Does the Stock Market Overreact?
Read the paperopens doi.org in a new tab
What they found
Borrowing from psychology research on overreaction, the authors formed portfolios of the 35 worst-performing NYSE stocks over the previous three to five years ('losers') and the 35 best ('winners') and tracked them for the next three years, repeating this from 1926 to 1982. The past losers beat the market by about 20% cumulatively over the following three years while the past winners lagged, and most of the loser rebound came in January. This was one of the first papers to argue that a systematic behavioral bias shows up in prices.
What you can use
- Over horizons of three to five years, extreme losers tend to bounce and extreme winners tend to fade.
- This is the opposite of momentum, which operates at 3 to 12 months; the horizon of your signal determines its sign.
- The reversal was concentrated in January and in small stocks, so tax-loss selling and illiquidity are part of the story.
Caveats
Returns are gross of the high costs of trading small, beaten-down stocks. Later work (Fama-French, Ball-Kothari) argued the effect is largely a value or risk effect and is sensitive to how returns are measured. Long-horizon only.
Tags: mean-reversion, overreaction, long-horizon, behavioral
Summaries are our own reading of the paper, not the authors' words. Educational only, not advice. Discuss it in Book Club.