On This Page
Key Takeaways
- The efficient market hypothesis says market prices reflect the information available to participants, so market value tends to track intrinsic value.
- Efficiency is described in three forms: weak form covers past prices and volume, semi-strong form adds all public information, and strong form adds private information.
- Weak-form efficiency concerns historical market data; semi-strong efficiency concerns public information. Evaluate abnormal returns against the selected risk and expected-return model.
- An event study prices the news: the stock’s 7.6% close-to-close return minus the 0.5% benchmark leaves a 7.1-point abnormal return around the report.
- Separate gross return, benchmark exposure and execution costs when evaluating a trading process.
What is the efficient market hypothesis (EMH)?
The efficient market hypothesis (EMH) is the theory that market prices reflect the information available to participants. Its return implication is stated relative to the expected return of a selected risk and pricing model: if prices already reflect an information set, trading on that information should not earn systematic abnormal returns. Researchers describe it in weak, semi-strong and strong forms.
Weak, semi-strong and strong form efficiency
Weak-form efficiency includes past prices and trading volume. Semi-strong efficiency adds public information such as earnings, filings and economic releases. Strong-form efficiency adds private information.
Formula
Abnormal return = actual return over the event window minus expected return from the benchmark model- A market-model calculation uses an intercept and market coefficient estimated for the chosen specification. Compare actual and fitted returns over the same event window.
- Abnormal returns that appear only at the announcement and then stop drifting are consistent with semi-strong efficiency.
- A drift that continues after the announcement is the pattern a trader tests: measure it against the same benchmark and after costs, then check whether a risk exposure explains it before calling it an edge.
Worked Example
A stock closes at $80 before an after-hours earnings report, opens at $86 and finishes the next session at $86.10. The index gains 0.5% over that same close-to-close window. Use a market model with intercept 0 and beta 1.
- Market-model return over the matched window: 0 + 1 × 0.5% = 0.5%, equivalent to $80.40 from the prior $80 close.
- Stock close-to-close return: $86.10 / $80 − 1 = 7.6%.
- Abnormal close-to-close return: 7.6% − 0.5% = 7.1 percentage points.
- The opening gap is $6, or 7.5%. A buyer at the $86 open earns only the next $0.10: $0.10 / $86 = 0.12% through the $86.10 close, before costs.
Most of the illustrated stock-price change occurs between the prior close and the $86 opening price. The full close-to-close abnormal return is 7.1 percentage points. An entry at the open participates only in subsequent price changes.
Abnormal returns vs ordinary price changes
Prices can change through the passage of time and required expected returns as well as through news. Efficiency does not require prices to remain fixed whenever no announcement occurs.
Raw outperformance and abnormal return answer different questions. A strategy with greater exposure to a rewarded factor can earn more raw return without earning an abnormal return under the model that prices that factor.
Event study example: an earnings gap
The stock closes at $80 before the announcement, opens at $86 and closes the next session at $86.10. The index returns 0.5% over the same close-to-close window.
With market-model intercept zero and beta one, the benchmark return is 0.5%. The stock returns 7.6%, giving an abnormal return of 7.1 percentage points.
The $6 opening gap is a 7.5% stock-price move. To split out its abnormal overnight piece, subtract the index’s return over the same overnight hours.
A buyer at the $86 open misses the $6 gap and earns the move to $86.10, 0.12% before costs. The chart shows when each price change happens: most of the news is in the price by the open.
How market efficiency is tested against a trading rule
An efficiency test also tests the expected-return benchmark. Specify the information available at the decision time, the execution rule, risk exposure and costs.
Post-event drift is the pattern to test: when a stock keeps moving after the news, measure that drift against the benchmark and after costs before trading it. Manager performance and historical-price rules require their own matched comparisons.
Market depth, information access, research coverage and trading frictions affect how information reaches prices. Information reaches prices more slowly where coverage is thin and frictions are high, so test a news-driven rule in those markets first.
Limitations of the efficient market hypothesis
- Check the benchmark before claiming an edge. An abnormal return like the example’s 7.1 points means either the market missed information or the benchmark left out a risk, so test the result against a second model before trading it.
- Judge efficiency market by market. It is defined for a specific information set, and depth, access, coverage and frictions differ across securities, so a rule that works in thin names needs its own test in liquid ones.
- Strip out the easy explanations first. Before trading an anomaly, subtract trading costs, adjust for risk and count how many rules you tested to find it; what survives all three is a candidate edge.
- Separate luck from skill. Chance and higher factor exposure both produce outperformance over a period, so require the result to repeat out of sample before sizing up.
Bottom line
The efficient market hypothesis is a framework for asking whether a trading edge is real: does a rule earn returns beyond what its risk and information set explain? Use it to test strategies against a matched benchmark, after costs, using only the information available at each decision time.
Clear answers
Frequently Asked Questions
Sources and Further Reading
How entries are sourced, calculated and updated: editorial methodology.