Demand for alternatives to traditional equity indexing has been growing, as investors seek cost-effective ways to outperform benchmark indexes, finds SEI’s latest commentary.
Report authors Stephen Beinhacker, CFA, managing director, Global Head of Equity, and Greg McIntire, CFA, U.S. Large Cap Portfolio Manager, believe these new approaches offer significant promise. But there are some challenge.
Construction-based approaches: a logical next step
A handful of providers have proposed methods for lowering the volatility and periodic uncertainty of single-factor portfolio returns. Beinhacker and McIntire believe it holds promise for enhancing returns with lower tracking error, less volatility and a better chance of avoiding periods of prolonged underperformance. As a result of these desirable characteristics, they expect the industry to trend in this direction in the coming years.
Read: Fixed income still a good option
Construction-based approaches seek to diversify into a portfolio of factor exposures in order to both smooth out the experience associated with single-factor strategies and produce better risk-adjusted returns.
“For example, in our analysis of the Fama-French data on momentum, size and value from November 1926 through March 2013, we found that a construction-based approach resulted in better information ratios (excess return relative to tracking error) than standalone factors. We obtained these results using two construction methods—equal weighting, which allocates equally to each factor, and risk parity, which allocates to the factors in proportions that balance each factor’s contribution to portfolio volatility.”
Read: Faceoff: REITs versus the real thing
The justification for equal weighting is straightforward. It is simple, assumes no conviction-based views, and has the benefit of bringing diversity to the construction process through equal weighting of capital. A key drawback is that it may result in suboptimal allocations, either in terms of return or risk.
Risk parity, on the other hand, seeks to diversify factor allocations by their expected contribution to portfolio risk or volatility.
“According to our analysis, the results for the equal- and risk-parity-weighted portfolios were a marked improvement on the standalone factor portfolios. Relative returns were positive, tracking error was lower, and the depth and duration of periods of underperformance were substantially improved.”
Developing an active-indexing implementation
The authors followed this up with another study in which we applied an expanded stable of factors to the Russell 1000 Index, using return data from October 1998 through March 2013.
“We again compared equal-weighted and risk-parity construction methods, applying them within each return factor category and then across factor categories. The stability factor, which tends to be associated with lower-volatility stocks, was employed contextually. When expected portfolio volatility was high, exposure to stability was increased in order to bring volatility back into line with the broader market. In this backtest, the risk-parity-construction approach produced more desirable results overall.”
Read: Choose all-caps for growth
As a result of their internal research, they prefer integration of multiple return factors, using a risk parity approach to portfolio construction. This technique can be employed as a carve-out within an active strategy, or offered as a standalone alternative-index vehicle. While single-factor implementations are capturing their share of investor attention, Beinhacker and McIntire believe their cyclicality can be challenging for a typical investor to endure.