Rethinking Resolution Readiness: Learning from Experience and Sharpening Focus

Introduction

Thank you to the Chamber of Commerce for inviting me to speak with you today. Over the past several months, the FDIC has devoted significant time to considering potential improvements to our approach for resolving failed banks. Today, I will provide an update on several of these workstreams.

IDI Rule

For the vast majority of failures managed by the FDIC, the failed bank is relatively small and simple, and the FDIC has a runway leading up to failure. In other words, the bank is a small institution with a small number of business lines, a relatively simple and homogenous balance sheet, and a limited number of platforms and systems with small data sets that can be reconciled relatively quickly. And prior to failure, there is typically a period of weeks or months to gather information from the bank and market the institution under a traditional weekend-sale model. In these cases, ex ante resolution planning is generally unnecessary, as the FDIC is able to obtain and process the information needed once it becomes clear that failure is approaching.

By contrast, advance planning is particularly valuable when neither of those characteristics is present: the institution is not small and simple, and the FDIC has a short or nonexistent runway. Additionally, larger institutions are more likely to fail with shorter runways, given greater public scrutiny and a higher likelihood of liquidity-induced failures.¹ In these cases, analyzing and understanding a bank's systems, infrastructure, and data in order to market and sell the institution is very difficult in a short period of time, given the volume and complexity of data sets and IT systems.

While advance planning in such scenarios is valuable, the FDIC's historical approach to resolution planning for large insured depository institutions (IDIs) needs to be fundamentally reexamined, reoriented, and rationalized. As I have described in the past, I am skeptical of the value of requiring institutions to prepare lengthy narrative plans discussing proposed resolution strategies and hypothetical failure scenarios.² Instead, our focus should be on maximizing an optimal resolution outcome while being prepared in the event that option proves unavailable.

In that vein, we have been working to propose a new approach that would contain two related parts. First, the IDI Rule would be significantly slimmed down to focus more narrowly on key operational information most pertinent to the FDIC's ability to execute an orderly resolution. Second, the FDIC would establish a new "resolution readiness adjustment" to our deposit insurance assessment framework for larger banks, which would allow any bank subject to the large bank scorecard to qualify for a downward adjustment to its quarterly assessment if the bank (1) demonstrates its ability to populate a virtual data room (VDR) in a short period of time and/or (2) provides the FDIC with temporary access to the bank's third-party service provider(s) and/or internal system(s) to enable the FDIC to build IT infrastructure to quickly access data in the event of failure. Fundamentally, an ability to quickly populate a VDR and/or providing the FDIC with advance systems access is likely to lower the cost of a bank's failure, and therefore should result in a lower deposit insurance assessment.

Assessments

In addition to proposing to add a resolution readiness adjustment to the assessments framework, we are also considering additional changes to the existing assessments regulations.

A perennial challenge that deposit insurance funds face is that bank failures are not evenly distributed over time and can be highly correlated. As a result, losses may be very modest for extended periods, and then explode during crises. This creates a challenging balancing act in determining the appropriate size of the Deposit Insurance Fund (DIF or Fund). On the one hand, we want to build up the DIF to minimize the likelihood of needing to procyclically raise assessments during a crisis, as occurred during both the 1980s crises³ and the 2008 financial crisis.⁴ On the other hand, we are mindful of the costs of assessments, which, among other costs, effectively take funds away from lending and investing in the real economy and divert them to financing the federal government.⁵

Following the 2008 financial crisis, which plunged the DIF into negative territory, the FDIC spent a number of years rebuilding the Fund, finally exceeding the statutory minimum reserve ratio⁶ of 1.35 percent in the third quarter of 2018.⁷ Since then, the DIF fell back below the statutory minimum in 2020 due to the unprecedented COVID-era stimulus, which resulted in a dramatic increase in the denominator of the reserve ratio. Subsequently, the FDIC raised assessments by two basis points across the entire industry in 2022,⁸ followed shortly thereafter by substantial losses for the DIF due to the 2023 bank failures.⁹ The reserve ratio returned above 1.35 percent last year and currently sits at 1.43 percent, which is the highest the reserve ratio has been since the 1960s, though well below the reserve ratio that would have been needed to have avoided turning negative during the 1980s or 2008 crises.

With this context in mind, we expect to propose several changes to the assessments framework. First, we plan to raise and index the threshold for banks subject to the large bank scorecard, which is currently set at $10 billion, so that the threshold better reflects the scale, complexity, and risk profile of institutions for which the large bank scorecard is designed. Second, we plan to reduce assessments, recognizing the progress that has been made in growing the Fund. For banks subject to the small bank scorecard, we expect to reduce the assessment rate by a full two basis points. For banks subject to the large bank scorecard, we expect to reduce the assessment rate by a smaller amount across the board, but large banks would be able to achieve a comparable overall reduction if they opt in to the resolution readiness adjustment, which would not be available for (and would be unnecessary to offer to) small banks. In any event, we expect to continue to build up the DIF and proceed towards the FDIC's long-term target of 2 percent, at a pace that resembles that of the pre-2022 trajectory.

Finally, we have also been working on modernizing the large bank scorecard, which establishes the risk-based formula that determines assessments for larger banks and has not been materially updated since 2011. I expect any potential amendments to the scorecard would be proposed at a later date than the other changes mentioned above.

Part 370 Recordkeeping Rule

In 2016, the FDIC finalized a new set of IT and recordkeeping requirements for banks with more than two million deposit accounts, codified at Part 370 of the FDIC's regulations.¹⁰ The rule requires institutions to configure their IT systems so they are able to calculate the insured and uninsured amount in each deposit account, and maintain information needed by the FDIC to determine deposit insurance coverage with respect to each deposit account.

Prior to the Part 370 rulemaking, in 2008, the FDIC finalized a separate rule, codified at 12 C.F.R. § 360.9, which provides insight into a bank's funding and promotes providing deposit information in a standardized format.¹¹ Today, when an institution crosses two million deposit accounts, it "graduates" from § 360.9 to Part 370.

A key motivation behind Part 370 was that, at the time, there were significant concerns about the FDIC's capability to process very large volumes of deposit account records quickly enough after a bank failure to support timely deposit insurance determinations. In other words, the original problem was, in large part, an FDIC systems-capacity issue. Part 370 intended to mitigate that concern by shifting substantial data and systems obligations to the largest banks. Consequently, covered institutions were required to build and maintain internal capabilities to support rapid deposit insurance determinations.

Since then, however, the FDIC has significantly enhanced its claims processing and deposit insurance determination capabilities, particularly following system modernization work implemented in 2021. We have recently conducted several tests to verify the ability of our systems to ingest, process, aggregate, and determine deposit insurance coverage for larger volumes of accounts. The FDIC also has plans to further expand those capabilities going forward.

These improvements in our internal capacity raise the question of whether the FDIC should continue to require large institutions to build and maintain separate insurance determination systems. The FDIC is actively considering an alternative approach that would, for institutions currently subject to Part 370, replace Part 370 with a modified version of § 360.9 that would (1) preserve certain streamlined requirements related to maintaining depositor records in a standardized format, while (2) relieving banks of the requirements to build and maintain independent insurance determination systems, which has proven operationally burdensome. Ultimately, the FDIC is responsible for making deposit insurance determinations in the event of failure; the FDIC maintains existing systems to apply deposit insurance rules, process depositor data, and coordinate the release of funds; and it seems sensible that our rules would reflect that.

Qualified Financial Contracts (QFC) Regulation

Another resolution-related rulemaking that we have been reviewing closely is the FDIC's Part 371 QFC recordkeeping rule.

The QFC rule¹² requires certain banks in "troubled condition" (generally a CAMELS composite rating of 3 or worse) to maintain comprehensive data related to QFCs.¹³ Once an institution is deemed to be in troubled condition, the rule provides an institution with 270 days¹⁴ to come into compliance. However, firms have generally struggled to come into compliance within this timeframe, and many have taken well over a year. Given that large banks may fail rapidly after becoming, or without ever becoming, deemed "troubled" for purposes of the QFC rule, this framework is problematic. Furthermore, a bank's condition may improve to the point at which it is released from the rule soon after making the investments to come into compliance, further calling into question the utility of the rule.

On top of that, the data that the FDIC eventually does receive is of marginal value to achieve the objective of the rule. The purpose of the rule is to equip the FDIC with information to decide whether to terminate a bank's QFC positions in the first 24 hours following a failure. But the data provided does not yield that type of insight; for example, it fails to provide insight into the impact of terminating positions on the value of associated assets or the bank's franchise value, or any broader knock-on effects. In 2023, the FDIC faced the decision of whether to repudiate QFCs following the failures of Silicon Valley Bank (SVB) and Signature Bank. Although the QFC portfolios were relatively small, and neither bank was subject to Part 371,¹⁵ the detailed information required by Part 371 would not have been actionable in deciding whether to terminate positions.¹⁶

We are actively considering revamping the rule to require a narrower set of information that banks can realistically produce in a short period of time, that is less of a data dump and more targeted to inform our decision-making. Additionally, the FDIC is considering whether we can leverage QFC transaction data that is already reported to various data repositories maintained by other agencies.

Resolution Contracting

A critical part of resolution readiness is ensuring that the FDIC can leverage top private sector services as part of the resolution process to maximize recoveries and minimize losses to the DIF. However, the FDIC's current procurement process presents significant challenges for resolutions, where speed and expertise are critical. Many procurement efforts take far too long, and top firms either are excluded from participating due to inflexible FDIC policies or choose not to participate because the process is too difficult, rigid, lengthy, or opaque.

One example is the FDIC's contracts for financial advisors who advise on marketing and selling failed institutions. The FDIC currently has a roster of several firms approved to provide services. While the FDIC has had positive experiences with some of these firms in the past, many of the most highly regarded firms have either not been invited to participate due to internal FDIC policies or chose not to participate because of the costly, time-consuming process. Given the critically important role such advisors play in a large bank failure, the FDIC plans to issue a new solicitation to a broader set of firms in the coming weeks.

More broadly, to address these issues, we have been working on a modified procurement process for resolution-related contracts. The goals are to increase competition, reduce timelines, and allow for greater flexibility in the overall process, so that we can contract with top industry participants and reduce overall costs to the DIF.

Least Cost Test

Under the FDI Act, the FDIC is required to choose the resolution option that is least costly to the DIF.¹⁷ The only exception to this least cost requirement is the systemic risk determination, which is generally only available when large banks fail, and provides a complete exception from the statutory least cost requirement.¹⁸

Over the last three years, there have been three failures in which uninsured depositors took losses. The difference between the winning insured deposit-only bid and the lowest cost all-deposit bid for the three failures were $754,000, $1.2 million, and $3.6 million.¹⁹ By contrast, the current estimate for covering uninsured deposits at Silicon Valley Bank is $16.6 billion. To put that in perspective, the cost of choosing transactions that covered uninsured deposits at all of those three small banks would have cost less than one twentieth of one percent of the cost of covering uninsured deposits at SVB, and less than one 200^th of one percent of the DIF's net worth.

While this state of affairs raises larger questions related to deposit insurance reform and the FDIC's emergency authorities, and the FDIC has been engaged with congressional offices on such reforms, a smaller, more targeted reform that Congress could consider in the interim is providing a de minimis exception to the least cost requirement, which would enable the FDIC to choose a resolution option that is not the least costly option, if either on a dollar or percentage basis, the difference is very small.

Primarily, this would help mitigate the perception of a two-tier deposit insurance regime, where uninsured depositors take losses at small banks but not large banks, by allowing the FDIC to choose an all-deposit bid when the difference in cost to the DIF is essentially a rounding error in the overall cost of bank resolutions. Imposing losses in such cases can have a material impact on a local community, and on community banking as a business model, while saving tiny amounts for the DIF. In addition, there are certain costs to the DIF that the FDIC cannot consider as part of the least cost analysis, such as future cost to the DIF due to contagion, and a de minimis exception could enable the FDIC to consider such costs in certain limited circumstances while still imposing fiscal discipline on the agency.

Nonbank Participation in Failed Bank Bidding Process

We also continue to take steps to increase the participation of private capital from outside the banking sector in the failed bank bidding process, in order to increase competition in the process and reduce costs to the DIF. This can take the form of a nonbank (1) setting up a shelf charter to purchase a failed back outright; (2) partnering with bank bidders in an "alliance bid," which can allow a smaller bank to bid on a larger institution; or (3) purchasing pools of a failed bank's assets at the time of failure.

In furtherance of these objectives, the FDIC (1) rescinded the 2009 Statement of Policy that established a number of restrictions on the ability of private investors to participate in failed bank purchases;²⁰ (2) has been engaged with the OCC and Federal Reserve on establishing an emergency exception to create a rapid shelf charter after a sudden failure;²¹ and (3) has been conducting a pilot²² to qualify a group of non-IDI investors to bid on asset pools from certain failed banks at the time of failure and be eligible for seller financing. We expect to open the pre-qualification process later this year to additional nonbanks that satisfy certain criteria. The intent of this pre-qualification process is to enable the FDIC to sell assets more quickly and increase overall recoveries.

Conclusion

The topics discussed today constitute several areas upon which we have been focused to improve our resolution preparedness. In addition, we continue to (1) work with the Federal Financing Bank to establish a permanent facility to quickly monetize receivership assets²³ and (2) explore reforms to the Title I resolution planning process, among other areas.

Thank you for your time today.