The Elephant in the Room

“Too Big to Fail” Financial Institutions Must Be Addressed

If conservatives truly want to end the policy of bailouts, they must stop ignoring the problem of “too big to fail” financial institutions, writes David Min.

Senate Banking Committee Chairman Sen. Christopher Dodd (D-CT) speaks at a press conference on Capitol Hill. (AP/Harry Hamburg)
Senate Banking Committee Chairman Sen. Christopher Dodd (D-CT) speaks at a press conference on Capitol Hill. (AP/Harry Hamburg)

A key point of contention in Sen. Chris Dodd’s financial regulatory reform bill, which will proceed to the floor for a vote when the Senate reconvenes in April, is how it addresses the problem of “too big to fail” financial institutions and the bailouts those firms received in the fall of 2008. The Dodd bill, reflecting the approach articulated by the Obama administration in its regulatory reform plan, attempts to tackle this issue by imposing heightened risk oversight on large financial companies, including higher capital requirements, increased reporting and examination, resolution authority, and the levy of a tax meant to finance a resolution authority fund in case one of these financial institutions fails in the future.

Many conservatives, including American Enterprise Institute Visiting Scholar Peter Wallison and Cato Director of Financial Regulation Studies Mark Calabria, have bitterly criticized the administration’s approach. They argue that by regulating certain financial institutions for systemic risk, we identify them as “too big to fail,” creating expectations that the government would bail them out in a future financial crisis, thus shielding their investors from losses. Investors therefore see an effective government guarantee on their investments in these companies, and are willing to accept a lower return on their investments as a result.

So the conservatives’ argument goes like this: Regulation of large financial institutions endows them with a lower cost of capital, giving these companies significant advantages over their smaller competitors, which in turn means the answer is to simply to “let them fail.” In this way, conservatives content that “too big to fail” is a political problem that can be solved by simply disavowing the prospect of future bailouts.

While this line of thinking is intrinsically appealing in its simplicity, it reflects a dangerous ignorance of what happened in the financial crisis of 2008 and of what it means to be “too big to fail.” The “let them fail” crowd mistakenly assumes that the “too big to fail” status—and the likelihood of bailouts it signifies—is conferred by government regulation, when it is actually a pre-existing condition that government regulation is designed to address.

In short, these conservatives are confusing the treatment—the regulation of systemic risk—with the disease, which clearly is “too big to fail.”

Such confusion is dangerous. Our financial system today is especially vulnerable to the type of banking panic that triggered the Great Depression. And the unfortunate reality is that the unsupported failure of any of the very large financial institutions—the ones dubbed “too big to fail”—would spark such a banking panic, causing massive damage that would reach far beyond Wall Street. Based on past experiences with banking panics, we would likely see many trillions of dollars of wealth destruction, as money market funds, pension funds, and other institutional investors saw their assets devalued overnight. The drastically reduced wealth of U.S. households coupled with a lack of available credit could mean a decade or more of economic contraction.

It is simply fantastical to believe that any responsible political leaders facing the choice between bailouts and economic armageddon would choose the latter. That is why the conservative argument that all we need to do is “let them fail” next time is so worrisome. Even if you believe that bailouts are an absolutely unacceptable policy, a 21st century repeat of the Great Depression is an equally unacceptable outcome. By ignoring the conditions that prompted the bailouts and pretending that “too big to fail” is not a real problem, these conservatives are doing a disservice to everyone. They dumb down the debate and make it more difficult to pass real solutions that might actually reduce the likelihood of future bailouts—or worse.

This elephant in the room—“too big to fail” financial institutions—is inclined to become destructive every so often, and we have two options to prevent it from causing major damage. We can either shrink the elephant down to a smaller size, or we can put strong restraints on it and watch it closely. Ignoring the elephant, or arguing that the problem is the expensive tranquilizers we bought to keep it in check during its last rampage, is not constructive.

Defining too big to fail and bank panics

To address this problem, we first need to define “too big too fail” and how the problem can implode our financial system. “Too big to fail” is best understood as a bank panic problem, and has arisen as the result of two developments in the global financial markets over the past several decades. The first development was the tremendous growth of a “shadow banking system” operating outside of the rules that have governed depository banking since the Great Depression. This shadow banking system essentially performed the same functions as banking—attracting short-term investments and using them to finance long-term loans—but did so through the use of entities that were not depository banks, and the use of financing instruments (such as mortgage-backed securities, commercial paper, or short-term repurchase agreements) that were not deposits. Because of this nonbank, nondepository structure, the shadow banking system, which grew to an estimated $10 trillion in size, fell outside the rules and protections of the regulated banking system.

The second development was the concentration of risk within the shadow banking system, such that a small number of financial firms were and are responsible for the vast majority of its liabilities. Before the 2008 crash, the five major U.S. investment banks had a combined balance sheet size of approximately $4 trillion, and this may have understated the true level of liabilities they were holding. Witness the recent revelations about failed Wall Street investment bank Lehman Brothers, which raises questions about the extent to which shadow banks offloaded balance sheet risk through the use of dodgy transactions.

These shenanigans obviously contributed to the market crash in the fall of 2008, but was this market crash appropriately viewed as the onset of a bank panic? Here’s how Council of Economic Advisers Chair Christina Romer describes a bank panic:

A banking panic arises when many depositors lose confidence in the solvency of banks and simultaneously demand their deposits be paid to them in cash. Banks, which typically hold only a fraction of deposits as cash reserves, must liquidate loans in order to raise the required cash. This process of hasty liquidation can cause even a previously solvent bank to fail.

As Romer reminds us, prior to the banking reforms implemented by the Glass Steagall Act of 1933, the U.S. financial system experienced banking panics of increasing frequency and intensity every 5 to 10 years, culminating in the 1930-1933 bank panics that triggered the Great Depression. And as scholars across the ideological spectrum note, these crises occur because banking systems are prone to taking on too much risk during good times—known as procyclicality—and are highly vulnerable to bank runs because their liabilities (deposits) are short term and can be withdrawn at any time, whereas their assets (loans) are long term and don’t trade in liquid markets, and are thus difficult to liquidate.

As a result, unregulated banking systems tend to experience severe boom-bust cycles, where too much credit risk in good times leads to high credit losses during bad times, which in sparks a run on the banking system by spooked depositors. During particularly severe cycles, these banking panics can result in systemic insolvency, wiping out even healthy well-run banks due to a lack of sufficient funds to meet the obligations of the banking system. As Romer and others, such as Nobel Prize-winning economist Paul Krugman and Yale finance professor Gary Gorton, clearly explain, uncontained banking panics can destroy enormous amounts of wealth across the entire economy, crippling economic growth.

In the United States, we successfully addressed these structural problems with banking through the Glass Steagall Act and a number of other reforms that worked by:

  • Imposing strong prudential risk oversight on banks, thus reining in their propensity to take on excessive risk during credit expansions
  • Providing a government backstop against bank runs in the form of Federal Deposit Insurance Corporation depository insurance and various sources of government-backed liquidity, addressing the problem of bank runs 
  • Separating banking from riskier activities such as the underwriting and trading of debt and equity securities

But during the last several decades, a “shadow banking system”—one which operates through an alphabet soup of various conduits, transactions, and financial securities to finance lending—developed outside the regulatory reach of the Glass Steagall banking regime. This shadow system developed in no small part because of the actions and inaction of regulators and policy makers enamored with the idea of an unregulated efficient market. Because of the lack of regulation, shadow banks became heavily leveraged, taking on higher levels of debt to fund more risk taking. Increasing leverage is an easy way to make money when times are good, but heavily amplifies losses when times are bad.

Paul McCulley, the managing director of Pimco, one of the world’s leading bond funds, is credited with coining the term “shadow banks.” He describes them as “levered-up intermediaries without access to either FDIC deposit insurance or the Fed’s discount window to protect against runs or stop runs … [nor do they] have to operate under meaningful regulatory constraints, notably for leverage.” By “levered up” McCulley means these institutions hold only small amounts of capital against the high levels of debt raised to fund their shadow banking operations.

Because the shadow banking system lacks risk regulation or a liquidity safety net, it is highly susceptible to the bubble-bust cycle that historically plagues unregulated banking systems. This was what we saw during the last credit cycle from 2003 to 2008. During the height of the credit bubble in the mid-2000s, the shadow banking system, unrestrained by any prudential risk oversight, took on too much credit risk, reaching more than 40-1 leverage at some shadow banks such as Merrill Lynch (now part of Bank of America). As the bubble deflated, this risk translated into high credit losses, which caused investor confidence to deteriorate, causing a bank run to occur.

The susceptibility of the shadow banking system to strong procyclicality and banking panics is made more complex by the high degree of risk concentration in shadow banking. The largest financial institutions have enormous exposure to nondepository financing activities. Ordinarily, a severe banking panic develops when a large number of banks go insolvent, creating a full-fledged crisis among investors and depositors, all of whom all seek to cash out at once. But in the case of shadow banking, the high concentration of risk means that there are a small number of shadow banks so large that their failures could single-handedly cause a massive panic in the shadow banking system.

The upshot: Because the shadow banking system is dominated by a handful of financial institutions with liabilities equal to thousands of smaller banks, should any one of these financial behemoths fail it would single-handedly trigger such a panic.

This is the situation we faced in the fall of 2008, when the failure of Lehman Brothers—the smallest of the large “shadow banks”— led to a liquidity squeeze in the nondepository sector of the financial system, putting severe liquidity pressure on financial intermediaries throughout the system. As economists and financial market practictioners such as Nouriel Roubini, Bill Gross, Gary Gorton, Raj Date, and Mike Konczal (among many others) all note, this was basically a run on the shadow banking system—one that threatened to take down the entire global financial system and result in many trillions of dollars of overnight wealth destruction in pension funds, money market accounts, and insurance funds. The massive reduction in household wealth, coupled with the lack of credit that occurs after bank panics, would have had a tremendous toll on economic growth for the foreseeable future.

Thus, while conservatives such as Wallison are correct that being “too big to fail” confers a lower cost of capital from investors, they are quite wrong in claiming that it is the government’s designation of “too big to fail” that matters in this calculus. When it comes to “too big to fail,” investors are less concerned with government designations and more concerned with the answers to the following questions:

  • What would happen if this shadow bank failed?
  • Is a responsible government willing to accept those results?

The reason that “too big to fail” financial institutions are perceived as enjoying an implied guarantee and thus enjoy a cost-of-capital advantage is that investors see the answer to the first question as a massive banking panic causing trillions of dollars in wealth destruction and strongly retarding economic growth. And they see the answer to the second question as a simple “no.”

Whether the government acknowledges “too big to fail” or disavows it as a nonexistent problem is irrelevant. As long as large financial institutions can single-handedly implode the U.S. economy by going insolvent, they will be seen as enjoying a de facto government backstop.

Indeed, a recent paper by the Wharton School’s Todd Gormley, former chief economist for the International Monetary Fund Simon Johnson, and Changyong Rhee of Seoul National University bolsters this point. Their research finds that in South Korea, even in the face of a credible “no bailout” policy by the government, investors still make independent assessments that some firms are “too big to fail,” pricing that assumption into the debt they purchased from “too big to fail” financial institutions in that country. In other words, just because the government says it won’t bail out big financial firms during a crisis doesn’t mean that investors will believe them.

Protestations of the “let them fail” crowd aside, there’s no reason to think these investors are wrong. Not even President George W. Bush, the self-proclaimed cowboy president whose unerring belief in the power of unregulated markets was legendary, was willing to risk the mutually assured financial destruction that would have accompanied an uncontained run on the shadow banking system in the fall of 2008. And quite clearly, the markets already believe that “too big to fail” exists independently of government designations of firms as such, as reflected in comments made by legendary financiers such as Warren Buffett and Bill Gross, among others.

So if we reject the idea of pretending “too big to fail” doesn’t exist as an appropriate policy solution, what are our options?  There are two general policy approaches we can pursue:

  • Break up “too big to fail” institutions and thus eliminate the problem
  • Heavily regulate “too big to fail” financial institutions, thus reducing their leverage and the risk they pose to the financial system.

The Dodd bill takes a hybrid approach, imposing higher capital requirements, taxes, and other burdens on “too big to fail” shadow banks, which serve the purpose of both providing some deterrence to being so large, and reducing the risk posed by these financial institutions. And certainly, there are legitimate disagreements with this approach. Some analysts believe the Dodd bill needs to go further to increase capital requirements and force “too big to fail” financial institutions to internalize the systemic risk they pose. Others believe that we ought to break them up; too big to fail is too big to be consistent with a capitalist economy.

But whatever you may think about the substantive merits of Sen. Dodd’s proposal, it at least attempts to end the problem of bailouts by addressing the underlying causes. The “let them fail” approach, by refusing to acknowledge or remedy the problems that result in bailouts, would guarantee that future bailouts are the rule, rather than the exception. If conservatives truly want to end the prospect of bailouts, they should devise constructive solutions to the underlying weaknesses of the shadow banking system, rather than simply pretend these don’t exist.

The positions of American Progress, and our policy experts, are independent, and the findings and conclusions presented are those of American Progress alone. A full list of supporters is available here. American Progress would like to acknowledge the many generous supporters who make our work possible.