By the end of this article, we hope that readers will have a better understanding of:
- How to quantify analyst forecasts
2. What are earnings surprises
3. How to form profitable trading portfolios with earnings surprises
Analyst forecast error
Analysts may forecast any company performance or accounting metric. The most common forecast that they issue is the earnings per share (EPS), which is the company’s quarterly or annual profits divided by the number of shares outstanding. Like all forecasts, analyst EPS forecasts are rarely accurate to the dot, and we measure the difference between their forecasted and actual EPS and call it the analyst forecast error (AFR).
AFR is interesting to analyze. In Part 1 of this series, we mentioned how analyst conflict of interests lead to biased forecasts, and AFR is the measure used. A biased forecast is one where the analyst’s mean forecast error for a company tends to be positively or negatively errored. The incentive for this bias is revealed by investigating his social networks or subsequent appointments in these firms.
Despite the presence of AFRs, we are still able to get a reliable estimate of the expected EPS of a firm by taking an average of all its issued analyst forecasts. The errors should attenuate via averaging, and the resulting number is the consensus forecast. This consensus forecast plays a very important role in the stock market because it is the best estimate of the expected value of the firm, given all existing information about the firm, and evaluated by specialists (informed participants).
What are earnings surprises?
Under the efficient market hypothesis, share prices in the stock market always reflect all relevant information about the firms in real time. This theory implies that stock prices should move in a significant manner only when new information is revealed.
Since the consensus forecast has incorporated such relevant information, any deviations between the actual EPS and consensus forecast constitute new information. If the actual EPS is higher than the consensus forecast, the market re-evaluates the firm as having higher earnings power than expected. If the actual EPS is lower than consensus, then the market re-evaluates the firm negatively. The difference between the actual EPS and consensus forecast is called the earnings surprise.
Forming profitable trading portfolios with earnings surprise
Empirically, we see strong correlations between the magnitude of earnings surprise and stock price movements. This observation implies that the more positive the surprise, the higher the stock price movement after the announcement. The more negative the surprise, the larger the plunge in stock price. The bulk of the price adjustments happens very quickly within the day, then continues to drift in their respective directions over the next few days. This drifting phenomenon, called the Post Earnings Announcement Drift (PEAD), contradicts the efficient market hypothesis and is an anomaly that is still observed today. The most widely accepted explanation is that investors underreact to news due to adverse behaviors, especially in situations of large surprises.
The earnings surprise and its PEAD presents some profitable trading opportunities that do not require heavy investments into computing technology:
- For each quarter of earnings announcements, rank the magnitude of earnings surprises by the cut-off points from the surprises in the last quarter.
- If the stock lies in the top rank, buy it.
- If the stock lies in the bottom rank, short it.
- Continue to build your portfolio in this manner and hold it until the week before the next quarterly announcement.
Empirically, the returns from this strategy are significant. However, pursue this strategy only in markets that are efficient and have fair disclosure regulations, such as in the US. The Regulation Fair Disclosure was promulgated by the US Securities and Exchange Commission (SEC) on August 2000 and prohibits any firm from divulging material information selectively to external stakeholders. This regulation ensures that any material information regarding the firm is announced to everybody at the same time, and nobody can profit from the information ex-ante due to their privileged positions. Without this regulation, stock prices would have reacted before any information is officially announced, thus rendering the above-mentioned strategy useless.
Dr. Jack Hong is the co-founder of Research Room Pte. Ltd., a management consulting and advanced analytics company that delivers complex prediction and decision-making capabilities for commercial, government, and not-for-profit organizations. Dr. Hong has extensive research and commercial experience in applying advanced empirical science to drive business, financial and policy value chains. He is concurrently an adjunct faculty with the Singapore Management University (SMU).