The 72.5% output floor costs your bank the difference between your internal model RWAs and 72.5% of the standardized approach RWAs whenever your models produce a lower figure. In practical terms, if your internal ratings-based models calculate RWAs at, say, 50% of what the standardized approach would produce, the floor forces you to hold capital as if your RWAs were at least 72.5% of the standardized figure. For many banks with sophisticated IRB models, this means carrying meaningfully more capital than their models alone would require. The sections below break down exactly how that calculation works, which portfolios feel it most, and what your capital planning process needs to account for.

How is the 72.5% output floor actually calculated?

The output floor calculation compares two numbers: your bank’s total RWAs under internal models and 72.5% of the total RWAs calculated using the Basel IV standardized approach. If your internal model RWAs fall below that 72.5% threshold, your reported RWAs are floored at the higher standardized-based figure. The capital requirement then applies to whichever number is larger.

Here is how to work through it step by step:

1. Calculate your total RWAs using all approved internal models (IRB for credit risk, IMA for market risk, and so on).
2. Calculate your total RWAs using only the standardized approaches across all risk types.
3. Multiply the standardized RWA total by 72.5%.
4. Compare the two figures. If your internal model RWAs are lower than the 72.5% standardized figure, the floor applies and you use the higher number.
5. Apply your minimum capital ratio (for example, 8% Tier 1 capital) to the floored RWA figure.

One important detail: the floor applies at the consolidated group level, not instrument by instrument. This means a bank with strong internal models across most of its portfolio may still be pulled up by the floor if a few concentrated exposures drive a large gap between internal and standardized results. The aggregation point matters enormously when you are trying to understand where the real capital drag comes from.

It is also worth noting that the floor is being phased in gradually. Most jurisdictions are implementing it in steps between 2025 and 2030, which gives banks a transitional window. But the direction of travel is clear, and planning ahead of the full implementation date is far more cost-effective than reacting at the last moment.

Which loan portfolios are most exposed to the output floor?

The portfolios most exposed to the output floor are those where internal models produce RWAs significantly below what the standardized approach would assign. In practice, this typically means low-default portfolios with strong historical performance, including high-quality corporate lending, specialized lending, and certain retail mortgage books where banks have built highly optimized IRB models over many years.

The gap between internal and standardized RWAs tends to be widest in these areas:

• Corporate and institutional lending: Banks with long track records of low default rates on large corporate exposures often have IRB models that assign very low probability of default. The standardized approach uses fixed risk weights that do not fully reflect this history, creating a wide gap.
• Residential mortgages: In markets where house prices have been stable and default rates low, advanced IRB models can produce RWAs well below the standardized floor weights, particularly for low loan-to-value exposures.
• Specialized lending: Project finance, infrastructure, and real estate finance often carry standardized risk weights that are materially higher than what sophisticated internal models assign based on deal-specific cash flow analysis.
• Sovereign and financial institution exposures: Where banks have built granular internal models for counterparty credit risk, the standardized approach can produce notably higher RWAs.

The exposure is not just about the size of the portfolio. It is about the ratio between internal and standardized RWAs within that portfolio. A relatively small specialized lending book with a large model-to-standardized gap can contribute disproportionately to the floor impact at the group level.

What is the difference between the output floor and the capital floor?

The output floor and the capital floor are related concepts but operate at different levels. The output floor is a specific Basel IV mechanism that limits how much benefit a bank can take from internal models by setting a minimum RWA level at 72.5% of the standardized approach. The capital floor, in broader usage, refers to any regulatory minimum on the capital a bank must hold, including leverage ratio requirements and Pillar 2 add-ons.

Think of it this way: the output floor is a constraint on the inputs to the capital calculation, while capital floors more broadly are constraints on the output. The output floor works by adjusting the RWA denominator before you even apply your capital ratio. A capital floor, by contrast, might set a minimum absolute capital level regardless of how RWAs are calculated.

This distinction matters practically because the output floor can affect banks that are otherwise well capitalized under their own models. A bank might comfortably meet its Pillar 1 and Pillar 2 capital requirements under its internal model RWAs, yet still face a meaningful capital shortfall once the output floor pushes its RWA base higher. The output floor is therefore a constraint on model benefit, not just a safety net for undercapitalized banks.

Under Basel IV, the output floor replaces the earlier Basel I floor that many jurisdictions had maintained as a transitional measure. The new floor is more precisely calibrated and more consistently applied across jurisdictions, which is one reason its impact on individual bank capital positions can differ significantly from the old transitional floors some banks were already managing.

How does the output floor affect banks using internal ratings-based models?

For banks using internal ratings-based models, the output floor directly limits the capital relief those models can generate. IRB models are valuable precisely because they allow banks to assign risk weights based on their own data and experience, often resulting in lower RWAs for high-quality portfolios. The output floor caps that benefit at 72.5% of the standardized approach, meaning no matter how good your models are, you cannot fall below that threshold at the aggregate level.

The practical effects for IRB banks include:

• Higher capital requirements: Banks whose IRB RWAs sit well below 72.5% of standardized RWAs must hold more capital than their models alone would suggest.
• Reduced return on equity for low-risk portfolios: The floor effectively increases the capital cost of holding exposures where internal models assign very low risk weights, which can change the economics of certain lending activities.
• Complexity in capital allocation: Internal capital allocation models need to account for the floor at the group level, which makes it harder to simply allocate capital based on business unit IRB RWAs without considering the aggregate floor impact.
• Potential business mix changes: Some banks may find it economically rational to shift toward higher-risk, higher-return assets if the floor already forces them to hold capital equivalent to a standardized risk weight, regardless of actual credit quality.

Banks that have invested heavily in building sophisticated IRB models over many years face a particular strategic question: how much of that model investment still generates capital efficiency once the floor is fully phased in? The answer depends on the specific portfolio mix, but for many advanced IRB banks, the floor will reduce, though not eliminate, the capital advantage their models provide.

There is also a data and reporting dimension. Running parallel calculations under both IRB and standardized approaches, comparing them at the group level, and monitoring the floor impact on an ongoing basis requires robust infrastructure. Banks that rely on manual or batch-driven processes for these calculations will find it difficult to get timely visibility into their floor exposure, particularly as portfolios change and market conditions shift.

How should banks prepare their capital planning for the output floor?

Banks should prepare for the output floor by building parallel RWA calculations into their regular capital planning cycle, stress testing their capital position under the floored RWA basis, and identifying which portfolios contribute most to the gap between internal model and standardized RWAs. Capital plans that rely solely on IRB-based projections without accounting for the floor will overstate available capital headroom.

A structured preparation approach covers four areas:

Quantify your current and projected floor impact

Start by calculating the gap between your current internal model RWAs and 72.5% of your standardized RWAs, at the group level and broken down by portfolio. This gives you a baseline floor impact figure. Then project how that gap changes under different scenarios: portfolio growth, changes in credit quality, new business origination, and the phased implementation timeline. Knowing where you stand today and where you are heading is the foundation for everything else.

Integrate the floor into stress testing and scenario analysis

Stress tests that only run on IRB-based RWAs will give you an incomplete picture. Your stress testing framework needs to apply shocks to both the internal model RWAs and the standardized RWAs simultaneously, then recalculate the floor position under each scenario. This is particularly relevant for credit risk stress tests, where a deterioration in portfolio quality might actually narrow the gap between IRB and standardized RWAs in some portfolios while widening it in others.

Review your portfolio strategy in light of floor economics

Once you have a clear view of which portfolios drive your floor impact, you can have an informed conversation about whether your current portfolio mix still makes economic sense. This does not necessarily mean abandoning low-risk lending. It does mean pricing that lending to reflect the actual capital cost under the floored RWA basis and being realistic about return on equity projections for business lines where the floor significantly increases the effective capital charge.

Invest in the infrastructure to run calculations in real time

The output floor is not a one-time calculation. It changes as your portfolio changes, as market conditions shift, and as the phased implementation schedule progresses. Banks that can only see their floor impact through overnight batch reports will always be reacting rather than planning. Real-time or near-real-time visibility into both IRB and standardized RWAs, and the resulting floor position, gives treasury and capital management teams the information they need to act proactively.

This is exactly where modern Basel IV platforms make a measurable difference. Our credit risk module within the .NXT platform covers the full output floor calculation alongside both revised standardized and IRB approaches, giving you a live view of your floor position without waiting for the next batch run. If you are building out your capital planning process for Basel IV, getting the calculation infrastructure right early is one of the most valuable investments you can make.

Related Articles

• What is SREP and how does it affect a bank's capital planning?
• What is NSFR and why do regulators care about it?
• What is the role of expert judgment in scenario design?
• How does COVID-19 impact modern stress testing approaches?
• How does credit risk stress testing differ from market risk stress testing?

This content was generated with the help of AI and it may contain mistakes