Low-volatility investing is a great place for investors and researchers. For long-term investors, it can create a simple tilt that offers an attractive alternative to following market capweighted equity indices, generating at least the same return for two-thirds of the risk. For short-term investors, it may provide an interesting basis for various trade ideas. For theorists, it’s an anomaly that strikes at the heart of financial theory, because it’s very difficult to reconcile with the concept of a ‘risk premium’.
Since Robeco pioneered its low-volatility approach around 2006, the low-volatility market has grown enormously, as its risk-adjusted returns have been significantly better than the market cap-weighted indices. For example, at the end of 2013, there was about USD 13 billion in low-volatility ETFs, and an even larger amount in institutional funds. The question arises as to whether this growth is eliminating the low-volatility premium going forward.
The key thing to remember is that stock prices are determined ‘at the margin’ by pricesensitive investors. Discovering who exactly these investors are is quite problematic, and clearly isn’t as simple as taking a weighted average of investing styles. It is interesting to note that the influx of small cap and value funds has not significantly changed these premiums over the past 30 years. And perhaps even more striking is that index investing itself has not changed the equity ‘risk premium’, or the expected return above the risk-free rate.
Another aspect is why the premium exists in the first place. I think it’s safe to assume that active investors are still massively overconfident and have strong incentives to focus on highly volatile stocks like Tesla, Alibaba and Facebook. Pim van Vliet, David Blitz and I wrote a paper in the Journal of Portfolio Management in 2014 outlining several of the mechanisms behind this (see article 4 of this book), such as incentives for market participants, and I see no evidence of these mechanisms disappearing. These market participants are, to my mind, the ones who are allocating capital day-to-day and driving supply and demand.
At a theoretical level I think that the finance profession – indeed, the whole field of economics – has yet to really digest the importance of the volatility effect. That is, the low-volatility anomaly is not so much an anomaly, but a broad refutation of the canonical model of economic agents. This is because the standard utility function – increasing wealth at a decreasing rate – implies a consistent risk premium, which is a necessary and sufficient condition for this function, but does so ‘if and only if’ this premium exists. If the risk premium does not exist, utility is a completely different concept to the one economists have applied for the past 60 years. It is one thing to say people have quirky biases, but to remove the risk premium would relegate many seminal theoretical models to irrelevance.
As mentioned above, there are several reasons why people do not approach their investing decisions in a vacuum: people have pervasive biases, such as overconfidence, and society generates incentives that reward excessive risk taking. Many economists think some of these biases explain the low-volatility anomaly. In my books, most recently in ‘The Missing Risk Premium’, I argue that it is necessary to assume that investor preferences are relative – pervasive benchmarking if you will. This focus on relative wealth has many implications, such as why so many people are concerned about the prosperity of the ‘1%’ and feel hurt by any further increase in their wealth. Some say this concern is really about aggregate productivity or stability (e.g., Thomas Piketty’s ‘Capital in the 21st Century’), but I think a simpler and more accurate explanation is just that most people feel worse when others do better than them. On one level it seems too petty to be true, but they don’t call economics the ‘dismal science’ for nothing. We should accept human actions for what they are, mindful that ‘ought’ and ‘is’ are not always in sync.
Given the relative impotence of economists to explain or predict the big matters of the day, I think a healthy skepticism about the practical use of theoretical models presents a path to a better way. It is simply not true that economists have it all figured out and that people should start listening to them. A better understanding and thorough investigation of the low-volatility anomaly would lead to discoveries large and small, which makes it an exciting idea. This new collection of Robeco articles provides a useful contribution to the debate and offers plenty of discoveries and insights.