A multivariate analysis of the effect of bank holding company diversification into nonbank activities on the performance of holding company banks

The purpose of this research is to evaluate the impact of product diversification by bank holding companies on the performance of commercial banks in the United States. The major hypothesis tested is that the performance of congeneric holding company banks is different from the performance of independent banks or banks that are members of holding companies with only bank affiliates. The main hypotheses concerning performance differences are that congeneric holding company banks will be more business oriented and more risky than other commercial banks. The performance of congeneric holding company banks is expected to be different because it is believed that bank holding companies seek a portfolio effect in the acquisition and management of their affiliates and the impact of this diversification is reflected in the performance of the holding company banks. The hypothesis was developed by applying portfolio theory to the bank holding company. Tc test the hypothesis, a total sample of 147 commercial banks was selected from three distinct populations: 49 banks were affiliates of product diversified holding companies - one-bank holding companies with nonbank subsidiaries, 49 banks were affiliates of geographically diversified holding companies - multi-bank holding companies without nonbank subsidiaries, and 49 banks were not affiliated with a holding company. Fifty-eight ratios for each bank in the sample were computed from financial statement data. These ratios are believed to measure performance and were used as variables in the statistical analysis. The large number of variables relative to the sample size necessitated a reduction in the number of variables. Two different sets of variables were derived from the original set using principal components analysis and multiple regression analysis. Both of these variable sets were employed in the statistical analysis.