AN EXAMINATION OF THE INFLUENCE OF DEPOSIT BEHAVIOR ON THE PERFORMANCE OF COMMERCIAL BANKS' ASSETS*

Abstract

THE HYPOTHESIS THAT the deposit behavior of commercial banks is a significant determinant of commercial banks' asset distribution, and therefore of earnings performance, is developed and tested in this dissertation. Support for this hypothesis is found in banking literature which delineates primary, secondary, and investment reserves (i.e., assets) to which cash available is allocated on the basis of the banks' needs to meet possible withdrawals and loan demands. If this hypothesis were confirmed by banking practice, it might well indicate to less profitable banks that, by stabilizing deposit behavior, perhaps through better diversification of depositors, profits could be improved. As part of a Research Fellowship, the Federal Reserve Bank of Chicago provided data for testing this hypothesis. In testing the hypothesis, the performance of assets is measured for the year 1962 by the ratios of total operating revenues and net current earnings to total assets, each adjusted to exclude earnings which cannot be attributed to asset distribution. Excluded from consideration were service charge income on deposit accounts, other revenue, trust revenue, capital gains and losses, valuation reserves, and taxes. Comparison of the revenue and earnings ratios, which differ only by operating expenses, indicates that operating expenses do influence net current earnings. The influence of factors affecting total operating revenues does not necessarily carry through to net current earnings. Using data for a sample of thirty banks drawn from the cash grain area of Illinois, six independent variables purporting to measure the magnitude and duration of fluctuations and the trend of demand and time deposits were regressed on each of these two measures of asset performance. These independent variables were measured for the four years, 1959 through 1962, three years, 1960 through 1962, two years 1961 through 1962, and one year, 1962. Results of the multiple regressions indicate that these measures of deposit behavior in the periods studied were not significant predictors of asset performance as measured by either total revenue or earnings ratios. Measuring of independent variables for four, three, two, and one year periods made little difference. The data failed to support the hypothesis. The fact that some of the more stable banks in the sample were not lending up to their capacity, as measured by the independent variables, while other, less stable banks had substantially higher volumes of loans to total assets, contributed to this refutation of the hypothesis. Two possible conclusions stand out: