Recent evidence suggests that announcements of bank holding company acquisitions result in wealth transfers from the bidding to target shareholders. Empirically, this is demonstrated through findings of negative average abnormal returns to bank holding company acquirers and positive average abnormal returns to targets on announcement. Using a sample of acquisitions from the early 1990s—a period marked by the removal of significant geographic entry barriers—this paper reexamines the issue by applying a general statistical model to the event study framework to more precisely measure abnormal returns. In particular, we model returns according to the GARCH process to control for time-varying volatility. With respect to the unconditional distributions of acquirer and target abnormal returns, our findings are consistent with prior research. Further investigation into the conditional distribution of acquirer returns finds that, on announcement, interstate acquisitions using the purchase method of accounting actually increased shareholder wealth for acquirers (on average) by 1.44%. However, over a longer event horizon, the most important determinants of acquirer abnormal returns appear to be the relative size of the transaction and the method of accounting.