UNTIL RECENTLY, discussion of monetary policy has focused on the overall reaction of the banking system to changes in policy variables originated by the central bank. Relatively little attention has been given to the complex factors that govern an individual bank's response to changes in monetary policy. Thus, while theories of financial behavior are implicitly formulated in terms of individual decision-making units, the attention of monetary research has been directed towards the behavior of aggregates. With the possible exception of a recent study by Bryan and Carleton [1], the testing of banking theory has generally been conducted in terms of aggregate demand and supply functions for money [2, 12]. Although the use of aggregate data in empirical research is valid, there comes a stage in the empirical analysis where some more information on the behavior of individual units is needed. This study departs from the general trend in that it concentrates on the behavior of the individual bank. Its purpose is to obtain empirical estimates of the parameters of a simple model of banks' asset choice. The model focuses attention on short run changes in the composition of the investment portfolio. Specifically it is argued that the demand for liquid assets is a function of asset size, interest rates and deposit instability. Our purpose is to estimate the sensitivity of liquid asset changes in bank portfolio to changes in these exogenous variables.