The Federal Reserve is proposing changes to the Capital requirements of U.S. Banks. In this article I take a deep dive into revisions to the G-SIB surcharge, a key component of the regulatory capital framework introduced after the Great Financial Crisis of 2008.

Background

Michelle W. Bowman, Vice Chair for Supervision at the Board of Governors of the Federal Reserve, recently gave a speech at the Cato Institute on Capital Rules for the Real Economy (listen here). In this she outlines upcoming proposed rules (not yet published) to implement the final phase of Basel III in the U.S and I quote:

“The proposals will modify each of the four pillars of our regulatory capital framework for the largest banks: stress testing, the supplementary leverage ratio, the Basel III framework for risk-based capital requirements, and the G-SIB surcharge.“

In today’s article, I will look into the G-SIB surcharge.

G-SIB surcharge

Global Systemically Important Banks (G-SIBs) are subject to capital surcharges. Eight U.S. banks are G-SIBs and I have written articles covering these here and here.

The proposal to modify the G-SIB surcharge:

• Updates the coefficients, used to weight each of the components, (which have not changed since 2015) and re-aligns with the Basel Method (which calculates a lower surcharge)
• Revises the surcharge component for short term funding, originally intended to be 20% of the surcharge, but now roughly 30%.
• Moves from year-end values to averages, reducing incentives for year-end adjustments to balance sheets
• Reduces cliff effects and increases sensitivity to changes in a firms’s risk profile by assigning surcharges in increments of 10 basis points instead of the current 50 bps.

The detail above on the last two is sufficient and they appear on the face of it to be good improvements, so I won’t elaborate on these.

However the first two are much more involved and to understand the detail, we need to look at the data.

G-SIB Surcharges

The G-SIB surcharges currently in effect are based on financial metrics as of December 31, 2024, which are provided in FR Y-15 Snapshot reports published in November 2025 (see FFIEC NIC page).

ActrixFT Apps, have the FR Y-15 data for Bank Holding Companies and calculate the surcharge.

Showing surcharges for each U.S. G-SIB:

• JP Morgan Chase – 5%
• Goldman Sachs – 3.5%
• Citigroup – 3.5%
• Bank of America – 3.5%
• Morgan Stanley – 3%
• Wells Fargo – 1.5%
• Bank of New York – 1.5%
• State Street – 1%

U.S. Bank surcharges are determined by the higher of two methods; Method 1 (Basel methodology) or Method 2 ( Federal Reserve methodology) and the later has always been higher, which is what I have shown above.

The Basel methodology surcharges can be see in the OFR Systemic Risk Monitor:

Showing:

• JP Morgan Chase – 2.5%, as compared to 5%
• Bank of America – 2%, compared to 3.5%
• Citigroup – 2%, compared to 3.5%
• Goldman Sachs – 1.5%, compared to 3.5%
• Morgan Stanley – 1%, compared to 3%
• Wells Fargo – 1%, compared to 1.5%
• Bank of New York – 1%, compared to 1.5%
• State Street – 1%, the same

So if we take the proposal to “re-align with the Basel Method” literally, this would result in very significant decreases in the surcharges for 7 of the 8 U.S. G-SIBs.

(Note one of the key differences in the methodologies is that teh Federal Reserve method replaces the measure of substitutability with a measure of short term wholesale funding.)

G-SIB Coefficients

The proposal refers to “updating the coefficients”, lets look into these next for the current Federal Reserve method 2, using JP Morgan as the example.

Coefficients are specified for each component item and these are multiplied by the value of each item (e.g. Total exposures) and the Product column is then summed to get the G-SIB Score, in this case 984.52. This then results in a capital surcharge of 5%

The proposal will change (decrease) these coefficients, so as to bring the G-SIB Score (more) inline with the lower Basel (method 1).

Short term Funding

The proposal specifically mentions that the short term funding component will be revised as it is now almost 30% of the surcharge, while it was originally intended to be 20%.

For our JP Morgan Chase example above it is, 150.99 / 984.54 = 15%, so that is lower, but let’s check for all the U.S. G-Sibs.

Showing that these differ markedly between institutions, Wells Fargo and JP Morgan Chase the lowest at 13% and 15%, with Morgan Stanley, State Street and BNYM the highest at 50%, 56% and 59% respectively.

An overall percentage of 29% for all 8 institutions.

If the proposed revision just takes the form of a lower coefficient for the short term funding metric, then it will result in the greatest reduction to the overall G-SIB Score for Morgan Stanley, State Street and Bank of New York.

Given the actual surcharges, the most meaningful capital reductions from lowering the short term funding component are likely to be for Morgan Stanley and Goldman Sachs.

2027 G-SIB Surcharges

The proposed changes are expected in a few weeks and the calculations determined in November 2026to take effect in 2027.

We can estimate what the surcharges would be without these changes, as the ActrixFT Apps have FR Y-15 reports for December 31, 2025.

The surcharge increasing by 0.5% for each of the first five institutions in the chart above and remaining the same for the remaining three. JP Morgan Chase up to 5.5% and Goldman Sachs to 4%.

However as the proposals are expected to be implemented this will not happen.

More likely than an increase of 0.5%, we should see absolute reductions from current levels of between 0.5% to 2%, if we take the “inline with the Basel method” literally. So a significant revision to the existing gold-plated Federal Reserve method.

How Material?

How material are these surcharge reductions likely to be?

Let’s again take JP Morgan Chase as our example. From the latest FRY9C filing (as of Dec 31, 2025):

• Total Capital is $344 billion (of which Tier 1 is $307 billion)
• Total Risk Weighted Assets are $1,982 billion
• Total Capital Ratio is 17.35% (as 344/1,982)
• Reducing the Capital Ratio by 1% to 16.35%, means $324 billion in capital, a reduction of $20 billion
• Or keeping the Total Capital but keeping the same capital ratio of 17.35%, allows Total RWA to increase to $2,104 billion, an increase of $122 billion

Meaning in this eventuality JP Morgan Chase could decide to return $20 billion to shareholders (buybacks or dividends) or grow its business by $121 billion RWA. Either is positive for shareholders.

Let’s make some assumptions on the surcharge reduction and estimate the capital saving or RWA increase for each of the 8 institutions.

Showing each institutions possible Capital saving or Asset increase, the total for all 8 G-SIBs is $117 billion of Capital (9% reduction) or $777 billion of RWA (10% increase).

JP Morgan Chase, Goldman Sachs, Morgan Stanley, Bank of America and Citigroup benefiting that most, in order of capital reduction as a percentage of total capital.

The actual revisions will be different, not least because of the 0.1% increment instead of 0.5%, but should be ballpark close.

Also recall I have focused only on one of the four pillars with revisions and there are likely to be reductions from the other pillars e.g. stress capital buffer.

A net positive for shareholders of these institutions, that I assume is already somewhat reflected in their share prices.

That’s all for today.

In Summary

• The Federal Reserve is proposing revisions to U.S. Bank Capital requirements for each of the four pillars; stress testing, the supplementary leverage ratio, the Basel III framework for risk-based capital requirements, and the G-SIB surcharge
• The G-SIB surcharge is gold-plated for U.S. bank and the proposal will bring it inline with the Basel method, reducing the surcharge
• Specifically the coefficients that weight difference metrics will be revised and short term wholesale funding metric’s contribution to G-SIB scores will be reduced
• ActrixFT Apps provide data and insights into the numbers
• I use these to estimate possible capital savings for each of the U.S. G-SIBs

If you are interested in more data and details, please contact us.