Skewness in Returns: Why Crashes Outpace Rallies

Key Takeaways

• S&P 500 daily skewness typically runs from minus 0.3 to minus 1.2 because crashes compress losses faster than rallies build gains.
• Harvey and Siddique showed in 2000 that assets with more negative coskewness earn higher average returns as compensation for downside risk.
• Short-volatility and carry strategies are structurally negatively skewed and can show Sharpe ratios above 1 right up to a blowup event.
• Investors should pair Sharpe ratio with skewness and maximum drawdown to detect hidden tail risk before allocating to any strategy.

What It Is

Formally, skewness is the third standardized moment of a distribution:

skewness = E[ (r - mu)^3 ] / sigma^3

Where r is the return, mu is the mean, and sigma is the standard deviation. A symmetric distribution like the normal has skewness zero. Positive skewness means the right tail is longer; negative skewness means the left tail is longer.

Equity index returns are consistently negatively skewed at daily and monthly frequencies. The S&P 500 typically shows a daily skewness between minus 0.3 and minus 1.2 depending on the sample window. Individual stocks vary more widely, and many small caps are mildly positively skewed because of lottery-like upside.

The Intuition

Crashes are faster than rallies. The 1987 Black Monday drop, the 2008 Lehman week, and the March 2020 COVID selloff all compressed multi-month losses into single days. Rallies of comparable magnitude usually unfold over weeks. That speed asymmetry shows up as negative skewness in the return series.

The economic story behind it is structural. Deleveraging cascades, margin calls, and risk-limit unwinds all amplify downside moves. On the upside, there is no mirror mechanic. A rising market does not force short sellers to buy in at the same urgency that a falling market forces long sellers to sell.

Harvey and Siddique showed in 2000 that investors actually price this asymmetry: assets with more negative coskewness earn higher average returns, because investors demand compensation for bearing downside risk.

How It Works

You estimate skewness from a sample using the third central moment:

sample_skewness = (1/N) * sum( (r_i - mean) / sigma )^3

Like kurtosis, the estimate is noisy because cubing amplifies outliers. A single crash observation can shift the estimate materially. Practitioners often report rolling one-year or five-year skewness to track how the asymmetry changes.

Three observations dominate the professional use of skewness:

1. Strategy selection. Some strategies are structurally short-skew. Selling out-of-the-money puts, merger arbitrage, and many carry trades produce small gains most months and occasional large losses. Others, like trend following and long-volatility, are long-skew: frequent small losses punctuated by rare big wins.
2. Risk adjustment. Sharpe ratio is blind to skewness. A strategy with Sharpe 1.5 and skew minus 3 is not comparable to one with Sharpe 1.0 and skew zero. Omega ratio, Sortino ratio, and skew-adjusted Sharpe corrections exist for this.
3. Option pricing. The implied volatility skew in equity options (out-of-the-money puts priced richer than out-of-the-money calls) is the market's direct quote on expected negative skew in the underlying.

Worked Example

Consider two hypothetical hedge fund strategies with the same monthly Sharpe ratio of 1.0 and 10 percent annualized volatility.

Strategy A (symmetric): monthly returns are normally distributed around 0.83 percent with standard deviation 2.89 percent. Skewness is zero. The worst month in 10 years is roughly minus 8 percent.

Strategy B (short-skew put writer): wins 0.5 percent in 95 percent of months and loses between 5 and 25 percent in the other 5 percent. Mean return matches Strategy A. Realized skewness is roughly minus 3.5.

Both look identical on a Sharpe ratio scorecard. Over 10 years, Strategy A's worst drawdown is about 12 percent. Strategy B's worst drawdown can exceed 40 percent in a single stress event, because its return engineering packs downside into rare, large observations. The skew statistic would have flagged this before the blowup.

Common Mistakes

• Ignoring skew when comparing Sharpe ratios. A short-skew strategy can look superior for years before a single bad month erases the gains. Always pair Sharpe with skewness and maximum drawdown.

• Confusing sign conventions. Positive skew means the right tail is longer, which for a profit distribution is typically the favorable case. Some textbooks flip the framing when discussing loss distributions. Always check whether the variable is a return (gain positive) or a loss (loss positive).

• Treating skewness as time-invariant. Skewness shifts with regimes. Bull markets often look mildly negatively skewed; crisis periods deepen the asymmetry dramatically. A full-sample estimate averages these states and can be misleading.

• Using high-frequency skew to judge long-horizon strategies. Daily returns may be strongly negatively skewed even when monthly or annual returns are close to symmetric, because the central limit theorem pulls long-horizon distributions toward normality. Match the estimation frequency to the investment horizon.

Frequently Asked Questions

Q: What is skewness in returns in simple terms? It measures whether a return distribution is lopsided: negative skewness means the worst losses are more extreme than the best gains, while positive skewness means there are occasional outsized wins relative to typical losses.

Q: How does skewness in returns affect investment decisions? It exposes the hidden risk profile of strategies that look attractive on Sharpe ratio. A put-writing fund and a trend-following fund can have similar Sharpe ratios but opposite skew profiles, meaning very different drawdown scenarios in a market crisis.

Q: What is a real-world example of skewness in returns? A strategy that writes out-of-the-money puts wins 0.5 percent in 95 percent of months but loses 5 to 25 percent in the other 5 percent. Its annual Sharpe can exceed 1.0 for years while its cumulative skewness of minus 3.5 accumulates unrealized tail risk.

Q: How can investors use knowledge of skewness in returns? Measure the rolling skewness of every strategy before allocating, require that high-Sharpe strategies with negative skew demonstrate adequate drawdown controls, and build portfolio-level skewness into stress-test scenarios rather than assuming Gaussian returns.

Q: How is skewness in returns different from kurtosis? Skewness captures directional asymmetry: which tail is longer. Kurtosis captures the overall weight of both tails combined relative to the center. A strategy can have near-zero skewness but still blow up through fat symmetric tails, so both statistics are needed together.

Sources

1. Cont, R. (2001). "Empirical properties of asset returns: stylized facts and statistical issues." Quantitative Finance, 1(2), 223-236. https://www.tandfonline.com/doi/abs/10.1080/713665670
2. Fama, E. (1965). "The Behavior of Stock-Market Prices." Journal of Business, 38(1), 34-105. https://www.jstor.org/stable/2350752
3. Harvey, C.R. and Siddique, A. (2000). "Conditional Skewness in Asset Pricing Tests." Journal of Finance, 55(3), 1263-1295. https://doi.org/10.1111/0022-1082.00247
4. CFA Institute. Quantitative Methods curriculum readings. https://www.cfainstitute.org/programs/cfa-program/candidate-resources

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.