Algorithmic Trading and Forward-Looking MD&A Disclosures

• 成果类型: Article
• 署名作者: Thomas, Wayne B.; Wang, Yiding; Zhang, Ling
• 署名单位: University of Oklahoma System; University of Oklahoma - Norman; University of Houston System; University of Houston; University of Houston Downtown
• 刊物名称: JOURNAL OF ACCOUNTING RESEARCH
• ISSN/ISSBN: 0021-8456
• DOI: 10.1111/1475-679X.12540
• 发表日期: 2024
• 页码: 1533-1569
• 关键词: voluntary disclosure; differential information; SHAREHOLDER LITIGATION; CORPORATE DISCLOSURE; earnings; liquidity; propensity; management; search

摘要

This study examines how algorithmic trading (AT) affects forward-looking disclosures in Management Discussion and Analysis (MD&A) of annual reports. We predict and find evidence that AT relates negatively to modifications in year-over-year forward-looking MD&A disclosures. This evidence is consistent with AT reducing investors' demand for fundamental information, which reduces managers' incentives to supply costly forward-looking disclosures. Cross-sectional tests provide additional evidence that this negative relation is more pronounced for firms with larger earnings surprises and those with losses. We further validate our conclusion by demonstrating that investors' fundamental information searches are a channel through which AT affects forward-looking disclosures. The conclusion is robust to using the SEC's Tick Size Pilot Program as an exogenous shock to AT and to using alternative disclosure measures (e.g., tone revisions and number of sentences in forward-looking MD&A disclosures). Overall, our study demonstrates that AT is a contributing factor to regulators' concerns over the diminishing usefulness of forward-looking information in MD&A disclosures.