Passive perspective: Is smart beta worth the slice paid in transaction costs?
By Elisabetta Basilico 25 Sep, 2017 at 16:33
The effect of transaction costs on the real performance of smart beta strategies is a topic of huge importance for investors. This is because these strategies’ returns are not, unlike datasets often used by academia, gross of fees.
Asking important questions
The first researchers to deal with this issue were Andrea Frazzini, Ronen Israel and Tobias Moskowitz. In a 2014 paper, they attempted to answer two questions. First, do size, value, momentum and short-term reversal survive trading costs? Second, what is the largest amount that can be invested before the strategy becomes less effective?
The answers they found will offer reassurance to investors. With the exception of short-term reversals, the well-known factor premia of size, value and momentum add value after transaction costs. Moreover, the largest amounts that can be invested are fairly substantial: $259 billion (£191 billion), $273 billion, and $141 billion, respectively.
However, this study was criticized by academics and practitioners for the dataset used to assess the impact of transaction costs. Relatively short time series were covered and data were limited to larger stocks.
Return to roots
Last year Robert Novy-Marx and Mihail Velikov took the issue to the next level by evaluating a larger set of well-known anomalies over a longer time frame. The article, ‘A Taxonomy of Anomalies and Their Trading Costs’, examines the after-transaction cost performance for 23 different factor-investing strategies. It considers them over longer horizons and across various market capitalization classes, an improvement on other studies.
One caveat is the authors calculate transaction costs using the effective bid-offer spread measure proposed in a 2009 paper by Joel Hasbrouck. This does not account for the price impact of large trades, which can move markets and influence stock prices. As such, it should be interpreted as the cost faced by a small investor.
They attempt to answer the same questions as Frazzini, Israel and Moskowitz. First, what are the costs of trading the most important anomalies (referring to factor premia)? Second, what capital capacity does each of these strategies have before it becomes unprofitable to marginal trading?
A second opinion
Once again, there is good news for investors. The authors reach a similar conclusion to prior research: size, value and momentum still have positive returns after transaction costs. Second, they calculate $170 billion capacity for size, $50 billion for value and $5 billion for momentum.
The authors’ estimates generally agree with Frazzini et al. (2014) on size and value. But they come up with a much lower estimate for momentum, which aligns with a 2004 paper from Robert Korajczyk and Ronnie Sadka.
Comprehensive due diligence is a must when evaluating whether to invest in smart beta products. This should include considerations about the impact of transaction costs on the back-tested results presented by academia, as well as by the profession.
Elisabetta Basilico is a quant investment expert and consultant who specialises in what she terms ‘turning academic insights into investment strategies’. Follow her at academicinsightsoninvesting.com