Rob Arnott, CEO and chairman of Research Affiliates, has been warning investors and advisors about the pitfalls of smart beta and factor investing since he wrote a paper on the topic in February 2016. His argument is, if you look for good past performance and it’s based on rising valuations, that’s where smart beta can go horribly wrong. Rather, when factors are trading cheap relative to their own history, they tend to perform better—even if its past record is worse than others, Arnott said, speaking at the Inside ETFs conference this week in Hollywood, Florida.
“Good ideas trading cheap will be ideas that have disappointed recently,” he said.
At the moment, value is the factor trading cheap, especially in the emerging markets. To prove his point, Arnott showed a graph with dots representing the relative valuation of value relative to growth. When value’s a third of the price of growth or higher, it might work; it might not. When it’s below 0.25, or a four-to-one ratio, it almost always works and never hurts you badly. When it’s below 0.2, historically it has always worked by an average of well over 1,000 basis points per year, compounded for 5 years. Right now, value stocks are trading around then 0.25 range, he said.
“All factors exhibit a pattern in which, when they’re trading cheap, their performance is better,” Arnott said. Still, a lot of investors look to a factor’s historical performance as an indicator of how it will perform in the future. That’s trend chasing, Arnott said.
“Past returns are worse than useless. Many investors prefer comfort, chasing what’s popular and loved.”