FDIC: Speeches, Statements & Testimonies – 10/4/2023

Thank you for this opportunity to speak with you all today.¹

I think it’s probably fair to say it’s been an eventful last six or seven months
at the FDIC. Just two months after I arrived, we confronted the second, third, and fourth
largest bank failures in FDIC history. For better or worse, we made the decision to invoke
the FDIC’s emergency powers—the so-called “systemic risk exception”—to
facilitate the resolutions of SVB and Signature.

Now, that decision was, in my view, an admission that 15 years of costly reform efforts had
still not left us with an effective framework for resolving failed banks. But I don’t
think the solution is to pile on yet more prescriptive regulation or otherwise try to push
responsible risk taking out of the banking system.

Instead, we should try to accept—I mean truly accept—that bank failures, even large bank
failures, are an inevitable result of a dynamic and innovative economy. We should plan for
those bank failures by focusing on strong capital requirements and an effective resolution
framework as our best hope for eventually ending this practiced habit we have developed of
privatizing gains while socializing losses.

That is why I supported our proposal to require certain large banking organization to have
outstanding a minimum amount of long-term debt. That is also why I have kept an open mind on
proposals to enhance our regulatory-capital framework.

All that said, I was of course not able to get to “yes” on our July proposal to
implement the Basel Endgame standards.

And so I would like to take this opportunity to elaborate some points on my Endgame
statement.² First, I’ll explore further
a point I advanced that we have not provided an adequate rationale for some of our key
design decisions. Second, I’ll address some counterarguments I’ve heard to my
concern that the proposal will push intermediation out of the banking system. And finally
I’ll close with an open question on our long-term debt proposal.

I’m mindful of course that we have an open comment period on each, so I’ll limit
myself to clarifying my Endgame and long-term-debt statements. But my hope is that a
takeaway of today’s discussion is that important aspects of the Endgame proposal are
underdeveloped, they are contested, and they are exceedingly consequential. Hopefully that
helps make the case for interested parties to take seriously their opportunity to submit
comments. I know many are already doing that.

This debate is an important one. It should be a vigorous one.

Faith in Basel

So, as to the first issue, we the regulators need to make a case for our rules. That’s
especially true if we’re going to do something that will have a real impact on our
economy. I’m not sure we did that for some key aspects of our Endgame proposal.³

Now, I don’t think the regulators are intentionally trying to hide the ball. Instead,
what’s unique here is that we’re implementing standards developed by a third
party—the Basel Committee—and the Committee itself offered little to no explanation for some
of its decisions. That then leaves us simply guessing, unable to defend or perhaps even
understand important aspects of the Basel standards we’re trying to implement. In
short, and I would put some emphasis on this, the proposal amounts to a big leap of faith in
the Basel Committee.

I’ll get to some examples in a moment, but before that, perhaps it’s worth
exploring whether this should matter. Some questions:

First, do commenters have a meaningful opportunity to comment on those aspects of our
Endgame proposal that do not have much in the way of explanation?

Second, what, if anything, do these gaps in the Endgame standards suggest about the Basel
process itself? Did the Committee perhaps leave some important work undone? And if so, how
did that happen?

Third, what, if anything, should the U.S. regulators do to fill in these gaps? Should we try
to prompt the Committee to go back and show its work? Or maybe, perhaps more to the point,
if we cannot discern a convincing rationale, should we consider deviating from the Basel
standards to develop our own approach that we can actually defend?

To make this more concrete, I’ll offer up a few examples of where the
Committee’s Endgame standards might be rather under-developed.

I already raised in my Endgame statement some questions about the Committee’s approach
to operational-risk capital. Just to recap, the Committee actually acknowledged that its
approach would result in overcapitalization of banks with high-fee revenues.⁴ The Committee
even proposed a fix for this issue.⁵ But it then dropped
that fix from the final standards without explanation. That leaves us trying to defend an approach for
operational-risk
capital that its own Basel authors have said does not work for high-fee-revenue banks.

Similarly, the specification of the formula and the sizing of the coefficients have a
significant impact on operational-risk-capital requirements. The Committee made considerable
changes to the formula in the final standards, again with no explanation. That leaves us
unable to defend an important formula.

Another example is the profit-and-loss attribution test, or what is called the “PLA
test.” A bank using internal models to determine its market-risk-capital requirement
would use the PLA-test metrics to measure the extent to which its risk-management models
align with its front-office models. Depending on the test metrics, each trading desk would
fall into a “green,” “amber,” or “red” zone. A desk in
the amber zone would have metrics that suggest some moderate divergence between the models.
That would then trigger an additional capital requirement for the desk called the “PLA
add-on.” A desk in the red zone would be disqualified from using its models. And so,
the quantitative thresholds for this traffic light approach really do matter in determining
the market-risk-capital requirement.

Given its importance, you might assume there is ample theory and data to back up the Basel
Committee’s calibration of these thresholds. But, as you have probably already
guessed, there is precious little in the public domain as to how the Basel Committee came up
with these thresholds.

I have heard some suggestions that the Basel working groups calibrated these thresholds
using simulated data and that the Basel Committee knew it had more work to do here. Maybe
there’s something to that. But we’re all left guessing here, including even the
U.S. regulators. It’s interesting that in the 2018 consultative document, the
Committee said that “upon finalisation of the traffic light approach into the market
risk standard, the Committee will continue to monitor the effectiveness of the finalised
calibration of the thresholds to ensure their appropriateness.”⁶ Perhaps this sense that the thresholds are something of a
work in progress is why the UK regulator has proposed a delayed implementation of the PLA
test of at least one year.⁷

This list could go on . . . .

With respect to the securitization framework, it would be helpful to have more insight into
the theory and evidence supporting the extent to which it departs from capital neutrality,
in particular the past quantitative impact study assessing the p value of 1.

It is also unclear to me whether commenters have enough data to assess the sizing of the
risk weights and correlation factors for the sensitivities-based method for market-risk
capital.

There are some open questions in my mind around the liquidity horizons for modellable risk
factors.

The rho of 0.6 that provides limited diversification benefits for certain non-modellable
risk factors could merit some more public-domain discussion.

More generally, the whole approach to diversification benefit under the market-risk
framework could merit more theoretical and evidentiary support.

Now, some might respond that the lack of transparency is nothing new, that the regulators
always have had to more or less pick numbers out of the air when setting risk-based capital
requirements. It might be true that past practice does sometimes look like that. But I am
skeptical that was good practice to begin with, even if it was understandable given the data
limitations of the time. I also would submit that things have changed in ways that make it
more important that we try to explain our calibration decisions.

Increases in regulatory-capital requirements since the financial crisis—in particular the
adoption of the stress capital buffer, GSIB surcharge, and eSLR—might have increased the
frequency and extent to which the regulatory capital for any particular exposure exceeds a
bank’s view as to the economic capital required for that exposure. For these
exposures, that could tend to make regulatory capital more important to each bank’s
bottom line. Perhaps even more importantly, that means regulatory capital could drive
standardization in banks’ business models and balance sheets, and perhaps even give
regulators considerable influence over the pricing and allocation of investable funds.

The practice of giving reasons for our calibration decisions becomes all the more pressing
in a world in which our regulatory-capital requirements matter more for at least two
reasons.

First, we are more likely to get capital requirements right—or at least to get them less
wrong—if we are more transparent as to how we calibrated the requirements. Well-developed
rationales check the understandable inclination to work backward from some gut sense as to
the right level of capital, a gut sense that might be motivated less by evidence and more by
recency bias or other extraneous concerns. Well-developed rationales also give researchers
and stakeholders an opportunity to show what we got wrong.

Second, transparency as to our calibration methodology can help foster legitimacy and
consensus. Given the importance of capital requirements to banks and society at large, there
is likely to be more scrutiny and even controversy engendered by changes in these
requirements. Well-developed rationales add legitimacy by dispelling any notion that the
changes are arbitrary. Where consensus fails, well-developed rationales clarify where we
disagree, which in turn can focus research efforts to eventually bridge those
disagreements.

And so, summing up this point, I hope commenters will take a close look at any less
well-developed aspects of the Endgame proposal and help us think through what we should do
next to ensure we get this right.

A Level Playing Field and the “Race to the Bottom”

Next, I thought I might spend a few minutes trying to clarify a point I made in my statement
regarding the financial-stability implications of the proposal.

Some have argued that the Endgame proposal could, if finalized, push intermediation out of
the banking system to nonbanks. To the extent those nonbanks are subject to less regulation,
that dynamic could tend to increase risk to financial stability. Others have then
counter-argued that we shouldn’t let this concern put us on a “race to the
bottom,” and instead we should revisit the regulation of nonbanks.

Perhaps folks are talking past each other; I don’t think anyone is actually arguing
for a “race to the bottom.”

So, to try to clarify the point, the premise is that we should aim to foster a level playing
field across similarly situated market participants, taking into account differences in risk
profiles associated with different funding and business models. The question, then, is
whether our Endgame proposal is consistent with that objective.

And on that point, I’m left struggling to see how we can work to harmonize
requirements across banks and nonbanks when we have sometimes not offered a calibration
rationale for the bank requirements, and indeed some aspects of the Endgame proposal
arguably might even be contrary to the available evidence. How are other regulators supposed
to regard our work product here?

I think the proposal’s approach to mortgages makes this point more concrete.

First a bit of history. We tend to forget that Fannie and Freddie have not always dominated
the mortgage market. The GSEs did not begin to grow rapidly until we began increasing bank
capital requirements in the 1980s. Because the GSEs’ capital requirements remained
unchanged, the GSEs had a competitive advantage in holding mortgage-related risk. The
natural incentive was for banks to take a mortgage that had a 50% risk weight, pay a
guarantee fee to a GSE, and get back a GSE-guaranteed mortgage-backed security that had a
20% risk weight. And so capital arbitrage drove rapid growth in the GSEs. The GSEs’
market share increased from 10% in the early 1980s to 25% by the end of that decade, and
then to 44% by the end of 2003, culminating in the bailouts of the GSEs in 2008.⁸ Today, following the increase in bank
capital requirements after the financial crisis, the GSEs now account for more than half the
market.⁹

In September 2020, FSOC took a look at this dynamic. FSOC then issued a statement that
argued that credit-risk-capital requirements at the GSEs that are less than that of banks
would concentrate risk at the GSEs. According to FSOC, alignment of credit-risk-capital
requirements across the GSEs and banks would mitigate the associated risks to financial
stability. To that end, FSOC “encourage[d] FHFA and other regulatory agencies to
coordinate and take other appropriate action to avoid market distortions that could increase
risks to financial stability by generally taking consistent approaches to the capital
requirements . . . .”¹⁰

The Endgame proposal was a great opportunity to take up that recommendation. It would have
been fairly easy to implement on that recommendation. The U.S. regulators played a key role
in developing the Basel Committee’s more risk-sensitive and empirically informed risk
weights, and, so, the Committee’s approach was quite familiar to us. Where the
Committee ended up also lines up fairly well with the credit-risk-capital requirements FHFA
adopted for the GSEs in 2020.¹¹

And so the stars were aligned for us to actually implement FSOC’s recommendations by
adopting the Endgame risk weights for mortgages. But instead we didn’t even
acknowledge FSOC’s review. And then we went further down our own road by adding a
large 20-percentage-point surcharge to where the Basel Committee ended up.

Compounding the enigma here, we offered no loss history or other evidence to support the
sizing of the surcharge.

So, bringing this back to the original point: Sure, lower capital requirements outside the
banking system should not lead to a “race to the bottom.” But how could FHFA or
any other regulator align its capital requirements with those contemplated in the Endgame
proposal when our proposed approach to mortgages appears to be so at odds with the available
evidence?

That’s a big open question I think commenters should look at. And it has nothing to do
with a “race to the bottom.”

Long-Term Debt: Market-Structure Implications

Finally, I would like to close with a few words on our long-term debt proposal.¹²

As I said at the beginning, we should plan for bank failures by focusing on strong capital
requirements and an effective resolution framework as our best hope for eventually putting
an end to our bailout culture. In the hopes of moving closer toward that goal, I generally
supported the proposal to require certain large banking organizations to have, like their
GSIB counterparts, outstanding a minimum amount of eligible long-term debt with an aim to
enhancing the resolvability of those banking organizations.

All that said, as I mentioned at the time of our vote, I do have some open questions about
whether the proposal could have some unintended consequences.
¹³

One unique feature of our proposal is that regional banks and certain foreign banking
organizations would be required to issue long-term debt out of both the top-tier U.S. parent
company and the bank subsidiary. U.S. GSIBs, in contrast, are required to issue long-term
debt out of only the parent. U.S. GSIBs do preposition some resources at the bank subsidiary
through the internal issuance of debt by the bank subsidiary, but that is done on a bespoke
basis developed through dialogue with its regulators.¹⁴

This difference in approach means that regional banks would have less flexibility than the
U.S. GSIBs to preposition resources throughout the banking organization. That, in turn,
would suggest that regional banks could have less flexibility—and perhaps more
difficulties—in developing resolution plans that preserve the franchise value of their
non-bank businesses.

This is perhaps not an issue today, as most domestic covered banking organizations generally
have a bank-centric business model. But I will be eager to see how commenters assess this
risk that the proposal that could lock in regional banks’ current business models or
even incentivize regional banks to pull more activities into their bank subsidiaries.

This would be particularly a problem to the extent it has the unintended consequence of
shielding the U.S. GSIBs from competition with our regional banks.

Conclusion

To close, thank you again for this opportunity to elaborate on some of the points I made in
my statements on our Endgame and long-term debt proposals. These are exceedingly
consequential rulemakings.

As I’ve said, we’re more likely to get these proposals right—or at least to get
them less wrong—if we hold ourselves accountable for giving our reasons for where we end up.
To the extent we’ve not done that, I hope commenters will let us know, and also share
views on how we should try to fix those gaps, even if that means developing our own U.S.
implementation that deviates from the Basel standards in some respects.

Thank you, and I’ll look forward to any questions.