October 04, 2023

The Role of Research, Data, and Analysis in Banking Reforms

Governor Michelle W. Bowman

At the 2023 Community Banking Research Conference sponsored by the Federal Reserve System, the Conference of State Bank Supervisors, and the Federal Deposit Insurance Corporation, St. Louis, Missouri

Watch Live

It's a pleasure to be back in St. Louis for the annual Community Banking Research Conference. I am grateful for the continued partnership between the Federal Reserve, the Conference of State Bank Supervisors (CSBS), and the Federal Deposit Insurance Corporation (FDIC), and I would like to thank the conference organizing committee for putting together this important forum, now in its 11th year. This conference provides an excellent opportunity for policymakers to pause and reflect on the impact of current U.S. bank regulatory policies on the financial system. At the same time, we should also consider how alterations to these policies will influence the future of the broader banking system.

We get a view into these questions—the effectiveness of past policy choices and insights that can help shape future policy reforms—from several aspects of this conference, including, of course, the research that will be presented and from the results of the CSBS Annual Community Bank Survey. We also gain insight from the experience and perspectives of our participants: your questions and commentary throughout the next few days can provide important context about the intended and unintended consequences of past policy decisions. And the combined perspectives from the wide range of experts gathered here—the researchers, regulators, policymakers, and bankers—will provide us with valuable insights into how to better establish and calibrate bank regulatory and supervisory policy in the future.

Research, data, and analysis are essential to thoughtful bank regulatory reform. These tools can be used to identify issues that must be addressed or remediated; they can help us evaluate which elements of the current bank regulatory framework may be effective or ineffective, and they can help us craft reforms with a clearer understanding of the intended and unintended consequences. Together, these tools can help us develop and implement appropriate regulatory reforms that do not impair the long-term health or impede the future growth of our diverse U.S. banking system, enabling us to avoid the pitfalls of reforms that fail to accomplish these important goals.

Today I will begin by discussing some of the research around community banks and the role they play in the U.S. banking system. I will then discuss how this research can inform policy, including by helping us be more sensitive to the unintended consequences of reform, and discuss how this perspective can be applied to the evaluation of bank mergers and acquisitions. Finally, I will mention a few additional areas where I think research could be valuable as an input to the rulemaking process.

The Role of Research, Data, and Analysis in Rulemaking
Good research leads to good policy. This statement will likely sound familiar to many of you, as it was how Jim Bullard, the former president of the Federal Reserve Bank of St. Louis, would describe one of the key purposes of this conference in past years. Over the years, many of the conference speakers and presenters have echoed that same sentiment since this conference first launched in 2013.

In my remarks today, I will discuss the importance of using an evidence-based approach to bank regulatory policy, which has been a long-standing theme and goal of this conference. While research can provide valuable insights to inform revisions to bank regulatory policy, research is particularly important when it comes to regulatory efforts that directly affect community banks, including changes to modernize the Community Reinvestment Act regulations, or reforms to the supervisory standards and expectations for community banks. Research also plays an important role in helping us understand policy reforms that affect community banks indirectly, such as the potential that regulatory requirements designed to apply only to larger banks could eventually be pushed down to smaller and community banks.

Research and evidence-based rulemaking can insulate the banking system from wide swings in policy over time. The banking system faced significant challenges this year, triggered by the failures of three large banks in the early spring. As I noted at the time, the policy choices made in responding to these failures will have important, long-term consequences for the U.S. banking system.1 The potentially far-reaching consequences of policy choices we make during this time highlight the importance of clearly identifying the problems we are attempting to address using data to craft solutions that are effective, targeted, and efficient. Bank failures demand scrutiny, but bank failures alone do not justify wholesale revisions to the bank regulatory framework. Before we undertake reforms intended to address issues that led to bank failures, we need to develop a comprehensive understanding not only of those root causes, but also of the costs and unintended consequences of potential reforms. Research can protect against over-reactive regulation, especially that which is not efficient, calibrated and tailored to address the actual risks and challenges facing the banking system.

My insistence upon being guided by evidence does not imply that I am opposed to regulatory reforms, but rather that policymakers should be expected to show their work. The banking system is not perfect, and policymakers should continually ask themselves if there are ways to improve regulation and supervision. I am always open to considering evidence-based proposals that address known deficiencies and shortcomings in our regulatory and supervisory framework, but any such proposals must stop short of interfering with or stepping into the role of managing the financial institution.

I am very pleased that this research and policy conference has continued to thrive and grow over the 11 years since its inception. The research work you all do, especially your efforts to develop data and analysis to inform policy, is a critical input to the rulemaking process.

While research is a foundation of prudent regulatory policy, the community banks that operate under this framework are the reason we are gathered together at the Saint Louis Federal Reserve Bank. So, let's build upon that foundation to highlight the role of community banks in the U.S. banking system.

The role of community banks in the U.S. banking system
As is necessary and appropriate when speaking at a research conference, let's start with the research. Research can help reinforce a statement that policymakers often make, which is that banks of all sizes—including community banks—are a critical part of a healthy and vibrant banking system. Today, there are around 4,200 community banks in the United States.2 What role do these community banks play in the current U.S. banking system and in the communities they serve?

While my discussion could not possibly do justice to the vast body of research addressing the important role of community banks in providing credit and supporting their communities, I would like to highlight a few research findings that will likely resonate with community bankers and researchers.

A classic 2002 paper from economist Allen Berger and his co-authors looked at how banks process information to make credit allocation decisions, with the goal of evaluating the differences between small and large banks.3 The paper provided empirical evidence that small banks have a comparative advantage over large banks in making small business loans due to their specialized ability to effectively produce "soft information" on smaller firms and make effective credit decisions based on this information. In contrast, the paper finds that larger banks are at a comparative advantage when lending to larger firms that can produce more "hard information" that can be easily verified and documented. This research validates and confirms that community banks leverage the power of "relationship banking," and in doing so, are an important source of credit to a critical segment of the economy. This finding also helps explain why community banks were able to more effectively and on a large scale implement the Paycheck Protection Program during the pandemic in comparison with large banks.

As another, more recent example, a paper published last year by economist Elizabeth Berger provides additional context to the role of community banks and the policy issues currently being debated. This paper focuses on the role community banks play in the creation of service sector jobs. We've seen the important role of service-oriented jobs for job growth the U.S., as our economy has moved away from manufacturing- and production-based employment over the course of many decades. The paper looks at both the extent to which small businesses' physical access to a local, relationship lender supports growth in service sector jobs and explores whether this growth is the result of access to local banking institutions. She found that service sector job creation is not only supported by relationship lending from local community banks but also depends on smaller, relationship lenders for growth. Her key finding is unambiguous: "finance [providers] must be local to facilitate service job creation."4

These research findings will certainly not surprise community bankers, state bank regulators, or those who directly see the positive impact of community banks on their local communities. But equally important from a policymaker's perspective is what these papers tell us about the impact of a community bank exiting its local community and the surrounding market. What happens when a small business has no local, community banker that takes the time to understand its business, whose business model is based on developing long-term banking relationships? Would small businesses be better served if their banking relationships are with large, non-relationship-based banking institutions? Or if the only alternative source of credit for small businesses were from nonbank lenders that are not subject to the same regulatory oversight as banks?

Understanding these issues can provide important context, not only to understand the current banking landscape, but also to evaluate the merit of reform proposals. Reforms that result in impairing the future of the community banking industry or the elimination of community bank charters to mergers or acquisition because of over-regulation or excessive compliance burden can have a significant, adverse impact on small business formation in local communities given the hurdles new small businesses face when seeking credit. And, while it might be tempting to assume that alternative sources of credit would arise to fill a gap left by community banks, this is not a given. For example, while credit unions compete with banks in many banking markets, credit unions are not, by design, by structure, and by regulation, small business lenders.

Some of the research that will be presented later today and tomorrow explores different aspects of community banking and the role community banks play in the current banking market. I look forward to learning more about this research and how it can serve to inform current and future bank regulatory reform efforts.

Implications of the evidence for regulatory policy
The research I mentioned earlier has important implications for policymakers, as it suggests that factors that contribute to a reduction in the number and lending footprint of community banks could lead to a number of negative impacts, including credit constraints for small businesses and small business start-ups. This issue could be especially acute when a market—an underserved market—is left without a community bank as a source of credit. The research highlights that a banking desert does not only affect individuals, but also can have a devastating impact on small businesses, local job formation, and local economies.

One important influence on banking market competition is bank mergers and acquisitions (M&A), including the important role of regulators in reviewing and approving these transactions. While one might naturally assume that approving mergers would always lead to a reduction of community bank lending, this result may not always be the case, as in markets viewed as less attractive or undesirable to large or regional banks, where the only option for community banks to survive may be to consolidate with other community banks in the same geographic area. Surely, some level of consolidation is preferable to the closure of community banks and the perpetuation of banking deserts or zombie banks.

At this conference last year, I described what I called the "New Landscape for Banking Competition" and stressed the urgency of updating and modernizing the framework the federal banking regulators and the Department of Justice use to evaluate the competitive effects of proposed bank mergers.5 I offered several commonsense proposals that should guide our thinking about how we evaluate bank mergers in the future and underscored how calibrating our guidelines correctly is critically important for community banks, given the evidence we have on their role and impact in providing credit in local communities.

Among those ideas was a more systematic inclusion of nonbank competitors and credit unions in our competitive landscape. I also proposed factoring in digital banks, and those more traditional banks who have established a strong national digital presence, in our competition analysis. These institutions have proven to be very effective at attracting deposits from anywhere in the country and therefore clearly compete with banks in urban and rural markets across the U.S.

Why is this important?

  • First, in our current analytical framework, more than 65 percent of banking markets, as currently defined, would be considered noncompetitive or "concentrated" in the language of antitrust, yet the intense competition for banking products and services has only grown, principally from nonbanks. We need to reconcile that.
  • Second, in our current framework, only 5 percent of the currently defined banking markets are considered highly competitive or "unconcentrated." In fact, today there are 19 states in the U.S. that do not have a single market that would be classified as highly competitive, and only 2 states have one-quarter or more of their markets designated as highly competitive.6

In our current system, many of the competitors and new entrants that we do have (nonbanks, fintech firms, credit unions), are systematically ignored. The baseline assumption in our current framework is that credit unions do not compete with banks, and yet, just last month we saw five announced acquisitions of community banks by credit unions.7 Despite this trend, after a community bank has been acquired by a credit union, the resulting credit union is no longer viewed as a baseline competitor with other community banks in the market. I think it is fair to question whether this view is consistent with reality.

We have also seen significant changes in how consumers access banking products and services and observed the migration of many types of loans from the banking sector to nonbanks, including 1–4 family residential mortgage loans. In fact, the top three mortgage lenders in the country are nonbanks. Even the largest U.S. banks are behind these nonbanks when it comes to mortgage originations.8 But we do not consider these entities as competitors when evaluating a merger involving a community bank.9

Under our current framework, it appears as if regulators have closed the door to a broad range of potential mergers that are perceived to raise competitive concerns in rural markets, by using an artificially narrow lens to evaluate competition concerns. Many merger proposals are regularly forced to go through an extensive, lengthy, and ultimately costly review process. As we've seen in recent months, this can result in extending the processing of the underlying merger, sometimes by as much as a year. While there is a lag in the reporting of applications data, data from 2022 show that the average processing time for applications has increased significantly from 2019 through 2022 for banks of all sizes.10 Information published by the Federal Reserve suggests that this increase was primarily the result of the Board's concurrent consideration of several relatively large and complex M&A proposals requiring Board action. Data through the first half of 2023 showed some improvement in average processing times, but the volume of approved M&A applications were down substantially from a year ago, while the number of M&A applications withdrawn in the first half of 2023 has already exceeded the total number withdrawn in all of calendar year 2022.11

Reported information highlights that community banking organizations also experienced increased processing time for M&A proposals from 2021 to 2022, with the average time increasing from 63 days to 72 days for banks under $1 billion in assets, and from 51 days to 96 days for banks from $1 billion to $10 billion in assets.12 While this average processing timeline improved slightly for community bank M&A applications in the first half of 2023, the most recent semiannual report on banking applications acknowledges that there were fewer M&A applications by community banks in the first half of 2023 as compared with the first half of 2022.13 Applications of any size can raise novel issues or concerns, but this should raise concerns with policymakers.

While the cause of such delays may vary—for example, receiving an adverse comment on a proposal can also result in a longer application processing period—lengthy review periods can have significant adverse consequences to firms and the banking system. Long delays can lead to a number of negative outcomes, including loss of key staff, pauses on new investment and development of new consumer and business products, expensive delays in systems integration, reputational risk, and shareholder losses. We've also seen that when a merger is either approved or denied after a lengthy process, the institutions involved are generally weaker than they would have been had a determination been made in a more reasonable time frame. This is not the recipe for a responsible regulatory system or a healthy banking system.

Of course, there are alternatives to this view of competition in rural markets. As noted in the research I cited earlier, when looking at the competitive effects of a potential bank merger, it might be more appropriate to look at the extent to which small businesses in a market would have choices that include a community bank when seeking credit, rather than looking only at the overall share of deposits an institution would have in a given banking market.

The research sends a strong signal that preserving local relationship-based community banks is vital to local economies. This suggests that a regulatory review framework should not be designed to obstruct mergers that result in stronger, more diversified community banks, with relationship business models. Clearly, these policies need to change, unless the goal is to facilitate further consolidation of banking products and services into the hands of a few national players.

This research has implications beyond just bank merger policy. For example, while bank regulatory reform proposals may consider the incremental additional burden on regulated institutions, this approach may be too narrow a lens, particularly for community banks. The cumulative burden of regulations coupled with extensive supervisory guidance can be a significant headwind for community banks, and local communities and customers served by community banks often bear the consequences.

Future Areas of Research
After nearly five years serving on the Board of Governors in the seat designated for someone with community banking or state bank supervisory experience, I am inspired by the potential of research to shape policy to reflect the practical reality. Many of you in this room have contributed important research that has energized policy discussions and debates. I am confident that this year's lineup of research papers will spark similar conversations.

Year after year, researchers have presented evidence that a banking system with diverse business models and sizes is better for business formation and leads to better penetration of financial services products across all industry sectors. You've demonstrated direct benefits to small businesses (including agriculture-based businesses) and to nontraditional mortgage borrowers when smaller, local lenders operate in a market.14 You've shown how community banks provide a window into local economies and local industries and provide an overall view of the health of the U.S. economy.

As we look past this conference, I'd like to suggest a few areas that warrant further research and debate.

Reconsideration of the community bank asset threshold
First, our regulatory system is very complex, and one thing that adds to this complexity is the often-inconsistent definitions we use, including the definition of a "community bank." These definitions are important—they are used to clarify which requirements and rules apply to a certain type of institution. Policymakers should strive to avoid unnecessary complexity or confusion around these thresholds.

Most commonly, when we refer to a community bank, we are referring to an institution with less than $10 billion in total consolidated assets. This threshold was referenced in several sections of the Dodd-Frank Act as a way to differentiate requirements among banks and was explicitly referred to in the subsequent Economic Growth, Regulatory Relief, and Consumer Protection Act.15 Of course, the reality is that our regulatory system applies different rules to different institutions across a wide range of asset sizes. For example, the Community Reinvestment Act regulation includes asset thresholds well below any common understanding of what a community bank is. It simply makes sense that the definition would be rationalized across the bank regulatory framework.

Are these asset size thresholds properly calibrated, and are the impacts, costs, and benefits to institutions and to customers when banks cross these different thresholds rational? Are these thresholds creating the right incentives to promote prudent lending while appropriately balancing risk?

We also need a better framework to understand whether we should move beyond simple asset size thresholds when tailoring our regulatory rules to the risks that different banks pose based, perhaps not on size, but on business model. Further, is it possible to move beyond asset size thresholds and still apply our dual banking system's tailored supervisory and regulatory approach in a way that is clear and transparent to all stakeholders?

Again, research continues to show the benefits that relationship-based commercial lenders provide to communities. I believe it is of paramount importance to develop a clear and transparent framework for defining who these lenders are so we can tailor our regulations appropriately.

Insured and uninsured deposits in bank funding models
A second area that deserves additional research is how banks fund themselves, and whether the deposit infrastructure—including deposit insurance—could be improved, including to make the banking system more resilient during times of financial stress. As we saw earlier this year, the speed at which customers can move money using modern technology is unprecedented. Much like our competition framework, we need to ensure that the framework that we put in place today supports the banking system of tomorrow and not the banking system of yesterday.

With respect to deposit insurance reform, the FDIC laid an important foundation when it released its report on the Options for Deposit Insurance Reform in May 2023.16 The framework shared the FDIC's views of the pros and cons of either providing unlimited deposit insurance coverage or targeted coverage based on different account types. The report suggested that targeted coverage might be the best option for maintaining financial stability and protecting depositors. We need to ensure that our deposit insurance infrastructure can support banking sector stability in the face of the challenges posed by today's technology.

This report, however, should not just serve as a call for policymakers to react to a narrow scope of options developed in reaction to a bank run crisis but should be seen as a call to the research community to provide evidenced research and analysis on these and other options. I see an opportunity for extensive stakeholder engagement and public dialogue on the topic of deposit insurance and the evolution of deposit practices. In my conversations with community bankers this spring, I learned that a large number of concerned customers called their local community banker with questions about deposit insurance. Community bankers have direct experience and insights into the benefits and challenges of our current deposit insurance infrastructure. Their insights can complement research on this topic and can bring a valuable and realistic practitioner's perspective to this significant matter of public policy.

Conclusion
As I reflect on the past seven years of my participation in this conference, first as a state bank commissioner, and now as a member of the Board of Governors, I am struck by how the approach taken by this conference is, in some ways, a smaller scale model of what I see as an effective approach to policymaking:

  • bring researchers and practitioners together to discuss important policy issues that are grounded in research, and
  • leverage the insights and experiences from a wide range of stakeholders to improve our understanding of both the key issues and of the policy options.

I'd like to commend all who have gathered today for participating in person to discuss the research and to learn from the work and experiences of one another. This conference contributes a wealth of insights and evidence that lead to better policy choices and can help us understand where future challenges to the banking industry might emerge.

Our call to action coming from this conference, in my view, is to resist the urge to act in a reactionary way when addressing policy matters. Instead, policymakers should choose to exercise restraint and patience, consider the evidence, and strive for well-calibrated policies that fully incorporate the costs, benefits, and impacts of reform and complement, rather than complicate or contradict, the existing regulatory framework.


1. Michelle W. Bowman, The Evolving Nature of Banking, Bank Culture, and Bank Runs, speech at the 21st Annual Symposium of Building the Financial System of the 21st Century: An Agenda for Europe and the United States, European Central Bank, Frankfurt, Germany, May 12, 2023. Return to text

2. See Federal Deposit Insurance Corporation, Quarterly Banking Profile, Second Quarter 2023 (PDF), at 14. Return to text

3. Allen N. Berger, Nathan H. Miller, Mitchell A. Petersen, Raghuram G. Rajan, and Jeremy C. Stein, Does Function Follow Organizational Form? Evidence from the Lending Practices of Large and Small Banks (PDF), National Bureau of Economic Research, Working Paper 8752 (Cambridge, MA: NBER, February 2002). Return to text

4. E. A. Berger, "Finance in the New U.S. Economy: Local Finance and Service Job Growth in the Post-Industrial Economy," Journal of Financial and Quantitative Analysis, September 9, 2020, 57(5), 1987–2021, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3718458. Return to text

5. Michelle W. Bowman, The New Landscape for Banking Competition, speech at the 2022 Community Banking Research Conference, sponsored by the Federal Reserve, the Conference of State Bank Supervisors, and the Federal Deposit Insurance Corporation, St. Louis, Missouri, September 28, 2022. Return to text

6. Reports of Condition and Income (Call Reports) and the Competitive Analysis and Structure Source Instrument for Depository Institutions (CASSIDI). CASSIDI combines data from the FDIC's Summary of Deposits and banking market definition information into a comprehensive analysis tool for use by regulatory agencies and banking organizations for performing pro-forma competitive factors analyses on potential bank mergers. Return to text

7. Jim Dobbs and Ken McCarthy, "Two Credit Unions Are on a Bank Shopping Spree," American Banker, September 1, 2023, https://www.americanbanker.com/creditunions/news/two-credit-unions-are-on-a-bank-shopping-spree. Return to text

8. Consumer Financial Protection Bureau, Home Mortgage Disclosure Act data, 2022. Return to text

9. Greg Buchak and Adam Jorring, Do Mortgage Lenders Compete Locally? Implications for Credit Access (PDF), working paper, July 2021. Return to text

10. See Banking Applications Activity Semiannual Report, July 1–December 31, 2022 (PDF), (10, 1), table 2 (May 2023), (showing that for mergers and acquisitions proposals considered by the Board, the average processing time for applications increased from 60 days in 2019 to 87 days in 2022). Return to text

11. Banking Applications Activity Semiannual Report, January 1–June 30, 2023, (10, 2), table 2 (September 2023) (showing 46 M&A applications approved and 12 applications withdrawn in the first half of 2023, compared to 87 M&A applications approved and 4 applications withdrawn in the first half of 2022, with 11 withdrawn M&A applications for the entire calendar year of 2022). Return to text

12. Banking Applications Activity Semiannual Report, July 1–December 31, 2022 (PDF), (10, 1), table 4. Return to text

13. Banking Applications Activity Semiannual Report, January 1–June 30, 2023 (PDF), (10, 2), table 4. Return to text

14. Daniel Rabetti, Non-Information Asymmetry Benefits of Relationship Lending (PDF), working paper, Coller School of Management (Tel Aviv: Tel Aviv University, August 2022), Claire Brennecke, Stefan Jacewitz, and Jonathan Pogach, Shared Destinies? Small Banks and Small Business Consolidation (PDF), working paper, July 2020. Return to text

15. Dodd-Frank Act, P.L. 111-203, §§ 165(h)(2)(A), (i)(2)(A), 723(a)(3), 763, 1025(a), 1026(a); Economic Growth, Regulatory Relief, and Consumer Protection Act, P. L. 115-174, §§ 201 (defining a qualifying community bank as one with total consolidated assets of less than $10 billion), 203 (giving relief to community banks from the Volcker rule). Return to text

16. Federal Deposit Insurance Corporation, Options for Deposit Insurance Reform, (Washington: FDIC, May 1, 2023). Return to text

Last Update: October 04, 2023