Clearly, an agreed upon classification for smart beta strategies and ETFs would benefit the end investor. In fact, we are working on an internal classification of smart beta that is rooted in buying behavior and aligned to factor premia.
One topic of debate is to include sector funds that employ alternative weighting mechanisms. For instance, a sector fund that employs an equal weighted methodology is likely not targeting the size factor, but instead weighting in a non-market cap weighted fashion to potentially get a better representation of a particular market segment. At the very least, sector based alternative weighted strategies deserve their own category, as the buying behavior is less about the factor and more about the sector.
Based on our initial work, smart beta ETF assets under management now approach over $300 billion [2], which, in my opinion, is far from any sort of bubble given the size of global equity markets ($43 trillion) [3] and the fact that not all factor strategies will have the same methodology, and therefore, holdings.
FA: Can factor cyclicality be used to generate alpha?
Bartolini: Successful core factor investing must account for cyclicality of factor performance. Multi-factor approaches may provide diversification benefits and offer the potential for improved consistency in performance over a long-term investment horizon. While not “alpha” per se, as these are premia available in the market, multi-factor strategies can potentially address cyclicality and enhance performance.
With respect to factor timing, the difficulty of timing factors has been well-documented [4], given the uncertainty of exogenous elements (macro, risk, sector, etc.) affecting their behavior and the complexity of the underlying relationships. Some have applied a fundamental evaluation measurement to factor valuations to time when factors are rich or cheap; however, this may lead to another form of value investing cloaked as factor timing. A fundamental view may work for a longer time horizon to tilt the portfolio one way or another; however, timing factors with precision is difficult in the short term. As a result, it is quite difficult to do from a top-down perspective by rotating among single factor exposures (e.g. single factor ETFs).
Therefore, any sort of timing may be best left to active managers to focus on at the single stock level and apply factor tilts that are dynamically shifted to account for a myriad of the exogenous factors. It likely should be part of the process, but not the process. After all, understanding the interdependencies of macroeconomic and market behavioral influences on factor premia is indeed at the heart of the active quantitative process.