Basel III Monitoring June 2025: Where the Capital Impact Sits

ByRegReportingDesk

On 24 March 2026 the Basel Committee on Banking Supervision published its Basel III monitoring exercise as of 30 June 2025, and the single number a capital planning team should read first is 1.7%. That is how much the fully phased-in final Basel III framework would add to the Tier 1 minimum required capital of the 101 large internationally active banks in the sample. Six months earlier the same figure was higher. The estimated impact fell by 1.1 percentage points for Group 1 banks and by 0.5 percentage points for the 29 global systemically important banks (G-SIBs), which the Committee attributes to implementation progress.

For anyone building a capital plan or recalibrating a buffer stack, the direction matters as much as the level. The aggregate cost of the finalisation is shrinking as jurisdictions switch their banks onto the final rules, and the residual increase is concentrated in two places: the output floor for smaller banks, and market risk for the largest. Read carelessly, the report reassures. Read as a benchmark against your own COREP output, it tells you where the last of the Basel III pressure actually lands.

This is a semiannual, anonymised aggregate built from voluntary submissions collected by national supervisors on 150 banks, 101 Group 1 (including the 29 G-SIBs) and 49 Group 2 banks. It assumes full implementation with no transitional or grandfathering arrangements, and it excludes Pillar 2 entirely. Those two design choices are the reason the exercise is a benchmark and not a forecast of your requirement.

Related reading: CRR3 Output Floor Phase-In 2026

What the June 2025 Basel III monitoring exercise measured

The monitoring framework tracks three things on a fixed cadence: the risk-based capital ratio, the leverage ratio, and the liquidity metrics. Banks split into Group 1 (Tier 1 capital above EUR 3 billion and internationally active) and Group 2 (everyone else in the sample). The June 2025 run covered 150 banks: 101 Group 1, of which 29 are G-SIBs, and 49 Group 2. Regional coverage is deep for the largest institutions, reaching 100% of the banking sector in some countries, and thinner and more variable for Group 2.

The headline figures worth pinning before anything else:

  • Reference date 30 June 2025; published 24 March 2026; compared against the December 2024 run.
  • CET1 ratio under the current framework: 13.9% for the unbalanced Group 1 sample (stable), 14.1% for the balanced Group 1 sample.
  • Fully phased-in final Basel III Tier 1 leverage ratio: 6.1% for Group 1, 6.0% for G-SIBs, 6.9% for Group 2.
  • Tier 1 MRC impact of full phase-in: +1.7% for Group 1, +0.8% for Group 2.
  • Regulatory capital shortfall under the final framework: EUR 1.0 billion across the sample.
  • Incremental TLAC shortfall: EUR 3.2 billion across 14 reporting G-SIBs.
  • Weighted average LCR: 135.0% (Group 1), 191.6% (Group 2). NSFR: 123.8% (Group 1), 134.2% (Group 2).

Two design features govern how you should read every one of those numbers. The estimates assume full implementation of the final standards as of the reference date, with no grandfathering. And the report does not reflect any Pillar 2 requirements. The exercise measures the Pillar 1 minimum under the finished rulebook, which is one part of the total capital a supervisor expects a bank to hold.

The 1.7% headline, and what sits underneath it

For Group 1 banks the Tier 1 MRC would rise by 1.7% once the final standards are fully phased in. The Committee decomposes that: risk-based requirements add 2.6%, and the leverage ratio requirement falls by 0.9 percentage points, netting to the 1.7%. Within the risk-based increase, the dominant driver is market risk at +1.7%.

The regional spread is wide enough that the aggregate hides the story. Group 1 banks in Europe see a +1.0% impact, the rest of the world sees +3.9%, and the Americas see a negative impact of 1.7%. A single global average of 1.7% describes almost none of the participating banks precisely. If your institution benchmarks itself against the headline rather than against its own regional cohort, the comparison is close to meaningless.

The fall versus December 2024, minus 1.1 percentage points for Group 1 and minus 0.5 for G-SIBs, comes from implementation progress. As banks move onto the final rules in their own jurisdictions, the estimated gap between where they sit today and where the final framework puts them narrows. The report is explicit that June 2025 is the first period in which European Union banks reported under their national final Basel III standards for all risk types except market risk. As more of the rulebook goes live, the measured increment keeps shrinking, because the increment is a distance to a moving target, not a fixed surcharge.

The output floor is where Group 2 banks feel it

The Group 2 picture inverts the Group 1 one. The overall 0.8% impact on Group 2 Tier 1 MRC is built from a 3.4% increase in risk-based measures, driven mainly by the output floor at +3.6%, partly offset by a 2.6 percentage point reduction in leverage ratio MRC. For smaller internationally active banks that rely on internal models, the output floor is the binding change.

The output floor caps the benefit a bank can take from internal models by setting modelled risk-weighted assets at no less than 72.5% of the standardised calculation once fully phased in. Under the Basel timeline it started at 50% on 1 January 2023 and rises annually to 72.5% on 1 January 2028, with an optional national cap on the resulting RWA increase during the transition. Because the monitoring exercise assumes the final state, it prices the 72.5% floor in full, even for banks still sitting at a lower transitional level. The Committee flags this directly: banks that have otherwise fully implemented the final standards can still contribute to the measured MRC change through remaining phase-in arrangements such as the output floor level.

This is where the global benchmark and your live reporting diverge most. The Basel schedule is not the schedule most reporting teams file against. EU banks apply the output floor under CRR3 from 1 January 2025 with its own transitional path, and other jurisdictions have set different dates and different interim floors. The monitoring number tells you the destination; the number in your COREP return tells you where you are on the road to it. For the EU-specific mechanics and the templated pre-cap and post-cap presentation, the CRR3 output floor phase-in has its own operational detail that the global exercise deliberately abstracts away.

Market risk drove the increase for the largest banks

Credit risk still dominates the capital stack. Non-securitisation credit risk covers 70.7% of total Group 1 MRC on average, with corporate exposures alone at 31.2%. Operational risk sits at 15.2%, down from a peak of 18.1% at end-2016 as Great Financial Crisis losses fade out of the loss series. Securitisation has roughly halved as a share of MRC since 2011.

Yet the thing that increased the Group 1 requirement was market risk. The +1.7% market risk contribution is the visible edge of the Fundamental Review of the Trading Book working through the standardised and internal model approaches. For teams standing up FRTB reporting, the monitoring result confirms where the marginal capital is being added at the top of the size distribution, which is a useful cross-check when you reconcile your own trading-book numbers against expectation. Our guide to CRR3 FRTB market risk reporting covers the boundary and template mechanics that sit behind that single percentage point.

A common misreading here is to treat the market risk contribution as evidence that trading books are riskier than they were. The contribution measures the change in required capital from the new methodology applied to June 2025 balance sheets, holding behaviour fixed. The report makes no assumption about how banks might reprice, hedge, or shrink positions in response. It is a methodology delta, not a risk-appetite signal.

The shortfalls are small at aggregate, and that is the trap

Banks in the sample reported a regulatory capital shortfall of EUR 1.0 billion under the final Basel III framework. Against a Group 1 CET1 stock that has grown 152% since June 2011, from EUR 1,203 billion to EUR 3,029 billion, that is a rounding error at the level of the system. The temptation is to conclude the finalisation is fully absorbed and move on.

The shortfall figure sums only the gaps at the specific banks that fall short, netted against nothing, because a bank in surplus cannot offset a bank in deficit. A EUR 1.0 billion aggregate can sit inside one or two institutions while the rest of the sample runs comfortable surpluses. The aggregate tells a supervisor the system is close to fully capitalised against the final rules. It tells an individual bank nothing about its own position, which is exactly why the exercise is published as a benchmark for analysis rather than a scorecard.

The TLAC line carries the same caution. Applying the 2022 minimum TLAC requirements together with the current framework, 14 G-SIBs reporting TLAC data showed an aggregate incremental shortfall of EUR 3.2 billion. That is a resolution-capacity gap concentrated in a subset of the largest banks, and it is distinct from the going-concern capital shortfall. Teams that manage loss-absorbing capacity should read it alongside their own resolution stack, separately from the going-concern number; the EU equivalent framework is set out in our MREL reporting guide.

Leverage ratio: the constraint that softened the blow

The fully phased-in final Basel III Tier 1 leverage ratio came in at 6.1% for Group 1 banks, 6.0% for G-SIBs, and 6.9% for Group 2. On the balanced data set, leverage ratios remain lower in Europe (5.0%) than in the Americas (5.7%) and the rest of the world (7.0%).

The leverage ratio did something counterintuitive to the headline: it reduced measured MRC, by 0.9 percentage points for Group 1 and 2.6 percentage points for Group 2. That happens because the leverage ratio is a non-risk-based backstop. As risk-based requirements rise under the final framework, fewer banks find the leverage ratio to be their binding constraint, so the leverage overlay contributes less incremental capital. The reduction changes which constraint binds; the leverage standard itself is unchanged. For a planning team, the operative question is which of the two is binding for your book at each phase-in step, because that determines whether the next output floor increment actually costs you capital or is absorbed under an already-binding leverage requirement.

Liquidity looks comfortable, and the ratios are still averages

Every bank in the sample reported an LCR above 100% at end-June 2025, so there was no LCR shortfall, against an EUR 18.3 billion shortfall spread over three banks just six months earlier in December 2024. The weighted average LCR was 135.0% for Group 1 and 191.6% for Group 2, and the Group 1 average edged up 0.8% on lower net outflows. Every bank also reported an NSFR above 100%, with weighted averages of 123.8% for Group 1 and 134.2% for Group 2.

The banks in the monitoring sample did not experience the LCR drops that some institutions outside the sample saw during earlier turmoil, which is worth remembering before reading the averages as a statement about the whole banking system. A 135% weighted average summarises survivors of a voluntary sample; it sets no floor and guarantees nothing about any single balance sheet under stress. The LCR, NSFR and ALMM reporting your team files month to month is where the entity-level picture lives, and it is the one a supervisor will hold you to.

Reading the exercise into your capital plan and buffers

This report works best as an external reconciliation point for your own numbers. Three habits keep that honest.

First, benchmark like against like. Compare your fully-loaded final Basel III position against the regional Group 1 or Group 2 cohort that matches your size and model status. The global 1.7% is the wrong comparator. A European Group 1 bank should be reading +1.0%, a rest-of-world bank +3.9%, and a Group 2 model bank the +3.6% output floor line. Using the wrong comparator produces a false sense of either comfort or alarm.

Second, remember what the exercise leaves out. It excludes Pillar 2 and it excludes the combined buffer requirement, so it can never describe your total capital demand. Your binding number stacks Pillar 1, the Pillar 2 requirement, the Pillar 2 guidance, and the combined buffer of the capital conservation buffer, any countercyclical buffer, and the systemic buffers. The monitoring result feeds the Pillar 1 corner of that stack only. Buffer calibration decisions belong in your ICAAP, where the combined buffer requirement stacks on top of the Pillar 1 minimum the monitoring exercise measures. The exercise reaches none of that.

Third, treat the shrinking impact as a planning signal, not a reprieve. The 1.7% is falling because implementation is advancing, so the marginal capital cost of each remaining phase-in step is what you plan against, chiefly the output floor climbing toward 72.5% and the FRTB market risk framework bedding in. The remaining distance closes on your jurisdiction’s timetable, and the number that matters for your 2026 and 2027 plans is the increment left on your own path, verifiable only from your own returns.

Frequently Asked Questions

Is the Basel III monitoring exercise the same data I submit in COREP?

No. The monitoring exercise is a separate, voluntary and confidential submission collected by national supervisors for aggregate analysis, and it assumes full implementation of the final standards with no transitional arrangements. Your COREP return reflects the rules actually in force for you at the reporting date, including any transitional output floor level. The two will not match, and they are not meant to.

Does the +1.7% mean my bank’s Tier 1 requirement rises by 1.7%?

Only by coincidence. The 1.7% is the average change across 101 Group 1 banks under full phase-in. The regional figures range from minus 1.7% in the Americas to +3.9% in the rest of the world, and the Group 2 figure is +0.8% driven by a different component. Your own impact depends on your model status, portfolio mix, and which constraint is binding for you.

Why did the estimated impact fall compared with December 2024?

Because the measured impact is a distance that banks are steadily closing. As they implement the final rules in their own jurisdictions, the gap between their current position and the final framework narrows, so the estimated increment shrinks. The Committee attributes the 1.1 percentage point fall for Group 1 to implementation progress.

Does the report account for the output floor still being in transition?

It prices the fully phased-in 72.5% output floor regardless of where a bank sits on its national transitional path. The Committee notes that banks with otherwise full implementation can still contribute to the measured MRC change because of remaining phase-in arrangements, output floor level included. That is why the monitoring number runs ahead of what most banks currently report.

Is the EUR 3.2 billion TLAC shortfall the same as an MREL shortfall?

No. The TLAC figure applies the 2022 Basel TLAC standard to 14 reporting G-SIBs and measures resolution capacity, not going-concern capital. MREL is the EU implementation of loss-absorbing capacity and is calibrated and reported under its own framework. A bank can meet one and fall short on the other, so they are tracked separately.

The report says liquidity is comfortable. Can I rely on the 135% LCR as a sector floor?

It is a weighted average of a voluntary sample and sets no floor. Every bank in the sample cleared 100% at end-June 2025, but the report itself notes that some banks outside the sample experienced LCR drops during earlier turmoil that sample banks did not. Use it as context; use your own ALMM and LCR returns for the entity-level position.

Where does market risk fit if credit risk is still 70% of the requirement?

Credit risk remains the largest block of MRC, but it was market risk that drove the increase at Group 1 level, contributing +1.7% as FRTB works through. The share of the stack and the driver of the change are two different questions, and the report answers both separately.

Key Takeaways

  • The final Basel III framework adds 1.7% to Group 1 Tier 1 MRC under full phase-in, down 1.1 percentage points from December 2024 as implementation advances.
  • Benchmark against your regional cohort, not the global average: Europe +1.0%, Americas minus 1.7%, rest of world +3.9%, Group 2 +0.8%.
  • For Group 2 model banks the binding change is the output floor at +3.6%; for Group 1 it is market risk at +1.7%.
  • The exercise assumes the fully phased-in 72.5% output floor and excludes all transitional arrangements, so it runs ahead of your live COREP position.
  • It also excludes Pillar 2 and buffers, so it measures the Pillar 1 minimum only, never your total capital requirement.
  • Aggregate shortfalls are small (EUR 1.0 billion capital, EUR 3.2 billion incremental TLAC across 14 G-SIBs) but concentrate in specific banks and cannot be read as an all-clear.
  • Liquidity cleared across the board: all banks above 100% LCR and NSFR, weighted average LCR 135.0% for Group 1, versus an EUR 18.3 billion LCR shortfall six months earlier.
  • Plan against the increment left on your own phase-in path: the output floor rising to 72.5% by 1 January 2028 under the Basel timeline, on your jurisdiction’s actual dates.

Sources and References

What to reconcile before the next monitoring run

The next semiannual exercise will carry the December 2025 reference date, and the measured impact will almost certainly fall again as more of the final framework goes live, particularly across EU banks now reporting under national final standards for all risk types except market risk. Before then, the productive work is reconciliation: line up your fully-loaded final Basel III position against the regional cohort that matches your bank, isolate the output floor and FRTB increments still ahead of you on your own timetable, and confirm that your Pillar 2 and buffer assumptions carry the load the monitoring exercise never counts. The report gives you the destination for the system; your returns give you the distance you still have to cover.

Disclaimer: The information on RegReportingDesk.com is for educational and informational purposes only. It does not constitute legal, regulatory, tax, or compliance advice. Always consult your compliance officer, legal counsel, or the relevant supervisory authority for guidance specific to your institution.

Similar Posts