Korva is a scroll of the internet’s best ideas.
The 10% average almost never happens
Average versus actual
Ask most people the S&P 500's average annual return and the answer is 10%. Ask how often a year actually lands on 10%, and the answer is pretty much never. The market overperforms in some years and underperforms in others, and it averages out to that figure.
Mean, median, mode
That makes 10% the mean, and definitely not the median or the mode. Anyone expecting 9 or 10% in a given year is expecting a number that almost never arrives.
The hidden range
Averages hide the range, the way a river that is five feet deep on average hides its deep spots, and many people have drowned crossing one. A norm, an average, or a result that was studied once can mislead anyone who treats it as the thing to expect.
Why does the S&P 500 average not match what you would earn?
The simple average of yearly returns is higher than the compound growth rate whenever returns swing. A fall of 50 percent followed by a gain of 50 percent averages to zero, yet it leaves an investment 25 percent lower. The compound rate is called the geometric mean, and the gap between the two grows as volatility grows. Analysts call this gap volatility drag.
How widely have yearly S&P 500 returns swung?
Including dividends, the index rose about 53 percent in 1954 and fell about 37 percent in 2008. Over the full record since 1926, only a small number of years have landed within a point or two of 10 percent. Most years are far above or far below it. The long-run figure only appears after many years of ups and downs are averaged together.
Why do market crashes happen more often than a bell curve predicts?
Mathematician Benoit Mandelbrot studied cotton prices in 1963 and found that extreme moves occurred far more often than a normal bell curve would allow. Later work found the same fat tails in stock returns. A bell curve treats a drop of five standard deviations as nearly impossible, yet markets have produced such drops several times. Risk models that assumed a normal distribution therefore understated the odds of large losses.
More from My First Million
