In the first article of the smart beta series, we covered the importance of avoiding false discoveries by having a solid framework to evaluate which factors stand up after rigorous analysis.
Recent research (Asness et al., 2015; Harvey et al., 2016 and Dimson et al., 2017) points to value, momentum and low volatility as 'robust' risk premia, worthy of being implemented in an investor’s portfolio.
However, a common and important question that investors need to answer is whether to implement one single factor (for instance to give a value tilt to the portfolio) or opt for a multi-factor solution (such as a fundamentally weighted low volatility and value combined tilt to the portfolio).
Multi factor in vogue
The latest survey conducted by FTSE Russell on 200 global asset owners documents a clear trend: 64% of respondents who are currently implementing a smart beta index are using a multi-factor strategy. That is more than triple the rate in the 2015 survey!
The popularity of multi-factor indices is backed by rigorous academic research. The ins-and-outs of smart beta investing have been documented and a number of implications have been reported.
First, while factors like value and momentum have solid empirical evidence of adding risk adjusted returns to a portfolio, they do not work all the time.
For example, Daniel and Moskowitz (2016) document strong relative underperformance (in their words 'crashes') of momentum.
Similarly, in 'Two Centuries of Price Momentum,' Geczy and Samonov document significant time variation in momentum payoffs.
For instance, from 1800 to 2012, there were three decades with negative annualised returns. When examined over ten year rolling windows, the periods increase to seven.
Academics studying factor premia and investors putting money at stake in these strategies experience the passage of time differently.
No time like the present
This phenomenon is commonly referred to as 'time dilation' in physics, and in psychology, it is referred to as 'present bias'.
Present bias simply means that we place more value on the present time compared to the past or the future.
For investors, we generally observe that their present bias extends to around three years. They are incapable of withstanding three or more years of relative underperformance before selling, at precisely the wrong time.
Second, a study published in the Journal of Finance in 2013, documents a very interesting and desirable property: value and momentum are negatively correlated!
Specifically, the authors find that a naive 50/50 combination of value and momentum in each market and asset class outperforms either value or momentum implemented independently.
The power of diversificiation
Apparently the value and momentum characteristics of stocks are diversifiers and can be used to reduce portfolio risk.
An example of this powerful combination has been documented in Japanese equities. In the February 2012 issue of the Journal of Portfolio Management, Cliff Asness published 'Momentum in Japan, the exception that proves the rule'.
Here, he shows that the well-known weakness of momentum not working in Japan is actually a strength. In fact, because momentum has a strong negative correlation with value, it works from a diversification and risk management perspective for Japanese investors.
In conclusion, the evidence suggests that combining robust factors that are uncorrelated may prevent investors from suffering through extended periods of poor relative performance when exposed to single factors only.
About Elisabetta Basilico
Dr Elisabetta Basilico CFA is a quant investment expert and consultant who specialises in what she terms ‘turning academic insights into investment strategies.’ She helped several private and institutional clients achieve stable returns and financial prosperity. Follow her at academicinsightsoninvesting.com