Common stock repurchases and agency costs of free cash flow

The purpose of this thesis is to test Jensen's hypothesis that common stock repurchases reduce agency costs of free cash flow by returning excess cash flows to shareholders. Two-day cumulative prediction errors are shown to be higher for firms identified as likely to have high free cash flow than for firms without high free cash flow for a sample of common stock repurchase announcements consisting of open market authorizations, intra-firm tender offers, and dutch auctions. This result is consistent with the argument by Jensen that common stock repurchases reduce agency costs of free cash flow. Corroborative evidence is sought by examining repurchase announcements made in conjunction with adoptions of Employee Stock Ownership Plans (ESOPs). Because of the unique interaction between the ESOP trust and the firm, all else equal, the cash payout of the ESOP repurchase is lower than the cash payout under the non-ESOP repurchase. Prediction errors for ESOP adoption-related repurchase announcements are found to be unrelated to high levels of free cash flow. This supports the hypothesis that ESOP adoption-related repurchases are less effective in reducing agency costs of free cash flow because they return less cash to shareholders. In a pooled analysis, however, differences in prediction errors between firms with and without high free cash flow are not significantly greater for the non-ESOP group than for the ESOP adoption-related group. The evidence suggests that returning cash flows to shareholders through common stock repurchases reduces agency costs of free cash flow, but, conclusive corroborative evidence can not be provided by the examination of ESOP adoption-related repurchase announcements.