Earnings announcement return extrapolation

We propose that extrapolative beliefs about earnings announcement (EA) returns may contribute to the understanding of EA return patterns. construct a theoretically motivated measure investors’ expectations based on stock’s recent history returns. then show this explains cross-sectional variation in stock and investor behavior around EAs. Stocks expected have high returns, according our measure, experience predictable increases prices before EAs decreases afterward. These patterns are economically significant: investors buy (sell) portfolio is long firms with short low pre-EA (post-EA) period earn daily five-factor abnormal 16.1 bps (18.3 bps). Using individual trades data institutional trading, we find more likely purchase stocks consistent at least subset forming