Bernie Sanders wants to break up the largest banks. He blames them for the financial crisis and accuses them of overcharging customers. It’s not really clear that breaking up banks would reduce the risk of another financial crisis. Small banks fail, too, and can bring down other banks with them. But the accusation that banks gouge customers is a more interesting and complicated one. If Sanders is right about it, however, breaking up the banks will not solve the problem. The real villains are not the banks but their owners—the respected mutual funds and institutional investors who are not (yet) in his sights.
For quite some time, depositors have complained about banking fees. Banks sometimes charge you a fee to open and maintain a checking account; they often charge people fees for using ATMs managed by other banks. A great deal of controversy has surrounded “overdraft fees,” charged when customers write a check for an amount larger than the balance in their checking account. Critics argue that banks make money on overdraft fees because customers underestimate the risk of bouncing a check, which can occur just because of the timing that banks use to record credits and debits; banks in this way make money on mental lapses that everyone is prey to.
The controversy over fees has always puzzled economists. It’s strange that people complain about fees but not about the bank’s failure to pay more than 0 percent interest on checking accounts. The fees also reflect banks’ costs. Someone who never bounces a check or uses ATMs owned by other banks saves his bank some money—and so he is rewarded by being spared the fees that others must pay.
Moreover, Econ 101 tells us that when a market is competitive, the sellers will set prices at just enough to cover their costs. Although there are some huge banks like JP Morgan and Citi, the market in banking is, in fact, quite competitive. In most places—and certainly in densely populated cities—you can choose among dozens of banks; you can also use online banks. The competition among banks should drive down prices and ensure that depositors are compensated fairly. It should also ensure that banks efficiently choose whether to pay depositors in the form of interest or low fees.
However, a new paper by a group of economists casts doubt on this happy story. José Azar, Sahil Raina, and Martin Schmalz argue that the long-standing assumption that the market for banking services is competitive is based on a defective measuring rod. This traditional measuring rod is called the Herfindahl-Hirschman Index. The index is just a fancy way of measuring the amount of competition in an industry based on the market shares of the various participants. Imagine a small town with 100 banks, each of which owns 1 percent of the market. The HHI index will be very low, reflecting the assumption that each of the banks must compete vigorously for customers by offering them good service and low prices. If one of the banks buys the other 99, the HHI index will shoot through the roof—we now have a monopoly. If, instead, the market consolidates to two or three large banks and a handful of small banks, the HHI index will be somewhat less high, indicating that competition exists but is not robust. Prices will fall somewhere between the competitive and monopoly level.
Because numerous banks exist in most markets, the HHI for banks is quite low, and that is why economists have thought of the banking industry as competitive, and the prices as presumptively fair. However, Azar and his co-authors make an important observation: While many banks exist, different banks are frequently owned by the same entities. If 100 banks exist, but they are all owned by just a few institutional investors, then competition may be low rather than high.
Imagine, for example, that your neighborhood is served by six banks—JP Morgan, Bank of America, Citigroup, Wells Fargo, U.S. Bancorp, and PNC Bank. You might think that these banks—which happen to be six of the the largest U.S. banks—would compete vigorously for your deposits. However, it turns out that the biggest shareholders of these banks are, roughly, the same companies: BlackRock, Vanguard, State Street, Fidelity, Wellington, and Berkshire Hathaway. These institutional investors earn profits by owning shares in other companies and managing shares owned by their clients.
While six banks would—if owned by different groups of people—compete vigorously with one another so as not to lose market share, the story changes when the six banks are owned by the same groups of people. The institutional investors all gain if the banks don’t compete—that is, if the banks act as if they were a single monopolist and charge prices higher than the competitive price.
And, in fact, the authors show that in markets with a high degree of common ownership of banks by institutional investors, customers get a worse deal. They receive lower interest rates for CDs, and they are charged higher fees for maintaining their deposit accounts. Terms are worst in the two markets with the most banks—California and New York—and that’s because in those markets common ownership of the banks by institutional investors is high. Not only that: Over more than a decade, across the country fees have risen in parallel with the rise of bank ownership by institutional investors.
These results mirror those of another paper written by Azar, Schmalz, and a different co-author named Isabel Tecu. As I discussed in Slate (with the economist Glen Weyl), that paper found that common ownership of airlines by institutional investors caused airline ticket prices to increase despite the existence of numerous airlines that serve the same routes.
In combination, the two papers show that mutual funds and other institutional investors—which have long been considered a benign force in the corporate world—may have a dark side. The benign theory was that institutional investors, by aggregating the stock investments of millions of people, can use their power over corporations to punish managers who make dumb decisions, overpay themselves, and waste corporate assets. Each individual investor does not possess enough votes or economic power to influence managers, or the time and attention to determine whether the managers act well or poorly. The growth of institutional investing over the past several decades was seen as a solution to what economists call agency costs—the inability of the principles (shareholders) to discipline their agents (the managers).
The Azar/Schmalz view is deeply subversive. It suggests instead that the growth of institutional investing has been driven, at least in party, by monopolization. This is actually not a new idea. In the 19th century, industrialists used trusts to control different companies in an industry. Oil companies would not compete against one another because they were all owned by a common entity—a trust—which directed them to overcharge customers. The appearance of competition was maintained while a shadowy trust pulled the strings. Our antitrust law has that name because monopolization originally took place through such trusts. (Europeans use the term “competition law.”) Mutual funds and institutional investors may be the modern-day version of the trust system.
How do the institutional investors cause the firms they own to refrain from competition? This remains a bit of a puzzle. It is unlikely that the boss of BlackRock calls up the CEOs of the banks and airlines and tells them what prices to charge. One possibility is that the bank CEOs don’t need to be told. They know that their compensation is influenced by the preferences of the institutional owners, who may be displeased if they see signs of excessive competition. Schmalz argues that several institutional investors voted down an effort by the hedge fund Trian to obtain seats on DuPont’s board last year, because Trian sought to make DuPont more competitive. Why would the institutional investors oppose a more competitive DuPont? Because DuPont’s main competitor is Monsanto, which the institutional investors also owned. Schmalz notes that while DuPont’s price fell when Trian’s venture was defeated, Monsanto’s shot up 3.5 percent.
If all this is true, breaking up the big banks wouldn’t reduce banking fees. As long as the institutional investors remain major shareholders of the small banks that emerge from the wreckage, the small banks will have every incentive to overcharge customers. Sanders also argues that the ATM fees that banks charge should be capped, but this will just cause banks to gouge customers in other ways—for example, by offering them lower interest rates for their deposits and CDs, or scrimping with lousy service. At the same time, we don’t want to bust up the institutional investors because they perform a genuine public service by making it easy for ordinary people to diversify their investments across companies.
Weyl and I argued in Slate that Congress could pass a law that permitted mutual funds to buy shares in firms in different industries but not more than one company in a single industry. This would allow mutual funds to obtain nearly all the benefits of diversification for their investors while eliminating their incentive to reduce competition. Bloomberg’s Matt Levine called our proposal a “wonderful mad” idea—he just meant “mad”—but, as Harvard law professor Einer Elhauge argues, roughly the same outcome could be achieved through existing antitrust law (albeit, I would add, in a maximally slow, expensive, and clumsy manner). Regulating institutional investors is not as crowd-pleasing as breaking up big banks, but it is more important. Sanders should take note.