The earning-announcement effect on the stock price movement is one of the most puzzling issues in the region of financial research as no firm-specific performance measure is more widely analyzed than accounting earning information. Many earlier authors find stock price under-reaction to earning-announcement and attempt to give possible explanation to this anomaly. However, there are also scholars who claim that their empirical evidence is indicating an over-reaction of stock price to earnings.
There are many evidences consistent with the under-reaction of stock price to earnings announcement. Post-earning-announcement drift has been confirmed in almost 20 literatures. The anomalies of “Value Line Enigma” (Copeland and Mayers, 1982) as well as “PE effect” (Basu, 1977) are also partially due to post-earning-announcement drift. Many literatures support the explanation of shifts in the risks of firms with extreme surprise to post-earning announcement (Bernard and Thomas, 1989). Bernard and Thomas (1990) suggest that post-announcement drift exists because of the failure of stock price to reflect fully the implications of current earnings for future earnings. Research design flaws are partially reason for the drift.
Among recent evidence, some authors focus on the effect of investor's reaction on the post-earnings-announcement drift. Bartov, Radhakrishnan and Krinsky (2000) find that the drift is lower for companies with higher proportion of institutional investors. Later on, Mikhail, Walther and Willis (2003) demonstrate that companies with more experienced analysts have smaller drift. Companies with higher arbitrage risks have stronger drift (Mendenhall, 2004). Moreover, other evidences examine the relationship between the post-earning-announcement drift and firm characteristic.
By contrast, Debondt and Thaler(1987) and Ou and Penman(1989) provide evidence consistent with the stock price over-reaction to earning-announcement. However, Ball and Kothari (1989), Klein (1990) challenge the conclusions by DeBondt and Thaler(1987), which are confirmed by DeBondt and Thaler(1990) and Chopra, Lakonishok and Ritter(1991). But the evidence from Chopra, Lakonishok and Ritter (1991) is uncertain, too early to arrive at conclusion, to represent the over-reaction to earning-announcement since it seems to contradict to the conventional theories of market efficiency and asset pricing models.
The structure for the rest of this paper is as follows. Section 2 is literature review. Section 2(2.1) describes the evidence of post-earning-announcement and existing literature explanation to under-reaction to earnings reports, while section 2(2.2) gives plausible interpretation of over-reaction to earnings. Finally, a conclusion and some future research proposal are included in the last section.
Evidence And Existing Explanation Of Stock Price Under-Reaction To Earning-Announcement
Abundant empirical studies document that stock price exhibit a significant variation with the earnings announcement. Since it is documented by Beaver (1968), the phenomenon of abnormal return surrounding the earnings announcement day has to be started. Ball and Brown (1968) firstly investigated the post-earning-announcement drift. The same results are also documented by Jones and Lizenberger (1970), Watts (1978) and Latane and Jones (1979).
Then in the 1980s, consistent results are found. Patell and Wolfson(1984) provide evidence that abnormal return are largest within the first 30 minutes after the announcement becoming public, especially within the first 5 to 10 minutes. Foster, Olsen and Shevlin (1984) constructed10 portfolios on the size of earnings surprise and analyzed that Positive (negative) abnormal returns for positive (negative) earnings and cumulative average return (CAR) of positive (negative) surprise continue to grow (decline) even after the information became public. At least 15 studies obtain the same conclusion.
However, according to discussions by Ball (1978), Foster, Olsen and Shevlin (1984), and Bernard and Thomas (1989), limitations and methodological problems may bias the empirical results, challenging the previous explanations to the post-earnings announcement drift. But Bernard and Thomas (1989) find significant post-earnings announcement drift even after controlling for the research design flaws observing a sample of 85,000 observations in NYSE and AMEX stocks over 1974-1986 periods.
Larger abnormal return may exist when the way changes, in which the research approach is designed. Forster, Olsen and Shevlin (1984) show absolute magnitude of the drift is larger for smaller firms. Later on, Freeman (1987) as well as Ball and Kothari (1991) demonstrate that the timing and the magnitudes of stock price reaction to earnings announcements are dependent on form size. Recently, similar results are verified by Jegadeesh and Livnat (2006) reporting that the post-announcement drift is significant for small firms but not for large firms.
To explain the causes of the post-earning-announcement drift, many studies existing attempt to give a plausible explanation considering a variety of variables.
Plausible risk is an explanation for post-earning-announcement drift, indicating that it is of failure to completely control for firm's risk, which means that the risk is higher for firms with good news, and risk is lower for firms with bad news, than the compared portfolios.
However, Chen, Roll and Ross (1986) found no evidence to demonstrate that long positions in stocks with good news during the post-announcement period were risky along any of 5-dimensions using arbitrage pricing theory.
Moreover, Bernard and Thomas (1989) assessed the possibility of the risk-based explanation to post-earning-announcement and provided evidence as skepticism to the risk-based explanation.
Firstly, Bernard and Thomas (1989)tested the Ball-Kothari-Watts hypothesis supporting the risk-based explanation and found evidence that such risk shifts do occur in the predicted direction though the Ball-Kothari-Watts methodology, and part of the risk shift persisted into the post-announcement period. But such shifts were only 8 to 13 percent as large as necessary to completely explain the magnitude of the post-earning-announcement drift. Secondly, empirical evidence from Bernard and Thomas (1989) showed 13times positive predicted abnormal return during period of 1974 to 1986 by observing the SUE portfolios. SUE is the difference between the predicted earnings and the actual scaled by the historical standard deviation of the forecast errors, based on the model proposed by Forster (1977). And the evidence indicates that any risk of the SUE portfolios had not surfaced in the form of loss on matter what the macroeconomic conditions were. Thirdly, stocks with bad news would provide valuable hedge in excess of the cost the stocks' risk. Bernard-Thomas (1989) proposed that the SUE strategy can be risky only when the loss during 1974 to 1986 is anomalous or only when the SUE portfolios risk increases the losses, which were not found during 1974-1986.
But combined studies by Latane, Jones and Rieke (1974) and Rendleman, Jones and Latane (1982) as well as the Bernard and Thomas (1989) based on different methodologies showed that positive abnormal return can be generated in each year during 1965-1986 for SUE portfolios and concluded that post-earnings announcement drift can be partially explained by risk shifts.
But results from Ball, Kothari and Watts (1990) showed much larger difference in risk over 9 months after earning announcement between top deciles and bottom deciles earnings performance than the estimated difference by Bernard and Thomas for the first 3 post-announcement quarters, which cannot supply significant implications on the extent to the post-announcement drift explained by risk due to used annual earnings data magnifying power of firms' risk premium.
Another explanation for post-earning announcement is market frictions, including transaction cost and implementation problems.
Since Bernard and Thomas (1989) obtain abnormal return for SUE portfolios with no implementation problems, at least prior to consider transaction costs and short-sale restraint, it is likely that transaction costs could not offset the abnormal return from SUE portfolios. According to Stoll (1991), the abnormal return during the post-earning-announcement drift is much larger than the roundtrip transactions costs, for example, 1.2% for individual and 0.75% for institutions. The costs could be higher for aggressive traders bearing a higher share of the bid-ask spread even if the transaction costs are not as large as necessary to offset anomalous return form post-announcement drift. Therefore, the market frictions indeed influence investors' action.
Besides the explanations above, Bernard and Thomas (1990) and Wiggins (1991) tested the possibility of the explanation that the post-earning-announcement drift is caused by the failure to fully reflect the current earning implications for future earnings. Summarized evidence was consistent with hypothesis that stock price reflected partially a naïve earning expectation: that future earnings were similarly earnings for the comparable quarter of the prior year. Their findings are as follows. Firstly, evidence provided obstacles to argument that methodological errors of forecasting expected return model could fully explain post-earning-announcement drift. Secondly, evidence summarized here showed that market inefficiency is the result of stock price failing to reflect the earnings information immediately, indicating delayed reaction hypothesis.
Recently, studies about the drift have investigated the effect of factors in terms of the investors' reaction and some other financial factors, attempting to explain the effect of those factors associated with various drift levels on the post-earning-announcement drift.
According to Han (2000), the drift could be smaller for large companies than for small companies basing the arbitrage risk model, since the small firms have a more opaque environment for delivering information. Moreover, Bartov, Ranhakrishnan and Krinsky (2000) showed that firms with higher proportion of institutional investors have lower drift as institutional traders were more sophisticated and less likely to under-react than individuals. And according to Mikhail, Walther and Willis (2003), smaller drifts belong to companies with more experienced analysts, who were more likely to completely incorporate earning information to their own estimation rather than less experienced analysts. Additionally, Jegadeesh and Livant (2006) found evidence that the drift depend on the contemporaneous magnitude of the revenue surprise. What is more, they also document that revenue surprise indeed predict more persistent future earnings growth and even after controlling for past earnings surprise, earnings growth is positively related to past revenue surprise.
However, Hirshleifer, Myers and Teoh (2008) provide evidence showing that individual investors' reaction cannot explain the post-earnings-announcement drift.
Overall, there is still no plausible explanation with identification to fully interpret the post-earning-announcement drift.
Stock Price Overreaction To Earning-Announcement
By contrast, other literatures are documented to suggest over-reaction to earning announcement.
In 1987, DeBond and Thaler provide evidence that stocks with previous negative performance would outperform stocks with positive prior performance using data for 4 years （1970-1973 and 1980-1983）,which show that losers outperform winners by 37 percent during test period. This is the first study to include the over-reaction to earning information. Then Ou and Penman (1989) showed consistent results with DeBond and Thaler. They construct portfolios based on “Pr” measure, which is the outcome of comprehensive fundamental analysis, and observe data from 1973-1974 as well as 1983-1984 to conclude that stocks of firms with high Pr would outperform the stocks of firms with low Pr, just like losers outperform winners documented by DeBondt and Thaler.
However, Ball and Kothari (1989) argue that the abnormal return is not obvious when controlling for the risk shifts based on the DeBondt and Thaler methodology.
But Chopra, Lakonishok and Ritter (1991) argue that underestimation of abnormal return from Ball and Kothari by assuming that risk premium for stock is the same as the excess return of the market portfolio. And they also find that after controlling for size, abnormal return are obvious in January and larger for smallest deciles firms but nearly zero for large firms, as though over-reaction happens solely without segment of market dominated by institutions traders. Another finding is that why the market would be astonished by the subsequent earning information is because the losers' expectation for future earning is pessimistic, which seems a correction of the previous over-reaction.
But doubt on over-reaction to earnings still remains. Zarowin (1989) construct portfolios of previous winners and losers using earning information annually to find that over-reaction hypothesis holds when analyzing market-adjusted returns but the hypothesis break down when controlling for risk and size. Moreover, Klein (1990) provides evidence supporting under-reaction rather than the over-reaction.
Recently, empirical evidence provided by Tawatnuntachai and Yaman (2007) using data from 986 firms during 1983-1998 show that there is no overreaction to earnings announcement. But Mahani and Poteshman (2008) investigate investors' reaction to earnings on value and growth stocks to find that unsophisticated investor over-react to past news on underlying stock by entering option position that load up on growth stock related to value stocks till earnings announcement, which may affect the stock price reaction overly to earnings.
Overall, over-reaction to earning announcement is still a controversial topic. Over-reaction does not hold with restriction of the CAPM model and risk shifts, but when the restriction on CAPM is released, the over-reaction hypothesis holds Continual studies may investigate the over-reaction with various situations that can affect the research results.
Reconcile Evidence For Over-Reaction To Earnings With Evidence For Under-Reaction To Earnings
There are several ways to reconcile the evidence for over-reaction with the conflicted evidence for under-reaction.
Firstly, over-reaction in the stock market may be caused by various factors, including random deviations of price from the intrinsic stock value, but not be due to earning information. And little evidence found to show direct relationship between abnormal return and over-reaction to earnings, even if DeBondt and Thaler (1987) as well as Chopra, Lakonishok and Ritter (1991) provide evidence of over-reaction. And even Klein (1990) found evidence more consistent with under-reaction.
Secondly, stock price with under-reaction to earnings could be found to over-react to earnings solely under situation too complicated to captured by a partition on previous earning changes, which explaining the difference between research designs, for example, research designs based on prior earnings changes indicate under-reaction (Bernard and Thomas, 1989) while research designs on basis of previous return suggest over-reaction (Chopra, Lakonishok and Ritter, 1991).
Thirdly, the market reaction to earnings information may defy a simple characterization as under-reaction or over-reaction. The abnormal return during the post-earning-announcement may carry on moving in the same direction if the stock price reflect naïve earnings expectations described by Bernard and Thomas (1990), indicating that the drift seems to be a completely initial under-reaction. Moreover, if the stock prices reflect the trend of earnings to attract a level less extreme than the earnings originally announced, a reversal of previous abnormal return would surface.
Conclusion And Future Research Proposal
In summary, evidences discussed above for the stock price reaction to earnings announcement are still controversial.
Many plausible explanations for the drift exist in terms of research design flaws, market frictions (Bernard and Thomas, 1989), failure to control for risk (Bernard and Thomas, 1990 and Wiggins, 1991), investors reaction (Bartov, Ranhakrishnan and Krinsky, 2000), but debate on the initial incomplete reaction to earnings as the cause of post-earning-announcement drift still remains and no fully explanation exists. Other existing studies provide evidence both for and against the stock price over-reaction to earnings information. For instance, no evidence showed to indicate over-reaction to earnings (Ball and Kothari, 1989), while Mahani and Poteshman (2008) show evidence to support over-reaction. It is still too early to give conclusion to the phenomenon of over-reaction.
Researches above use data from different markets and periods, which make evidence not as valid as necessary. Additionally, there are no evidence could explain why transactions irrelevant for earning-announcement in stock market cannot eliminate the drift, which leave much potential space for future research. And evidences above as threat to market efficiency hypothesis exclude some factors uncovered by the empirical research, which may shade the over-reaction or under-reaction to earnings. For instance, transaction environment in terms of macroeconomics and microeconomics are quite different across industries and significant different between matured and emerging markets. Therefore, there are remains potential for future research in the area of the subject.
Moreover, as global integration becomes the primary tendency in stock markets, interaction among stock market is more and more intense and fast. For investors, it is necessary for them to consider the emerging situations such as new technology used in the process of transaction prior to make trading strategy in stock markets.
Ball, R. “Anomalies in Relationship Between Securities' Yields and Yield-Surrogates”, Journal of Financial Economics (June-September, 1987), pp. 103-126.
Ball, R., and P. Brown, “An Empirical Evaluation of Accounting Income Numbers”, Journal of Accounting Research (Autumn, 1968), pp. 159-178.
Ball, R., and S.P. Kothari, “Nonstationary Expected Returns: Implications for Test of market efficiency and serial correlation in returns”, Journal of Financial Economics
(November, 1989), pp. 51-74.
Ball, R., and S.P. Kothari, “Security Returns Around Earnings Announcements”, The Accounting Review (forthcoming, 1991).
Ball, R., S.P. Kothari, and R. Watts, “The Economics of the Relation Between Earnings Changes and Stock Returns” Working Paper, University of Rochester (December, 1990).
Bartov, E., S. Radahkrishnan, and I. Krinsky. “Investor Sophistication and Patterns in Stock Return after Earnings Announcement”, Accounting Review, vol.75, no. 1 (January, 2000), pp. 43-63.
Basu, S. “Investment Performance of Common Stocks in Relation to Their Price-Earnings Ratios: A Test of the Efficient Markets Hypothesis”, Journal of Finance (June, 1977), pp. 663-682.
Beaver, W.H., “The Information Content of Annual Earnings Announcement”, Journal of Accounting Research (Supplement, 1968), pp. 67-92.
Bernard, V., and J. Thomas, “Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium?”, Journal of Accounting Research (Supplement, 1989), pp.1-36.
Bernard, V., and J. Thomas, “Evidence that Stock Prices Do Not Fully Reflect the Implications of Current Earnings for Future Earnings”, Journal of Accounting and Economics (December, 1990), pp. 305-341.
Brown, L., and Han, “Do Stock Price Fully Reflect the Implications of Current Earnings for Future Earnings for AR1 Firms?”, Journal of Accounting Research (2000), vol. 38, no.1 (Spring): pp. 149-164.
Bodie, Z., A. Kane, and A. Marcus, Investments, Irwin, Homewood, IL (1989).
Chen, N., R. Roll, and S. Ross, “Economic Forces and the Stock Market”, Journal of Business (July, 1986), pp. 383-404.
Chopra, N., J. Lakonishok, and J. Ritter, “Performance Measurement Methodology and the Question of Whether Stocks Overreact”, Working Paper, University of Illinois (January, 1991).
Copeland, T., and D. Mayers, “The Value Line Enigma (1965-1978): A Case Study of Performance Evaluation Issues”, Journal of Financial Economics (Vol. 10, 1982), pp. 289-321.
DeBondt, W.F.M. and R.H. Thaler, “Does the Stock Market Overreact?”, Journal of Finance (July, 1985), pp. 793-805.
DeBondt, W.F.M. and R.H. Thaler, “Further Evidence of Investor Overreaction and Stock Market Seasonality”, Journal of Finance (July, 1987), pp. 557-582.
DeBondt, W.F.M. and R.H. Thaler, ”Do Security Analysts Overreact? ”, American Economic Review (May, 1990), pp. 52-57.
Foster, G., “Quarterly Accounting Data: Time Series Properties and Predictive-Ability Results”, The Accounting Review (January, 1977), pp. 1-21.
Foster, G., C. Olsen, and T. Shevlin, “Earnings Releases, Anomalies, and the Behavior of Security Returns”, The Accounting Review (October, 1984), pp. 574-603.
Freeman R.N., “The Association Between Accounting Earnings and Security Returns for Larger and Small Firms”, Journal of Accounting and Economics (1987): pp. 195-228.
J.N. Myers, L.A. Myers and S.H. Teoh, “Do individual investors Cause Post-Earnings-Announcement Drift?”, The Accounting Review (December, 2008), vol.83, issue 6.
Jegadeesh, Narasimhan, and Joshua Livnat, “Revenue Surprise and Stock Returns”, Journal of Accounting and Economics (2006).
Klein, A., “A Direct Test of the Cognitive Bias Theory of Share Price Reversals”, Journal of Accounting and Economics (July, 1990), pp. 155-166.
Kochari, S.P., “Capital Market Research in Accounting”, Journal of Accounting and Economics (September, 2001), vo;.31, nos. 1-3: pp. 717-724.
Latane, H., C. Jones, and R.D. Rieke, “Quarterly Earnings Reports and Subsequent Holding Period Returns”, Journal of Business Research (April, 1974), pp. 119-132.
Latane, H., and C. Jones, “Standardized Unexpected Earnings---1971-1977”, Journal of Finance (June, 1979), pp. 717-724.
Livnat, Joshua, and Richard Mendenhall, “Why is the Post-Earnings-Announcement Drift Larger for Surprises Calculated from Analyst Forecasts?” Working Paper, New Youk University (2004).
Mendenhall, R., “Evidence on the Possible Underweighting of Earnings-Related Information”, Journal of Accounting Research (Spring, 1991), vol.29, no.1: pp. 170-179.
Mendenhall, R., “Arbitrage Risk and Post-Earnings-Announcement Drift”, Journal of Business (October, 2004): pp. 875-894.
O. Tawatnuntachai and D. Yaman, “Do investors over-react to earnings warnings?”, Review of Financial Economics (2007): pp.177-201.
Ou, J., and S. Penman, “Financial Statement Analysis and the Prediction of Stock Returns”, Journal of Accounting and Economics (November, 1989a), pp. 295-330.
Ou, J., and S. Penman, “Accounting Measurement, Price-Earnings Ratios, and the Information Content of Security Prices”, Journal of Accounting Research (Supplement, 1989b), pp. 111-144.
Patell, J. and M. Wolfson, “The Intraday Speed of Adjustment of Stock Prices to Earnings and Dividend Announcements”, Journal of Financial Economics (June, 1984), pp. 223-252.
Rendleman, Jr., R.J., C.P. Jones, and H. A. Latane, “Empirical Anomalies Based on Unexpected Earnings and the Importance of Risk Adjustments”, Journal of Financial Economics (1982), pp. 269-287.
R. S. Mahani and A. M. Poteshman, “Overreaction to stock market news and misevaluation of stock prices by unsophisticated investors: Evidence from the option market”, Journal of Empirical Finance (2008): pp.635-655.
Stoll, H., “Measuring Equity Trading Costs”, Working Paper, Vanderbilt University (1991).
Watts, R.S., “Systematic 'Abnormal' Return after Quarterly Earnings Announcements”, Journal of Financial Economics (June-September, 1978), pp. 127-150.
Wiggins, J.B., “Do Misperceptions about the Earnings Process Contribute to Post-Announcement Drift?” Working Paper, Cornell University (1991).
Zarowin, P. “Does the Stock Market Overreact to Corporate Earnings Information?” Journal of Finance (December 1989), pp, 1385-1400.