Source: S&P Dow Jones Indices LLC, Factset. Data from Dec. 31, 2000 to Dec. 31, 2019. Past performance is no guarantee of future results. Chart is provided for illustrative purposes.
SkewnessIn a normal distribution, observations array symmetrically around their average value. The cross-sectional distribution of stock returns is not normal, but rather is positively skewed – i.e., the average value is greater than the median. This isn’t surprising – after all, a stock can only go down by 100%, while it can appreciate by much more than that. Large positive values pull the average level of the distribution above the midpoint.
We can observe this in Exhibit 5, which plots the distribution of cumulative returns for the constituent stocks of the S&P 500 for the last 20 years. The median return was 52%, far less than the average of 239%. Exhibit 6 shows similar results for the S&P Europe 350, where the 36% median lagged the 136% average. Notably, the positive skew in equity returns demonstrated by Exhibits 5 and 6 is not simply an artifact of a small number of highly-skewed years: in the 29 years between 1991 and 2019, e.g., the average S&P 500 stock outperformed the median 25 times.16
If stock returns were normally distributed, stock selection would be no harder than a coin flip; a randomly-chosen stock would have an even chance of delivering above-average performance. When the distribution is skewed, selection becomes much harder. Of the 1010 stocks that were part of the S&P 500 between 2000 and 2019, only 267 were above average. The probability that a randomly-chosen stock would deliver above-average performance, in other words, was 26%, not 50%. When fewer stocks outperform, active management is harder.
https://www.suerf.org/publications/suerf-policy-notes-and-briefs/the-growth-of-indexing-what-is-happening-and-why/
