This paper considers the aggregate performance of the banking industry, applying a modified and extended Bennet (1920) dynamic decomposition of bank return on equity as an illustration. The aggregate performance of any industry depends on the underlying microeconomic dynamics within that industry - adjustments within banks, reallocations between banks, entry of new banks, and exit of existing banks. Bailey, Hulten, and Campbell (1992) and Haltiwanger (1997) develop dynamic decompositions of industry performance. We extend those analyses to derive a Bennet decomposition that includes their decomposition as a special case. We also extend the decomposition, consider geography, and implement decomposition on a state-by-state basis, linking that geographic decomposition back to the national level. We then consider how deregulation of geographic restrictions on bank activity affects the components of the state-level dynamic decomposition, controlling for competition and the state of the economy within each state and employing fixed- and random-effects estimation for a panel database across the fifty states and the District of Columbia from 1976 to 2005.