Glass–Steagall: legislation, limits and loopholes

Section 16 prohibited national banks from purchasing or selling securities except for a customer's account (i.e., as a customer's agent) unless the securities were purchased for the bank's account as "investment securities" identified by the Comptroller of the Currency as permitted national bank investments. Section 16 also prohibited national banks from underwriting or distributing securities.[2]

Section 16, however, permitted national banks to buy, sell, underwrite, and distribute US government and general obligation state and local government securities. Such securities became known as "bank-eligible securities."[2]

Section 5(c) of the 1933 Banking Act (sometimes referred to as the fifth Glass–Steagall provision) applied Section 16's rules to Federal Reserve System member state chartered banks.[3]

Section 20 prohibited any member bank of the Federal Reserve System (whether a state chartered or national bank) from being affiliated with a company that "engaged principally" in "the issue, flotation, underwriting, public sale, or distribution" of securities.[4]

Section 21 prohibited any company or person from taking deposits if it was in the business of "issuing, underwriting, selling, or distributing" securities.[5]

Section 32 prohibited any Federal Reserve System member bank from having any officer or director in common with a company "engaged primarily" in the business of "purchasing, selling, or negotiating" securities, unless the Federal Reserve Board granted an exemption.[6]

As explained in the article on the Banking Act of 1933, if the 1933 Banking Act had not been amended, it would have required all federally insured banks to become members of the Federal Reserve System.[21] Instead, because that requirement was removed through later legislation, the United States retained a dual banking system in which a large number of state chartered banks remained outside the Federal Reserve System.[22] This meant they were also outside the restrictions of Sections 16, 20, and 32 of the Glass–Steagall Act.[23] As described below, this became important in the 1980s when commentators worried large commercial banks would leave the Federal Reserve System to avoid Glass–Steagall's affiliation restrictions.[24]

Although Section 21 of the Glass–Steagall Act was directed at preventing securities firms (particularly traditional private partnerships such as J.P. Morgan & Co.) from accepting deposits, it prevented any firm that accepted deposits from underwriting or dealing in securities (other than "bank-eligible securities" after the 1935 Banking Act's "clarification"). This meant Section 21, unlike the rest of Glass–Steagall, applied to savings and loans and other "thrifts," state nonmember banks, and any other firm or individual in the business of taking deposits.[5] This prevented such "depository institutions" from being securities firms. It did not prevent securities firms, such as Merrill Lynch, from owning separate subsidiaries that were thrifts or state chartered, non-Federal Reserve member banks.[25] As described below, this became important when securities firms used "unitary thrifts" and "nonbank banks" to avoid both Glass–Steagall affiliation restrictions and holding company laws that generally limited bank holding companies to banking businesses[26] and savings and loan holding companies to thrift businesses.[27]

Section 21 was not the only Glass–Steagall provision that treated differently what a company could do directly and what it could do through a subsidiary or other affiliate. As described in the Banking Act of 1933 article, Senator Glass and other proponents of separating commercial banks from investment banking attacked the artificiality of distinguishing between banks and their securities affiliates.[28] Sections 20[4] and 32[6] of the Glass–Steagall Act, however, distinguished between what a bank could do directly and what an affiliated company could do.[29]

No bank covered by Section 16's prohibitions could buy, sell, underwrite, or distribute any security except as specifically permitted by Section 16.[2] Under Section 21, no securities firm (understood as a firm "in the business" of underwriting, distributing, or dealing in securities) could accept any deposit.[5]

Glass–Steagall's affiliation provisions did not contain such absolute prohibitions. Section 20 only prohibited a bank from affiliating with a firm "engaged principally" in underwriting, distributing, or dealing in securities.[4] Under Section 32, a bank could not share employees or directors with a company "primarily engaged" in underwriting, distributing, or dealing in securities.[6]

This difference (which would later be termed a "loophole") provided the justification for the "long demise of Glass–Steagall" through regulatory actions that largely negated the practical significance of Sections 20 and 32 before they were repealed by the GLBA.[30] The fact Sections 16, 20, and 32 only restricted Federal Reserve member banks was another feature that made the Glass–Steagall Act less than "comprehensive" and, in the words of a 1987 commentator, provided "opportunities for banking institutions and their lawyers to explore (or, perhaps more accurately, to exploit)."[31]