Both smart beta strategies exploit the weakness of the cap-weighted index—its natural tendency to overweight companies that are overpriced by the market. Therefore, it is not surprising that both strategies consistently and materially outperform the cap-weight strategy. Nonetheless, there is a substantial difference in the performance of the fundamentally weighted and equally weighted strategies: the former outperform the latter in almost all countries, with an average return advantage of 1–2%.
An argument commonly proposed in favor of equal-weight strategies is the significantly higher diversification relative to the Fundamental Index strategy. However, diversification is not the ultimate goal in itself; rather, it is a means to achieving the desired risk characteristics. And, as can be seen from Table 1, the volatility and tracking error of the equal-weight indices are in the same ballpark as those of the fundamentally weighted indices. It may be surprising that the much broader diversification of the equal-weight strategy does not ensure materially better risk characteristics. However, most of the benefits of diversification can be achieved with a relatively small number of stocks; reducing concentration further results only in marginal improvements.2
It is important to emphasize that much of the outperformance of the fundamental- versus equal-weight strategy is due to the selection effect. Capitalization as a criterion to select stocks for the equally weighted strategy favors the high price, potentially overpriced, stocks. The equally weighted strategy further exacerbates the potential problem of favoring overpriced stocks by giving them a significant weight in the portfolio. If we were to equal-weight the entire universe, the performance difference between the fundamentally and equally weighted strategies would shrink materially. In this case however, the equal-weight index would have a significant weight in many extremely small stocks, further increasing the market impact costs
The costs of running an index-based strategy are largely composed of fees and market impact. Under competitive pressure, there is no reason to expect the fees to be very different; more importantly, the fees are likely to fall as the assets under management (AUM) grow. Beyond a certain level of investment, the market impact of trading represents the dominant share of transaction costs.
In estimating market impact, we follow the approach proposed in Aked and Moroz (2013), which postulates a linear per-security price impact and aggregates it across all the securities in the index. This approach predicts that:
(a) the market impact on performance (in bps) is proportional to the aggregate AUM invested in the strategy by all investors; and
(b) the scaling factor depends on both turnover of the strategy (primarily the turnover resulting from additions and deletions) and the “tilt,” or the degree to which index weights deviate from a trade volume-weighted index.
To obtain an estimate of the costs, then, we need to specify the amount of assets. In the following table, we show how the performance looks after adjustment for various amounts of AUM. An equal-weight strategy has a larger turnover and a more pronounced tilt and, therefore, its net performance falls off with asset size much faster than a fundamental strategy.
Table 2 presents the impact of transaction costs on the performance of cap-weighted, fundamentally weighted, and equal-weighted indices at several levels of global AUM; the assets are allocated to the countries in proportion to the market cap in order to create a meaningful cross-country comparison of costs. Because the market impact model we utilize needs to be calibrated to a specific index, we standardize all results to correspond to a 50 bps market impact for the $2 trillion in the U.S. market capitalization-weighted strategies. A more comprehensive version of Table 2 is presented in the Appendix.