The reasons for the divergence of IPO lockup agreements

Most initial public offerings (IPOs) feature share lockup agreements, which prohibit insiders from selling their shares for a specified period of time following the IPO. However, some IPO firms agree to have a much longer lockup period than other IPO firms, and some are willing to lockup a much larger proportion of shares. Thus, the primary research question for this study is: "What are the reasons for the divergence of the lockup agreements?";The two main hypotheses that this dissertation investigates are the signaling hypothesis based on information asymmetry, and the commitment hypothesis based on agency theory. This study uses methods that have not been applied by previous studies in the literature relating to IPO lockups.;First, I directly use IPO firms operating performance as a proxy for firm quality. The results show neither a negative nor a strong positive relationship between lockup length and firm operating performance. Thus, based on operating performance, the evidence does not support the agency hypothesis while showing weak support for the signaling hypothesis.;I then examine the long-run returns for IPO firms with different lockup lengths. I find that firms with short lockup lengths have much better long-run returns than firms with long lockup lengths. Therefore, the results reject the signaling hypothesis while supporting the agency hypothesis. This dissertation further contributes to the IPO long-run underperformance literature by showing that firms with a high agency problem have much worse long-run returns than those with a low agency problem.;Finally, I investigate the short-term stock returns around lockup expiry. Generally, I find that firms with short lockup periods experience better stock returns around lockup expiry than firms with long lockup periods, though the returns are not significantly different from one another. Overall, I conclude that the results reject the signaling hypothesis while partially supporting the agency hypothesis. In addition, I show that firms with high agency problems have much worse stock returns than those with low agency problems around lockup expiry, even though the agency variable is not significant in the regression analysis.