Benchmarking Most Assets to the S&P 500 Is a Mistake
February 28, 2017
Some investors use the S&P 500 as a reference for a wide variety of other return-seeking and diversifying assets, from international and emerging-market stocks to hedge funds and other alternative investments. We think judging apples by their similarity to oranges is a recipe for bad outcomes.
This idea seems simple and intuitive: If the S&P 500 is available at a low cost, other risky assets should do at least as well to justify their fees. “Why else would you take money from US stocks to invest in those other assets?” you might ask.
We agree that using the S&P as a benchmark for US large-cap equity portfolios is sensible, as long as the performance period is long enough to be statistically meaningful. But comparing any asset apart from US large-cap stocks to the S&P 500 will lead to consistently bad outcomes, our research suggests. A simple example highlights why using the S&P 500 as the benchmark for another asset class can lead investors astray.
In the Display below, column A shows the total return of the S&P 500 for each year from 2007 through 2016. The compound annualized return of the S&P 500 over the 10-year period was 6.9%, with a standard deviation of 18.9% annually.
Column B lists the same return series in reverse order. Obviously, it has identical returns and risk over the full 10-year period. Also obviously, there are enormous differences between the returns in column A and column B: Column B “underperformed” column A by more than 38 percentage points in 2015, but “outperformed” by almost 13 points in 2014, and so on. These large swings reflect the relatively low correlation between the two series: The 0.3 correlation between columns A and B is fairly close to zero, which indicates no correlation.
An investor benchmarking asset B to the S&P 500 would probably throw in the towel after one of the years of deep underperformance.
A better strategy would be to buy more of the weakly correlated asset after a year of terrible relative performance. Look at the rightmost column, which shows the performance of a portfolio that maintains a 50% weight in both A and B, rebalancing between them each year. (For this simple example, we assume that we can rebalance without transaction or tax costs.)
Over the 10 years, rebalancing across A and B returned 7.7% a year, with a 15.3% standard deviation. That’s a significant return pickup with less risk than A or B had over the full period—a truly fabulous result.
How is such a great outcome possible? The key is the low correlation between A and B, which allows the rebalanced mix of the two return streams to compound at a much steadier rate than either A or B. In other words, the big annual differences between A and B are good. They provide a valuable diversification benefit: They reduce risk.
Most investors accept that they should own bonds, as well as stocks, to obtain stability and the diversification benefit of owning assets with extremely low or negative correlations to each other. Historically, an allocation to non-US stocks or to hedge funds and other alternative investments has provided less stability than an allocation to bonds, and diversified US stocks less, but the diversification benefit has still been valuable.
The asset classes available vary widely in their risk and return characteristics, and their correlations with either other. In practice, it’s not easy to gauge the appropriate allocation to the various assets available, in order to get the best combination of return and risk for your specific needs. That’s why most investors can benefit from thoughtful advice. In our view, good advice should also cover how to assess the performance of the whole portfolio and its components, as well as when and how to adjust the asset mix over time.
The views expressed herein do not constitute research, investment advice, or trade recommendations and do not necessarily represent the views of all AB portfolio-management teams.