The Post-Earnings Announcement Drift: Why Earnings Surprises Often Keep Moving Stocks After the News

Overview

The post-earnings announcement drift, or PEAD, is one of the most important findings in anomaly research: stocks with positive earnings surprises tend to keep outperforming after the earnings release, while stocks with negative surprises tend to keep underperforming.

That is the core result. Not “the market instantly prices everything.” Not “earnings news is old news the moment it is released.” PEAD says the opposite: the market often adjusts to earnings news slowly, not all at once.

Reviews of the anomaly still describe it exactly this way, and Ball and Brown’s original line of research remains foundational more than fifty years later.

This chart shows a stock that reports a positive earnings surprise. The initial gap on the announcement day is significant, but the drift continues for weeks afterward. That post-announcement continuation is PEAD in action — the market did not fully price the news on day one.

This is the most important PEAD evidence chart. Stocks are sorted into ten buckets by the size of their earnings surprise, then average 60-day forward returns are measured. The pattern is clear: the larger the positive surprise, the higher the subsequent return. The larger the negative surprise, the lower the subsequent return. The relationship is monotonic and economically meaningful.

This chart separates total post-earnings return into two parts: the day-0 reaction (the initial gap) and the subsequent drift. For both positive and negative surprises, the drift component is comparable in magnitude to the initial reaction itself. That means almost half the total information impact happens after the announcement, not on it.

A simple way to define PEAD is with an earnings surprise variable. The raw surprise is the difference between reported earnings and the market’s expectation. Often this is standardized into standardized unexpected earnings (SUE) by dividing by a scale factor such as forecast dispersion or historical earnings volatility.

The basic prediction is: when SUE is high, future returns tend to be higher. When SUE is low, future returns tend to be lower. That is PEAD in one line.

Bernard and Thomas helped establish this as a major anomaly, and later surveys still describe PEAD as the continuation of returns in the direction of quarterly earnings surprises.

At first glance, PEAD should not exist. A simple efficient-markets view would imply: earnings surprise ⇒ immediate repricing. But PEAD says the actual pattern is closer to: earnings surprise ⇒ initial reaction ⇒ continued drift.

That is why it became so famous. It directly challenged the idea that public information is always reflected in price immediately. Modern reviews still frame PEAD as a long-running anomaly inconsistent with a fully instantaneous adjustment to public earnings news.

A clean way to think about PEAD is: earnings news → partial investor response → slow updating → continued price drift.

Why might that happen? One explanation is underreaction. Investors, analysts, or institutions may revise their beliefs too slowly after a surprise. Another is friction: limits to arbitrage, trading costs, firm size, liquidity, and short-selling constraints can all slow full adjustment.

The literature contains support for both delayed reaction and market-friction stories, with PEAD often larger where trading is more difficult or attention is weaker.

This is where PEAD becomes especially important. Traditional momentum says: past winners tend to keep winning. PEAD says something more specific: stocks with positive earnings surprises tend to keep drifting up.

So PEAD can be viewed as one important micro-foundation for momentum: price momentum ≈ slow reaction to fundamental news. Not entirely, but significantly.

That is one reason PEAD matters so much. It ties momentum not just to charts, but to the market’s sluggish processing of earnings information.

A natural question is: if PEAD is well known, why does it not disappear? The main reasons usually given are implementation frictions, limits to arbitrage, and the fact that crowding is not the same as full elimination.

PEAD has been documented for decades and is still discussed as persistent, though its size can vary over time and across markets. Ball’s retrospective on Ball and Brown explicitly says the drift continued decades after the original discovery, and later reviews note evidence in multiple international markets as well.

A simple workflow looks like this: earnings release → measure surprise → rank stocks → hold winners / avoid losers.

A basic score is just the SUE itself. Then rank stocks cross-sectionally and focus on the top bucket. For example: score ≥ 90th percentile for positive-surprise names, or score ≤ 10th percentile for negative-surprise names.

A more practical version usually adds filters such as liquidity, minimum price, trend confirmation, and sector strength. This matters because raw academic PEAD strategies often look cleaner on paper than in live trading once costs and slippage are included.

This is an important distinction. A simple earnings-gap rule buys if the stock gaps up on the announcement. PEAD is broader. It says the information is not exhausted on day 1. The anomaly is about the post-announcement window, not just the opening reaction.

So the real claim is: strong surprise ⇒ continued abnormal return over subsequent days or weeks. That is why PEAD is more than an event-day setup. It is a drift phenomenon.

PEAD is not uniform. Research often finds stronger drift in places where information processing is slower or arbitrage is harder, such as smaller or less liquid stocks. Studies also link the drift to small-investor behavior and to institutional constraints.

That gives a useful intuition: less efficient corner of market ⇒ larger delayed reaction. This is one reason the anomaly is often more interesting outside the biggest, most-followed names.

Like most anomaly-based strategies, PEAD is not smooth. It tends to struggle when earnings expectations are already fully reflected, when macro shocks dominate firm-specific news, when implementation costs eat the edge, or when crowding compresses the drift.

So the anomaly can be real and still be hard to monetize perfectly. The literature repeatedly notes that PEAD is economically meaningful before transaction costs, but implementation matters.

A straightforward implementation captures the clean academic logic of PEAD well. The steps are: compute earnings surprise, keep only liquid names, rank stocks cross-sectionally, buy the top bucket, avoid or short the bottom bucket, and hold for a fixed post-event window.

The most interesting part of PEAD is that it says fundamental news is not always absorbed instantly. That is subtle.

A lot of factor research is price-based: momentum, relative strength, breakouts. PEAD says one driver of those moves may be slow incorporation of earnings information. That makes the anomaly conceptually rich. It connects accounting information, analyst expectations, investor psychology, and price formation.

The post-earnings announcement drift is one of the strongest examples of a market rule that links fundamentals directly to future returns.

Stocks with strong earnings surprises often keep moving in the same direction after the announcement. The reason is not that earnings are magical. It is that investors and institutions often do not fully process the news immediately, which creates a delayed adjustment in price.

That is why PEAD remains one of the most important bridges between accounting information and momentum investing. It shows that what matters is not just whether a company beats expectations, but how long it takes the market to fully absorb that information.