Skewness is a term that describes the distribution of returns for a specific security, asset class or portfolio, and is often utilized when combining asset classes or fund managers as part of the portfolio construction process. When considering a portfolio of investments, one with zero skewness would have a perfectly symmetric distribution of returns around its mean (think of a typical bell curve), and this is essentially the goal of many investors – to eliminate any large “left-tail” risk, or risk of a large drawdown. But some investors look at skewness as a favorable characteristic, particularly positive skewness that can be used to offset other investments that have negative skewness.
In their white paper The Taming of the Skew, PhDs Brendan Hoffman and Kathryn Kaminski, both of Campbell & Company, investigate techniques investors can employ to get the skewness they’re looking for.
Positive / Negative Skewness
Most investments and portfolios have negative skewness. This means they have most of their returns above the mean, with fewer but larger spikes below the mean. The S&P 500 tends to go up over time, so most of its returns are clustered above the mean – but it occasionally has pronounced spikes to the downside, which is what gives it “left tail asymmetry,” another name for negative skew.
Investments with positive skewing are more difficult to find, but they can help offset negative skew in a way that can be beneficial to portfolios. Positively skewed investments are those that typically perform well during times of crisis and are typically thought of as “crisis alpha” or “flight to safety” assets. This category of investments includes investments such as U.S. Treasury securities and managed futures strategies.
Strategies that aim to provide constant risk targeting (“CRT”), such as risk parity, are designed to have as little skewness as possible. In the long run, CRT portfolios have the best risk-adjusted returns, as measured by the Sharpe ratio.
Diversifiers vs. Complements
If CRT portfolios have the best long-run, risk-adjusted returns, why would investors want to target either positive or negative skewing? Mr. Hoffman and Ms. Kaminski explain by delineating between diversifiers and complements:
- A diversifier aims to improve a portfolio’s overall Sharpe ratio;
- A complement aims to improve a concentrated portfolio by exploiting conditional correlation.
Diversifiers could thus be considered strategic, while complements are more tactical. Since diversifiers aim to improve long-term, risk-adjusted returns, they aim for as little skewness as possible. Complements, by contrast, try to manage skewness to take advantage of particular situations.
Mr. Hoffman and Ms. Kaminski examined three portfolios in conducting their research:
- Constant risk targeting (“CRT”), which aims for zero risk variation, has the highest long-run Sharpe ratio, no skewness, negative or low correlation to equities, and can generate moderate crisis alpha.
- Signal risk targeting (“SRT”), which varies risk with average trend-signal strength, has the lowest long-run Sharpe ratio, positive and high skewness, negative and low correlation to equities, and moderate crisis alpha.
- Equity risk targeting (“ERT”), which varies risk with equity volatility and trend / equity correlation, has a moderate long-run Sharpe ratio, positive and moderate skewness, negative and high correlation to equities, and high crisis alpha.
As Mr. Hoffman and Ms. Kaminski note, the SRT portfolio shows that having the highest skewness doesn’t necessarily generate the most crisis alpha: timing matters. Using trend-following as an example, different approaches to time-varying risk illustrate the difference between skewness by random chance and skewness by design.
The authors conclude their paper thusly: “In the end, skewness is simply an outcome; the ultimate decision of whether or not to vary risk over time depends on the investor’s objective: to diversify or to complement?”
For more information, download a pdf copy of the white paper.
Past performance does not necessarily predict future results.
Jason Seagraves contributed to this article.