We investigate the abnormal returns and market-based risk effects of four Federal Reserve Board decisions to allow bank holding companies to engage in investment banking through Section 20 subsidiaries. Positive abnormal returns for commercial banks were observed for initial, limited powers granted by the Federal Reserve. However, authorization to engage in underwriting corporate debt and equity and subsequent expansion of potential revenues from underwritings produced negative abnormal returns and increases in risk.

The purpose of this paper is to provide evidence on the following question: Are there more banking offices available per person to furnish consumer and business services in branch banking states than in unit banking states? This question is a central part of a broader issue of what limitations should be placed on the ability of individual banks to branch. Indeed, in a recent review of the literature dealing with the branching question, and prepared for the Senate Banking Committee (McIntyre Committee), Guttentag [8] stated: “One of the most pervasive arguments for branch banking is that branch banks provide more office facilities than unit banking.” Yet the available evidence on the question is sparse and existing research contains methodological difficulties which make the findings of questionable value.

The commercial banking industry has been buffeted by a variety of forces in recent years. Alternating periods of intense monetary restraint and the severity of the 1973–74 economic contraction (especially as it affected the real estate industry), huge losses on loan portfolios, a heavy commitment of funds to less developed countries on the part of a few major banks, and the failures of a number of individual banks have created considerable discussion about the stability of the banking system. Questions have been raised about the risk involved in committing funds to the securities of banking organizations. Moreover, the importance of these questions has been underscored for bank management by the necessity for many banking organizations to raise substantial amounts of external funds to prevent further depletion of existing capital ratios.

A substantial amount of scholarly effort in recent years has been devoted to the determination of the relationship between banking structure and performance. In general, the results of these studies indicate that banking structure affects both the price and quantity of banking services, but, for practical policy purposes, the impact of banking structure is quite small. Yet, the results of these studies have been inconclusive and contradictory to a substantial degree.

The recent repeal of the Glass–Steagall Act in the US has cleared the way for commercial banks to enter the securities underwriting business. Many of the concerns that resulted in the original passage of the Glass–Steagall Act, however, still exist. One of these is the possible conflict of interest a universal bank faces. This paper provides evidence on this issue from the experience of Canada following its removal of restrictions on chartered bank ownership of investment dealers. Both ex ante bond yield comparisons between commercial and investment bank underwritten issues and equity price reactions to bond issue announcements provide no evidence of a conflict of interest.

We provide evidence regarding the extent to which commercial banking organizations that have entered investment banking have adopted pay-performance compensation systems that are like those used by investment banks. We find that pay-performance sensitivities for these banks once they begin securities underwriting are very similar to the sensitivities for commercial banks that have chosen not to enter investment banking. We also find that pay-performance sensitivities for both types of commercial banks are less than for investment banks.