Addressing TBTF When Banks Merge: A Proposal
Bank mergers provide unique opportunities for TBTF policy reform.
Published September 1, 2005 | September 2005 issue
The growth of big banks through mergers has been a long-standing concern to many observers, including the Federal Reserve Bank of Minneapolis.1 Because the current policy framework adequately responds to concerns about reduced competition and the obligation
of banks to serve their communities, our focus is not on these
time-honored considerations.2 Instead, the hazard from the creation and growth of mega banks that worries us is the so-called too big to fail problem.
Unfortunately, existing policy does not adequately address the TBTF issue. We have offered a variety of proposals to correct this deficiency. Recently, for example, we detailed why we think TBTF is a growing problem and described the options that elected officials should consider to address the root causes of TBTF (see Too Big to Fail: The Hazards of Bank Bailouts, Brookings Institution Press, 2004). In this essay, we continue that tradition and suggest a reform that links the government's review of large bank mergers to efforts to better manage TBTF.
But first a little background—discussed in greater depth in our book—on why it is worth devoting resources to fix this problem. When the creditors of a bank believe it is TBTF, they come to expect government bailouts if the bank gets in trouble. Creditors recognize that policymakers would rather bail out a large, systemically important bank than take the chance that its failure would "spill over" and threaten other banks, financial markets or the economy as a whole. With less at risk, the creditor of the large bank has less incentive to monitor the bank and "discipline" it when it becomes more risky. For example, the creditor of the TBTF bank might provide the bank with more funds than he or she would if the creditor thought he or she would bear all the loss if the bank failed. The end results are banks taking on too much risk, which wastes economies' resources.
All else equal, we think the TBTF problem gets worse when large banks merge. Certainly a proposed large bank merger and acquisition (M&A) must pass legal hurdles and U.S. government scrutiny. This merger review process would be a natural time for bank supervisors to raise TBTF concerns and explain how they are addressing them. But the current review process neglects the TBTF issue altogether. We propose a fix, albeit one in nascent form, to remedy this weakness.
Our reform would require select federal agencies to report on their progress in addressing TBTF when large banks merge. Because this reform is linked to, but not necessarily part of, the current review process, the reform would neither reduce current analyses of bank mergers nor reduce the flexibility of the government in approving those mergers. Over time and after achieving some success in addressing TBTF concerns, our proposed reform could be more fully integrated into the review process, and we conclude with one idea to accomplish that (requiring merging banks to explain in their regulatory filings how they have complied with the various rules, regulations and policies supervisors have identified as making a bailout less likely). In any case, by introducing this reform proposal in a formative stage, we hope to encourage additional discussion and suggestions for reforming the current large bank merger review process.
The current merger review process
The Federal Reserve and the Department of Justice typically review M&As involving large banks.3 The DOJ's review focuses on traditional analysis of market concentration and structure. Simply stated, the DOJ first defines the markets in which the banks under review compete and then compares the market share of deposits and/or loans that the proposed combined banking entity would have under pre- and post-merger scenarios.4 Generally speaking, greater market share for a smaller number of banks means less competition. If the absolute level of market share held by a few banks exceeds a certain threshold or has increased too much, the DOJ may conclude that the combined entity would result in or tend to create a monopoly or substantially less competition. At that point, the DOJ may seek divestitures, or even act to prevent the merger.
The Federal Reserve's review also includes competitive concerns along these lines.5 But significantly, it goes beyond that scope in two ways (due to requirements of the Bank Holding Company Act). First, the Federal Reserve considers the financial and managerial resources and future prospects for merging institutions. If the acquiring bank is in weak financial condition and is of supervisory concern, all else equal, the Federal Reserve would prefer, and even require in some cases, that it concentrate on improving its own state before trying to expand.
Second, the Federal Reserve considers the effects of the proposed M&A on the convenience and needs of the community to be served. The latter criterion means, to a large degree, the institution's compliance with and performance under the Community Reinvestment Act. Congress passed the CRA to give banks an incentive to extend credit to borrowers across the geographies in which they operate. So, for example, a bank making a relatively high level of mortgage loans to lower-income or minority households in its community may be viewed as meeting the needs of the community. The Federal Reserve Board can set a variety of conditions (divestiture, commitments to future performance and so on) or even reject a merger if the merger does not meet standards for competitiveness, financial condition or community needs.
Proposed reform to address TBTF
The inclusion of a community needs assessment during a merger review clearly shows a desire to achieve broader public policy goals through the review. Moreover, that inclusion recognizes the unique opportunity offered by a bank merger. It is a time when legislators' and supervisors' (and even the public's) attention is focused on the merging institutions and the social and economic consequences of the merger.
The glaring policy omission is the lack of attention to TBTF issues, and we offer a potential fix. We propose that for mergers between two of the nation's 50 largest banks, the Federal Reserve, the Federal Deposit Insurance Corporation (FDIC) and the U.S. Treasury should report publicly on their respective efforts to address and manage potential TBTF concerns.6 Put another way, a proposed merger between large banks would trigger a specific reporting requirement for supervisory agencies.
Understanding how this proposal would work in practice requires a brief description of how large bank bailouts occur under current law: After a bank fails, insured depositors are paid off and other creditors receive the proceeds from sale of the bank's assets. Under landmark legislation passed after the banking crisis of the 1980s, Congress mandated that this resolution process be carried out in a fashion "least costly" to the taxpayer. Congress gave an opening, however, to allow for creditor bailouts that exceed the least-cost amount. To do so, however, the Federal Reserve, the FDIC and the U.S. Treasury must first find that failure of the bank would pose systemic risk (that is, costly spillover to the broader economy). In other words, a bailout must be preceded by these agencies invoking the "systemic risk exception."
Our proposal leverages off this legal framework. We propose that as part of their review of large bank mergers, these agencies describe steps taken over the past year to facilitate the resolution of large bank failures without resorting to the systemic risk exception. In essence, the agencies would be required to describe what they have done recently to reduce the likelihood of having to bail out collapsed banks. At heart, then, what we are proposing is TBTF reporting by bank regulatory agencies, triggered by proposed mergers between large banks.
Of course, there are numerous ways to structure this reform. Reporting could be triggered by mergers between any of the 100 or 75 largest banks instead of the top 50. This reform could also be implemented through voluntary coordination of the three agencies. Alternatively, elected officials may wish to enshrine the reform as a legislative requirement, perhaps to ensure that it lasts beyond the tenure of the supervisors who volunteered to put it in place. In either case, the type of analysis carried out under the reform should meet the needs of lawmakers.
For example, Congress may desire updates on supervisory actions and process changes that the agencies believe make invocation of the exception less likely (for example, simulations of supervisory response to large bank failures that do not lead to bailouts). In a similar vein, supervisors could review the regulatory changes made to better manage TBTF (for example, promulgation of rules that would make it easier to impose losses on creditors). Modifications to the resolution process that make it more likely that creditors take their hits could also be highlighted.
Note that this particular reform would not require agencies to disclose any institution-specific information about the banks under review. Rather, the disclosed information would be more general, applying to all banks with TBTF potential. Moreover, the TBTF analysis would not influence how the Federal Reserve and other agencies act on the specific merger proposal before them. The purpose of our reform is to address TBTF concerns in general rather than the specific merger at hand. This does not mean that agency descriptions of supervisory steps taken should be generic; on the contrary, they should provide concrete examples of specific actions implemented over the past year to address TBTF concerns.
Benefits of proposed reform
There is precedent for adding a consideration not related to competition to the government's analysis of bank mergers. But what justifies this specific proposal which focuses on supervisory responses to TBTF? And what are the benefits of enacting this reform?
First, the links between a review of large bank mergers and analysis of TBTF concerns are inherent and would exist even absent implementation of our reform. As noted, fears of spillovers underlie the provision of bailouts. Mergers between large banks would typically solidify or increase those fears. For example, the new, bigger bank resulting from the merger may have more and larger connections between it and other banks and financial markets, heightening the possibility and severity of a potential spillover. The new, bigger bank may also have an extremely wide scope of very complex operations. If supervisors find it difficult to fully comprehend a bank's operations, they will be more likely to trigger the systemic risk exception.
Second, the reform would direct bank supervisory resources to
high-value uses. Bank supervisors face many competing requests for their staff, time and attention. Supervisors surely face pressure to address the most tangible concerns and put out the fires of the day. The costs of TBTF, in contrast, result from excessive risk-taking that may not be obvious until failure is imminent. But the public benefits of addressing TBTF are likely to far exceed the gains from fixing the problem du jour.
Third, the analysis would provide valuable information to lawmakers and creditors. A bank merger focuses bank creditors on their status in ways that virtually no other action would. It is precisely at this point that bank supervisors should be explicit in warning creditors of TBTF banks that they are at real and growing risk of loss. Creditors would get that message if supervisors would specify the actions they have taken to make bailouts a more remote possibility.
Achieving this type of outcome is the rationale for linking supervisory reporting to an M&A rather than to a fixed timetable (for example, by December 31 of each year). Regular TBTF reports from the agencies concerned would be better than none at all. But a once-a-year report may not seem as salient or relevant as one presented to creditors, the public and legislators during the throes of a large bank M&A.7 Moreover, if supervisory reviews of large bank M&As become associated with efforts to reduce TBTF expectations, bank managers would have little reason to pursue mergers or acquisitions whose primary benefit is to reach or solidify TBTF status. Instead, mergers would reflect combinations that make "business sense."
At the same time, this reform avoids several potential pitfalls. First, it would not eliminate any of the analysis already undertaken as part of an M&A review. Reducing the scope of current review efforts would do nothing to address TBTF concerns, and we believe the existing framework responds adequately to concerns about competition and community service. Second, it would neither allow nor encourage rejection of proposed mergers based on TBTF concerns in the current period. In today's climate, preventing additional large bank mergers will not address the underlying reasons why creditors expect bailouts. Attention should focus instead on putting creditors at credible risk of loss, regardless of the merger's outcome.
The reform also avoids a third pitfall in that it doesn't require authorities to address TBTF concerns if they don't believe the evidence warrants such action. While there is a broad consensus that spillovers have the potential to be a serious problem, we recognize that not all bank regulators—indeed, not all Federal Reserve officials—have come to the same conclusions about TBTF that we have. Some believe that existing legislation and various internal efforts have made bailouts of large bank creditors a very remote possibility, and they are unlikely to see additional actions as necessary. Under the reform we propose, such views are easily expressed during the reporting process. If readers find their arguments persuasive, no pressure to take additional steps to address TBTF would arise.
Conclusion and next steps
We think there are several compelling reasons for linking a review of large bank mergers to a review of steps taken by authorities to reduce the likelihood of invoking the bailout exception. This link should reduce bailout expectations and the wasteful misallocation of economic resources that such expectations encourage. These benefits can be achieved while avoiding the costs that could arise from a less flexible reform that more explicitly ties approval of mergers to a TBTF review.
Because the merger process typically requires the applicant to respond to various issues, one could imagine expanding the reform already described to one more firmly embedded in the application process itself. For example, and in addition to the requirement for supervisory reporting, a large bank applying to merge with or acquire another large bank would describe compliance with the rules, regulations and policies that the agencies identify as reducing the chance that the systemic risk exception is invoked.8 Such a requirement would parallel in theory current assessments banks make of their CRA compliance, for example, in today's merger review process. This expansion makes sense, by definition, only after the agencies have made substantial progress in addressing TBTF concerns and if the agencies identify on an ongoing basis what banks should be doing to make orderly resolution possible.
This essay reflects a proposal in a formative stage. While we contend that bank M&A reviews present an opportune time to scrutinize TBTF efforts undertaken by bank supervisors, we welcome the reactions of those who feel otherwise as well as the suggestions others may have for refining our proposal. But whether TBTF concerns are addressed at the M&A review stage or at another time altogether, we strongly believe that much more must be done to diminish expectations that taxpayers will bail out failing banks. The cost of fixing TBTF problems after banks collapse is far greater than the price of addressing them before banks take on imprudent risk.
Helpful comments were provided by Dan Hanger and Niel Willardson.
Errors are our own.
1 We greatly simplify by using "bank" to generically refer to depository institutions, holding companies and so on, unless otherwise noted. For a thorough review of banking consolidation and its implications for policy, see Group of Ten, "Report on Consolidation in the Financial Sector," January 2001.
2 For example, even as national measures of banking concentration have increased, "measures of local market banking concentration [remain] virtually unchanged." See Alan Greenspan, "Banking," October 5, 2004.
3 More precisely, the Federal Reserve reviews M&As between bank holding companies or an M&A when a state chartered bank that is a member of the Federal Reserve System is the surviving entity. The DOJ has the authority to review all bank and holding company M&As. The Office of the Comptroller of the Currency (OCC), a bureau of the U.S. Department of the Treasury, reviews bank M&As when a national bank is the surviving entity. The Federal Deposit Insurance Corporation (FDIC) reviews bank M&As when a state nonmember bank is the surviving entity.
4In general terms, Congress also forbade authorities from approving an M&A if it would result in the acquiring bank holding more than 10 percent of U.S. bank deposits. This cap, which applies to M&As across state lines, was put in place by the Riegle-Neal Banking Act of 1994 (RN). Note that a bank can exceed the national cap if its deposit growth comes from a non-M&A source (that is, so-called organic growth). RN placed a similar cap on the state level of 30 percent of deposits but granted states the ability to pass more restrictive or less restrictive alternatives.
5See Laurence H. Meyer, "Mergers and Acquisitions in Banking and Other Financial Services," testimony before the Committee on the Judiciary, U.S. House of Representatives, June 3, 1998, for a discussion of the Federal Reserve's M&A review process.
6The FDIC insures customers' deposits at U.S. banks. The U.S. Treasury oversees most of the nation's largest banks through the OCC.
7The once-a-year reporting requirement could, however, back up our reform. For example, our reform could include a requirement that in a year when a triggering large bank merger does not take place, the required agencies would issue the mandated reports on steps taken to address TBTF before the end of the year.
8 It would be more difficult, however, to accept or reject a large bank M&A based on its direct effect on TBTF. Consider just one example. Relative to the factors currently considered in large bank mergers (for example, CRA performance or effect on market structure) there is not a consensus method for measuring the size and cost of TBTF. For a methodology that the government might use if it more formally considered TBTF-related concerns when evaluating a merger, see Edward Kane, "Incentives for Banking Megamergers: What Motives Might Regulators Infer from Event-Study Evidence?" Journal of Money, Credit and Banking 32 (August 2000): 671-701.