"Chicago Banking": Additional Comment

Abstract

In their study, Chicago Banking,' Schweiger and McGee treated branch banks and holding-company affiliated banks as equivalent for purposes of analysis. Among other criticisms of this study, Jacobs and Lerner objected to this grouping because company banking may be closer to unit banking than to branch banking in its operation.2 In his reply, Schweiger acknowledged this criticism, but did not offer any quantitative evidence to refute the statement.3 The purpose of this note is to furnish limited evidence that in certain areas of pricing, holding-company affiliates act more as offices of branch systems than as unit banks. Also, a tentative conclusion is made regarding the role that holding companies may play in integrating geographically distinct bank market areas. One stated advantage of a branchbanking structure compared with a unitbanking structure is that as a result of uniform pricing policies among branches of branch banks, geographically separate bank market areas tend to become more integrated.4 The same phenomenon need not be true in a unit-banking state even though holding companies operate in that state. Each holding-company affiliate operates under a different name, and customers of a holding-company affiliated bank may not be aware of the common ownership, or may not question differentials in prices even though they are aware of these differentials. Therefore, in practice, holding companies are in a better position than branch banks to charge different prices in different markets. The empirical question is, however, whether or not they do operate in this manner. This study was designed to ascertain whether holding companies do follow central pricing policies. All two-bank towns in Minnesota were selected for analysis. Three subsamples were then formed: (1) the first group includes one independent bank, randomly selected, from each town in which both banks were independent; (2) the second group is comprised of independent banks in two-bank towns in which one of the banks was a holding-company affiliate; and (3) the third group includes holdingcompany banks. If both banks in the town were affiliated with a holding company, one bank was randomly chosen. Two hypotheses were tested. First, if * Thanks are extended to Jerry Hollenhorst, Jim Duprey, and Richard Herder for their constructive comments.