Are Big Banks Necessary?
WASHINGTON — The drive to break up the big banks has won a surprising number of adherents from both sides of the political spectrum — everyone from Neel Kashkari, the Republican head of the Federal Reserve Bank of Minneapolis, to Bernie Sanders of Vermont, the Democratic socialist who has made it the heart of his campaign.
The issue is usually analyzed from a political standpoint, focusing on whether a breakup is possible and how it could be done.
But the topic raises a critical question that is seldom asked, much less answered: Is there a legitimate, coherent business case to be made for the largest and most complex banks to stay large and complex? Do megabanks serve a critical function that justifies their undeniable risk to the financial system?
In what could be seen as a Nixon-to-China moment, Minneapolis Fed President Neel Kashkari, a former Goldman Sachs executive and one of the architects of Treasury's bailout of the largest banks, said that breaking up the big banks and turning them into public utilities may be the only way to solve "too big to fail."February 16
In a speech in New York City, Sanders vowed to remove the ability of the Federal Reserve to pay interest to banks for their excess reserves, turn the credit rating agencies into nonprofits, allow the U.S. Postal Service to offer bank products, and cap ATM fees and interest rates for loans.January 5
Democratic candidate Hillary Clinton's financial reform plan is far less ambitious than other Democrats' proposals but would still be a tough sell in a Republican-controlled Congress.October 8
The answer is, unsurprisingly, difficult to determine and varies greatly depending on whom you ask. American Banker interviewed bankers, industry representatives, analysts, reform advocates, academics and others to attempt to break down whether the presence of large U.S. banks is still necessary for the economy.
Following are the arguments over what big banks bring to the table — and a look at whether those arguments make a case for keeping the banks intact.
Megabanks bring benefits through economies of scale.
There are more than 6,000 federally insured banks in the U.S., most of which are relatively small community banks with assets in the millions or hundreds of millions of dollars. Just eight domestic institutions are considered global systemically important banks, or G-SIBs, which are subject to higher capital, liquidity and prudential requirements.
Those are: JPMorgan Chase ($2.4 trillion in consolidated assets); Bank of America ($2.1 trillion); Citigroup ($1.8 trillion); Wells Fargo ($1.7 trillion); Goldman Sachs ($880 billion); Morgan Stanley ($834 billion); Bank of New York Mellon ($377 billion); and State Street ($247 billion).
Their size enables these banks to take advantage of certain markets that are characterized by their low margin — activities like payment and clearing services, triparty repurchase agreements, prime brokerage services or custody banking. To make a profit in those markets, a bank has to be very large to take advantage of the efficiencies that come with economies of scale, enabling the banks to deal in sufficient volume.
"Instead of Boeing and Airbus, could you have a series of community airplane producers? I think the answer is no, or you could but at extremely high cost," said Greg Baer, president of the Clearing House Association and a former executive of JPMorgan Chase and Bank of America. "So as you reduce the size of the firms, you're effectively asking them to shed economies of scale and scope, and that comes with a cost."
Karen Shaw Petrou, managing partner at Federal Financial Analytics, said those activities are "fundamental to the U.S. financial markets' functioning and infrastructure," but pointed out that they could mostly be performed by smaller banks in a disaggregated system. Such an arrangement, however, would be less efficient, and therefore more expensive.
"All of the efficiencies of the G-SIBs come with added risk," Petrou said. "You would have less risk" if the banks were broken up, "but it would be at considerably less efficiency."
Petrou also noted out that many of these activities do not inherently have to be performed by banks — they could almost as easily be performed by very large private equity firms or other financial services firms. If those firms were to take up such activities, however, they would likely have to be very large in order to take advantage of the same efficiencies. A bank breakup would move the risk to another part of the economic system — but it wouldn't eliminate the risk.
Marcus Stanley, policy director for Americans for Financial Reform, acknowledged that the big banks did have economies of scale, but said there's little data to demonstrate that the largest G-SIBs need to be as big as they are to take advantage of them. Perhaps a bank would need to have several hundred billion in assets to compete in certain markets, but academic research on the subject is heavily dependent on how the research is modeled.
"Say you're a derivatives and repo dealer operating on a global scale. That's something that you have to be big to do, but these are also businesses that would have collapsed without massive public support in 2008," Stanley said. "If I'm looking at the period when I'm making money, things might look good, but if I'm looking at the period when I'm taking the public bailout, not so good."
Dennis Kelleher, president of the public advocacy group Better Markets, argued that any legitimate market function that the big banks are performing would not simply disappear if those banks were made smaller and less complex. Perhaps they would be provided at greater cost, but that cost would likely be a better reflection of external costs posed to the public by the banks' risk to the taxpayer.
"I have no doubt that if there is a service that the market wants ... that that service will be provided by somebody," Kelleher said. "Will it be provided at the exact same price? Probably not. Am I worried about that? No."
Megabanks' global reach is critical to the U.S. economy.
Regardless of whether the G-SIBs' participation in repo markets or other activities requires or justifies their size, their size enables those firms to have presences and activities in markets across the globe — hence the "G" in "G-SIB."
Tony Fratto, a partner at Hamilton Place Strategies and former Treasury official in the Bush administration, said that distinction is the main difference between the largest banks and smaller ones — even super-regionals that may have hundreds of billions in assets. It is a key benefit to the giant multinational corporations and midsize companies that are growing and looking to export their products or break into international markets, he said.
"Because they're a little bit smaller and they're breaking into new economies — and their current relationship is with a larger regional bank — they don't want to go out and hire four different banks to do that. They want to find a bank ... and there are only a handful of banks [that] can do that," Fratto said. "Some are American, some are foreign, but it's not a long list."
The companies that hire the big banks share this view. David Hirschmann, president and chief executive of the Center for Capital Markets Competitiveness at the U.S. Chamber of Commerce, said his members' main priority when it comes to financial services is choice. Some prefer larger banks to smaller ones or vice versa, or prefer differently sized banks for different needs.
But having the option to choose a large bank is something that policymakers should want to preserve.
"There are unique capabilities that are well served by midsized banks, others that are well served by regional banks, but there are also things that you need globally capable banks" to accomplish, Hirschmann said. "The question of size is really the wrong question. It's a question of, How do we preserve the diversity of the system, and a system that will meet every need of every type of business in the United States?"
Bank critics, however, are not convinced. Kelleher said banks are active in generating research to demonstrate that they facilitate real commerce, but they don't provide enough hard evidence.
"There's no robust data set that can be independently analyzed that confirms a single claim made by a 'too big to fail' bank for some purported service or product that is desperately needed by somebody in the real economy that won't be provided if they were broken up," Kelleher said. "Zero. There's none."
Stanley acknowledged that the largest banks' global presence can facilitate commerce, but said that banks are overstating their case. The proportion of banks' exposures across international borders has exploded, he said, from roughly 13% of U.S. GDP in 2000 to 27% in 2008 (and 23% in 2015, reflecting a dip following the euro crisis in 2011). Over the same period, GDP growth has been more or less flat and the trade deficit has grown, he said, suggesting that the banks' expanding international reach may not be such a boon to growth.
"There's been this huge increase in international cross-border bank exposures just since the year 2000, and it's an increase that has happened much more rapidly than the increase in international trade," Stanley said. "Those cross-border bank exposures add to financial instability. I think it's somewhat questionable whether they're giving real economic benefits that correspond to that."
Megabanks serve as the "buyer of last resort."
It is a frequently reported statistic that three of the four largest U.S. banks — JPMorgan, Bank of America and Wells Fargo — have gotten bigger since the crisis. But it is also true that those banks acquired several failing banks, brokerages and financial services firms at the height of the housing crash, in many cases at the urging of the government itself.
Bank of America famously bought the mortgage originator Countrywide for $4.1 billion in 2007 and later bought the brokerage Merrill Lynch for $50 billion in 2008 (and nearly bought Lehman Brothers). JPMorgan bought the investment bank Bear Stearns in 2008 at the urging of federal officials and later bought most of the assets of Washington Mutual from the Federal Deposit Insurance Corp. receivership in 2008 — acquisitions that chief executive Jaime Dimon told Bloomberg last week may have been a mistake. Wells Fargo, meanwhile, bought Wachovia after federal officials had already brokered a deal to arrange a sale of the bank to Citigroup.
H. Rodgin Cohen, senior chairman at Sullivan & Cromwell, said that, after taking those acquisitions into account, the largest banks have shrunk considerably since the crisis. And those acquisitions were brokered in large part because the federal government was afraid of what would happen if those institutions failed and went into bankruptcy the way Lehman Brothers did in 2008, he said.
That makes the G-SIBs a source of financial stability rather than a source of uncertainty, he said. And 2008 was not the first time regulators have turned to the banks to provide that service — similar deals have been arranged in the Texas oil patch recession of the 1980s and the Ohio thrift crisis a few years earlier.
"A very important role that the big banks play is, in effect, the buyer of last resort when we have real fissures in the banking system," Cohen said. "Time and time again, they have provided support when there have been weaknesses in the banking system."
Other bank advocates have argued that the G-SIBs' bigness is precisely their source of strength, because having an expansive business with interests in a wide variety of products, services, markets and geographies acts like a natural hedge against volatility.
Baer said that diversity allows banks to weather downturns in ways that smaller firms can't.
"We don't have large firms only doing one thing," Baer said. "They are diversified geographically, and they are diversified in their product lines, and that makes them more stable. Diversification is the only free lunch in finance, and the largest institutions have that."
But Stanley noted that just because regulators sometimes appeal to large banks to buy smaller, troubled banks in times of stress doesn't necessarily mean that they should.
Having a bank buy another bank is easier than having a bank go into receivership and bankruptcy, but it is just taking the downside risk of a failed institution and diluting it within a larger and more systemically risky institution.
"There's no question that one of the bank resolution methods that was used in the crisis, and has been used at other times, is to find a big fish to swallow up the little fish," Stanley said. "So it's convenient for regulators ... but we've been doing that since the '80s. Is that a good thing? That because we can't resolve banks on their own or put them through bankruptcy, we're continuing to concentrate our financial system by gluing these failed banks onto an even bigger bank?"
There are also problems with concluding that the largest banks emerged from the crisis as more stable because of their size and diversity, Stanley said, because it fails to account for the government's explicit position that the largest firms would not be allowed to fail. Whether the G-SIBs would have escaped the crisis is impossible to know for sure, because unlike Wachovia and Washington Mutual, failure was not an option.
"The bigger, more diversified banks were just flat-out not allowed to fail," Stanley said. "It's very hard to disentangle this thing of who was going to fail and who was permitted to fail."
It's not clear where to draw the line of what's a megabank.
Last month, Kashkari became the latest regulator to ask whether large banks need to be broken up or forced to hold higher capital, making them into public utilities. His argument rested on the conclusion that the post-crisis supervisory regime wasn't adequate to address another crisis.
Within minutes of his remarks, several former regulators took issue with that presumption. Don Kohn, former vice chairman of the Fed, said he didn't know why the living wills, capital rules and other reforms put forward in the Dodd-Frank Act would not work in a crisis. And former Minneapolis Fed President Gary Stern similarly expressed skepticism that the reforms do not go far enough.
Many bankers share that skepticism. Baer said that Dodd-Frank's combined emphasis on higher capital, more guaranteed liquidity, and — most recently — a plan for bailout-free resolution plans as part of the living wills and Total Loss Absorbing Capacity rules have made the banking sector safer. The Fed's stress testing regime also demonstrates the banks' resiliency on an annual basis. If Kashkari or anyone else believes that is not enough, it is up to them to prove it.
"I think the onus is on those saying that the banks don't have enough capital to weather a financial crisis to show why they believe that to be the case, because all the evidence appears to be to the contrary," Baer said. "Even in the extraordinarily unlikely event that they did fail, they would fail in such a way that would not require taxpayer support and would not have systemic consequences."
John Dearie, acting chief executive of the Financial Services Forum, argued that there is little reason to assume that, despite five-plus years of regulatory and legislative progress in shoring up the financial system, more drastic measures are needed.
"There's been tremendous progress since 2009 to end 'too big to fail' and strengthen the banking system, and to simply drop out of the sky seven years down the road and say not enough has been done is unfair to the industry, to policymakers and to regulators, all of whom have been working terribly hard on this," he said.
If policymakers decided to break up the big banks, however, it's not clear how it should be done or where the line should be drawn. In the case of the largest G-SIBs, even if they were broken up into four or five separate banks, those new banks would still hold hundreds of billions in assets — far above the $50 billion threshold in Dodd-Frank for banks to be considered systemically important financial institutions.
While most policymakers believe that trigger is too low, there's no agreement on what the right level would be.
"That's our working statutory systemic designation," Petrou said.
Stanley said that proposals to break up the banks generally fall into two categories. One is something like a reintroduction of the Glass-Steagall restrictions on combining commercial and investment banking, which would set a clear limitation on activities. The other is a formal cap on size.
"Neither of these things are weird ideas from Mars — both have track records," Stanley said. "We had Glass-Steagall for decades, and the usual size cap people talk about is something in the multiple hundreds of billions of dollar range — $500 billion or something like that. That's still a very large bank."
But Fratto said the idea of a formal cap on size for the banking industry alone is unfair and problematic — especially since no one is entertaining calls for a similar cap on other businesses dominated by very large firms like aerospace, telecommunications or technology. Regulators would be better served by enforcing and revising the regulatory framework already in place.
"It just doesn't make sense to have these rigid ideas for what is required," Fratto said. "That is not to dismiss the other point of safety and soundness. ... I get that goal, it's just there are better ways to achieve it than the blunt instruments of 'Cap their size, turn them into utilities.' It would have been like after the Titanic hit the iceberg, we all just decided, 'All right, no more big boats.' "