Breaking Up the Big Banks Would Presumably Be a Big Disaster

Bernie Sanders had quite the interview with New York Daily News a few days back. Part of the interview entailed his plan to break up the big banks, the ones that are labeled "too big to fail".  Essentially, the "too big to fail" theory postulates that certain [financial] institutions are so large that the failure of one or two large banks will bring down the whole system. To mitigate this contagion effect, Sanders would like to chop up the larger banks into smaller, less connected institutions to minimize the effect of institutional failure of a certain financial institution. Only a couple months ago did the President of the Minnesota Federal Reserve Neel Kashkari announce that he was going to create a plan by the end of this year to break up banks that are "too big to fail" (TBTF), so at least this time, there is some independent support of a Sanders platform policy. The question here is whether breaking up big banks would work or if it is another one of Sanders' overly simplistic idea to solve something as complex as financial market failures. Let's bring up some of the finer points of the problems faced with breaking up the bigger points. While I will use hyperlinks to various sites for citation purposes (also read Clearinghouse presentation for more information on big banks), an October 2014 policy report from the Bipartisan Policy Center [BPC] (also on the Brookings Institution website) on breaking up big banks (which is well worth the read if you're particularly interested in the topic) will provide a sizable amount of content:

• Breaking up big banks is no guarantee of preventing financial crisis. While TBTF is a plausible theory, the underlying rationale for determined whether a bank is TBTF has never been demonstrated as true. Looking back at the Great Recession, financial institutions, whether large or small, would have collectively had the same incentives to operate with too little capital or liquidity (BPC, p. 35). Paul Krugman brought this up in an article last week lambasting Sanders for his naïveté on the issue. As Krugman pointed out, predatory lending was carried out by smaller, non-Wall Street institutions such as Countrywide Financial, and the crisis itself was centered on Lehman Bros., a small "shadow bank." Bear Sterns, Washington Mutual, and Wachovia were also smaller financial institutions that were central to the financial crisis, and "breaking them up" would not have stopped the overindulgence in risky mortgages. Other institutions, such as AIG, Freddie Mac, and Fannie Mae were not banks, but greatly contributed to the Great Recession. We also have to recall that the Great Depression was not caused by big banks, but a flurry of small bank failures. 
• Size matters. It might be politically expedient to malign "evil corporations" or billionaires, but the truth is that there are advantages that come with larger banks. Large banks come with economies of scale, which is the cost of unit per output decreases with scale since fixed costs are more spread out with each unit of output. Economies of scale is helpful for platform creation and developing human capital. A larger scale means being able to underwrite a large bond, loan out larger sums of money, and expand customer base. Globalization has also meant that being able to contend with international regulatory compliance and develop a more nuanced information-technology infrastructure (BPC, p. 19). Banks can also provide a wider range of [complementary] services, including financing, foreign exchange, risk management products (e.g. derivatives, and other operational services (BPC, p. 20). As another example, the widespread usage of the ATM machine was made possible because of large banks (BPC, p. 22). Larger banks also help the customer since the costs of infrastructure, technology, and capital expenses are spread out over a larger customer base. All of these benefits are either easier to come by with larger banks, or can only come into fruition with larger banks.    
• How big is "too big?" There is no objective way to determine what is "too big" in terms of asset size. Until we can measure the costs and benefits of breaking up banks more effectively, there is no way to determine what is "too big (BPC, p. 35)." And while we're on the topic of size, if we compare banking to other markets, banking is relatively not concentrated, especially when you compare it to pharmaceuticals, automobiles, and computers. Also, when comparing American banking to that of its foreign counterparts, the United States' banking market is far less top-heavy. When looking at the assets of the United States' five largest banks, it is less concentrated than any of the other G-7 nations, and is less concentrated the the G-20 average (BPC, p. 31). The banking assets to GDP ratio in this country is also lower than the U.S.' foreign counterparts. 
• Transition costs. For one, there would be a loss of customer focus. Transitioning would be all about internal reorganization. Customer relations would also be interrupted. How would you untangle the network of assets and liabilities interwoven throughout the global economy? You would have to renegotiate millions of contracts, not to mention the litigation caused by the decisions made during the transition (BPC, p. 37). Another point to be made: We don't put a profit cap on other companies such as Google, Apple or Wal-Mart and declare them "too big to fail." Much like any other company, if you tell a company they can only make so much profit, I can tell you right now that is going to perversely affect how banks act. Given the desire to make up for the loss in market share, more would venture in the world of "shadow banks (BPC, p. 38)," or better yet, increase the cost of doing business or cutting jobs. And remember that when talking about job loss, we wouldn't be just talking about big-whig financiers, but working-class tellers, loan officers, secretaries, administrators, security guards, and janitors. We're talking about a financial sector with over 5.7 million people, so while some upper management would lose jobs, a lot of cuts in pay and pink slips would hit the working class for which Sanders purports to advocate.
• Domestic assets only? Even if you have determined what is "too big," there is another question: will this regulation cover international assets? If it covers both domestic and international assets, then you are encouraging a more insular market. If it covers just domestic assets, you are encouraging capital flight (BPC, p. 36). Neither are conducive for the financial sector.

Dodd-Frank, with its hundreds of pages of financial regulation, make it difficult for banks to become too big. The U.S. Government Accountability Office (GAO) published a report in December 2015 showing how Dodd-Frank adversely affects community banks and credit unions because compliance is so difficult. One can argue that such high level of compliance actually has made big banks bigger than ever. It can also be argued that during the Great Recession, the larger banks had the economies of scale to acquire the smaller, failing banks that helped make sure that a recession didn't turn into a depression. Both theories are plausible and not mutually exclusive, that's for sure.

We can discuss whether capital rules should be more stringent, whether we should create incentives to shift the cost to investors so that bad doesn't turn into worse, how we can contain liquidation processes, or whether the bigger banks should have less tax exemptions. However, one thing is clear: breaking up big banks would not do the United States economy any favors. Breaking up big banks would impose costs on banks, which would be passed down to U.S. companies and consumers. Less available credit that would cost more to access and a larger trade deficit would be but two major costs imposed upon the United States economy. Also, financial regulators are a long way away from figuring out what causes systemic risk. Capriciously dismantling the financial sector to fulfill some populist whims is something the global economy can ill-afford.