Asymmetric Inefficiency in the Market Response to Non-earnings 8-K Information

• 成果类型: Article
• 署名作者: Li, Xing; Tan, Qin
• 署名单位: Xi'an Jiaotong University; City University of Hong Kong
• 刊物名称: CONTEMPORARY ACCOUNTING RESEARCH
• ISSN/ISSBN: 0823-9150
• DOI: 10.1111/1911-3846.12756
• 发表日期: 2022
• 页码: 1389-1424
• 关键词: ANNOUNCEMENT DRIFT; cross-section; FULLY REFLECT; stock-prices; SHORT SALES; news; INVESTMENT; analysts; disclosure; anomalies

摘要

This paper examines the pricing efficiency of 8-K filings for events other than earnings announcements. Since these filings provide timely information that is material to investors and explain variations in quarterly returns to a degree similar to other disclosures, understanding how the stock price absorbs their information is important for investors, regulators, and academics. By testing the statistical correlation between the immediate stock returns in response to these filings and subsequent stock returns before, during, and after the forthcoming earnings announcement, we find evidence of investor overreaction to good news but underreaction to bad news in the immediate window. Essentially, the price increases too much for good news but fails to decrease enough for bad news, resulting in overpricing for both. Most of the correction for this overpricing occurs in the period leading up to the forthcoming earnings announcement, while the rest happens during the announcement. Drawing on Miller (1977), we further illustrate that, in the presence of short-sale constraints, increase in investor disagreement spurred by interpretation difficulty is the most likely mechanism for the observed overpricing. We fail to find sufficient evidence in support of alternative mechanisms, including managerial disclosure strategies, analyst optimistic bias, and retail investor participation. This asymmetric mispricing for non-earnings 8-Ks contrasts with the symmetric mispricing commonly found for other types of disclosures, where investors either systematically underreact or overreact to public information. Our results could broadly speak to the pricing of other public information that is inherently difficult to interpret.