A Proper Accounting of Glass-Steagall

Over the past two decades in the United States and Europe, elite decision-making has hit an extended rough patch. Fairly or not, a lot of important public policy decisions have turned out horribly for the best and brightest. Whether that involved running a stratospherically levered financial institution hurtling into the financial crisis; or the government encouraging the sale of mortgages to people who could hardly afford to pay them back under any realistic scenario; or the creation of a single currency union without any effective fiscal or political union; or invading a country with little thought given to what comes next—elites have proven themselves at best all too human or, worse, inclined to pursue the counterintuitive and abstract with little apparent concern for real-world consequences. No doubt many will quibble, but to locate the starting point of these last twenty mostly regrettable years of elite decision-making, we could do worse than to begin with the repeal of the Glass-Steagall Act in 1999 (more precisely, the 1999 repeal of the prohibition under the Glass-Steagall Act of insured banks being affiliated with firms like investment banks that engage in securities underwriting),¹ and the failure to regulate the pervasive use of derivatives by commercial and investment banks. Taken together with U.S. housing policy under the Clinton and Bush administrations, these three policy failures accounted for the alarming increase in leverage in the overall economy and the enormous growth of large financial institutions in the years preceding the global financial crisis.

The government’s policy response to the financial crisis and the ensuing banking crisis has been a mixed bag of executive branch enforcement actions and legislative fixes. In 2008, the government put both Fannie Mae and Freddie Mac into conservatorship, where they remain today, and mortgage underwriting standards were strengthened. Under the Obama administration, the Treasury, Department of Justice, and other agencies were very active in reaching monetary settlements with the major banks for precrisis malfeasance. Additionally, in 2010, Congress passed the Dodd-Frank Act. As a result, many of the gaps in clearing and margin requirements for derivative contracts have been addressed, and new restrictions have been placed on banks trading on their own balance sheet. These policy and administrative actions, however, failed to address one key component of the crisis: the biggest banks have become even bigger in the intervening ten years.

In an effort to compensate for the government’s unwillingness to limit the size of large financial institutions and to address the ad hoc approach by the government during the 2008 crisis, Dodd-Frank created a new and most unusual regulatory power for the government. In what almost seems like a practical joke played on unsuspecting senators, Dodd-Frank authorizes something called the Orderly Liquidation Authority (OLA) that permits the FDIC, with the approval of the Treasury secretary and Federal Reserve Board of Governors (by a two-thirds supermajority), to preemptively seize and wind down any systemically important financial institutions (SIFIs) that are “in danger of default,” and to subjectively decide which claims to prioritize and at what price. More will be discussed below, but just to assure the reader’s good sense, the OLA represents some of the most wishful public policy thinking to ever come out of Washington, D.C.

But we are getting ahead of ourselves. To understand how we arrived at this point, we must revisit the rationale for the Glass-Steagall Act in the first instance as well as its limitations before finding a path forward beyond Dodd-Frank—one that addresses the size and complexity of large financial institutions and manages the systemic risks they pose to the economy in a fair and commonsense fashion.

The Glass-Steagall Act and Its Disputed Legacy

The Glass-Steagall Act (the “Act”) refers to specific sections of the Banking Act of 1933, in particular to sections 16 and 21, which separated commercial banks from investment banks. Glass-Steagall was passed in response to the failure of the Bank of the United States in 1930 and the abuses that Congress discovered in its investigation into National City Bank’s underwriting practices leading up to the Great Depression of 1929. Popular disillusionment with speculators and the national banks led to the passage of the Banking Act, which also saw the creation of the FDIC.

Congress was concerned that commercial banks, generally, and member banks of the Federal Reserve System, in particular, had both aggravated and been damaged by the stock market crash because of their direct and indirect involvement in trading and ownership of stocks. Glass-Steagall henceforth prohibited commercial banks—banks that receive deposits subject to repayment—from going into the investment banking business. Section 16 of the Banking Act differentiates between what is permissible and prohibited for a commercial bank on the basis of the type of securities involved rather than the activities performed. It stipulates that a national bank “shall not underwrite any issue of securities of stock” and that purchasing and selling stock shall be solely for the account of its customers and “in no case for its own account.”

The hazards that Congress had in mind were not limited to the obvious danger that a bank might invest its own assets (and its clients’ deposits) in corporate securities, thereby exposing their (now guaranteed) clients’ bank deposits to unwarranted risks. As the U.S. Supreme Court explained in its 1971 ruling in Investment Co. Institute v. Camp—in which it overturned a decision by the Comptroller of the Currency to authorize a national bank to sell its customers a stock fund created and maintained by the bank—Congress was also concerned about the “subtle hazards” that occur when a commercial bank goes beyond the business of acting as fiduciary and enters the investment banking business. “This course places new promotional and other pressures on the bank which in turn create new temptations,” Justice Potter Stewart wrote for the Court’s majority. Such temptations may end badly and public confidence in the bank could be damaged. And since public confidence is essential to the solvency of the bank, the bank may be tempted to throw good money after bad to “shore up” the original investment. Furthermore, national banks could be tempted to make their balance sheets and credit facilities more easily available than they would otherwise be to those companies with whom they maintain an investment banking relationship. Or, national banks might simply be tempted to make loans to customers with “the expectation that the loan would facilitate the purchase of stocks and securities”—the very stock and securities that the national bank or its affiliates underwrote. It is the prevention of these subtle conflicts of interest that forms the regulatory intent behind Glass-Steagall.²

Glass-Steagall has been criticized since the day it was passed into law. The core challenge to the Act is its emphasis on defining impermissible activities that are often hard to distinguish from other permitted activity. The trust approved by the Comptroller of the Currency in Camp involved the purchase of corporate equities in a pooled vehicle. Glass-Steagall did not prohibit a national bank from pooling trust assets; nor did it limit a national bank’s ability to work on behalf of customers to purchase equities. It was the combination of the two that the Court said was explicitly prohibited. This is a very fine line and points to the serious limitations of the the Act. For example, the Act did not contemplate that new products like money market funds would become commonplace with check redemption privileges that erased the distinction between them and NOW (negotiable order withdrawal) accounts. And yet, competitors created these types of accounts, which encroached on the banks’ turf. So, as critics rightfully pointed out, while Glass-Steagall was designed to keep commercial and investment banks separate, it was emphatically not meant to shield other segments of the financial marketplace from competition.

Amid calls to address these shortcomings over the years, numerous banking laws were passed, and boundaries redrawn, but repeal remained the ultimate goal of many. Banking interests and academic studies in the late 1980s and 1990s began to question the Act’s very premise of separating commercial banks from investment banks. George J. Benston, author of the seminal work The Separation of Commercial and Investment Banking, argued in a 1994 article that “the claim that universal banks are more risky than specialized banks, while plausible in some ways, is not borne out of experience.” He cites the savings and loans crisis of the 1980s, which was sparked when the Federal Reserve hiked short-term interest rates to 12 percent in 1979 while these institutions had sold fixed mortgages with substantially lower interest rates. This disaster cost taxpayers close to $150 billion. It was one that was “exacerbated by the U.S. prohibition against nationwide branching,” which made depository institutions susceptible to local factors, like steep declines in the price of oil or land.³

The same was true during the Great Depression. Of the over nine thousand banks that failed during the Great Depression, the vast majority were single-office banks located in small-town America. “Historical experience,” Benston concluded, “tends to show that bank involvement in securities actually helps to stabilize an economy in times of financial crises.” Though 26 percent of national banks failed during 1930–33, only a small fraction of the sixty-two banks which operated securities affiliates failed.

Randall Kroszner, former governor of the Federal Reserve, and Raghuram Rajan, former governor of the Reserve Bank of India, drew a similar conclusion in their 1993 paper, “Is the Glass-Steagall Act Justified?”⁴ In the pre-Glass-Steagall era, national banks were prohibited under the National Banking Act of 1864 from handling common stocks. To get around this prohibition, national banks incorporated affiliates under state corporate charters. The 1920s saw a dramatic increase in the number of national bank affiliates and their involvement in nonbank activity like securities underwriting. The authors’ hypothesis was that if the concerns that motivated the passage of Glass-Steagall were well founded, evidence should exist that the securities underwritten by affiliates underperformed the securities underwritten by investment banks. But the evidence suggests the opposite: only 11 percent of affiliate-underwritten issues defaulted versus 28 percent of investment bank–underwritten issues during the period. Kroszner and Rajan concluded that the market was rational enough to assess the quality of underwritten issues and properly discounted any conflicts of interest that might have existed. They argued that there were intangible factors, like reputation risk, that mitigated perceived conflicts of interest by elevating the underwriting standards of national bank affiliates.

A Failure of Imagination: From Repeal to Crisis

Although marshalling important facts, do these academic studies ultimately suffer a failure of imagination? After all, universal banks (or SIFIs in today’s regulatory nomenclature) only appear in the marketplace following Glass-Steagall’s repeal. While in 1994 Benston could confidently conclude that “neither theory nor evidence support the assumption that limitations on banking . . . either were or are likely to be effective in reducing risk-taking,” the world would only have to wait a few years for the necessary empirical evidence to dispel any doubt that the absence of structural limitations on the size of banks could indeed precipitate a system wide financial meltdown. Only two years following Glass-Steagall’s repeal, moreover, Wall Street was hit with the so-called analyst scandal which exposed how equity research teams at the major investment and commercial banks, whose research was presented as independent, did little more than carry water for their firms’ investment bankers, who sought to win lucrative merger and IPO deals. This subtle conflict of interest led to $1.4 billion in total fines.

Alan Greenspan, longtime chairman of the Federal Reserve, shared a similar belief that bank executives were best suited to manage their own self-interest and would not assume risks that imperiled their interests or those of their shareholders—notwithstanding the increasing size and complexity of the largest banks. His faith extended to the derivatives market where, in the late 1990s, he, Robert Rubin (then secretary of the Treasury), and Rubin’s deputy Lawrence Summers were successful in keeping this market a private one between counterparties: they shut down efforts by the CFTC (Commodity Futures Trading Commission) to compel the use of independent clearinghouses for the trading of derivatives.⁵ In October 2008, he was forced to admit his faith was mistaken: “Those of us who have looked to the self-interest of lending institutions to protect shareholders’ equity, myself included, are in a state of shocked disbelief,” Greenspan told the House Committee on Oversight and Government Reform.

While self-interest is typically well suited to managing risk, large financial institutions are a different animal altogether. First, as they grow their balance sheets, they can become too complex for even the most experienced executives to understand. And, second, at a certain size, mismanagement or an old-fashioned run on the bank is no longer solely the shareholders’ problem. Large banks become society’s problem when they become too large to fail, as was the case in 2008—a twist on the old joke that if you owe the bank a little bit of money, it’s your problem, but if you owe the bank a lot of money, it’s theirs. And this may well be the single largest benefit to Glass-Steagall. Despite the impracticality of defining permissible activities (as opposed to, say, limits on bank leverage or proper margin requirements), the Act has proven “a better regulator than the anti-trust and other laws for curbing excessive aggregation of banking power in only a few financial institutions,” in the words of Roberta Karmel, a former commissioner of the SEC.⁶

But many former critics of Glass-Steagall will not readily concede the point. Larry Summers has been adamant that nostalgia for Glass-Steagall is mistaken:

Virtually everything that contributed to the crisis was not affected by Glass Steagall even in its purest form. Think of pure investment banks Bear Stearns and Lehman Brothers, or the government-sponsored enterprises Fannie Mae and Freddie Mac, or the banks Washington Mutual and Wachovia or American International Group or the growth of the shadow banking system. Nor were the principal lending activities that got Citi and Bank of America in trouble implicated by Glass Steagall.⁷

This is almost certainly true in a technical sense, but misses the larger point of Glass-Steagall’s ability, for sixty-six years, to control the growth and size of national banks. Yes, the repeal of Glass-Steagall’s prohibition against insured banks owning affiliates did not lead inexorably to risky loans financed with insured deposits. The real culprits included a number of incentives for subprime borrowers to receive mortgages with unrealistic terms—with these mortgages, often guaranteed by Fannie and Freddie, bundled in complex debt structures (CDOs). This activity was often accompanied by fraudulent underwriting by Wall Street commercial banks, their affiliates, and other institutions. But the repeal did allow insured banks to grow substantially in size and to own affiliated investment banks or other entities whose failure, while nominally at arm’s length, posed the kind of reputational risk that Justice Stewart warned of in the Camp decision. To take but one example: between September 2007 and July 2008, well before the meltdown at Lehman, Citibank was forced to transfer over $100 billion in off-balance-sheet liabilities onto the bank’s balance sheet and record over $7 billion in losses. Further, the use of derivatives created a web of entanglements between all the banks (commercial and investment) and other large financial institutions (such as AIG) that made a mockery of whatever walls were thought to exist in a post-Glass-Steagall regulatory environment. Together, these factors contributed to the stupendous growth of the large banks, and thus placed the entire financial system in a precarious state.

As Ben Bernanke explained to the Financial Crisis Inquiry Commission in 2010:

As a scholar of the Great Depression, I honestly believe that September and October of 2008 was the worst financial crisis in global history, including the Great Depression. If you look at the firms that came under pressure in that period . . . only one . . . was not at serious risk of failure. . . . So out of maybe the 13, 13 of the most important financial institutions in the United States, 12 were at risk of failure within a period of a week or two.

Therefore it is deeply misleading to claim, as Peter J. Wallison of the American Enterprise Institute does, that “insured banks got into trouble in the financial crisis by buying and holding MBS backed by subprime and other low-quality mortgages” on their balance sheets, and not because the prohibition of insured banks owning affiliates was repealed.⁸ With expanding balance sheets and swelling off-balance-sheet liabilities, the largest banks were quickly becoming too complex for senior executives to comprehend and manage.⁹ The big four banks’ balance sheets increased from $3 trillion to $8 trillion between 2003 and 2008 alone, while their derivatives exposure exceeded $200 trillion on a notional basis at the end of 2008.¹⁰ This was a direct result of repealing Glass-Steagall and the passage of the Gramm–Leach–Bliley Act, also known as the Financial Services Modernization Act of 1999.

The Post-Post-Glass-Steagall World

In the aftermath of the financial crisis, Congress and the Obama administration chose a three-pronged approach to manage the size and appetite of large financial institutions. First, the U.S. Treasury permanently extended the conservatorships of Fannie and Freddie while restructuring their balance sheets away from fixed-income arbitrage and subprime, raised the G-fees they charge to guarantee mortgages, and bolstered mortgage underwriting standards. Second, Congress passed the Dodd-Frank Act, which prohibited banks from trading on their own balance sheets (the so-called Volcker Rule), mandated that most derivative contracts be standardized and traded through clearing houses with margin requirements similar to futures contracts, created the Orderly Liquidation Authority (OLA), and tweaked the bankruptcy code to handle future failures of large financial institutions. Third, the U.S. Treasury, Department of Justice, and a myriad of other agencies fined the banks repeatedly for crisis-era misconduct (over $110 billion has been collected from the six biggest U.S. banks) and postcrisis mishaps.

But the elephant in the room remains. The big four U.S. banks continue to grow in size despite penalties, fines, and declining returns on capital. They now control close to 40 percent of total bank assets. In many ways, having fewer and larger banks serves the interests of the government and the banks. Heightened regulation is the devil they both know and an effective barrier against more regulation or increased competition.¹¹ On the other hand, a future financial crisis will necessitate a repeat of the banking bailouts and emergency measures, the august goals of the OLA notwithstanding. In fact, the OLA’s existence may make the next time even more destructive.

Under Title II of Dodd-Frank, the OLA empowers the FDIC, once it receives permission from the Treasury Secretary and Federal Reserve, to seize control of any SIFI that poses a significant risk to the financial stability of the United States before or after the company declares bankruptcy. Once empowered, the FDIC is expected to set up a so-called single point of entry (SPOE), appointing the FDIC as receiver of the top-level holding company and allowing all of its subsidiaries (the commercial bank, investment bank, broker-dealer, insurer, etc.) to continue operations. With capital provided by the U.S. Treasury, the FDIC would serve as the debtor-in-possession and would establish a bridge financial company, where it could then transfer the assets and some of the seized firm’s liabilities at the FDIC’s sole discretion. The new entity could be capitalized by converting a prearranged class of debt structured to convert into equity. Newly restructured, with the backing of the FDIC, the firm should in theory be able to access the financial markets to fund operations—unless of course the seizure triggers a market panic as the FDIC subjectively decides the value of various debtor claims.

Paul Singer, founder and CEO of Elliott Management, a large hedge fund based in New York City, has been an outspoken critic of the government’s response to the financial crisis and, in particular, the creation of the OLA. In a letter to Chairman Crapo of the Senate Committee on Banking in April 2017, Singer warns that the OLA’s subjective nature would allow government officials, post-seizure, “to have wide latitude to establish bridge companies, shift assets to them with little due process, and then treat creditors differently even if they hold the same priority of claims.”¹² Such wide latitude will undermine “the ability of investors to confidently assess how their claims will be handled in a distressed situation.” This creates a “serious threat that creditors will be spurred to sell their debt instruments as soon as there is any indication (even a rumor) that an institution is troubled.” In a repeat of 2008, this could accelerate the crisis and increase panic among investors, the exact opposite of the behavior that the government wants to encourage. But it is precisely the kind of wishful thinking that we have seen from policy makers over the last two decades, with all the very bad and very predictable consequences that follow.

In a rebuttal of Singer’s claims, dozens of prominent law professors and financial scholars wrote to the Senate and House Committees on Banking urging the chairmen to retain the OLA. Congress has recently examined replacing the OLA with the Financial Institution Bankruptcy Act as a means to address some shortcomings to Chapter 11. “Repealing OLA would leave bankruptcy courts with the entire responsibility in a crisis for handling restructurings in ways that they have never done before.”¹³ The scholars argue that current Chapter 11 case law suffers from “four limits”—international coordination, planning, coordinated response, and liquidation provision—and the OLA, with some modest improvements, is best equipped to address these limits, in their opinion.

Yet Singer, a successful investor who held significant claims in the Lehman estate, argued that the bankruptcy courts worked just fine in 2008 with one exception: “The cross-default provisions common to bilateral derivative contracts.”¹⁴ So in lieu of the complex and highly subjective OLA framework, why not simply make the appropriate tweaks to the existing process?

Early indications are that the Trump administration will support keeping the Orderly Liquidation Authority in place.¹⁵ Without any alternative approach, it could hardly do otherwise. Given the size of the large banks and the unwillingness of Congress to place absolute limits on how big any single bank may become, the OLA may be best understood as providing a more structured legal cover for the government to bail out the financial system one firm at a time during the next crisis, thus providing regulators some legislative justification to act during a moment of widespread panic. But this leaves the U.S. financial system firmly with one foot in a Glass-Steagall world (prohibition of certain types of activities) and one foot in the post-Glass-Steagall world (no effective antitrust mechanism in place to limit the size of the banks). It is as if the government has decided to double down on the worst aspects of both eras. The new regime streamlines the bailout process, while ignoring at least one significant contributor to the crisis—the concentration of assets within a few institutions.

The large banks seem to be content with all of this. Bank executives, in one way at least, receive the best of both worlds: no limitation on the size of their assets, coupled with an implicit bailout guarantee. Incumbency has its advantages, after all. Yet it is unclear how shareholders will react. AIG, the large insurer that was bailed out by the government in 2008, has been pressured by activist investor Carl Icahn to shed many of its assets in order to avoid a SIFI designation. Shareholders, whose short-term interest is usually to increase banks’ ability to take risks in order to drive returns, presumably hope to increase AIG’s flexibility and returns on equity by reducing its size. Such pressure from shareholders may be the sole remaining counterbalance against further concentration of assets within the largest banks, but shareholders are divided on this issue. Bank executives and the government, on the other hand, are largely in agreement that size is a competitive and structural benefit. This bodes well for the perpetuation of the status quo.

Nevertheless, an opening will exist following tax reform, and the Trump administration would be wise to seize it to address the biggest remaining issue from the financial crisis: the sheer concentration of assets in the country’s largest financial institutions. Any effective policy would seek to limit the kind of micromanaging of Dodd-Frank and address instead the three pillars to every banking crisis: the absolute size of the largest banks; the amount of borrowed money or leverage the largest banks employ on those assets; and the quality or underwriting of those assets. The question, of course, is whether such a simple, commonsense approach to policymaking can garner the necessary support in an era where the counterintuitive and abstract—not to mention financial industry lobbying efforts—hold such a grip on the thinking of our elites.

This article originally appeared in American Affairs Volume II, Number 1 (Spring 2018): 29–41.

Notes