ETF Strategy Summit
October 15-16, 2018 | Hyatt Regency | Dallas, TX
Benjamin Lavine – 3D Asset Management
The Importance of Factor Blends
Benjamin Lavine serves as the Chief Investment Officer and sits on the executive management and investment committees at 3D Asset Management. Ben works closely with 3D’s investment committee to refine and enhance the investment architecture underlying 3D’s ETF managed portfolios. We recently spoke with Ben, who will be speaking at our ETF Strategy Summit (Oct. 15 – 16 – Dallas), as he shared his thoughts on how to invest in factors.
ETF Strategy Summit: Since a multi-factor ETF may shift between factor, style and sector exposures as it rebalances, how should advisors think about implementing one within a client portfolio?
Ben Lavine: Most rules-based ETFs will reconstitute semi-annually and rebalance quarterly although some multi-factor ETFs will reconstitute quarterly to account for higher turnover factors such as momentum. Regardless of reconstitution frequency, the underlying factor exposures should not change too significantly over time unless there is a change to the multi-factor model itself. Many multi-factor ETFs also optimize their risk exposures versus a cap-weighted index like the S&P 500 or Russell 1000 Index, so investors can be reasonably assured that they will be invested in a strategy anchored to broad market risks. Hence, multi-factor ETFs can serve as a close substitute for traditional cap-weighted ETFs to provide core market exposures, but the devil is always in the details, so make sure to analyze both basket design and implementation and how multi-factor ETF candidates interact with other components within an investment program.
ETF Strategy Summit: Are there specific factor combinations or blends that investors should try to avoid? Which pairs or combinations make the most sense in today’s market environment?
Ben Lavine: Factor combinations or blends can run afoul of data mining resulting in a model with the best performing factors combined in the best performing manner over the testing period. The ETF provider should be ready to explain the economic or intuitive rationale for, say, why a dividend factor should be combined with a low volatility factor. Most investable factors can be combined into two categories: value and sentiment. Many multi-factor model ETFs will combine factors from these two categories because they have historically blended well resulting in performance where the ‘whole’ is greater than the ‘sum-of-the-parts.’ There is some ongoing debate over whether low volatility is an investable, alpha-generating factor or is better implemented as a risk control. For instance, several multi-factor ETFs use volatility as an initial screen to weed out high volatility stocks from the starting universe before generating factor scores. Some recent multi-factor ETFs use a dynamic approach to time factor exposures based on a backward assessment of fundamentals (i.e. factor valuations) and/or a forward assessment of the economic backdrop with the idea that certain factors perform better depending on the stage of the cycle. If investors believe that we are about to enter a recession, historically, you would have wanted to avoid ‘value’ and increase exposure to more defensive factors such as ‘yield’ and ‘quality.’
ETF Strategy Summit: Does it matter if factors are accessed actively or passively?
Ben Lavine: Probably not, but the ‘rules’ underlying a passive implementation provide greater transparency and more assurance that an ETF will follow its intended mandate. In addition, passive ETF managers will more likely use the basket creation/redemption to minimize realized capital gains distributions so as to minimize the tax impact from periodic rebalancings. On the other hand, active implementation of factors, unconstrained by a tracking benchmark, enables an ETF manager to (potentially) create a basket with more targeted factor exposures versus a passive implementation as the ETF manager can rebalance much more frequently – an important aspect for higher turnover factors like momentum.
ETF Strategy Summit: How has sector neutralization impacted the performance of factor ETFs thus far in 2018? Should advisors seek single factor or multi-factor ETFs that are sector neutralized, or employ more unconstrained factor products?
Ben Lavine: Group neutralization (whether sector, country or even risk buckets) can make a huge difference in factor performance over the short-term such as what was experienced in 2017 when the technology sector ran ahead of the major sectors. In 2018, it’s been a mixed bag for neutralization. The technology sector had been handily outperforming the other sectors throughout the first quarter but then reversed in April when energy and industrial cyclicals took over sector leadership. However, the impact from group neutralization tends to decline over longer periods of time. In addition, group neutralization has generally not worked for sentiment factors like momentum because much of momentum’s performance comes from both sector and stock-specific exposures. On the other hand, group neutralization has worked better for fundamental-based factors such as value. At the margin, group neutralization can help reduce factor volatility by removing group-level effects, but it can also reduce factor efficacy. Advisors should always consider how a single or multi-factor ETF fits within a broader investment program. A factor ETF that does not neutralize groupings may make more sense if it is offsetting group level bets from other strategies held in the program.
ETF Strategy Summit: Are ETFs and strategies that cycle through factors more desirable than factor blends?
Ben Lavine: The jury is still out on the efficacy of factor timing and whether it can produce consistent excess returns versus a static model. Some investors believe that factors exhibit momentum tendencies over a 12-18 month period, similar to the momentum effect at the security level. Others believe that factor exposures should be determined on where we are in the macroeconomic cycle. We believe that there are periods when factors look exceptionally overbought or oversold, but these are infrequent moments at best; hence, investors will probably be better served with a strategic factor blend.
ETF Strategy Summit: Thanks Ben. We look forward to hearing more of your thoughts at the ETF Strategy Summit October 15 – 16 in Dallas.