Asymmetric Learning from Prices and Post-Earnings-Announcement Drift

• 成果类型: Article
• 署名作者: Choi, Jaewon; Thompson, Linh; Williams, Jared
• 署名单位: University of Illinois System; University of Illinois Urbana-Champaign; Yonsei University; University of Texas System; University of Texas El Paso; State University System of Florida; University of South Florida
• 刊物名称: CONTEMPORARY ACCOUNTING RESEARCH
• ISSN/ISSBN: 0823-9150
• DOI: 10.1111/1911-3846.12477
• 发表日期: 2019
• 页码: 1724-1750
• 关键词: FULLY REFLECT; stock-prices; Cash flows; conservatism; INFORMATION; perception; anomalies; RISK

摘要

Motivated by research in psychology and experimental economics, we assume that investors update their beliefs about an asset's value upon observing the price, but only when the price clearly reveals that others obtained private information that differs from their own private information. Specifically, we assume that investors learn from the price of an asset in an asymmetric manner-they learn from the price if they observe good (bad) private information and the price is worse (better) than what is justified based on public information alone. We show that asymmetric learning from an asset's price leads to post-earnings-announcement drift (PEAD), and that it generates arbitrage opportunities that are less attractive than alternative explanations of PEAD. In addition, our model predicts that PEAD will be concentrated in earnings surprises that are not dominated by accruals, and it also predicts that earnings response coefficients will decline in the magnitude of the earnings surprises.