This study attempts to investigate the impact of changes in bank governance on bank performance in Pakistan. Governance changes entail the privatization and restructuring of state-owned banks, and the merger and acquisition of private and foreign banks. Using the concept of frontier efficiency, we adopt an empirical framework that allows us to study the impact of all governance reform variables in the same model. First, we estimate a stochastic cost frontier model using unbalanced panel data on commercial banks for the period 1991-2005. Second, we decompose banks' total factor productivity (TFP) change into its different components, using the estimated frontier. In general, the results show an improvement in banks' cost efficiency following changes in bank governance. We note that governance changes bring about an improvement in banks' TFP vis-à-vis that of banks that did not undergo governance changes. We find a declining trend in TFP change (TFPC), which could be a consequence of the banking industry's increased profitability. We also note that bank selection for governance changes has a mixed effect on TFPC, while bank consolidation seems to be more effective in improving TFPC.