Breaking Up the Big Banks: Is Anybody Thinking?

Editor’s Note: Peter J. Wallison is a senior fellow at the American Enterprise Institute. This post is based on an article by Mr. Wallison; the full article, including footnotes, is available here.

Breaking up the biggest banks is said to have growing support in Congress, but the idea’s supporters—even those who are respected commentators—do not appear to have given it any deep thought. Without any serious discussion, it should come as no surprise that the idea has bipartisan support among the American people. But Martin Baily of Brookings, always levelheaded in his judgments, calls it “nuts.”

Obviously, for the United States to break up its largest banks would be a very consequential step with significant implications for our economy and financial system. Before proceeding, we should have a reasoned debate on the costs and benefits. Instead, what we have had thus far is a surprising chorus of commentators calling for breaking up the banks without seeming to give any attention to the most elementary issues such a step would entail.

This article will lay out some of those issues. These are not technical matters; they are the simple, first-order questions that ought to occur immediately to anyone who supports the idea of breaking up the largest banks—and they have been largely ignored. Ultimately, this is a depressing commentary on how our discourse on important matters of financial regulation and financial structure has descended—in this era of 24/7 media and instant reaction—to the level of slogans and bumper-strip opinionating.

This is not to say that breaking up the largest banks into smaller entities is necessarily a bad idea; I will not argue for or against it in this Outlook, although I will outline some of the more obvious pros and cons. What is clear, however, is that—whatever its long-term benefits—such an action would be highly disruptive to the economy and the financial system. Millions of existing relationships between banks and their individual or company clients would have to be renegotiated; lines of credit that were possible with large banks but not with smaller ones would have to be terminated; employees of large banks engaged in activities that smaller banks would not be able to pursue would have to find other things to do; US companies operating abroad that rely on the assistance of US banks may have to find that assistance, if available at all, from foreign banks. This is only a small list of the revolutionary changes in the US financial system that would have to attend the breakup of the largest US banks.

Moreover, it is reasonably clear—shown by their continuing profitability—that the largest banks are performing valuable financial and other services. The idea of breaking up successful businesses is always troubling unless—as in the case of, say, the old Standard Oil trust—they are substantially harming the economy, an argument that no one has made. Indeed, the argument that the largest banks caused the financial crisis when they stopped lending to one another after the Lehman bankruptcy demonstrates their centrality to the smooth functioning of the global economy.

To be sure, numerous benefits are said to flow from breaking up the biggest banks. Among them are preventing a large bank failure that might bring on another 2008-like financial crisis, assuring that these large and complex institutions can be effectively managed, and reducing the risk-taking and funding advantages that result from their status as too-big-to-fail (TBTF) institutions. But the costs never seem to be taken into account.

Of all these reasons, the most important by far is the possibility that the failure of any of the largest banks might present a danger to the stability of the US financial system—the reason they are said to be TBTF. That, in my view, is the only reason serious enough to warrant the disruption and the risks that would accompany a significant downsizing of the largest banks. Accordingly, I will focus on the TBTF issue and the questions that must be considered—at a minimum—before we proceed with something as consequential as a breakup effort.

This discussion will be framed by the following five questions, for all of which the proponents of breaking up the largest banks should have answers.

1. Is TBTF real? Before we decide to break up the largest financial institutions, we should be satisfied with the idea that the failure of a large financial firm will actually cause instability in the financial system.

2. What size would a bank or other financial institution have to be before it would not be considered TBTF? It would be foolish to go through the disruption associated with breaking up the biggest banks only to find that their smaller progeny are still being rescued.

3. What kinds of institutions should be broken up? If it makes sense to break up the big banks, then why not the large insurance companies, finance companies, and holding companies of various kinds? If we recoil from that idea, why are we treating banks differently?

4. What would it mean for economic growth if, because of downsizing, the largest banks were no longer able to carry out the functions that they perform in the US and global economy today?

5. Why is the Dodd-Frank Act inadequate? The Dodd-Frank Act has imposed enormous regulatory costs on the US financial system, all for the purpose, ostensibly, of preventing another financial crisis caused by the failure of a large financial institution. Calls for the breakup of the largest banks, then, many of them from avid supporters of Dodd-Frank, are implicit statements that the act has not achieved its essential purpose. Why, then, should it not be repealed?

Is TBTF Real?

Asking whether TBTF is real is an appropriate preliminary question because the notion that the failure of a large financial institution could bring on another financial crisis—the underlying rationale for calling a financial firm TBTF—has never been demonstrated. Is it sensible to proceed willy-nilly with a potentially painful downsizing of the largest banks when we are not even sure that the thing we fear is anything more than a figment of the regulatory imagination?

That the largest banks are TBTF is a plausible theory, but no more than that. It has a basis in reality because, in the past, including during the recent financial crisis, regulators have acted on it, taking extraordinary actions to rescue large financial institutions because they feared that the failure of these firms would cause a serious disruption, or even a collapse, of the financial system. Thus, the term “too big to fail” is shorthand for the more descriptive phrase “too big in the opinion of regulators to be allowed to fail.”

Of course, regulators’ belief that the failure of a single institution will shake the foundations of the financial system does not make it so. Would the system have come apart if Bear Stearns, AIG, Wachovia, and Washington Mutual (WaMu) had been allowed to fail? Although many government officials and others make this assertion, we do not actually know the answer, simply because Bear and WaMu were sold to JPMorgan Chase, Wachovia was sold to Wells Fargo, and AIG was rescued with a cash infusion from the Federal Reserve. The one large financial institution that was actually allowed to fail was Lehman Brothers, and that event produced ambiguous evidence for the notion that large nonbank financial institutions are TBTF.

The evidence from Lehman’s failure is ambiguous because when the firm failed it did not drag any other financial firm down with it, although the TBTF theory posits that this should have happened. To be sure, in the panic that ensued after Lehman’s bankruptcy, a money market mutual fund called the Reserve Primary Fund “broke the buck” (was unable to meet the requirement that each of its shares could be redeemed at the value of one dollar) because of the losses it suffered in holding Lehman commercial paper. The losses for the shareholders of this fund eventually totaled about one cent for each share. If this had happened in an otherwise stable market, it is highly unlikely that it would have produced runs on other funds. However, in the panic that ensued after Lehman, the Reserve Fund’s loss set off runs on other funds.

Outside the money market fund industry, however, other financial institutions of all sizes were unaffected by Lehman’s failure, although the panic caused banks to hoard cash because of fear that their depositors and other creditors would seek to withdraw funds or draw on lines of credit. None of the institutions rescued after Lehman—Wachovia, WaMu, and AIG—were made insolvent or unstable or had to be rescued because of exposure to Lehman. The fact that Lehman did not drag these others down with it—even in the midst of a panic—strongly indicates that the failure of a large financial institution is not likely to set off a systemic collapse if the market is otherwise stable at the time of the failure.

Events that surrounded the failure of the large investment bank Drexel Burnham Lambert in 1990 further support this position. Drexel was the fourth largest investment bank at the time, roughly equivalent to Lehman’s relative size in 2008, and failed at a time that the other firms were not considered unstable or insolvent. The Drexel bankruptcy created no serious knock-on effects for others. Accordingly, in trying to determine whether or not TBTF is a valid idea, the context in which a failure occurs may be the most important fact. Some evidence exists—primarily from what followed the losses at the Reserve Primary Fund—that TBTF is real when the market is already in a panic mode, with many firms looking weak or insolvent. But when the market is stable, regulators are likely to be overreacting if they proceed with a rescue. Since financial panics are extremely rare—the last one was in 1907—assuming that banks and other large financial firms are always TBTF appears to be a serious policy error. Thus, the supposed dangers of TBTF ought to be examined carefully before, as a prophylactic measure, we think about breaking up the largest banks; we may be configuring the financial system to deal with a remote possibility, while at the same time stifling the economic growth that will improves peoples’ lives.

What Size Makes a Firm Not Too Big to Fail?

Those who would break up large banking organizations are seldom asked what size they think would take a bank out of the TBTF category. It would be surprising if they actually had an answer. The fact is that no one—least of all the proponents of a breakup—seems to have any idea what size a bank must be to avoid characterization as TBTF. To use the terminology of former defense secretary Donald Rumsfeld, this is a “known unknown.”

The Dodd-Frank Act arbitrarily sets $50 billion in assets for banking organizations as the TBTF threshold, but this is absurdly low. It is not credible that the US economy would suffer a systemic breakdown if a $50 billion bank failed. And it would make no sense to endure the economic and employment disruption that would be associated with reducing the four largest banks—JPMorgan Chase, Citi, Bank of America, and Wells Fargo, representing about $6 trillion in assets—to 120 separate $50 billion institutions.

More significantly, there are three principal reasons that no one can answer the “what size” question. First, as outlined above, the decision on whether a financial institution like a bank is TBTF depends entirely on a judgment by an institution’s regulators, and of course no one can know how the regulators in office when a bank is failing will make that determination.

Second, the decisions of the relevant regulators will be substantially affected by conditions in the financial market at the time a failure occurs. If the market is healthy and there is little doubt about the financial condition of other financial institutions, investors and creditors will not believe it necessary to withdraw their funds from institutions they perceive as healthy, even if a similar institution fails. However, they are likely to panic and be inclined to run if many institutions look shaky at the same time. This is exactly what happened when Lehman failed in 2008, at a time when investors and creditors were anxious about the condition of many financial institutions. Lehman’s failure triggered a panic that caused many banks to hoard cash to meet demands for withdrawals from their depositors and creditors. The fact that banks refused to lend to one another during this period was the principal and most frightening characteristic of the financial crisis. When the market is close to panic, regulators will be more likely to consider an otherwise marginal firm TBTF.

Third, regulatory discretion cannot be limited in advance by arbitrarily establishing an allowable size for a bank. As the economy grows, the significance of a bank’s size will be affected by the size of the financial system as a whole. In 1984, the Continental Illinois Bank—then a $40 billion institution—was considered TBTF, and its creditors were bailed out by the Fed. Today, a $40 billion bank would not be considered a threat to the stability of the $13 trillion US economy. Thus, the Dodd-Frank Act notwithstanding, any arbitrary number set as the allowable size for a non-TBTF bank could never be accurate. Under these circumstances, it is difficult to understand what commentators might be thinking when they recommend that large banks be broken so that they are not TBTF. They cannot have any idea what size a non-TBTF bank should be, and even if they did, their estimate would not be the correct several years later, nor necessarily reflect the judgment of the bank’s regulator when the bank is about to fail. The surprising thing is that lawmakers and commentators of all levels of expertise routinely announce their support for breaking up the large banks when they cannot really have any idea what the resulting size should be.

A good example of this is a speech by Richard Fisher, president of the Dallas Federal Reserve Bank, in November 2011, in which he suggested that an international conference could decide what size banks should be allowed to attain. That is like predicting the weather on a particular day in 2020. A convocation of regulators will not be able to decide by consensus, years in advance, whether or not a financial institution on the brink of failure will be TBTF.

Similarly, it would be truly absurd to endure the enormous controversy and financial disruption necessary to break up the largest banks, only to find that when one of the newly independent smaller institutions gets into financial difficulties, it is rescued anyway. Until someone can say definitively what size will not tempt regulators to rescue banks, breaking them up because they are TBTF seems like a fool’s errand.

Break Up Nonbanks?

The Financial Stability Oversight Council (FSOC) is a new agency established under the Dodd-Frank Act, consisting of all the federal financial regulators and headed by the secretary of the Treasury. As authorized by the act, the FSOC is currently in the midst of examining large nonbank financial firms—insurance companies, hedge funds, finance companies, securities firms, and private equity firms, among others—to determine which are suitable for designation as “systemically important” because their failure may create instability in the US financial system. Once that designation is made, Dodd-Frank requires that these firms be turned over to the Fed for what the act calls “stringent” regulation. The Fed in turn has made clear that it will impose basically the same rules on these nonbank firms as it will apply to banking organizations.

Thus, nonbank firms that have been designated as systemically important are in the same regulatory category as banks because both are deemed to represent the same danger of a systemic breakdown. They will also be subject to resolution by the Federal Deposit Insurance Corporation (FDIC) under Title II of Dodd-Frank in essentially the same way that banks are resolved. Accordingly, if we believe it is necessary to make banking organizations smaller to avoid TBTF, we should be willing to do the same for insurance firms and all the others the FSOC is currently examining. As many commentators have noted, “If you’re too big to fail, you’re just too big.” There does not appear to be any principled or logical basis under Dodd-Frank for treating banks and nonbanks differently.

Still, there have been no calls for breaking up large nonbank firms. The reason for this is difficult to discern. It’s not a function of size; the largest banks tend to be somewhat larger than the largest insurance companies, but again, no one has any idea what size would make a financial institution not TBTF, and the Dodd-Frank Act designates $50 billion as the size at which a bank may be TBTF.

It is noteworthy in this connection that Senator David Vitter (R-LA) and Representative Scott Garrett (R-NJ) recently introduced legislation in the Senate and House that would eliminate the authority of the FSOC to designate nonbank financial institutions as systemically significant. The bill applies only to nonbank financial firms and does not attempt to modify the Dodd-Frank Act’s determination that all banking organizations with more than $50 billion in assets are systemically important. If proponents of Dodd-Frank oppose this legislation because they believe that nonbank financial firms should continue to be designated as systemically important firms regulated “stringently” by the Fed, they should be prepared to explain why those institutions should not also be broken up into smaller units. Thus far, no one has attempted to explain either why large nonbank financial institutions that have been labeled as systemically important should not be broken up or why they are different from banks in this respect.

What Would It Mean to Break Up the Biggest Banks?

If someone is willing to suggest breaking up the biggest banks without having any idea what size they should be, how much thought could have been given to the disruptions in the economy that would follow from such an act? As far as I can determine, no one who has advocated a large bank breakup has attempted to assess the consequences in any realistic way.

This is not the place to list all those consequences, but here are a few of the most obvious questions that breakup supporters should be able to answer:

Trillions of dollars flow daily through the New York clearinghouse payment system, which is operated by the largest banks. Could the clearinghouse international payment system continue to operate effectively if the largest banks were broken up into many smaller institutions?

US corporations operating around the world use US banks to transfer funds, make payroll, furnish short-term liquidity and cash management services, provide data and data transfers, make available letters of credit to facilitate trading in goods and services, and offer standby lines of credit for those periods when their clients cannot access the securities markets. How would globally active US firms be affected if US banks were no longer large enough to operate globally?

The risk-management activities for hundreds of thousands of firms around the world are managed by the world’s largest banks through interest rate swaps, currency swaps, commodity swaps, and credit default swaps. Hundreds of trillions of dollars in derivatives are involved. What would happen to the world’s risk-management system if US banks were no longer able to participate in these activities?

There are very few exchanges for fixed-income securities—bonds and notes issued by governments and private firms to finance their activities. These instruments can be considered liquid only because the largest banking organizations make markets in them by offering to buy and sell from a portfolio they hold. This assures traders and investors there will be a buyer when they want to sell, and a seller when they want to buy, and substantially reduces the liquidity risk associated with owning most fixed-income securities. How would this market function without US banks, and what would happen to the costs of issuing new debt securities if US bank capital and market-making capacity were removed from the market? An indication of the answer to this question is the large number of protests from foreign governments (among other issuers) about the possibility that the Volcker rule might restrict US bank market-making. Breaking up the banks so that they are too small to make markets would go well beyond the Volcker rule—even adversely affecting the trading of US government securities, which were exempted from the Volcker rule.

The four largest US banks employ more than 1 million people, and the top ten employ almost 1.5 million. If all these banks were downsized substantially, they would have to stop performing a lot of their current functions. What would be the effect on the US economy and the many communities in which these banks operate if they had to close branches and shut down operations?

We could ask many more questions about the breakup of the largest banks. After the “what size” question, those who advocate breaking up the largest banks should also be able to answer at least the most basic questions about consequences.

What’s Wrong with Dodd-Frank?

Proposals to break up the biggest banks or financial institutions reflect a conclusion that the reforms embodied in the Dodd-Frank Act are not sufficient to prevent the adverse consequences of TBTF. Since the whole purpose of Dodd-Frank was to prevent another financial crisis, it is more than a bit ironic that many of the same people who saw no deficiencies in Dodd-Frank now advocate going further—breaking up the largest banks—because Dodd-Frank has not apparently achieved its supposed purpose.

I certainly will not defend the Dodd-Frank Act, but anyone who wants to break up the big banks should be able to explain why the act’s provisions are insufficient. These provisions include the act’s determination that every banking organization larger than $50 billion is systemically important—that its failure could produce instability in the US financial system. The act then subjects these institutions to “stringent” regulation by the Fed—likely to include highly problematic exposure limits—and imposes a range of additional restrictions, provisions, and rules, such as enhanced capital requirements, restrictions on exposures to one another, capital surcharges on the most systemically important institutions, new liquidity requirements for all banks, annual stress tests run by the Fed, and the Volcker rule restricting proprietary trading. In addition, there is the Orderly Liquidation Authority under Dodd-Frank’s Title II, which gives the secretary of the Treasury the authority to seize any financial firm and turn it over to the FDIC for resolution. This, too, was intended to prevent the adverse consequences of the failure of a financial institution that is TBTF and thus was said by the act’s supporters to have eliminated TBTF.

All of these measures were, in theory, designed to prevent the failure of the largest banks from causing instability in the US financial system. The calls now for breaking up these banks seem to be an admission that the much-vaunted Dodd-Frank Act will not do the one thing it was designed to do—prevent the failure of large financial institutions from causing another financial crisis. For those of us who favor the repeal of Dodd-Frank, it is good to know that others have recognized its failure to meet even the central goal of its framers, but if this is so, the advocates for breaking up the big banks should join in an effort to repeal the act.


It is not a good time to be a big bank in America. A recent article in American Banker noted broad support at both political conventions for breaking up “mega banks” and observed that the popularity of “what was once a radical suggestion implies trouble for megabanks that have proven vulnerable to scandals and large errors over the past summer.” A media campaign to vilify banks and bankers is not surprising; neither is populist support for radical steps after years of a media campaign. What is surprising, however, is the number of supposedly sophisticated commentators who allow themselves to be quoted in favor of something that is roughly equivalent to jumping off a cliff without knowing the depth of the water below.

Both comments and trackbacks are currently closed.


  1. jp99
    Posted Saturday, September 29, 2012 at 6:27 pm | Permalink

    Interesting discussion of the detail implications of breaking up the big banks – akin to “how many socially useless bankers can dance on the point of a needle?”

    “Banking” will continue to explode on a regular basis until we get the political will to nationalize all of them and treat them like a regulated utility that is there to serve the people. Let’s get on with it and debate that – nationalization – rather than the pointless angst over the current failed “private enterprise” model.

  2. TAL
    Posted Monday, March 21, 2016 at 6:52 pm | Permalink


    Right on! Enough with this endless, worthless, academic diatribe that goes nowhere & solves nothing. Get on with the solution you recommend!