the fund providers aren’t very careful about structuring the fund procedures.
Just as USO has proven to be a product that can be exploited by smarter traders at the expense of naïve investors, smart beta ETFs may be exploited by smart money traders to capitalize on rigid ETF processes. Here’s the problem in a nutshell; factor investing (the original name for smart beta) categorization is serially correlated and forecastable. That’s not to say that stock returns are forecastable – they aren’t (at least in the short term). But firm characteristics are forecastable and they tend to operate in trends over time as well.
Smart beta ETFs generally focus on a single factor category and usually not size since it’s too widely used already (plus the size factor has had an atrocious decade which limited enthusiasm for the category). Those categories are value, momentum, profitability/quality, and volatility/betting-against-beta. Because ETFs are often organized to exploit only a single factor and are generally long-only, figuring out which stocks an ETF is going to buy simply requires determining the rebalancing procedures of the fund and the stocks which either have moved factor categories or are about to move factor categories.
Neither process is particularly difficult. The obvious criticism of the strategy is that smart beta ETFs are still a small part of the market so it’s hard to see how their investment decisions lead to large moves in stock price. That’s true, especially for large cap companies, but it misses three important points.
First, the smart beta category is growing rapidly and becoming more influential over time. Thus the strategy of front-running such ETF is likely to become more profitable over time.
Second, for small less well-known firms, smart beta ETFs can be an important contributor to overall investment demand. Figuring out which companies are most reliant on smart beta ETFs is a fairly trivial task.
Third, even if a long-only front running strategy fails to lead to a significant price change, the worst case scenario is that the investor is holding a portfolio of stocks that score highly on factor dimensions which means that the portfolio should outperform the broader market over the long-term. In essence, front-running factor ETFs is a win-win; either the investor succeeds in generating a short-term return, or they get an attractive portfolio to hold for the long-term. The chart below shows that factor investing can significantly outperform passive investing over time.
None of this is to say that factor investing is a poor investment strategy or that it is not appropriate for ETF vehicles. But it behooves ETF providers to come up with the most effective methods of portfolio construction, and doing this requires an effective understanding of factor category forecasting. On the other hand, smart traders should be just as interested in learning more about the topic.
Mike McDonald is a PhD in finance and a university professor in the subject. He also runs a consulting company doing work on quantitative investing, big data, and machine learning for a variety of financial firms, asset managers, institutional investors, and government regulators. Prior to getting his PhD, Mike worked for a major Wall Street bank and one of the top hedge funds. Comments, questions, and concerns are always welcome – email Mike at M.McDonald@MorningInvestmentsCT.com or visit his firm’s website.