Trends in Organizational Form and their Relationship to Performance: The Case of Foreign Securities Subsidiaries of U.S. Banking Organizations

Debate about the effects of permitting U.S. commercial banks to expand their range of activities has intensified in recent years. Some observers worry that banks with access to a federal safety net have strong incentives to use new opportunities to take greater risks and increase their likelihood of failure at possible cost to the FDIC and taxpayers. Others fear that the safety net might give banks a competitive advantage relative to nonbank rivals. A key element of this debate is whether a holding company structure does a significantly better job of mitigating against these potential problems than a bank subsidiary alternative and should be made mandatory for banking organizations that want to engage in nontraditional activities. Unfortunately, hard, current empirical evidence on the benefits and costs of alternative structures generally is lacking. The purpose of this paper is to provide this sort of evidence. In the study, annual financial data for the 1987–1997 period for an unbalanced panel of foreign securities subsidiaries of U.S. banking organizations are used to investigate two questions: What factors influence how bank holding companies organize securities activities when they have a choice? And are the observed differences in organizational form related to significant differences in key measures of subsidiary performance? This sort of study is possible because U.S. banking organizations can and do engage in securities activities through subsidiaries of the bank as well as holding company affiliates. These subsidiaries also file financial reports with bank regulators. A probit model is used to empirically identify important factors influencing structural choice. Univariate and multivariate statistical techniques are used to determine whether or not differences in subsidiary structure are related to differences in subsidiary risk, funding costs, and efficiency. Simultaneity is investigated to a limited extent. In brief, the empirical results do not support the position of the holding company proponents. Safety net subsidy incentives don’t appear to be a primary determinant of structural choice. The evidence does not indicate that bank-owned securities subsidiaries tend to be more risky than holding company securities subsidiaries. Bank securities subsidiaries also do not appear to enjoy any funding advantage relative to holding company subsidiaries. These two results are particularly noteworthy because section 23A and 23B restrictions on intracompany funding currently do not apply to transactions between banks and their direct and indirect bank subsidiaries. Finally, some evidence indicates that bank subsidiaries tend to be more efficient.