This chapter will examine the regulation of the Big Five investment banks in the context of the changes which took place in the structure of banking after the repeal of the Glass-Steagall Act and the introduction of the European Union's Consolidated Supervision Directive in 2004.

Immediately after the financial crisis, various reasons were found for the failure of so many banks, and indeed for the collapse of Lehman Brothers. This is despite the obvious fact that the major investment banks were stand-alone investment banks. One of the most popular scapegoats was the repeal of the 1933 Glass-Steagall Act, which had forbidden commercial banks, their holding companies and affiliates from undertaking investment banking activities, while investment banks were forbidden from accepting deposits and thus acting as if they were commercial banks. The barriers imposed by the Act remained unchallenged until the 1980s, when both commercial banks and investment banks began to offer a wider range of services, including for households, assisting in stock purchases and sales. They also began to offer corporate financial services to businesses, such as the private placement of securities issuances or assistance with mergers and acquisitions. Investment banks encouraged larger corporations to raise capital through bond issuances, rather than through loans from commercial banks. These changes continued throughout the 1980s, when in 1987, the Federal Reserve first allowed commercial banks to undertake a limited amount of underwriting of corporate securities. The limits were further relaxed in the 1990s. ‘Finally, Citicorp's 1998 acquisition of the Travelers Group, which encompassed a diverse portfolio of insurance operations and a major investment bank, Salomon Smith Barney, which the Federal Reserve allowed on an interim basis, provided the imminent impetus for Congress to act.’¹

The Gramm-Leach-Bliley Act 1999 did not ‘repeal’ the Glass-Steagall Act in its entirety. It is important to be clear about what exactly it did and did not repeal. Only if that is understood is it possible to see that any diagnosis of the causes of the financial crisis which relies on that interpretation will lead to the wrong conclusions about the remedies.

Relevance of the Gramm-Leach-Bliley Act 1999

Resolving many years of controversy about the nature of financial competition and regulation, the 106th Congress passed the Gramm-Leach-Bliley Act (GLBA) by overwhelming majorities in November 1999, and President Clinton signed the legislation a few days later. The Act is very comprehensive, addressing affiliations of banking, insurance, securities firms and regulation of the resulting organizations, securities and insurance regulation, financial privacy and the modernization of the Federal Home Loan Bank System, amongst many other issues. The Act also requires a number of regulatory agencies to develop new regulations to implement it.

What exactly did the Gramm-Leach-Bliley Act repeal?

The view that the Glass-Steagall Act was repealed by the 1999 Act is entirely misconceived.² The Glass-Steagall Act is still applicable to banks insured by the Federal Deposit Insurance Corporation (FDIC), and still prevents such banks from underwriting or dealing in securities. It left intact section 16 of the Glass-Steagall Act, which prohibits banks from underwriting and dealing in securities. Under section 16, ‘The business of dealing in securities and stock by the association shall be limited to purchasing and selling such securities and stock without recourse, solely upon the order, and for the account of customers, and for its own account, and the association shall not underwrite any issue of securities or stock.’³

Section 21 prohibits securities firms from taking deposits. GLBA only repealed sections 20 and 32 of the Glass-Steagall Act, which prohibited member banks from affiliating with organizations dealing in securities.⁴ It is important in this context to refer to the usual definition of a bank as an institution which takes deposits that can be withdrawn on demand, and makes loans, and which is chartered by the federal government or national or state authorities and insured by the FDIC. The Act comes with a warning attached to it: ‘Before concluding that the GLBA reduced banking regulations and complexity, however, note that the GLBA comprises 144 pages of text. While the new law eliminated many restrictions, it also maintained the practice of delineating allowable activities with a combination of numbing details and vague terms.’⁵

The Glass-Steagall Act was designed to separate banks from the risks which might be created by their affiliates, especially a holding company or any subsidiary or affiliate allowed to engage in securities trading, by strictly limiting transactions between banks and their affiliates so that the safety net of deposit insurance and access to the discount window was not extended beyond banks to their holding companies or their non-bank affiliates, and to protect banks' financial position from the risks taken by their affiliates. The aim was to allow the holding company or even a bank securities affiliate to fail, without endangering the bank. The Act continues to prevent banks from underwriting (that is, assuming the risk that an issue of securities will not be fully sold to investors) and dealing (holding an inventory of securities for trading purposes).

The Glass-Steagall Act did not prevent banks from making investments, that is purchasing and selling securities acquired for investment, nor did it prevent banks from buying and selling whole loans or from securitizing loans. They were allowed to buy and sell securities based on assets, such as mortgages, but they could not deal in or underwrite mortgage-backed securities (MBSs), a restriction which remained in place even after the GLBA, with the only difference being that the GLBA now allowed banks to be affiliated with firms engaged in underwriting or dealing in securities, including MBSs. Glass-Steagall allowed banks to deal in and underwrite US government securities, the securities of Fannie Mae and Freddie Mac⁶ and the general obligation bonds of states and municipalities, mostly where these were guaranteed by the government, although Fannie Mae and Freddie Mac's bonds were not guaranteed. They were, however, generally believed to be guaranteed by the government, which turned out to be true when the two companies collapsed in 2008. The point here, however, is crucial to the subsequent debate and the introduction of the Volcker rule, which prohibits banks and their affiliates from engaging in bond trading on their own account. The GLBA repealed only the provisions of the Glass-Steagall Act which referred to affiliations of banks. It did not allow banks to do anything that they were previously prohibited from doing. The ‘repeal’ of the Glass-Steagall Act had no effect whatsoever on the ability of banks to engage directly in the risky business of underwriting and dealing in securities.

The 1956 Bank Holding Act established the terms and conditions under which a company can own a bank, authorizing the Federal Reserve to adopt regulations to uphold the Act. Most banks in the USA are owned by bank holding companies, and all are supervised by the Federal Reserve Bank. The percentage of banks owned by BHCs rose steadily from the 1980s to 2012, reaching 80 per cent of all banks. Under the 1956 Act a BHC was allowed to engage directly in or establish or acquire subsidiaries that engage in non-banking activities determined by the Federal Reserve to be closely related to banking, such as mortgage banking, consumer and commercial finance and loan servicing, leasing, asset management and financial and investment advisory services. In other words, banks can engage in securities trading for only a very limited category of securities, primarily government securities or those backed by the government, but a BHC or an investment bank is not subject to these restrictions. A BHC is a business corporation which controls a bank. It is not specifically chartered like a bank and is not allowed to take deposits, nor does it have automatic access to the Federal Reserve's discount window, participation in the nation's payments system or deposits insured by the FDIC. BHCs are regulated by the Federal Reserve, whereas most banks and almost all the large ones are regulated by the Office of the Comptroller of the Currency. State chartered banks are regulated by their home state regulators and the FDIC at the federal level.

The Act also allows investment banks to set themselves up as FHCs. Politicians and others do not seem to understand the role of investment banks. They facilitate mergers and acquisitions, undertake reorganizations of companies, broker trades for institutions and private investors; create capital for other companies, underwrite new debt and equities, and act as an intermediary for securities issuers and the investing public. Typically, an investment bank also provides guidance to issuers regarding the issue and placement of stock. They assist corporations to obtain debt financing by finding investors for corporate funds as well as pre-underwriting counselling and continuing advice after the distribution of securities. The size of an investment bank is also an asset, since the more connections a bank has, the more likely it is to make a profit by matching buyers and sellers. Even before the GLBA, investment banks were allowed to trade and hold mortgage-backed securities, credit default swaps, derivatives, and collateralized debt obligations. The increase in investment banks' trading portfolios resulted from their increased capital base, as a result of their becoming publicly held companies, as opposed to being corporate partnerships, an arrangement which predated the GLBA.

Investment banks did not wish to set themselves up as FHCs, although they sometimes complained about the ability to do so of the larger commercial banks, as measured by the total assets on their balance sheets, and their willingness to use these to attract clients and to provide low-cost funding by attracting deposits. The Glass-Steagall Act had prevented commercial banks from being affiliated with investment banks. After the GLBA was passed, commercial banks could change their BHCs into FHCs, and so could engage in investment banking activities through their non-bank subsidiaries. No domestically headquartered investment bank, especially not the Big Five investment banks, wished to become an FHC, because they would then have had the Federal Reserve Bank as their over-arching regulator.

Under functional regulation, federal and state regulators regulate banking activities, securities regulators regulate securities and insurance regulators regulate insurance. Even immediately after the passage of the Act, the authors recognize the problems in this approach to regulation.

The Act also ‘exempted certain bank activities that have a “securities” component from regulation by the Securities and Exchange Commission’. The lengthy list of examples includes various traditional banking activities, such as commercial paper, private placements, asset-backed securities and derivatives. The Federal Reserve set out its Framework for Financial Holding Company Supervision a year later.

Limits are imposed by law on the ability of banking regulators to impose capital requirements on a functionally regulated securities firm. Functional regulation where financial companies increasingly offer a wide range of financial services is fraught with problems, as many regulatory authorities, including those of the UK, have discovered. The USA was soon to find that the functional approach would face problems, due to the EU and its introduction of consolidated supervision.

How did the Federal Reserve regulate and supervise Financial Holding Companies?

In its Supervisory and Regulatory Letter of 2000, the Federal Reserve set out its guidance and the purpose of its supervisory oversight of FHCs.¹⁰ The Federal Reserve is

As part of its supervision, ‘the Federal Reserve will develop strong relationships with senior management and boards of directors of FHCs’¹² and access to timely information from them, including assessing the centralized risk management and control processes in order to understand the overall risk profile and determine how risks are being controlled on a consolidated basis.

The Federal Reserve, apart from its own supervisory activities, also works closely with other regulators, depending on information received from them, since the Act also preserved the role of federal and/or state banks, securities and insurance regulators as regulators of the various companies in the FHC. The Federal Reserve became the ‘umbrella’ supervisor for any FHC owning a bank; under its ‘streamlined supervision’ remit, the Federal Reserve was limited in its day-to-day authority to oversee functionally regulated non-banking subsidiaries of these holding companies. Safeguards for financial safety and soundness applied through the new holding, with existing firewalls being extended to provide further protection for banks within FHCs. The Act envisaged functional regulation, that is, by activity rather than the institution. In spite of its reliance on the functional regulators, the important point here is that the Federal Reserve is ultimately responsible for the consolidated supervision of the BHCs and FHCs.

This examination of the laws governing banks and holding companies and their affiliates is crucial. It makes it entirely clear that the GLBA did not remove any of the provisions of Glass-Steagall forbidding commercial banks from engaging in underwriting and dealing in securities, which remain in force. The financial crisis cannot be attributed to a departure from the Glass-Steagall Act, a conclusion which would amount to a failure to take into account the banking laws prevailing at the time of the crisis, and to understand their implications.

Did the Gramm-Leach-Bliley Act cause the crisis?

Many have argued that the GLBA caused the crisis to a large extent because they thought it repealed the Glass-Steagall Act in its entirety. Richard Parsons, a former director of Citigroup, made such a claim two days after he stepped down from office, saying that to some extent, the crash of 2007 and 2008 had been a result of its repeal, ‘because we haven't gotten our arms around it yet. I don't think so because the financial services industry moves so fast’.¹³ Others have argued that banks were undermined through their connection with investment banking, which many regard as ‘casino’ banking. Vince Cable, the UK Business Secretary, described the UK's concerns that ‘retail banking is being tied up with investment banking which some people call “casino” banking’. This was ‘the government's anxiety about the future stability of banks’.¹⁴

Such a view bears some relationship to the Dodd-Frank Act para 619, known as the Volcker rule after the former Chairman of the Federal Reserve Bank and President of the President's Economic Recovery Advisory Board, which prohibits insured commercial banks and their affiliates from engaging in ‘proprietary trading’ of all securities except US government debt by all ‘bank-related entities.’ The Volcker rule, on the one hand, is thus more restrictive than the Glass-Steagall Act. On the other hand, it allows commercial banks to engage in underwriting, making markets and hedging, provided they are acting solely for their customers or for their own hedging transactions. The distinction proved very difficult to define clearly, causing extensive delays in its implementation by the regulators, so that the final rules were not completed until 13 December 2013, with the date for full implementation by national banks being July 2015. Governor Tarullo admitted that the

That difficulty may prove to be insurmountable.

The trouble with this analysis and the proposed solution in terms of the Volcker rule, which clearly has been very difficult to articulate and even more difficult to enforce, is that it misses the point. As noted above the GLBA's repeal solely of the affiliation provisions of the Glass-Steagall Act did not permit banks to do anything that they were not previously prohibited from doing. The ‘repeal’ did not have any effect whatsoever on the ability of banks to engage directly in underwriting and dealing in securities.¹⁶ It is necessary to look elsewhere for the causes of the collapse of some of the banks and their financial weakness. The GLBA did not alter that. Where banks failed or got into difficulties, it was because their investments were unwise and the banks did not exercise due diligence, relying on the rating agencies to do their work for them. With the encouragement of successive Administrations, the banks lent to those who were not in a position to repay the loans, underwriting conditions were abandoned, loans were sold to Fannie Mae and Freddie Mac, so that the discipline of retaining loans on their books disappeared. In addition, banks invested in mortgage-backed securities that pooled subprime and other low-quality mortgages. More exotic securities were created out of the subprime loans, such as collateralized debt obligations (CDOs). When house prices began to falter in mid-2006, subprime borrowers began to default and mortgage-backed securities fell in value, because they had been over-valued in the first place by the rating agencies. The failures were amongst the banks and thrifts, large investment banks and the two government sponsored-entities, Fannie Mae and Freddie Mac, all of which were too highly leveraged.¹⁷

What about the ‘Big Five’ investment banks?

None of this affected the Big Five investment banks, Goldman Sachs, JP Morgan, Merrill Lynch, Bear Stearns and Lehman Brothers. Understanding exactly what effect the repeal of the Glass-Steagall Act had on the American banking system is to see that it was not only irrelevant as far as the financial crisis is concerned, but also that it did not affect the status of these stand-alone investment banks. All the Act did was to give the Big Five investment banks the opportunity to continue as stand alone banks without becoming FHCs which would have brought them under more effective regulation and supervision.

The SEC's focus has always been on the securities broker-dealer, for which the key rule is rule 15c3-1 of the Securities and Exchange Act 1934, the ‘net capital rule’ which is intended to ‘protect customers and other market participants from broker-dealer failures and to enable those broker-dealers to liquidate in an orderly fashion without the need for a formal proceeding or financial assistance from the Securities Investor Protection Corporation’. Rather than considering the appropriate means of supervising a financial conglomerate, the SEC merely sought to adapt its existing rules without a specific focus on the holding company or the fact that the holding company was an investment bank.

None of that was of any use to the investment bank holding companies, operating in the European Union. EU Directive 2002/87/EC requires financial conglomerates operating in the European Union or the European Economic Area (EEA) to have a single supplementary supervisor to provide oversight of the group, in particular with regard to solvency and risk concentration, intra-group transactions, internal risk management processes at the conglomerate level, and fit and proper management. Article 3 of the Directive set out the thresholds for identifying a financial conglomerate, and Articles 6–17 set out the way in which supplementary supervision was to be exercised.

The Directive required any conglomerate to have a single lead supervisor in place by 1 January 2005. The supervisor had to provide a level of oversight equivalent to that provided by European regulators. SEC Director Annette Nazareth, in her testimony to the House Committee on Financial Services, stated that

It might lead to higher capital and risk control requirements than an EU-based firm or the creation of a sub-holding company in the EU. Hence, the five investment banks had to obtain an appropriate level of supervision at the level of the holding company or undertake reorganization of companies operating in the EU, which would have been costly and time-consuming.

As a result, the SEC formally adopted amendments to the 1934 Securities Exchange Act in August 2004. These amendments permitted the non-bank affiliated holding companies of the US broker-dealers the alternative of ‘voluntarily’ committing themselves to having the SEC as supervisor. They then became ‘consolidated supervised entities’ (CSEs) and continued to operate in the EU and the EEA. All of the Big Five US investment banks became CSEs.

Regulations for the consolidated supervised entities

The new regulations for the five stand-alone investment banks are set out in the Federal Register, although the term ‘investment bank’, for the ultimate holding company, does not appear there.¹⁹ It is worth examining the way in which the SEC adapted its long-standing rule, 15c3-1, the net capital rule, in order to establish a voluntary, alternative method of computing capital for certain broker-dealers. The new rules came into force on 20 August 2004. They consisted of ‘alternative net capital requirements for broker-dealers that are part of consolidated supervised entities’,²⁰ which included allowing the broker-dealer to use mathematical models to calculate net capital requirements for market and derivative-related risk. A broker-dealer using the alternative method of computing net capital would have to hold a higher level of net capital. If the tentative net capital should fall below $5bn, the Commission would have to be informed immediately. The SEC would then have to consider whether appropriate remedial action should be taken.

In addition, the SEC required firms to maintain an overall Basel capital ratio at the consolidated holding company level of not less than the Federal Reserve's 10 per cent well-capitalized standard for bank holding companies.²¹ The broker-dealer had to have in place comprehensive internal risk management procedures for market risk, credit risk, liquidity, legal and operational risk. To be able to use the alternative calculations, the holding company had to provide the SEC with extensive financial information, including financial, operational and risk management information, group-wide allowable capital and allowances for market, credit and operational risk calculated in accordance with Basel II standards, along with many other reports on a monthly, quarterly and annual basis. The broker-dealer's ultimate holding company and its affiliates also had to agree to group-wide supervision.

These rule changes were designed to ‘help ensure the integrity of the broker-dealer's risk management (and other) procedures’. At the same time, the ‘ultimate holding company’ had to provide information about ‘its own financial and operational conditions’ as well as the information about risk exposures from its senior risk managers. It also had to comply with rules regarding the implementation and documentation of a comprehensive, group-wide risk management system, which identified and managed market, credit, liquidity, legal and operational risk. The holding company had to compute the group-wide allowable capital and allowances for the specified risks in accordance with Basel Committee on Banking Supervision requirements.

In the course of presenting the rule changes for CSEs, the SEC noted that the ultimate holding companies might own many other companies, including both broker-dealers and non-broker-dealer companies operating on a global basis. (Indeed, they all owned broker-dealers, which were already supervised, as separate subsidiaries by the SEC.) Any one of these could become insolvent and so affect a broker-dealer's access to short-term funding, or the broker-dealer's capital might be diverted to prop up an ailing firm. The more sophisticated broker-dealers had already been pressing the SEC to allow them to use models to measure risk and compute capital levels along the lines of those being adopted by the more advanced banks under the (then) new capital adequacy rules. They would have to agree to their holding company being supervised on a group-wide basis, which meant additional reporting, record-keeping and the examination of all its entities, including the parent company and affiliates, by the SEC.

The final rules are convoluted indeed. Their purpose is to help the Commission ‘to maintain the integrity of the securities markets, by improving the oversight of broker-dealers and providing an incentive for broker-dealers to implement strong risk management practices … and to reduce costs for broker-dealers by allowing very highly capitalized firms which have developed internal risk-management practices’ to use them. The broker-dealer may only use the alternative method of computing net capital, if it maintains tentative net capital of at least $1bn and net capital of at least $500m. Since the broker-dealer may take smaller deductions for market and credit risk, then the company must notify the Commission if its tentative net capital falls below $5bn, although in some cases the SEC may exempt the firm from this requirement.

In applying for permission to use the alternative method of computing net capital, a broker-dealer has to provide a description of their internal risk control system and how the firm intends to use it for deductions of net capital. The broker-dealer may use a VaR model for calculating market risk, provided he uses the multiplication factor to help provide capital during periods of market stress. In a further concession, the SEC agreed that broker-dealers need not limit their VaR calculations to securities for which there is a ‘ready market’. Broker-dealers were permitted to use internal calculations to determine counterparty credit risk weightings without any specific maturity adjustment factor, although Basel did specify a maturity adjustment. The Commission had proposed an additional charge for credit risk, if the broker-dealer's aggregate current exposure for all counterparties for unsecured exposures was greater than 15 per cent of its net capital, but when the industry objected, this suddenly turned into 50 per cent.

These examples have been set out in some detail as they illustrate the failings of this attempt by the Securities and Exchange Commission to regulate financial conglomerates. The above quotations from the Federal Register of Monday 21 June 2004, have been referenced partly because they reveal the extent to which the SEC modified the rules in response to pressures from the industry, and sometimes in an arbitrary fashion. The SEC's focus is entirely on changing the rules for broker-dealers, and do not appear to be concerned with the supervision and overall evaluation of a financial conglomerate, despite the reporting requirements. By contrast, the 2002 EU Directive's whole emphasis is on the need for group-wide supervision, having noted the accelerating pace of consolidation in the financial services industry and the intensification of links between the financial markets. Supervisors of different sectors of the financial services industry and of various member states should be able to establish a coordinated approach so that an overall prudential assessment can be made,

The EU Directive introduced a series of rules for the supplementary supervision of financial conglomerates, especially capital adequacy, intra-group transactions and management with a single co-ordinator selected from the relevant member state. The group's ‘own funds’ could be used more than once to provide capital adequacy for the parent company and also for a subsidiary. It was up to the regulatory authorities in each member state to make sure that at the level of the financial conglomerate that the capital adequacy requirements were properly met. The methodology to be used in calculating the appropriate level is set out in an Appendix to the Directive. However, the capital required at group level is not available to bail out one of the member companies. The US ‘source of strength’ doctrine, where the bank holding company and/or subsidiaries were expected to support an ailing deposit-taking subsidiary, is not part of the EU Directive.

By contrast, although the SEC recognizes that large broker-dealers are owned by holding companies which also own other entities involved in financial services worldwide, the emphasis does not seem to be on the supervision of the financial conglomerate as a whole. That is in spite of the requirement that the holding company must have a group-wide internal risk management system and would make periodic reports to the SEC, where the SEC's supervision would consist of analysing reports and records provided by the CSE of the broker-dealer. The CSE rules do not allow for the role of a co-ordinator to facilitate the supervision of other companies in different sectors of the financial services industry, nor would the SEC examine any functionally regulated affiliates of the broker dealer.

In evidence to the Committee on Financial Services on 22 May 2002, Annette Nazareth declared that the Commission and the EU approaches to ‘group-wide supervision’ were based on the same principles:

This suggests that the SEC had adequate tools and statutory backing for taking on the consolidated supervision of the Big Five investment banks, and would develop appropriate procedures when it published the final rules. That is why the voluntary CSE scheme is focused entirely on the effects of the actions of the holding company and its affiliates on the ‘safety and soundness’ of the broker-dealer. This was because

SEC Deputy Director Robert Colby then describes the liquidity standards the CSEs were required to meet:

Later in the same presentation he added that ‘the CSE regime is tailored to reflect the reliance of securities firms on mark-to-market accounting as a critical risk and governance control.’ The supervisory regime consisted of regular monthly meetings with senior market and credit risk managers, including a ‘granular system of limits’, ‘articulates to each business or desk the risk appetite of senior management’. These meetings also reviewed the performance of models and aggregation tools and the risk reporting and analytics prepared for senior management. SEC staff discussed the amount and nature of liquid assets held by the holding company with the CSE treasury managers and with the financial controllers to review the financial results. In particular, SEC staff examined the results of the firm's internal price testing procedures, intended to validate the marking-to-market of complex and illiquid products.

Colby's testimony paints a picture of efficient and thorough supervisory activities on behalf of Commission staff. However, further examination in the next chapter indicates that all was not as it seemed. The SEC's inability to carry out effective supervision was to be revealed all too soon.