A biologist, a chemist, and a statistician are out hunting. The biologist shoots at a deer and misses five feet to the left. The chemist takes a shot and misses five feet to the right. The statistician yells “We got ’em!”
Among global regulators, “capital neutral” has become popular shorthand.[1] Easily said, it sounds reasonable and measured, particularly when mentioned by people who are reasonable and measured. Start peeling the “capital neutral” onion, though, and one discovers that there is really nothing at its core. If the goal of policymakers is to stimulate economic growth while maintaining a safe and sound banking system, then the right path requires an honest look at the costs and benefits of every capital requirement that is long overdue, and the pursuit of “neutrality” is only an unhelpful detour.
Background and Definitions
It is important to note that there is a major, though subtle, difference between (1) a comprehensive or holistic review of capital requirements and (2) a “capital neutral” approach to assessing capital requirements.
Holistic
A holistic review of capital requirements seeks to ensure that each requirement is calculated appropriately but removes overlaps or inconsistencies among them — for example, ensuring that a given risk is not capitalized twice, or not capitalized at all. Such a review is long overdue, as the Global Financial Crisis produced myriad new or heavily revised capital requirements: new minimum capital requirements and buffers, new leverage capital requirements, new stress-test-based capital requirements, new capital surcharges for the very largest banks and a comprehensive overhaul of the risk-weighting framework for loans and trading activities. Furthermore, almost all of these were developed with an emphasis on “conservatism” – that is, a level of capital above what is warranted by data and analysis — without analysis of the cumulative impact of those add-ons.
The result of this approach is frequently a duplication of requirements for the same risk. For example, large U.S. banks are required by both the Federal Reserve’s stress capital buffer and the Basel III endgame changes proposed in 2023 to the risk-weighting framework to hold capital to cover both operational risk and market risk under extreme, stressed conditions. The stress capital buffer includes the assumption that a bank’s largest derivatives and securities financing transactions counterparty defaults, while the Basel endgame proposal makes basically the same assumption.
Capital neutral
A “capital neutral” approach to reforming requirements is different, however. It predetermines the desired outcome of any comprehensive or holistic review and therefore makes it a charade. A risk that is double-counted must remain double-counted, unless some other risk is given a higher risk weight or some new risk is found to require capital. Doing so at this point necessarily requires ignoring fifteen years of post-crisis history that policymakers should be using to conduct any holistic review.
What Exactly Does “Capital Neutral” Mean?
The idea of “capital neutral” reforms raises a host of conceptual and practical problems.
First, as an aside, there is the question of “capital neutral as of when?” In the United States, the Federal Reserve Chairman first introduced the concept of capital neutrality using a different phrase – that capital levels are “about right” – in 2019.[2] At that point, aggregate required capital for the eight GSIBs was 18 percent lower, on average, than it is today. Nothing in the intervening six years – whether that be real-world experience or academic theorizing – has occurred to suggest that his assessment erred on the low side, and there is considerable evidence to suggest the contrary. Thus, those espousing a capital-neutral view that felt that capital levels were “about right” in 2019 should support actions to reduce capital requirements by 15-20 percent, in line with capital levels at that time.
As noted, this point is an aside, as it takes “capital neutral” as a given, and a rational approach. As described below, that point should not be taken and is not rational (or even legal).
First, at a macro level, “capital neutral” does not appear to mean leaving each individual bank’s capital requirements unchanged, but rather reverse engineering changes to ensure that, together, they produce no aggregate increase or decrease in requirements. That means some banks will see increases and others decreases — that is, it means picking winners and losers. And for any bank to win — that is, receive a decrease in its capital requirements — another bank must lose — that is, see an increase in its capital requirements. The choice of winner and loser, and the size of the increases and decreases, becomes predominantly subjective and more than likely a matter of political horse trading.
Of course, this phenomenon played out in the original, failed Basel III endgame proposal in the United States, albeit at a level far above “capital neutral.” Then Vice Chair for Supervision Michael Barr stated in a speech on September 10, 2024, that the proposal would increase capital requirements for GSIBs by 9 percent on average.[3] That estimated average change included a wide range of impacts on individual banks – based on industry estimates, the proposal would have increased capital requirements for U.S. G-SIBs between 4 and 23 percent.[4]
Second, at a micro level and in a risk-based capital regime, “capital neutral” means that if research and historical data demonstrate that the risk weight for one asset is too low, then it can be lowered – but only if the risk weight for another asset is arbitrarily raised to offset the effect, or if another part of the capital framework (e.g., the stress capital buffer framework or the GSIB surcharge methodology) can be adjusted to increase capital requirements elsewhere. Put more simply, it means that for each thing that is fixed, something else must be made wrong (or more wrong).
A good example of the problems with micro capital neutrality is presented by the Federal Reserve’s GSIB surcharge. In 2015, the Federal Reserve based the surcharge on the higher of two different methodologies, one based on the Basel Committee’s framework that measures each bank’s systemic activities on a relative, market-share basis over time, and another that it created and based on each bank’s “systemic” activities relative to those of the entire global banking system as of a fixed date. It acknowledged that the requirements resulting from its “fixed approach” “could be influenced by factors unrelated to systemic risk such as general economic growth”[5]; in other words, economic growth would inflate the score even as a firm’s systemic importance remained constant. It thus pledged to “periodically reevaluate the framework to ensure that factors unrelated to systemic risk do not have an unintended effect on a bank holding company’s systemic indicator scores.”[6] For political reasons – that is, because it would reduce capital requirements for the largest banks– the Federal Reserve has never done so. But a revision now appears up for consideration.
So, how would this work for a “capital neutral” adherent? At a macro level, the Federal Reserve could make the adjustment “capital neutral” by raising capital requirements for banks not subject to the GSIB surcharge (i.e., regional and mid-size banks) in an amount equal to the decrease for GSIBs. This would be both irrational and unfair. At a micro level, the Fed could nullify any decrease in the GSIB by raising some other capital requirement that applies to GSIBs–say, by raising the capital conservation buffer only for GSIBs, or engineering a model change in the global market shock of the stress test, which applies only to them. Of course, this would be irrational and make the entire exercise pointless. But that is what “capital neutral” dictates.
As another example, some have suggested that “capital neutral” should apply to leverage ratio reform: in other words, to the extent that a binding leverage ratio is recalibrated downward, then risk-based capital ratios should be raised to reach the same level of capital requirement. But this again means assuming that the leverage ratio, not the risk-based ratio, produced the most accurate level of capital (even though the entire basis of changing that leverage ratio was that it was flawed), and that the latter has to be contorted to reproduce the effect of the former.
What all of these contortions reflect is a necessary belief that the parts that add up to the whole are simply unimportant, and that a subjective sense of what each bank’s overall capital should be can and should simply override any thoughtful design of the methodologies behind the risk weights for loans and trading assets, the Federal Reserve’s stress test, and the GSIB surcharge. There is a germ of truth to this: while the Basel risk-weighting framework for credit, market and operational risk were also once based on historical losses, they have all grown increasingly detached from any historical experience – they have effectively become the product of supervisors’ personal heuristics, not data. Similarly, it is an open secret that the GSIB surcharge was not the product of a studied empirical analysis but rather was reverse-engineered to produce a desired outcome for the very largest banks.[7]
But even with these problems, the alternative — targeting a predetermined, total level of capital in the banking system rather than measuring capital needs at each firm and adding them up — is far worse. And of course there is no basis —other than a leverage ratio that regulators have rightly declaimed as a poor measure of risk — to allocate that total, predetermined amount of capital to each bank.
Of course, by its very terms, any approach to capital reform predicated on capital neutrality would be illegal under U.S. law. Regulators are not permitted to make arbitrary adjustments to regulations; they must be based on available data and sound analysis.[8] They must also be subject to notice and comment, but “capital neutral” makes comment fruitless, as the end result has been predetermined — the only subject of public debate is which set of arbitrary choices to make to achieve that end result.
Put another way, regulators would have to demonstrate why the predetermined, aggregate “capital neutral” capital level is more appropriate than the result of adding up each of the parts of the capital stack – Basel minimums, the GSIB surcharge, the capital conservation buffer or stress capital buffer. Of course, if they made that case, then each of the components could be eliminated, and a lot of time and money saved.
[1] See Bloomberg News (May 16, 2025): “A high-level ECB taskforce will propose capital neutral measures to simplify Europe’s banking rulebook by the end of the year, European finance ministers were told this week, as the debate intensifies over the bloc’s regulatory burden.”
[2] See:
- Jerome Powell, Chairman of the Federal Reserve (February 27, 2019): “I think that the overall level of our capital, particularly at the largest firms, is about right.” https://www.govinfo.gov/content/pkg/CHRG-116hhrg35694/html/CHRG-116hhrg35694.htm
- Jerome Powell, Chairman of the Federal (July 31, 2019): “Okay. So, on the first, I would say that I view the level of capital requirements and the level of capital in the system as being about right.” https://www.federalreserve.gov/mediacenter/files/fomcpresconf20190731.pdf
- Randal Quarles, Former Vice Chair for Supervision of the Federal Reserve (Oct. 2, 2018): “…said he thinks bank capital levels are ‘roughly about right,’ adding,‘As I said before, Chairman Powell has also said, I think, that the capital levels, the total loss absorbency capacity in our system is roughly about right.’” https://www.govinfo.gov/content/pkg/CHRG-115shrg33409/html/CHRG-115shrg33409.htm
- Michael Hsu, Former Comptroller of the Currency (2021): “Despite a once-in-a-lifetime pandemic, the banking system remains healthy. Key measures of financial strength – capital and liquidity ratios – are strong.” https://democrats-financialservices.house.gov/uploadedfiles/hhrg-117-ba00-wstate-hsum-20210519.pdf
- Michael Hsu, Former Comptroller of the Currency (2023): “OCC-supervised banks in the aggregate have strong levels of regulatory capital and healthy levels of profitability, while maintaining sufficient liquidity buffers.” https://www.congress.gov/event/118th-congress/senate-event/LC74296/text
[3] Michael Barr, The Next Steps on Capital, (Sept. 10, 2024): https://www.federalreserve.gov/newsevents/speech/barr20240910a.htm
[4] “Large-Cap Banks 2025 Outlook,” Barclays, 6 January 2025.
[5] Board of Governors of the Federal Reserve System, Regulatory Capital Rules: Implementation of Risk-Based Capital Surcharges for Global Systemically Important Bank Holding Companies, 80 Fed. Reg. 40982, 49085 (Aug. 14, 2015) (final rule).
[6] Id.
[7] This occurred first at the Basel Committee, then again in the U.S., after the Federal Reserve decided it preferred to an even higher GSIB surcharge outcome.
[8] Importantly, such data and analysis should not be a selective portion of outdated academic studies of what constitutes optimal capital. First, those studies generally suggest that capital levels are too high (see https://bpi.com/what-is-the-optimal-level-of-bank-capital/). Second, those studies again talk about overall levels of capital across the banking system in general and have nothing to say about how to calibrate capital requirements among individual banks.
