Expense preference in banking

The level and change of expense preference behavior in the banking industry over years 1977 through 1986 is studied. Expense preference behavior is one type of agency cost that occurs when managers make decisions in their own best interests rather than in the best interests of the stockholders. Expense preference behavior is encountered when managers are not the owners of the firm and therefore managers face temptations to spend resources for their benefit rather than for the productive benefit of the owners. Expense preference behavior can take many forms. One form includes excess numbers of staff. Other forms include excess expenditures on salaries, furnishings, and entertainment.

The motivation for using the years 1977 through 1986 is based on the changes that occurred in the banking industry over this period. The period is one in which the banking industry became more competitive. Competition increased both between banks, and between banks and other financial service firms. In addition, the period spans economic boom, bust and boom. A recession serves to accelerate the impact of increased competition on expense preference behavior.

The theoretical framework on which the study of expense preference rests is agency theory. Agency theory is discussed by several authors in the context of the theory of the firm.

Variables such as type of Federal regulator, multibank holding company membership, and level of competition are used in tests to determine their impact on the level of and change in expense preference behavior by bank managers over the 1977 through 1985 period. The tests are based on a model of perfect competition using a Cobb-Douglas production function. The model requires the use of the price of labor input, the price of capital input, and a measure of the size of a bank's output to be used as control variables to which each of the three test variables are added. Furthermore, banks are divided into large and small subsamples to accommodate differences in economies of scale.

The results show that the regulator, multibank holding company membership, level of competition, and the number of employees utilized affect the level of bank expenses. Differences in the results are found between banks in the large and small subsamples. Most significant is the finding that for large banks, expense preference behavior declined over the test period to the point where it may no longer be of major concern in this subsample.