Regulators should rethink the way they assess bank mergers

With President Biden’s appointees to bank regulatory agencies moving into place, there soon will be a long overdue reconsideration of policies toward bank mergers. Banking was one of many sectors specifically identified in Biden’s executive order of last summer, which called for a government-wide effort to promote competition in the American economy.^[1] However, realizing public benefits from a retooled approach to bank mergers requires more than just an instinct to get tougher (though more rigor is certainly warranted). What’s needed first is an analysis by the involved agencies of the dynamics of the financial services industry that will enable them to assess specific bank merger applications with more of an eye to where the industry is heading, and not just where it has been.

The bank merger approval process is complex, both procedurally and substantively.  Most bank mergers must be approved under both the Bank Merger Act^[2] and the Bank Holding Company Act.^[3] The federal banking agency (the Comptroller of the Currency, Federal Deposit Insurance Corp., or Federal Reserve) that is the primary regulator of the acquiring or surviving insured depository institution approves the merger under the Bank Merger Act. The Federal Reserve, as regulator of all bank holding companies, must approve the merger under the Bank Holding Company Act. To complicate the process further, the Antitrust Division of the Department of Justice retains its usual antitrust jurisdiction and thus can sue to prevent the merger even if the banking agencies have approved it.

Substantively, the standards for approval in the two banking laws are very similar. Each requires the banking agencies to disapprove the merger if it would have anti-competitive effects.  But each also includes other factors for the banking regulators to consider, and they can provide an independent basis for disapproving the merger (or, less typically, for approving the merger despite potential anti-competitive consequences). The competition standard is a condensed version of antitrust statutes, prohibiting mergers that would create a monopoly, substantially lessen competition, or otherwise be in restraint of trade. The other factors for consideration include the financial and managerial resources of the banks, the convenience and needs of the community, the effectiveness of the banks in combatting money laundering, and risk to the stability of the U.S.  banking or financial system.

There are many comments one could make on current bank merger policies. I will highlight two: First, the competition analysis of the banking agencies has not kept up with structural and technological changes in the industry. The current analysis can lead both to ill-advised mergers being approved and, quite possibly, to mergers that should be allowed being discouraged by the banking agencies.^[4] Second, the financial stability factor added by the 2010 Dodd-Frank Act is being applied in a haphazard fashion. It needs to be built out into a rigorous framework that provides guidance both to agency staff and banks in assessing the financial stability effects of potential mergers. I illustrate these points by reference to two recent Federal Reserve orders approving transactions involving very large banks – the 2019 merger of BB&T and SunTrust to form the $500 billion Truist,^[5] and the 2020 acquisition of E*Trade by Morgan Stanley that pushed it over the $1 trillion asset level.^[6]

Competition Analysis

One peculiarity of merger analysis under the banking laws is that it does not include all actual, much less potential, competitors. The Federal Reserve, for example, uses insured deposits in relatively small geographic areas as the basis for calculating market concentration, which is in turn used to evaluate the impact on competition of a proposed merger. This made sense decades ago because of three salient features of the financial system. First, a great majority of Americans depended almost exclusively on insured depository institutions for their transaction accounts and their credit needs – mortgages, car loans, and other forms of credit, as well as for their transaction accounts and their principal form of savings. Second, banking was very much a local market, in large part because of longstanding legal restrictions on branching, but also because households and smaller businesses had to bank in person. Third, the credit products offered by banks in a given geographic area were generally very similar to those offered by competitors.

While one can still see the vestiges of this market structure today, much has changed.  Nowhere is change more apparent than in residential mortgage lending. On the eve of the Global Financial Crisis in 2007, banks and their affiliates accounted for roughly three quarters of residential mortgage originations. Today independent mortgage companies account for nearly 70% of mortgage originations. These companies have led the way in streamlining the mortgage application process by moving much of it online. They have grown to take advantage of scale economies, including those associated with the digital platforms they have created.

The cumulative effect of these changes is that the residential mortgage market is moving toward being truly national. There are, to be sure, features of this market that distinguish it from other lending, beginning with the existence of government-sponsored enterprises that provide a steady stream of funding to the independent mortgage originators by purchasing their conforming mortgages. Hence deposits as a source of funding and a balance sheet stabilizer matter much less here than in other kinds of lending. But the advantages of scale, and specifically of the scale associated with sophisticated financial technologies, are operative in most other areas. Nonbank fintech companies, mostly doing business online in much of the nation, are already significant players in unsecured personal and small business loans. And, on the other side of the balance sheet, banks are engaged in a delicate balance of co-ventures and competition with fintech and Big Tech companies offering the digital payments services that are increasingly popular with consumers.

Thus, the related factors of scale economies, technological innovation, product differentiation, and the continued incursion of non-banks are driving lending towards larger markets, though at very different paces (and, perhaps, end points) for different forms of credit.   The obvious implication for merger analysis appears to be that the assessment of competitive effects needs to be broadened to take account of non-bank lenders, as well as online competition from banks lacking a physical (i.e., deposit-taking) presence in local markets. This has been the centerpiece of the argument offered by bank trade associations, with the accompanying message that merger analysis need not be tightened, because there is more competition than suggested by existing market measures.

There is undeniably a good deal to this point. But the implications for competition analysis are mixed. For one thing, the amount of non-bank and non-locally-based competition varies across forms of lending. Thus, while a merger between two banks might not result in competitive harm for some bank products, it could do so for others. There is also the possibility that a significant part of a customer base does not have effective access to online banking services. If so, there is an argument that these customers comprise a sub-market that could be adversely affected by a proposed merger.

More generally, the factors driving the evolution of the banking industry may also have mixed implications for overall regulatory merger policy. For example, if some banking markets are evolving from local to regional, then a merger that raises only modest concerns when looking at a series of local markets might nonetheless be part of a trend creating a regional market in which actual and potential competition will be reduced. If concentration levels in local markets remain roughly the same, but four national banks have displaced community and smaller regional banks as the dominant players in all those markets, is the competitive environment really unchanged?^[7] A version of this point was raised in public comments submitted to the Federal Reserve in connection with the 2019 BB&T merger with SunTrust. The Fed noted the comment in a footnote but did not respond to it, even though the two banks competed in 81 local markets in the Southeast.^[8] The Fed’s competition analysis was focused exclusively on the dozen local markets in which the combination of deposits in the two banks’ existing branches would create a presumption of anticompetitive effect under the merger guidelines.

Whether the BB&T/SunTrust merger should have passed muster under the competition standard of the Bank Merger and Bank Holding Company Acts is uncertain. What is clear is that the analysis provided by the Federal Reserve did not address key questions in making that determination – from whether there was effective online competition to whether a region was a relevant geographic market for at least some banking products and services. Moreover, the factor most often cited by the merging banks themselves – increased scale to increase investment in technology – was mentioned only in passing by the Fed.

An analysis of the competitive effects of a merger that doesn’t take account of the stated motivation for the merger is almost surely missing much of what is going on in the relevant markets. The Fed, and the other banking regulators, should develop and explain to the public their view on the key issues that will determine the competitive structure of the industry, beginning with the changes and trends I have noted. There will be contestable points in this analysis, to be sure, as well as some difficult judgment calls. There will, for example, almost certainly be mergers among regional banks that would diminish competition in at least some products and geographic areas, but that may provide the scale necessary to enable the resulting bank to compete with very large banks and with less regulated fintech firms. And even though many mergers among small banks may be justified as providing the necessary scale for effective digital or other forms of competition, this will not always be the only important consideration. In such cases, a high-level view of industry dynamics will need to be combined with a case-specific inquiry into such issues as how persistent the loss of competition is likely to be and precisely what economies would be realized by the merger. The agencies’ analysis may well be controversial, perhaps challenged simultaneously from both sides. But at least it will move beyond the anachronistic approach revealed in the approval of the merger that produced Truist, and begin grappling with what is actually occurring in the financial services industry.

Non-Competition Factors

The preceding discussion focused on some current and evolving characteristics of financial services markets that should affect the approach taken by the Justice Department and the banking agencies in evaluating the competitive effects of proposed mergers. The banking laws require the agencies (but not the Justice Department) to consider other factors in acting on merger proposals. They include “the financial and managerial resources and future prospects” of the banking organizations involved in the merger, as well as the impact on financial stability, the “convenience and needs of the community to be served,” and the effectiveness of the merging institutions in combatting money laundering.

Some of these factors are intended to shine an especially bright light on certain ongoing statutory obligations of banks. For example, when an acquiring bank has significant shortcomings in its compliance with anti-money laundering or consumer protection laws, there is a good argument that its “managerial resources” should not be diverted to the integration of its target bank until it has remediated those shortcomings. Application of these factors requires judgment in their exercise (i.e., how significant must past compliance issues be for regulators to deny or delay approval of a merger), but not much in the way of an analytic overhaul.

The factor of increased “risk to the financial stability of the United States banking or financial system,” on the other hand, remains analytically underdeveloped more than a decade after it was added by the Dodd-Frank Act. The Federal Reserve Board order approving Morgan Stanley’s acquisition of E*Trade contained only a perfunctory analysis. It simply listed a number of factors that could affect financial stability (e.g., the size of the resulting banking organization) and noted the change in those factors that would be produced by the acquisition. Then, with very little reasoning, the order jumps to the conclusion that the acquisition would not result in “meaningfully greater” financial stability risks. The order’s only mention of a quantitative metric for assessing the merger, such as the Herfindahl-Hirschman Index used in antitrust analysis, was the “G-SIB score,” which is used by the Fed to calculate capital surcharges for the most globally systemically important banks (hence “G-SIB”). The order noted that the acquisition, through which Morgan Stanley’s balance sheet increased by over 7%, resulted in a G-SIB score increase that was “only” about 2%.

There was no explanatory framework detailing the dynamics through which the Fed anticipates that financial stability risks could be increased, no explanation of why a 2% increase in the G-SIB score was acceptable, and certainly no basis for anyone reading its order to understand what kind of transaction would raise material financial stability concerns.

The absence of a reasonably well-developed and well-explained approach to financial stability analysis in bank mergers was not of great concern so long as there were no mergers among very large regional banks or significant acquisitions by G-SIBs. But the last few years have seen several deals that did call for rigorous financial stability assessments. Because it has jurisdiction over the holding companies of the banking organizations, the Fed needs to elaborate its framework that both provides guidance to banks considering mergers and allows the public to understand the Fed’s policy choices. Ideally, this framework should include a metric that functions as a screening mechanism, to create at least mild presumptions that a specific merger does or does not raise financial stability concerns. If the Fed chooses to use its existing G-SIB methodology, it should explain how, precisely, that methodology is relevant for the merger approval process. It should specify what kind of increase in that score raises presumptive concerns (with, one would assume, a sliding scale that was more sensitive to mergers involving banks that are already G-SIBs). And, as with competition analysis, the development of an appropriate metric and analytic framework should incorporate an assessment of how the financial services industry is changing.

Conclusion

The U.S. banking industry presents an odd picture in 2022. Returns on equity have lagged those of many other sectors since the Global Financial Crisis, itself caused in part by the excesses of some major banks. Yet, at least in the aggregate, the cost of financial services to average households has not declined noticeably in decades. On the basis of those two facts alone, one might surmise that there is a need both for more competition to drive innovation and for some consolidation to provide banks the scale necessary to fund and profit from that innovation.  Paradoxically, the policy prescription seems to be both more and less competition. With a full and nuanced analysis that unpacks this paradox, the banking agencies and the Antitrust Division of the Justice Department can better evaluate the dynamic competitive implications of bank mergers – large, medium, and small. Similarly, a well-developed assessment of the changes taking place in the industry can provide a rigorous starting point for considering the important statutory factors of financial stability risks and “convenience and needs of the community to be served.”

Finally, the banking agencies – along with their colleagues at the Consumer Financial Protection Bureau, Securities and Exchange Commission, and other financial regulators – need to think more proactively about non-bank competition to prudentially regulated depository institutions. To the degree that non-banks enjoy a competitive advantage because they are not restrained from creating financial stability risks through excessive reliance on leveraged short-term funding, or because they are not effectively held to the same standards of consumer protection or anti-money laundering compliance as banks, then the competitive landscape for banks will be an uneven one.

^{[1] Executive Order 14036, Promoting Competition in the American Economy, July 9, 2021.}

^{[2] 12 U.S. Code §1828(c).}

^{[3] 12 U.S. Code §1842.}

^{[4] Generally speaking, banks will not proceed with a merger if they believe it will raise the presumption of anti-competitive effects in multiple local markets because of increases in market concentration, or because they have received signals from the banking agencies that disapproval is a significant possibility.}

^{[5] https://www.federalreserve.gov/newsevents/pressreleases/files/orders20191119a1.pdf.}

^{[6] https://www.federalreserve.gov/newsevents/pressreleases/files/orders20200930b1.pdf.}

^{[7] Though beyond the scope of this brief comment, there is a broader debate among industrial organization economists in which the coherence of market concentration analysis, much less its correlation with anti-competitive effects, has been questioned. For an analysis of these issues see Kaplow, Replacing the Structural Presumption, (forthcoming Antitrust Law Journal 2022).}

^{[8] The footnote is on page 9 of the Order, https://www.federalreserve.gov/newsevents/pressreleases/files/orders20191119a1.pdf: “Commenters expressed concern that the proposal would increase concentration in banking markets throughout the Southeast and Mid-Atlantic, including in Florida, Georgia, Maryland, North Carolina, and Virginia. Similarly, a commenter argued that the proposal would result in consolidated market power among only a few large banks. Commenters suggested that the reduced competition as a result of the merger would result in less favorable outcomes for consumers.”}

The Brookings Institution is financed through the support of a diverse array of foundations, corporations, governments, individuals, as well as an endowment. A list of donors can be found in our annual reports published online here. The findings, interpretations, and conclusions in this report are solely those of its author(s) and are not influenced by any donation.