Lipper Insight: The Various Types of Smart-Beta Indices

Smart-beta indices typically are broken into two types: economic and diversification. Economic smart-beta indices aim for a better representation of the business side of the economy by replacing market capitalization with different measures—such as profits, book to value, or revenue—of fund “size.” Diversification-focused smart-beta indices aim to provide an approximation of the optimal risk-return portfolio (the tangency portfolio of Modern Portfolio Theory). Diversification smart-beta indices are based on the recognition that market-cap-weighted indices are poor proxies of the tangency portfolio.

Related to diversification smart-beta indices are “deconcentration” indices. Deconcentration indices tend to reduce risk by spreading the invested dollars over the largest possible effective number of stocks (e.g., in an equally weighted index based on all of the constituents of the S&P 500). Another deconcentration method is to have stocks appear in the index on an equal-risk basis. Deconcentration indices do not exploit correlations across stocks—the traditional means of diversification. So, we say that, based upon the underlying philosophy of the deconcentration approach, the indices are not—or are not strictly—diversification smart-beta indices.

Now that we have briefly defined smart-beta indices, we illustrate the exposure smart-beta funds and their related indices have to equity risk factors. We adapt a recent article by Amenc et al.[i]  The risk factors Amenc et al. use are: beta, size (large-cap/small-cap), volatility (high/low), dividend yield (high/low), and momentum (high/low).

Using eight years of weekly data, Amenc et al. find that many smart-beta indices have a significant exposure to the size factor, others have a “tilt” toward the volatility factor and the beta factor, while other indices have a momentum tilt. So, all the smart-beta indices tested by Amenc et al. have very significant exposure to at least one (if not more) risk factors. Moreover, when we consider the percentage of variability in the excess returns of the cap-weighted index that is explained by differences in risk factors, it is clear that risk-factor tilts explain an important part but not all of the performance of smart-beta strategies; R-squares range from 50% to 90%.

REUTERS/Alistair Scrutton

While risk-factor exposures may or may not be important to investors in general, we suggest that the tilt of a smart-beta fund, be it to small-cap or value, be examined in light of any existing exposure to small-cap and/or value funds, i.e., does the investor wish to increase exposure to small-caps? Investors also need to keep in mind that market cycles will affect the performance of many smart-beta funds. Recently, small-cap and value smart-beta funds have done better than those having a momentum tilt. However, smart-beta funds with a momentum tilt would have done significantly better than either small-cap or value funds in the middle to late 1990s.

To understand why such risk tilts may exist, it is important to understand that smart-beta indices deviate from a standard market-cap index in terms of stock selection or stock weighting or both. For example, FTSE GWA and FTSE RAFI weight stocks by fundamental characteristics, including cash flow and book to value. FTSE RAFI uses these characteristics to select stocks as well, while FTSE GWA does not modify the specific stocks in the underlying cap-weighted index. Similar differences can be found among manufacturers of risk-based indices.

Different smart-beta indices try to achieve their given objective by using stock selection and/or a weighting scheme. Among weighting schemes we have already made a distinction between two types: diversification-based and economic- or characteristic-based. Diversification-based schemes consider not only the characteristics of the stocks but also the interaction that arises between stocks when they are combined. Characteristic-based schemes tend to ignore the interaction between stocks, so it is fair to say these schemes focus on stock selection.

In our next article on smart-beta funds we will focus on the difference between diversification-weighting and stock-selection schemes.