Three essays on Corporate Finance and Insider Trading

Abstract

In chapter one, we introduce IPO Rival Insider Selling (IRIS) as a novel predictor of IPO underpricing and long-run performance. We argue that incumbent industry rivals possess private, dispersed information about upcoming IPO competition and trade on these insights ahead of IPO events. Aggregating pre-IPO insider trading imbalances among IPO rivals, we construct IRIS, which captures dispersed competitive information that attention or capacity constrained underwriters may miss. We find that IRIS identifies which IPOs eventually outperform: higher IRIS strongly predicts greater first-day underpricing and superior long-run returns. Consistent with informationally constrained underwriters and delayed market efficiency, these effects are driven by non-obvious IPO rivals and are most pronounced when market sentiment (attention) is high (low). Our findings highlight that rival insider trading provides real-time signals about future competitive dynamics and IPO outcomes.

In chapter two, utilizing a sample of 1,899 M&A events from 1996 to 2020, we observe positive and significant abnormal trading volumes in the option market of target firms’ rivals, particularly in out-the-money options. Our analysis further explores the underlying reasons for these patterns based on two major theories from existing literature: the acquisition probability theory, which suggests that rivals of target firms experience abnormal returns due to the increased likelihood of future acquisitions, and the collusion theory, which asserts that horizontal mergers lead to enhanced market power and, consequently, abnormal returns for rivals. Our findings support the acquisition probability theory.

In chapter three, we investigate how investor attention to insider trading disclosures confers an informational advantage in stock market. Using a novel web log data from EDGAR (the SEC’s Electronic Data Gathering, Analysis, and Retrieval system) on Form 4 visits, we show that more sophisticated investors, especially those engaging in high-frequency monitoring of EDGAR, are more likely to identify the most informative insider trading filings. Form 4 filings that attract higher human attention lead to stronger subsequent abnormal returns. The attention-return effect is most pronounced when there is greater information asymmetry, such as indirect insider trading, smaller firms, low analyst coverage, and generally firms with limited public visibility. We also find that when Form 4 filings are viewed by users or IP addresses previously flagged for automated downloading (indicating Human + Machine “Hybrid” information processing), the return reactions are even more pronounced. This suggests a synergistic benefit to combining human judgment with machine power in processing insider information.