Market Fragmentation And Post Earnings Announcements Drift PDF Download

The stock market, according to the efficient market hypothesis, is informationally efficient in that prices instantly reflect all available public information. Prior financial literature on the study of the relationship between earnings announcements and their effect on the stock market reveals that there is a significant "drift" of a firm's cumulative abnormal return that occurs in the direction of its earnings surprise. This phenomenon is in contrast to how the efficient market hypothesis would expect the market to react to this new information. The prior studies on this topic were conducted in the 1980s - before the existence of both high-speed access to news via cell phone alerts and the increasing ability to trade quickly on new information via online brokers. This study attempts to test this "post-earnings announcement drift" on the current market to see if this phenomenon is still relevant in today's market and to see if it can be exploited. This study finds that there is still a post-earnings announcement drift that persists for the twenty-one days following earnings announcements. The cumulative abnormal returns continue to drift upwards for "good news" firms and continue to drift downwards for "bad news" firms for twenty-one days and may continue in the same direction after this period. This study also finds that a trading strategy that involves forming long portfolios of firms that beat earnings by the greatest magnitude (most positive earnings surprise) and also have the largest abnormal return on the day of the announcement and forming a short portfolio of firms that miss estimates by the greatest magnitude (most negative earnings surprise) and have the most negative abnormal return on the day of the announcement had an average annualized return of 20.343% over the ten year period starting in 2004 while the S & P 500 had an average annualized return of 9.1% over the same period.