Expectations Management and Stock Returns

We show that proxies for firms’ incentives to manage earnings expectations toward beatable levels contain strong predictive power for earnings announcement returns. Firms with stronger incentives to manage expectations predictably underperform before, and subsequently outperform during, their expected earnings announcement months. This predictable V-shaped pattern in prices yields strategy returns exceeding 1% per month and suggests that firms manufacture positive “surprises” in high attention periods by walking down investors’ expectations ahead of their announcements. We validate our proxies by showing that firms with stronger incentives display several intuitive patterns; they are more likely to beat analysts’ forecasts even when earnings decline, more often narrowly beat expectations than narrowly miss, and issue low-ball earnings guidance. Taken together, our findings provide novel evidence that expectations management contributes to the prevalence of earnings announcement premia, as well as quarterly return seasonalities synchronized across firms’ fiscal periods. JEL Classifications: G10, G11, G12, G14, M40, M41 ∗We thank Brad Barber, Audra Boone, John Campbell, Aiyesha Dey, Grant Farnsworth, Jeff Harris, Charles Lee, Dawn Matsumoto, Hao Zhang and seminar participants at MIT, the Securities and Exchange Commission (SEC), American University, Texas Christian University, Cubist Systematic Strategies, Arrowstreet Capital, and University of Massachusetts for helpful comments and suggestions. Corresponding authors: Jinhwan Kim, [email protected], and Eric So, [email protected]. Expectations Management and Stock Returns 1