Skewness and Kurtosis - Why Returns Aren't Normal

By EC Assets Research Team, Risk Research · Published 20 May 2026 · Updated 28 May 2026

Skewness and Kurtosis — Skewness measures the asymmetry of a return distribution and kurtosis measures its tailedness - the third and fourth moments beyond mean and volatility. Financial returns are negatively skewed and fat-tailed, which is why tools that assume normality (VaR, Sharpe, mean-variance) systematically understate crash risk.

Beyond Mean and Variance

The mean and the standard deviation describe a return distribution's centre and its spread - but they say nothing about its shape. Two portfolios can share an identical average return and an identical volatility while behaving completely differently in the tails. Skewness and kurtosis are the third and fourth moments that capture that shape: skewness measures asymmetry, kurtosis measures tailedness. For anyone who cares about crashes - which is everyone managing real money - they matter more than volatility.

Skewness = E[(r − μ)³] / σ³ Kurtosis = E[(r − μ)⁴] / σ⁴

A normal distribution has skewness 0 and kurtosis 3. The amount by which kurtosis exceeds 3 is called excess kurtosis, and positive excess kurtosis is the signature of fat tails.

Skewness - Which Tail Is Fatter

Skewness tells you which direction surprises tend to come from:

• Negative skew means a long left tail: frequent small gains and occasional large losses. Equity indices are negatively skewed - markets "take the stairs up and the elevator down." Selling options, carry trades, and most premium-harvesting strategies manufacture negative skew.
• Positive skew means a long right tail: frequent small losses and occasional large gains. Long options, trend-following, and venture-style bets are positively skewed - many small losses, the occasional huge winner.

Investors are not indifferent between the two. Most people dislike negative skew (the risk of a sudden large loss) and will pay to avoid it, which is why negatively-skewed strategies tend to carry a return premium - they are paid to hold the crash risk others shed.

Kurtosis - How Fat the Tails Are

Kurtosis measures how much of the variance comes from rare, extreme observations. A distribution with high excess kurtosis (leptokurtic) produces far more extreme events than a normal distribution predicts. Financial returns are emphatically leptokurtic: daily equity returns routinely show excess kurtosis well above zero, often in the high single digits or more.

The practical consequence is severe. Under a normal distribution a five-standard-deviation daily move should occur roughly once in fourteen thousand years. In real markets, five-sigma days arrive every few years. The model is not slightly wrong - it is wrong by orders of magnitude precisely where it matters most.

Worked Example

Imagine two funds, each reporting 8% annual return and 12% volatility. On paper - and in a mean-variance optimiser, or a Sharpe ratio - they look identical.

Fund A: skew near zero, kurtosis near 3. Its returns really are roughly normal; the 12% volatility tells the whole story. Fund B: skew −1.5, excess kurtosis +8. It posts steady positive months, then occasionally loses 25% in a week. It is short volatility in disguise.

Their Sharpe ratios match, but Fund B is carrying a hidden crash exposure that the headline statistics completely conceal. An allocator who looks only at mean and variance cannot tell them apart - and will be blindsided by Fund B.

[!key] Volatility tells you how much a portfolio wobbles on a normal day. Skewness and kurtosis tell you what happens on the abnormal day - and abnormal days are when portfolios are actually destroyed. A high Sharpe ratio with strong negative skew and fat tails is a warning, not a recommendation.

Why It Matters for Institutional Investors

• Hidden risk in smooth returns. The most dangerous strategies are those that convert tail risk into steady income - short volatility, credit, carry. They show high Sharpe ratios, mild volatility, and lethal negative skew with fat tails. Moments three and four are how you detect them before the blow-up.
• Model risk. Value-at-risk, Sharpe ratios, and mean-variance optimisation all assume (or behave as if) returns are normal. Under fat tails, VaR understates losses and the Sharpe ratio flatters the strategy. This is exactly why expected shortfall and explicit tail modelling exist.
• Manager selection and due diligence. Asking for the skewness and kurtosis of a track record - and stress-testing the worst historical month - reveals the true risk profile that the average and standard deviation hide.

[!warning] Treating returns as normal is the single most common and most expensive error in quantitative finance. Real markets have fat tails and negative skew, so any tool that assumes normality - VaR, Sharpe, mean-variance optimisation - systematically underestimates the risk of exactly the events that cause permanent loss.

References

1. Mandelbrot, B. (1963). The Variation of Certain Speculative Prices. The Journal of Business, 36(4).
2. Taleb, N. N. (2007). The Black Swan: The Impact of the Highly Improbable. Random House.
3. Cont, R. (2001). Empirical Properties of Asset Returns: Stylized Facts and Statistical Issues. Quantitative Finance, 1(2).
4. McNeil, A. J., Frey, R., & Embrechts, P. (2015). Quantitative Risk Management (2nd ed.). Princeton University Press.