Skip to main content
U.S. flag
An official website of the United States government
Dot gov
The .gov means it’s official. 
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.
Https
The site is secure. 
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
Speeches, Statements & Testimonies

Remarks by FDIC Vice Chairman Travis Hill at the Cato Institute on “Insights on the FDIC's Agenda”

September 21, 2023

It has been about six months since the high-profile failure of Silicon Valley Bank (SVB), and the brief banking turmoil that followed.  Since then, banking conditions have stabilized … but remain somewhat fragile.  In the second quarter, deposits decreased for a fifth consecutive quarter, but outflows have moderated following record-setting declines earlier this year.1 Net interest margin declined for a second straight quarter, but by less than expected, as banks continue to pay more for deposits and competition for funding remains strong.  For example, money market fund assets continue to set all-time highs at the same time that total bank deposits continue to fall, further pressuring deposit rates.2

Overall, industry conditions are encouraging, but significant uncertainty remains.  High rates may persist3 and continue to pressure the industry, and if rates do fall, the cause might be that an economic downturn — along with deteriorating credit quality — has finally arrived.

Meanwhile, the banking agencies are in the midst of an aggressive regulatory agenda – including both a number of items that were under consideration before March and a number under consideration in response to March.  While I think some response to the bank failures is warranted, I worry that an overreaction is underway, and that we are moving too quickly to impose a long list of new rules and expectations at a time when conditions remain precarious.

Today, I am going to share a few thoughts on several of the issues on our agenda.  Overall, I think that any policies we adopt should be balanced, thoughtful, and targeted, and that we should be mindful of both the aggregate impact of all the changes and the current economic environment.

Capital Standards 

To start, in late July, the banking agencies approved a proposal to substantially revise the capital requirements for large banks, which I view as two proposals in one.  One implemented the 2017 international Basel endgame agreement, which would substantially revamp how large banks calculate risk-weighted assets, and the second, completely unrelated to the first, undoes almost all of the tailoring of the capital framework for large banks.

With respect to the Basel standards, our capital rules for our largest banks are already meaningfully more conservative than those in other developed jurisdictions.4 We have already gold-plated the underlying Basel standard that exists today.  The proposal would further gold-plate the new Basel standard in a number of ways, at the same time that European jurisdictions appear to be deviating in the opposite direction.5 The result will be some combination of higher prices and less availability of products and services, and I hope there is openness to revisiting some of these choices as the process progresses.

Large Bank Resolution 

A month later, in late August, the FDIC Board approved several items related to large bank resolution.  One of the proposals would impose a long-term debt requirement on large regional banks, which I generally support.  The objective is to ensure that, if a large bank fails, there is a pool of resources that will always be available to absorb losses in front of the Deposit Insurance Fund and the depositor class, which would significantly reduce both the cost that is socialized across the industry and the tail risk to taxpayers.  There were several aspects of the proposal that I would have addressed differently, but I still think the proposal was worth issuing to receive comments, and my hope is that once we receive and review comments, there will be some willingness to seek consensus among Board members at all the agencies.

The presence of long-term debt would be helpful regardless of how a bank is resolved and regardless of the degree and extent of resolution planning.  The FDIC also proposed substantial changes to the IDI resolution planning rule.  While resolution plans can provide the FDIC with some useful information, and certain aspects of the proposed changes might be helpful, I think the proposal could have better focused on key areas of resolution planning, such as maximizing the likelihood of a weekend sale in the event of a regional bank failure.6

Bank Merger Policy 

Looking ahead, one topic that has been under consideration at the FDIC for the past couple years is bank merger policy.  If we reopen merger policy, I encourage regulators to keep a few principles in mind.

First, the U.S. banking sector and financial services industry more broadly are highly competitive.  While the total number of banks in the U.S. has shrunk considerably in recent decades, the U.S. still has more depository institutions than anywhere in the world,7 and U.S. banks also compete with thousands of other institutions that perform bank-like functions, including credit unions, fintechs, money market funds, retailers, technology companies, independent mortgage companies, private credit, and a range of other nonbank financial companies.

Banks and nonbanks are also no longer bound by geographical limits, as any bank with a website or a phone app can offer products to virtually any customer with a computer or a smartphone.  While not all banks compete nationwide, all banks in effect compete with those who do.  This is a notable contrast from when the bank merger framework was put in place decades ago, when banking was generally a much more local business and banks were heavily restricted in their ability to operate in different geographies.8

To the extent that regulators are concerned about consolidation, rather than make the merger process more difficult, we should instead try to address some of the underlying causes of consolidation, which includes the ever-rising cost of compliance, the steep challenges associated with technology adoption, and the dramatic decline in de novo activity since the 2008 financial crisis.

Additionally, the current merger application process is, in many cases, too long and too opaque.  I appreciate that the FDIC has an important role to play in reviewing mergers under the Bank Merger Act, and that some applications present challenging complexities that take time to work through, but it is in no one’s interest when banks, after publicly announcing a merger, spend months waiting for initial feedback and sometimes much longer before receiving a final decision – while employees and stakeholders wait in limbo.  The FDIC began to make improvements to the process under Chairman McWilliams, but there is still much more work to do, and I fear we have been moving backwards since her departure.

We should also be mindful that it is helpful for banks that are struggling in this rate environment to seek partners, and it is much, much better for a struggling institution to be purchased on an open bank basis rather than bought from the FDIC out of receivership.  One example of this was the announced merger between PacWest and Banc of California,9 and since then we’ve seen more potential interest in the creative reverse merger structure those institutions utilized.  With an industry adjusting to high rates, and possible credit problems around the corner, this feels like a bad time for a crusade against mergers.

Finally, policymakers have talked about a desire for more competition,10 but in the banking context, it’s worth considering to what end?  A primary way banks compete is on price – by offering higher yields on deposits and lower rates on loans.  This is good for consumers, who can earn more on savings and pay less to borrow.  But one way for a bank to get on an FDIC watch list is by paying too much on deposits compared to how much it earns on loans.  Banks also compete on loan terms – by offering longer maturities, less covenants, and other features that can make regulators queasy.  Another way banks compete is through innovation and efficiency, which over the long term can involve things like transitioning away from higher-cost branches and toward lower-cost websites, and engaging in third-party partnerships or banking-as-a-service.  Yet these are all types of activities that bank regulators currently view skeptically.  Of course, capitalism is built on competition, and competition has many benefits, but it appears that regulators want a merger policy that encourages more competition, yet dislike many of the things banks do when faced with stiffer competition.

Liquidity Rules 

Another policy item under consideration is potential changes to liquidity rules for large banks, particularly the liquidity coverage ratio (LCR).11

Following the 2008 financial crisis, when the primary liquidity sources that banks relied on became unavailable, the banking agencies implemented the LCR, based on the rationale that large banks needed to hold a stockpile of high quality liquid assets (HQLA) that could be sold or monetized in times of stress.  In 2023, we experienced a very different stress environment than that of 2008, in which banks had large pools of HQLA, but were limited in their ability to sell these securities, in part because selling would have forced banks to realize losses on securities that had lost value.12

Since the bank failures in March, when the FDIC has looked at banks under stress, the liquidity metric to which we have paid the closest attention is a comparison of (1) a bank’s total cash plus borrowing capacity at the Federal Home Loan Banks and the Federal Reserve to (2) its total uninsured deposits.  In other words, the question was not whether banks had sufficient liquid assets to survive a 30-day stress period, based on a sophisticated formula with lots of assumptions looking at the entire balance sheet; it was a much narrower question of whether banks had enough cash, or quick access to cash, to survive a near-instant exodus of all their uninsured deposits – similar to the SVB experience.

This liquidity metric bears some resemblance to Mervyn King’s “pawnbroker for all seasons” proposal from several years ago.13 Under King’s proposal, the discounted collateral value of a bank’s assets prepositioned at the central bank would be required to exceed its short-term liabilities.  In other words, banks would always need to have sufficient access to cash to satisfy a mass exodus of runnable liabilities.  While replacing the existing liquidity rules with such a proposal would be a significant leap today, it is worth considering how our liquidity rules could better reflect how supervisors on the ground view banks’ liquidity and how banks respond to liquidity needs under duress.

The 2023 experience raises questions at the core of what our liquidity rules are designed to achieve.  A challenge with any liquidity regime is that liquidity available today may not be available under stress.  The only liquidity source that will always be available in any environment is the central bank, because only the central bank has unlimited capacity to manufacture liquidity at any time.  However, in the U.S., a bank in deteriorating condition can lose access to primary credit at the Federal Reserve’s discount window, as occurred with First Republic in April.14 Meanwhile, a banking panic always has the potential to defy our peacetime assumptions about how liabilities will behave.  Even core, retail, insured deposits may withdraw from a bank if customers lose confidence.

I understand the impulse to reconsider aspects of our liquidity rules in light of lessons learned, but if we do, we should do so holistically.  For example, if we are going to change outflow assumptions for uninsured deposits to reflect the possibility that they may run more quickly than previously expected, we should also consider that in such an event, banks are unlikely to fire-sell their stockpile of HQLA in a matter of hours, and instead will more likely pledge all assets available to borrow against.

Overall, I think we should be hesitant to respond to our experience last spring, or any period of stress, by (1) modifying outflow assumptions only for types of liabilities that moved faster than expected, (2) narrowing the definition of HQLA only for types of assets that proved harder to sell than expected, and/or (3) layering on top of the existing liquidity rules a new liquidity metric to reflect the unique characteristics of the most recent stress event.  Instead, I encourage regulators to think holistically about reforms that reflect how banks actually behave in times of stress and are durable for the full spectrum of possible stress events.

On a related note, while some uninsured deposits, at risk of loss in the event of failure, moved extremely quickly this spring, perhaps the stickiest of deposits proved to be brokered certificates of deposit.  As an example, Silvergate Bank had no brokered deposits prior to the fall of 2022, but as the bank began experiencing outflows of crypto-related deposits, it replaced the departing deposits with brokered CDs.15 By the time the bank announced plans to self-liquidate in March, more than 98 percent of the bank’s non-brokered deposits had run off,16 while its brokered CDs all remained through maturity, and some remain on the balance sheet even now.  From the FDIC’s perspective, the main downsides to brokered CDs are that they are costly for the bank and have no value in resolution.  The flipside is that because the depositors have no relationship with the bank, earn high rates, are fully insured, and generally cannot withdraw before maturity, the deposits are extremely sticky, and the depositors are indifferent to whether the bank has a future or not.  Far from being “hot money,” these deposits are so cold they are virtually frozen in place.

Supervision 

On the supervisory front, supervisors have been reexamining several issues following the failures earlier this year, including interest rate risk, concentrations of uninsured deposits, liquidity risk management, contingency funding, and timely remediation of supervisory criticisms.  I generally agree these issues warrant attention, but we should do so with humility, and remember that examiners do not and should not manage banks or determine business strategies, and that bank supervision cannot and should not prevent all bank failures.

On the issue of timely remediation, the failures of SVB and Signature Bank both illustrate the point.  Over the course of several examination cycles, supervisors identified substantial weaknesses in the risk management programs at both banks, but the banks were given long runways and ultimately failed to remediate the underlying issues.  I agree that is unacceptable.

But as we consider ways to ensure timely remediation of supervisory issues, supervisors also need to consider ways to (1) complete exams and communicate findings in a more timely way and (2) better prioritize core safety and soundness risks.  As an example of the former, the FDIC’s report on the supervision of Signature details the extensive delays the FDIC experienced in completing exams and communicating exam findings to the bank.17 As an example of the latter, SVB had a long list of supervisory criticisms, but most were on topics unrelated to the core financial condition of the bank.  I think these issues all need to be addressed collectively: prioritizing the key issues that are fundamental to safety and soundness, and ensuring they are properly processed, communicated, escalated, and remediated in a timely way.  My fear is that instead we may escalate every supervisory criticism, with the result that core safety and soundness issues in some cases still find themselves at the back of the line.

Climate Change 

Speaking of the need to prioritize, another issue on the agencies’ agenda is climate change.

According to the National Oceanic and Atmospheric Administration (NOAA), there have already been 23 weather or climate events with damage exceeding $1 billion in the U.S. in 2023, the largest number on record, surpassing the previous record of 22 set in 2020.18

In my years at the FDIC, I have attended dozens of meetings discussing supervisory issues at particular banks, including healthy banks and unhealthy banks, banks on the verge of failure and banks very far from failure, banks on the coasts and banks far from the coasts, and many in between.  Yet never once have I ever heard a bank supervisor or FDIC staff member mention a climate event as causing stress at a particular bank.  This despite year after year of record- or near-record-setting damage attributable to climate events.

Banks have dealt with natural disasters and extreme weather events for centuries.  In the U.S., there is no record of banks ever failing because of climate-related events, and it has been extremely rare for banks to even suffer meaningful losses.19 Extreme climate events are tragedies for local communities, and for the bankers who live in and support those communities, but financially, banks often benefit in the aftermath, as demand for loans grows, recovery funds flow into the community, and economic activity rebounds.

In 2022, the FDIC proposed principles for managing climate-related risks and discussed plans to issue further guidance in the future.20 The proposed principles lay out a long list of steps the FDIC would expect institutions to take to manage climate risks, which in total would have the effect of significantly elevating the relative importance of climate risks in many aspects of a bank’s decision-making.  To the extent the principles are meaningful and more than a check-the-box compliance burden, the inevitable result will be that banks offer less credit, or charge more for credit, to consumers and businesses in communities that are most vulnerable to climate events, including those in low- and moderate-income areas.  As a general matter, when bank regulators declare something a safety and soundness concern, the expected result should be that banks will do less of it or charge more for it.

When a natural disaster or extreme weather event occurs in the U.S., the FDIC always issues guidance promising flexibility to enable banks to support their communities.21 What the FDIC does not do is tell banks to retrench out of concerns for their safety and soundness.  The proposed principles flip this idea on its head.  Rather than viewing banks as a source of strength to serve communities most vulnerable to climate change, the agencies instead are telling banks to worry about themselves.

Of course, past performance is no guarantee of future results, and the risk that a climate event or climate-related risk could someday result in bank failures is not zero.  But today , that risk is much lower than many other risks that have gotten much less attention over the past couple years, including, as an example, prior to March, the impact of rising rates.

Conclusion 

I will conclude with two final points.

First, I think the agencies are trying to do too much all at the same time – both with respect to the aggregate number of things on our agenda and with respect to individual agenda items trying to solve too much at once.  There needs to be more prioritization, more appreciation of the aggregate costs of doing all this simultaneously, and more consideration of the uncertain economic environment.  And what I described today is far from an exhaustive list of all items under consideration at the banking agencies,22 which is in addition to the robust agendas at other financial regulators, such as the Securities and Exchange Commission (SEC) and Consumer Financial Protection Bureau (CFPB).

And second, undoing the tailoring of our rules would be a mistake.  We benefit from having a financial system with banks of many different sizes and business models.  As institutions shrink in size and complexity, the relative benefits of more burdensome rules diminish – as banks’ relative systemic importance declines and resolution options broaden – while the relative costs of complying with rules increase – as banks have less economies of scale.  Eliminating rules and standards that differentiate among types of firms will ultimately lead to a reduction in differentiation among firms themselves, incentivizing further consolidation and homogeneity in its place.

Thank you for your time.

Last Updated: September 21, 2023