26:10
The strategy that keeps MAKING BILLIONS to INSTITUTIONAL traders: PEAD.
MatFinOg
Overview
This video explains the Post-Earnings Announcement Drift (PEAD) phenomenon, a well-documented market inefficiency where stock prices continue to move in the direction of an earnings surprise for several weeks after the announcement. The presenter, a former market maker and hedge fund manager, details the academic research behind PEAD, dating back to 1968, and demonstrates how to construct a simple, mechanical trading strategy based on this research. The video covers data sourcing, strategy implementation using free tools, and backtesting results, highlighting that PEAD remains a viable edge for traders willing to follow an institutional approach.
How was this?
Save this permanently with flashcards, quizzes, and AI chat
Chapters
- PEAD is a market inefficiency where stock prices drift in the direction of an earnings surprise for weeks after the announcement.
- If a company beats earnings, the stock tends to rise; if it misses, the stock tends to fall.
- This contradicts the Efficient Market Hypothesis, which suggests all public information is immediately priced in.
- Institutional traders have profited from PEAD for decades, while many retail traders are unaware of it.
Understanding PEAD reveals a persistent market anomaly that can be exploited for potential trading profits, challenging traditional notions of market efficiency.
If a company reports earnings significantly higher than expected, its stock price may continue to climb for several weeks following the announcement, rather than immediately settling at a new equilibrium.
- The phenomenon was first documented in 1968 by Ball and Brown, showing markets don't fully react to earnings on announcement day.
- Bernard and Thomas (1989) formalized PEAD into a trading strategy, showing a drift lasting about 60 days.
- Livvenel and Mandol (2006) modernized the methodology by comparing analyst forecasts to time-series forecasts.
- A 2021 review confirmed PEAD persists globally across over 200 papers spanning 53 years, indicating its robustness.
The extensive academic research provides a strong, evidence-based foundation for the PEAD anomaly, validating its existence and historical significance.
The paper by Bernard and Thomas demonstrated that sorting stocks by earnings surprise and trading the spread between top and bottom performers yielded positive returns for 41 out of 48 quarters, with the drift lasting approximately 60 days.
- Market participants process surprising news gradually, with institutional investors adjusting models and positions over weeks.
- Sell-side analysts revise estimates over days, while retail investors may react emotionally to headline numbers.
- Structural limitations, such as large fund exposure limits and transaction costs, prevent full arbitrage.
- These 'frictions' allow the inefficiency to persist, as it's not easily exploited without incurring significant costs or risks.
Understanding the reasons for PEAD's persistence helps explain why market inefficiencies can exist even when they are known, due to practical trading constraints and behavioral factors.
Large hedge funds may be hesitant to take massive positions in every single stock experiencing an earnings surprise due to risk management constraints and the potential for high transaction costs, thus leaving the drift exploitable.
- A trading strategy requires entry conditions, exit conditions, and position sizing.
- Entry condition: Use a 'concordant signal' where both the earnings surprise and the stock's immediate price reaction align (e.g., beat earnings AND stock goes up).
- Exit condition: Based on research, hold the position for approximately 60 trading days, without stop-losses or take-profits.
- Position sizing: Start with a fixed percentage of capital (e.g., 10%) per trade to manage risk and avoid excessive single-stock exposure.
This chapter outlines a clear, rule-based methodology for implementing the PEAD strategy, emphasizing a mechanical approach derived directly from academic findings.
If a company reports earnings above consensus and its stock price increases on the announcement day, a long position is initiated. If it reports below consensus and the stock price decreases, a short position is initiated. If the surprise and reaction disagree, no trade is taken.
- Essential data includes price history and earnings data (announcement date, EPS, consensus estimate, timing flag).
- Free data sources like Zacks.com can be used, though quality may vary compared to paid providers.
- The process involves gathering data, potentially using Excel, and then coding the strategy.
- Large Language Models (LLMs) can assist in quickly generating code for the strategy based on defined rules.
This section provides practical guidance on acquiring the necessary data and leveraging accessible tools, including LLMs, to build and test the PEAD strategy, making it actionable for learners.
Manually copying earnings data from a free website like Zacks.com into an Excel spreadsheet to create a dataset for analysis, including the earnings surprise and the announcement timing.
- The strategy was backtested on 20 large US stocks across various sectors over eight years.
- The initial test, using a concordant signal and a 60-day hold, yielded a positive profit with a reasonable drawdown and Sharpe ratio.
- Testing variations showed that removing the concordant filter increased returns but also significantly increased drawdown.
- Filtering out small surprises (below 5%) combined with the concordant filter actually reduced performance, suggesting filters can overlap negatively.
Backtesting provides empirical evidence of the strategy's viability and reveals crucial insights into the interaction of different filters, demonstrating the importance of rigorous testing.
An initial backtest on Google showed a profitable equity curve, validating the strategy's potential, with trades correctly identified based on positive earnings surprises and subsequent upward price movement.
- Analysis revealed that the short leg of the strategy consistently lost money, a documented asymmetry in post-2010 markets.
- Companies often pre-announce bad news, leading to negative reactions being priced in before the official release.
- Removing the short leg and focusing solely on long positions (using the concordant filter) resulted in the cleanest and most robust performance.
- This long-only strategy confirmed the existence of a modest but real risk-adjusted edge, consistent with academic findings.
This refinement highlights a key market asymmetry and leads to a more effective and simpler strategy, demonstrating how iterative testing and understanding market dynamics can improve trading systems.
By removing the short-selling component and only taking long positions on positive earnings surprises with positive price reactions, the strategy achieved a net profit of $146,000 on a $100,000 starting capital over the backtest period.
Key takeaways
- Market inefficiencies like PEAD persist because of gradual information processing, behavioral biases, and structural trading frictions.
- Academic research provides a robust foundation for identifying and exploiting market anomalies.
- A mechanical, rule-based trading strategy derived from research can be more effective than discretionary approaches.
- The 'concordant signal' (earnings surprise matches price reaction) is a crucial filter for a successful PEAD strategy.
- Post-2010 market dynamics show an asymmetry where positive earnings surprises tend to drift more reliably than negative ones.
- Testing variations of a strategy, especially interactions between filters, is essential to avoid degrading performance.
- Accessible tools and free data sources can be leveraged to implement and test institutional-grade trading strategies.
Key terms
Post-Earnings Announcement Drift (PEAD)Market InefficiencyEfficient Market Hypothesis (EMH)Earnings SurpriseConcordant SignalAcademic Finance LiteratureArbitrageBacktestingDrawdownSharpe Ratio
Test your understanding
- What is the core principle of the Post-Earnings Announcement Drift (PEAD) and how does it challenge traditional market efficiency theories?
- Why does the PEAD phenomenon continue to exist despite being documented and known to institutional traders?
- How can a 'concordant signal' be used as an entry condition for a PEAD trading strategy?
- What is the significance of the observed market asymmetry where positive earnings surprises tend to drift more reliably than negative ones?
- How can learners leverage academic research and accessible tools to develop and test their own trading strategies based on market inefficiencies like PEAD?