The Standardized Approach for credit risk under Basel IV (implemented in the EU as CRR3) is a regulatory method that assigns fixed, supervisory risk weights to different exposure categories to calculate Risk-Weighted Assets (RWA) for minimum capital requirements. Unlike the Internal Ratings-Based (IRB) approach, it does not rely on a bank’s own models. Basel IV significantly revised the original Standardized Approach, making it more risk-sensitive, tightening the use of external credit ratings, and introducing new rules for real estate, infrastructure, and other exposure classes. The sections below walk through the most important changes, how risk weights work in practice, and what the rules mean for your capital calculations today.

What changed in the Standardized Approach under Basel IV / CRR3?

Basel IV overhauled the Standardized Approach for credit risk to make it more granular and risk-sensitive than its predecessor. The revised framework, known as the Revised Standardized Approach (RSA), introduced new exposure classes, updated risk weight tables, tightened the conditions under which external credit ratings can be used, and added stricter due diligence requirements for banks that do rely on those ratings. Some of the most notable changes include:

• New and refined exposure classes: The revised framework expanded the number of exposure categories, adding dedicated classes for subordinated debt, equity, and infrastructure project finance, among others.
• More granular risk weights for retail and corporate exposures: Risk weights are now more differentiated based on the nature of the counterparty and the specific characteristics of the exposure.
• Revised real estate rules: The treatment of residential and commercial real estate was fundamentally restructured, moving away from a single flat risk weight toward a loan-to-value (LTV) based approach.
• Reduced reliance on external ratings: Jurisdictions that allow the use of external credit ratings must apply stricter due diligence, and floors were introduced to prevent risk weights from falling too low even for highly rated counterparties.
• Revised Credit Risk Mitigation (CRM) framework: The conditions for recognizing collateral, guarantees, and credit derivatives were updated and, in some cases, tightened.

In the EU, these changes were transposed through CRR3 (Capital Requirements Regulation 3), which entered into force in 2025 and applies from January 2025. The revisions directly affect how banks calculate minimum capital requirements and interact closely with the output floor introduced alongside them.

How are risk weights assigned under the Standardized Approach?

Under the Standardized Approach, risk weights are assigned based on the exposure class of the counterparty and, where permitted, the external credit rating of that counterparty. Each exposure is bucketed into a predefined category, and a supervisory risk weight percentage is applied to the outstanding exposure amount. The resulting figure is the Risk-Weighted Asset (RWA) that feeds into the capital ratio calculation. The main exposure classes and their general risk weight logic work as follows:

• Sovereigns and central banks: Risk weights range from 0% for highly rated sovereigns (AAA to AA-) to 150% for those rated below B-. Unrated sovereigns receive 100%.
• Institutions (banks and investment firms): Risk weights are tied to the credit quality of the institution, typically ranging from 20% to 150%, with specific rules depending on whether the jurisdiction permits the use of external ratings.
• Corporates: Rated corporates receive risk weights between 20% and 150% depending on credit quality. Unrated corporates typically attract 100%, though a dedicated 65% risk weight applies to “investment grade” corporates meeting specific criteria.
• Retail exposures: A 75% risk weight applies to qualifying retail exposures that meet defined criteria including granularity and low individual exposure limits.
• Defaulted exposures: These generally attract 100% or 150% depending on the level of specific credit risk adjustments applied.
• Equity exposures: A new 250% risk weight applies to most equity holdings, with 400% for speculative unlisted equities.

The key principle is that risk weights are not negotiated or modeled internally. They are set by the regulator and applied uniformly, which makes the Standardized Approach more straightforward to implement but less sensitive to a bank’s actual portfolio risk profile compared to the IRB approach.

What are the new real estate exposure rules under CRR3?

CRR3 replaced the old flat risk weight for real estate with a loan-to-value (LTV) based framework. This means the risk weight applied to a residential or commercial real estate exposure now depends directly on the ratio of the loan amount to the value of the property securing it. The lower the LTV, the lower the risk weight.

Residential real estate

For residential real estate, CRR3 distinguishes between exposures where repayment depends materially on the cash flows generated by the property (such as buy-to-let) and those where it does not (such as owner-occupied mortgages). For standard owner-occupied mortgages, risk weights under the LTV-based approach are applied on a graduated scale: 20% for LTV up to 50%, 25% for LTV above 50% up to 60%, 30% for LTV above 60% up to 80%, 40% for LTV above 80% up to 90%, 50% for LTV above 90% up to 100%, and 70% for LTV above 100%, per the BCBS Basel III finalisation framework (BCBS CRE20). For income-producing residential real estate, risk weights are higher across the LTV spectrum; the applicable risk weights at higher LTV bands should be verified against CRR3 Articles 124 and 126 for EU-specific calibration.

Commercial real estate

Commercial real estate follows a similar LTV-based structure, but with higher risk weights reflecting the greater volatility of commercial property values. The general commercial real estate risk weight sits at 60% up to 60% LTV, rising to 80% for LTV above 60%. For income-producing commercial real estate, the framework is more conservative still, with a 70% weight up to 60% LTV and 110% above that threshold. An important practical point: property valuations used to calculate LTV must meet specific independence and frequency requirements under CRR3. Banks cannot simply use origination-date valuations indefinitely. This adds a data management dimension to real estate credit risk calculations that was less prominent under the previous rules.

How does the output floor affect SA-CR calculations?

The output floor is one of the most significant structural changes introduced by Basel IV. It sets a minimum level for a bank’s total RWA calculated using internal models, expressed as a percentage of the RWA that would result from applying the Standardized Approach across the entire portfolio. The floor is set at 72.5%, meaning a bank using the IRB approach cannot report total RWA below 72.5% of what the Standardized Approach would produce. This directly elevates the importance of SA-CR calculations, even for banks that primarily use IRB models. If a bank’s IRB-based RWA falls below the floor, the bank must top up its RWA to the floor level and hold additional capital accordingly. In practice, this means every IRB bank now needs accurate, well-maintained Standardized Approach calculations running in parallel with its internal models. The output floor is being phased in gradually. In the EU under CRR3, the floor started at 50% in January 2025 and increases in annual steps, reaching the final 72.5% level by January 2030. Banks that have invested heavily in sophisticated IRB models may find that the floor constrains the capital benefit those models previously delivered, particularly for low-risk portfolios such as prime mortgages or highly rated corporate lending. From a calculation standpoint, the output floor does not change how individual SA-CR risk weights are assigned. It operates at the aggregate RWA level, comparing total IRB RWA against total SA RWA. But it does mean that errors or inefficiencies in SA-CR calculations can have direct capital consequences, even for banks that do not primarily use the Standardized Approach.

What’s the difference between the Standardized Approach and the IRB Approach for credit risk?

The Standardized Approach uses fixed, regulator-defined risk weights applied uniformly to exposure classes. The Internal Ratings-Based (IRB) Approach allows banks to use their own internal models to estimate key risk parameters, producing risk weights that reflect the bank’s actual assessment of counterparty credit quality. The IRB approach generally produces lower RWA for high-quality portfolios, but it requires regulatory approval, substantial model infrastructure, and ongoing validation. The two approaches differ across several practical dimensions:

• Risk sensitivity: IRB risk weights vary continuously based on modeled Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD). SA risk weights are fixed buckets determined by external ratings or exposure class.
• Implementation complexity: The Standardized Approach is faster to implement and requires less internal modeling infrastructure. IRB requires approved models, data history, and ongoing backtesting.
• Regulatory approval: IRB requires formal supervisory approval before use. The Standardized Approach does not.
• Capital outcomes: IRB can produce meaningfully lower capital requirements for well-diversified, high-quality portfolios. But the output floor limits how much lower those requirements can go relative to the SA baseline.
• Scope restrictions under Basel IV: Basel IV restricted the use of the Advanced IRB (A-IRB) approach for certain exposure classes, including large corporates and financial institutions, where banks must now use the Foundation IRB (F-IRB) or the Standardized Approach instead.

Many banks use both approaches simultaneously, applying IRB where they have strong model coverage and the SA for portfolios where IRB approval has not been sought or is no longer permitted.

When did Basel IV / CRR3 Standardized Approach rules come into effect?

The Basel IV Standardized Approach for credit risk came into effect in the EU on January 1, 2025, through CRR3. This means that as of 2026, banks in the EU are already operating under the revised SA-CR framework and the initial phase of the output floor. The full output floor of 72.5% phases in by January 2030. Outside the EU, implementation timelines vary by jurisdiction. The UK published its own Basel 3.1 rules with a planned implementation date of 1 January 2027, per PRA PS1/26. Other major jurisdictions, including the US, have faced delays in finalizing their Basel IV-equivalent rules, and the US implementation timeline remains subject to regulatory and political developments as of 2026. For banks operating across multiple jurisdictions, this uneven rollout creates a practical challenge: SA-CR calculations need to reflect jurisdiction-specific rules, not just the Basel Committee’s original text. The differences between CRR3, the UK’s Basel 3.1 rules, and other national implementations can affect risk weights, eligible collateral, and CRM recognition in ways that require careful configuration at the calculation level. If you want to understand how your institution can manage the full scope of Basel IV credit risk requirements, including the Revised Standardized Approach, IRB calculations, output floor monitoring, and ICAAP stress testing, our Basel IV credit risk solution is built to handle exactly that. At ElysianNxt, we support the complete SA-CR and IRB framework within a single platform, so your teams can run both approaches in parallel, monitor output floor impacts in real time, and move from calculation to reporting without switching systems.

Related Articles

• What is an IRB model and how does Basel IV change the rules around it?
• How long does regulatory reporting typically take?
• How scalable are cloud-based regulatory reporting platforms for growing banks?
• How does technology improve integrated risk and stress testing processes?
• How does interest rate risk affect credit risk?

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