Skip to main content

Post-Earnings-Announcement Drift (PEAD)

Updated Aug 22, 2026 at 4:52pm

Research Draft Medium 923 words

After a company reports an earnings surprise — actual results well above or below what the market expected — its stock price tends to keep drifting in the same direction as the surprise for weeks to months afterward, rather than fully and instantly repricing on announcement day. A strong positive surprise is followed by further positive abnormal returns; a strong negative surprise by further negative ones. This persistence is widely interpreted as an under-reaction: the market absorbs the news gradually instead of all at once, which is exactly the behavior an efficient market is not supposed to exhibit. PEAD is one of the oldest and most robust documented anomalies in finance.

The evidence

The phenomenon traces to Ball & Brown (1968), An Empirical Evaluation of Accounting Income Numbers (Journal of Accounting Research, vol. 6, pp. 159–178). Working with annual earnings, they observed that prices continued to drift in the direction of the earnings surprise after the announcement — the foundational observation, though their annual, two-group design produced only a modest drift. Foster, Olsen & Shevlin (1984), Earnings Releases, Anomalies, and the Behavior of Security Returns (The Accounting Review, vol. 59, pp. 574–603), sharpened the result using quarterly data and unexpected-earnings models.

The canonical formalization is Bernard & Thomas (1989), Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium? (Journal of Accounting Research, vol. 27, pp. 1–36), and their 1990 follow-up. They ranked stocks by standardized unexpected earnings (SUE) — the gap between actual and expected earnings, divided by the standard deviation of past earnings (or forecast errors) so surprises are comparable across firms — sorted into deciles, and went long the top SUE decile and short the bottom. Per Bernard & Thomas (1989), the spread between the top and bottom SUE deciles was positive in 41 of the 48 quarters from 1974 to 1985. Reviews of their work report drift on the order of roughly 8–9% over the subsequent quarter for the long-short portfolio (before transaction costs), with the effect stronger in small firms. Bernard & Thomas (1990) further found that roughly 25–30% of the drift clusters in the three-day windows around the next earnings announcement — only about 5% of trading days — evidence that the market keeps being surprised in the same direction. The exploitable drift window is generally measured over the ~60 trading days (about one quarter) following the report.

Why it may exist

The leading explanation is under-reaction: investors fail, in Bernard & Thomas's phrase, "to recognize fully the implications of current earnings for future earnings" — they don't appreciate that earnings surprises are autocorrelated, so a beat this quarter predictably foreshadows the next. Competing and complementary accounts include limits to arbitrage (the drift is concentrated in smaller, less-liquid, less-covered stocks where high bid-ask spreads and short-sale frictions make the trade costly to enforce) and analyst-revision lag (forecasters update estimates only gradually after a surprise, dragging prices with them). The debate over whether PEAD is genuine mispricing or partly compensation for risk is long-running; the weight of evidence favors delayed price response over a pure risk premium, but it is not settled.

How a swing trader uses it

PEAD is the academic backbone of the earnings-gap continuation swing setup. The practical play: after a strong, high-SUE beat that gaps the stock up on heavy volume, treat the multi-day-to-weeks drift as a continuation tailwind rather than fading the move. Enter on the post-gap consolidation or a pullback that holds above the gap, sizing for a hold of days to a few weeks — the horizon over which the documented drift plays out. It pairs naturally with the earnings-gap setup already in this branch: the gap supplies the catalyst and the entry structure; PEAD supplies the statistical reason the move tends to continue. The mirror-image short applies to a strong miss that gaps down. Quality of the surprise matters more than its mere existence — the largest SUE deciles, on real fundamental beats, are where the historical edge concentrated.

Honest caveat

PEAD is well-documented and decades-old, but the edge has reportedly weakened over time. Studies including Chordia, Subrahmanyam & Tong (2014) and Martineau (2021) document that immediate announcement-day reactions have grown stronger while the lingering drift has shrunk — long-short drift estimates falling from roughly 5% in the 1980s–90s toward 3% or lower by the late 2010s in some reviews. Whether the cause is heavier arbitrage or declining persistence of earnings news is itself debated (some text-based studies still find robust drift). Critically, the drift lives disproportionately in small, illiquid, costly-to-trade names — after realistic transaction costs and slippage the net edge is far smaller than the headline gross numbers, and for a retail swing trader it is a probabilistic tilt, not a guarantee. Right-size it: use PEAD as a reason to favor continuation over reversal on a quality beat, not as a standalone signal to bet large.

Sources