Revisiting an Argument About Regional Banks' Uninsured Deposits
Plus, Vice Chair for Supervision Barr talks crypto
Given the fast-moving situation, I was planning to avoid writing about Silicon Valley Bank lest I write something that looks pretty dumb in hindsight. (Or even in foresight.) But a tweet from Brian Cheung of NBC News raised a point that made me go down a rabbit hole yesterday afternoon and all of a sudden here we are.
Specifically, Brian observed that SVB’s deposits are more volatile than those of your typical bank.
To try to put some concrete numbers around this, I looked at the December 31 call reports of non-GSIB banks with between $130 billion and $200 billion in total deposits.1
On Schedule RC-O Memorandum Item 1.b of their call report, banks are required to report the number of deposit accounts (excluding retirement accounts) with deposits of more than $250,000. SVB does look like a bit of an outlier here compared to banks with a similar level of deposits.2
Also notable is Memorandum Item 2 of Schedule RC-O, which requires banks to estimate their total amount of uninsured deposits.3 Looking at uninsured deposits as a percent of total deposits, SVB looks a little different here as well.
Finally, for silly comparison purposes only, here is the same data for a small sample of other banks. Do not read anything into the choices here, I was just trying to include a selection of different kinds of banks: a GSIB, a super regional larger than the regionals above, a boring $50 billion-ish assets bank, and a couple of crypto-involved banks that have been in the news recently.4
I am mostly sharing this because I thought it could be interesting, and not because I believe it supports or undermines any particular argument.
But if you are looking for broader implications, one thing it did make me think of was the Federal Reserve Board and FDIC ANPR last year suggesting that it might be appropriate for large regional banks to be subject to TLAC or other resolution-resource related requirements. One reason the agencies thought this could be necessary was an increase in uninsured deposits:
In addition, some large banking organizations have increased their reliance on large uninsured deposits to fund their operations over the past decade. These deposits may be less stable relative to insured deposits under conditions of firm-specific stress and resolution. Uninsured deposits comprise a significant portion of Category II and III banking organizations' funding base, standing at roughly 40% of total deposits as of the first quarter of 2022 as a group. While GSIBs also have high levels of uninsured deposits, the regulatory resolution framework that has been built up around them—including TLAC and long-term debt requirements—help to mitigate related risks.
As we discussed back in January, most commenters that addressed this argument did not find it compelling. For example, the ABA said:
Because a significant part of the growth in the 2019-2021 period that the Agencies cite remains in very liquid, low-risk assets at the affected banks, however, the implications for resolution preparation and management are much less concerning. […]
These factors are also very relevant in addressing the Agencies’ concerns about the increased proportion of uninsured deposits in non-G-SIB LBOs. The Agencies note that some non-G-SIB LBOs have increased their reliance on uninsured deposits to fund their operations over the past decade and that these deposits may be less stable than insured deposits under conditions of stress. Again, however, the consequences would be mitigated as long as the institutions’ asset mix reflects a disproportionately high component of low-risk, highly liquid assets. This factor means, first, that serious institution-specific stress situations are relatively less likely, and, second, that any uninsured deposit runoff is relatively more manageable, making failure less likely.
I raise this not to suggest that ABA is necessarily wrong,5 but rather to just set down a marker here that, given the events of recent weeks, uninsured deposits seem likely to continue to factor into the debate as to whether additional requirements for regional banks are necessary.
One final thought: if insured deposits are in fact going to be regarded as an indicator of risk justifying more stringent requirements, what does that imply for the tailoring framework, which currently does not directly take this factor into account? The agencies in their ANPR suggested that further sub-categorization might be necessary. I could see how the above data would support that view.
Three Quick Thoughts on Michael Barr’s Crypto Comments
Yesterday morning the Federal Reserve Board’s Vice Chair for Supervision Michael Barr went to the Peterson Institute for International Economics to deliver a speech on the Board’s approach to crypto-related activities.
The substance of the speech was consistent with recent statements from the federal banking regulators, but there were three ancillary things that were sort of new or otherwise worthy of comment.
The Downsides of a Light-Touch Approach to Small Banks
In the Q&A moderated by Anna Gelpern following the speech, Vice Chair for Supervision Barr noted that many of the banks that have found themselves in the worst situations recently have been relatively small.
Michael Barr: […] I think what we’ve seen with some of the smaller institutions that have gotten involved in providing traditional banking services to crypto-related companies is that the extent to which they are exposed to correlated risks was not well understood. And I think what we saw with what people call the crypto winter is that risks in the crypto sector are highly correlated. So if you have deposits from crypto-related entities, it is highly likely that those entities will suffer stresses in similar ways at the same time. And those stresses might cause wild swings in inflows of deposits and outflows of deposits. The experience of smaller institutions is that they have actually experienced that. […]
Anna Gelpern: Shouldn’t somebody have picked that up?
Michael Barr: Yeah, I do think if you look at the range, particularly of smaller institutions that were involved, we tend to have a very, I would say, light-touch approach to the smaller institutions. So there is more of an impetus on them to actually be paying attention to these new and novel risks. And we need to make sure that they understand them. There are obviously larger institutions that are exposed to these risks too, but the exposure tends to be a very, very small part of their balance sheet. So even if they experience the same deposit outflows, they are more insulated because of the diversification of their balance sheet. […]
Capitol Account yesterday suggested this might be a shot at Republicans who typically argue for this sort of light-touch approach. Maybe there is something to that, but decreasing the regulatory burden on small banks has generally been a bipartisan endeavor.6 So I heard this as less an attempt at point scoring and more just as a factual observation.
I do wonder what, if anything, it implies for other new or unproven business models a number of smaller banks have also been eager to adopt, such as a banking-as-a-service. I think it is good to allow banks to experiment, and I take seriously the concerns of those who worry that a heavier-touch approach might stifle innovation. Even so, it may be worth giving more thought to what can be done — while still taking a risk-based, tailored approach — so that small banks are not caught off guard by new and novel risks.
No Balance Sheet Activities (Sort Of)
Elsewhere in the Q&A, Vice Chair for Supervision Barr addressed the ongoing discussions on crypto regulation at the international level. After noting that the Federal Reserve is an active participant on the Financial Stability Board, VCS Barr addressed the recently finalized Basel framework for crypto-asset exposures.
With respect to the Basel process, the Basel process has resulted for example in a framework for capital treatment for crypto-related activities. And that would also be consistent with our thinking about the approach in this area. We don’t currently, to our knowledge, have any firms that we regulate that have crypto on their balance sheet, and we’ve made it clear to firms that we don’t think that they should. We don’t think that in our current environment, our current state of governance and controls in the crypto sector, we don’t believe it would be a safe and sound practice for them to do so. So we don’t yet have the question of how much capital they should hold if they have [that exposure]. But those larger frames are consistent with how we are thinking about risks in this area.
It is true that Board-supervised institutions do not currently have “crypto on their balance sheet” in the way that banking regulators typically think about it.
But thanks to SAB 121, at least a few firms under the Board’s supervision do currently report crypto assets and liabilities on their balance sheet. And, so far, the Board has not provided public guidance as to what this means or how banks should treat these kinds of exposures for capital purposes.
This is not the biggest deal in the world, and everyone knows what the Vice Chair for Supervision meant, so this is not intended to be a gotcha, but even so it stood out to me because glossing over this point here was in keeping with Chair Powell’s congressional testimony earlier this week in which he worked hard to punt SAB 121 back to the SEC. It may have started there, but now for better or worse it is in the banking regulators’ court.
He’s Putting Together A Team
In his speech, Vice Chair for Supervision Barr stated that the Board intends to enhance its supervision of crypto-related activities, including by “creating a specialized team of experts that can help us learn from new developments and make sure we're up to date on innovation in this sector.”
I am not positive this is related, but in case of interest the Board a few days ago began soliciting resumes for an “Associate Director, Novel Activities and Innovation Policy” in the Division of Supervision and Regulation.
Reporting to the senior associate director, the associate director has responsibility for the direction and administration of divisional activities related to the changing landscape and risks in the financial system, specifically fintech and crypto-asset related activities and to strengthen the Federal Reserve’s supervision and regulation of these activities. The associate director collectively oversees three sections to include Innovation Policy, Novel Activity Analytics and Novel Activity Monitoring. The associate director will also oversee the System team executing exam work for the Novel Activity Program.
It is not clear where specifically this role is going to sit within the broader S&R org chart.
Thanks for reading! Thoughts, challenges, criticisms are always welcome at bankregblog@gmail.com
Most GSIB banks have deposits above this level, but I include the non-GSIB qualifier here because there are a couple of GSIB bank subsidiaries with total deposits (at least on a standalone bank basis) in this range.
Because this is data as of December 31, 2022, the BMO data is data for BMO Harris Bank alone and does not incorporate the effects of the Bank of the West acquisition.
The call report instructions tell banks they should estimate this number based on the $250,000 deposit insurance limit and the figures reported in response to Memorandum Item 1.
I chose to use September 30 data for Silvergate Bank because the data reported as of December 31, 2022 now clearly seems to be point-in-time data reported in the midst of a run that was already in progress.
There is also the point BPI made that even if, viewed in isolation, uninsured deposits make banks comparatively more risky than they were previously, this does not “take into account the aspects of the regulatory and resolution framework [currently applicable to regional banks] that mitigate related risks.”
For instance, in a 2016 podcast interview Barr said that he thought there was “room in areas beyond Dodd-Frank for community bank relief.” Barr also in 2018 criticized the EGRRCPA for including provisions that “that only benefit large banks” without providing “meaningful community bank relief.”