CRR3 revises the EU’s standardized approach to credit risk by introducing more granular, risk-sensitive risk weights across most major exposure classes, replacing the blunt categorizations that characterized the older framework. For banks using internal models, it also introduces an output floor set at 72.5% of the standardized approach RWA, which directly limits how much capital relief internal models can generate. The sections below walk through the specific changes, the exposure classes most affected, the impact on internal model users, what banks need to do now, and how CRR3 relates to the broader Basel IV framework.
What specific changes does CRR3 make to the standardized approach?
CRR3 overhauls the standardized approach to credit risk by replacing broad, flat risk weight categories with a more granular, risk-sensitive structure. The key change is that risk weights now depend more heavily on the specific characteristics of the borrower, the loan type, and, in many cases, the external credit rating, rather than applying a single percentage to an entire exposure class. Banks using the standardized approach can no longer rely on simplified bucketing.
The most significant structural changes include:
- Revised risk weights for retail exposures, splitting the category into regulatory retail, other retail, and transactors versus revolvers for credit cards
- New treatment for unrated corporate exposures, with a general 100% risk weight but specific carve-outs for investment-grade corporates
- Revised real estate exposure framework, moving away from a single residential property risk weight toward loan-to-value (LTV)-based bucketing
- New exposure class for equity, with differentiated risk weights depending on whether the holding is speculative, publicly traded, or held for strategic purposes
- Revised treatment of off-balance sheet items, with updated credit conversion factors (CCFs) replacing the older, more generous estimates
- Stronger due diligence requirements for external ratings, meaning banks cannot mechanically apply agency ratings without internal validation
The overall direction is toward more capital sensitivity. In practice, this means the same portfolio that generated a certain RWA figure under the old standardized approach may produce a noticeably different number under CRR3, depending on how the underlying exposures break down across the new categories.
Which exposure classes see the biggest risk weight increases under CRR3?
The exposure classes that see the most material risk weight increases under CRR3 are real estate, certain retail sub-categories, equity holdings, and off-balance sheet commitments. These are areas where the old standardized approach applied relatively flat or generous weights that did not fully reflect the risk profile of individual loans or positions.
Real estate exposures
Real estate is one of the most heavily revised areas. Under CRR3, residential mortgage risk weights are now tied directly to LTV ratios. Low-LTV loans benefit from relatively favorable weights, but as the LTV rises, the applicable risk weight increases significantly. Commercial real estate receives similar treatment, with income-producing real estate now distinguished from owner-occupied property and carrying higher weights in many cases. For banks with large property-secured loan books, this recalibration can have a meaningful impact on total RWA.
Equity and off-balance sheet exposures
Equity holdings face substantially higher risk weights under CRR3. Speculative unlisted equity can attract a risk weight of 400%, while other equity holdings not qualifying for preferential treatment are typically weighted at 250%. Off-balance sheet commitments also see increases, as the revised CCFs reduce the discount applied to undrawn facilities and contingent exposures. Banks that previously benefited from low CCFs on revolving credit lines or liquidity facilities will need to revisit those calculations.
How does CRR3 affect RWA calculations for banks using internal models?
For banks using internal ratings-based (IRB) approaches, CRR3 introduces an output floor that sets a minimum RWA equal to 72.5% of what the standardized approach would produce for the same portfolio. This means that even if a bank’s internal models generate lower capital requirements, the floor prevents the total RWA from falling below that threshold. The output floor is the single most consequential change for advanced banks.
Beyond the output floor, CRR3 also restricts the scope of IRB application. For certain exposure classes, including exposures to large corporates, banks, and financial institutions above defined size thresholds, the advanced IRB approach is no longer permitted. Banks in these categories must use either the foundation IRB or the standardized approach, which reduces the degree of model-driven capital optimization available.
The practical effect is that many banks with sophisticated internal models will see their effective RWA rise, even if their underlying credit risk has not changed. The output floor essentially anchors capital requirements to the standardized approach as a baseline, which is precisely why the revisions to that standardized approach matter so much. A higher standardized approach RWA means a higher floor, and therefore a higher minimum capital requirement.
Banks need to run parallel calculations under both the internal model outputs and the revised standardized approach to understand where the floor binds and by how much. This is not a one-time exercise. It needs to be embedded into ongoing capital planning and stress-testing processes.
What do banks need to do to be ready for CRR3 by 2026?
Banks need to complete several interconnected workstreams to be operationally ready for CRR3 in 2026. The regulation is live, and institutions that have not yet embedded the new rules into their risk infrastructure, data processes, and capital reporting are running out of time to close the gap.
The most important areas to address are:
- Data granularity: The revised standardized approach requires more detailed data at the exposure level. LTV ratios, borrower classification, rating status, and off-balance sheet commitment types all feed directly into risk weight determination. If your data infrastructure cannot reliably produce these inputs at scale, your RWA calculations will be inaccurate.
- Parallel RWA calculations: Banks using IRB must run standardized approach calculations alongside internal model outputs to monitor the output floor. This requires systems capable of running both methodologies simultaneously across the full portfolio.
- Capital impact assessment: Before the rules fully bite, you need a clear picture of how CRR3 changes your capital ratios relative to current levels. This means running the new risk weights across your actual portfolio and stress-testing the results under different scenarios.
- Model and configuration updates: Any system that calculates credit risk RWA needs to reflect the new exposure class definitions, LTV buckets, CCFs, and rating-based weights. Hardcoded legacy configurations are a real operational risk here.
- Regulatory reporting alignment: Output needs to flow correctly into your regulatory reporting submissions. Regulatory calculations and reporting work best when treated as separate disciplines, with calculation engines producing accurate, traceable outputs that feed into your preferred reporting framework.
The institutions that are best placed are those that invested early in flexible, configurable risk infrastructure rather than fixed legacy systems. If your calculation environment requires IT development every time a regulatory parameter changes, CRR3 will be painful. If your team can update risk weight tables and exposure classifications through a user interface, the transition is far more manageable.
How does CRR3’s standardized approach compare to the Basel IV framework?
CRR3 is the European Union’s legislative implementation of Basel IV. The standardized approach revisions in CRR3 directly transpose the Basel Committee’s updated credit risk standards into EU law, so the two frameworks are closely aligned in structure and intent. The key difference is that CRR3 includes EU-specific adjustments, transitional provisions, and, in some areas, more granular or modified treatments that reflect the European banking context.
Basel IV, finalized by the Basel Committee on Banking Supervision in 2017, set out the global standard for revised risk weights, the output floor, and the restrictions on IRB scope. CRR3 takes those global standards and adapts them for EU member states, adding elements such as the SME supporting factor and certain real estate transitional arrangements that are specific to the European regulatory environment.
For practical purposes, if you understand the Basel IV standardized approach, you understand the core logic of CRR3. The exposure class structure, the LTV-based real estate framework, the revised CCFs, and the output floor all trace directly back to the Basel IV text. Where CRR3 diverges, it typically does so to accommodate features of the European banking system or to phase in requirements more gradually.
Banks operating across multiple jurisdictions need to track both. Basel IV sets the global floor. CRR3 defines what EU-regulated institutions must actually comply with. Other jurisdictions have their own implementing legislation, and the timelines and specific calibrations vary. For EU banks, CRR3 is the operative document, but understanding the Basel IV logic behind it helps you anticipate how regulators will interpret ambiguous areas and how future revisions are likely to develop.
At ElysianNxt, our credit risk solution is built to handle the full revised standardized approach, IRB calculations, output floor monitoring, and the exposure class granularity that CRR3 demands, so your team can focus on the analysis rather than the configuration.
Frequently Asked Questions
What is the timeline for CRR3 implementation, and are there any transitional arrangements banks should be aware of?
CRR3 officially applies from January 1, 2025, with the output floor being phased in gradually through to January 1, 2030, starting at 50% and stepping up to the full 72.5% over that period. This phased approach gives banks time to adjust their capital planning, but it does not delay the need to build the underlying data and systems infrastructure — those need to be in place well before the floor reaches its final level. Certain EU-specific transitional provisions, such as those for real estate exposures and SME lending, also apply and should be reviewed carefully against your specific portfolio composition.
How does the SME supporting factor under CRR3 interact with the revised standardized approach risk weights?
The SME supporting factor is an EU-specific adjustment that reduces the effective risk weight on qualifying SME exposures by applying a scaling factor of 0.7619, partially offsetting the capital increases that the revised standardized approach would otherwise generate for this segment. It applies to both the standardized approach and IRB, and it is one of the key ways CRR3 diverges from the pure Basel IV text. Banks with significant SME lending books should model the supporting factor explicitly in their RWA impact assessments, as it can materially soften the headline capital increase for that exposure class.
What are the most common mistakes banks make when calculating LTV-based risk weights for real estate under CRR3?
The most frequent errors involve using stale or imprecise property valuations, misclassifying exposures between income-producing and owner-occupied real estate, and failing to update LTV ratios as property values change over time. CRR3 requires that valuations meet specific standards and that LTV be calculated at origination and monitored on an ongoing basis, meaning a static snapshot approach is not sufficient. Banks should also ensure their systems correctly apply the whole-loan approach versus the loan-splitting approach for LTV bucketing, as the choice of method can produce meaningfully different risk weight outcomes for the same loan.
If our bank currently uses the advanced IRB approach for large corporate exposures, what specifically changes under CRR3?
Under CRR3, the advanced IRB (A-IRB) approach is no longer permitted for exposures to large corporates with consolidated revenues above €500 million, as well as for exposures to institutions and other financial entities. Banks in this position must migrate those portfolios to either the foundation IRB (F-IRB) or the standardized approach, which removes the ability to use internally estimated loss-given-default (LGD) and credit conversion factors for those segments. This is a significant operational change that requires model decommissioning, data remapping, and regulatory approval processes, so banks affected by this restriction should have already initiated the transition planning.
How should banks approach stress testing under CRR3, particularly given the output floor interaction?
Stress testing under CRR3 needs to account for the dynamic interaction between internal model outputs and the output floor, because the floor can bind differently under stress depending on how standardized approach RWA moves relative to IRB RWA. In some stress scenarios, model-estimated risk increases may actually be partially absorbed by the floor if the standardized approach rises proportionally, while in others the floor may become less binding — understanding which regime applies under each scenario is critical for accurate capital adequacy projections. Banks should build stress testing frameworks that run both methodologies in parallel under each scenario and explicitly track where and by how much the floor is binding, integrating this into ICAAP and recovery planning documentation.
Does CRR3 change how banks need to treat unrated exposures, and how should we handle counterparties that lack external credit ratings?
Yes, CRR3 introduces a specific treatment for unrated corporate exposures, applying a general risk weight of 100%, but with an important carve-out for investment-grade corporates that meet defined qualitative criteria — these can receive a 65% risk weight even without an external rating. The key challenge is that banks must apply rigorous internal due diligence to substantiate the investment-grade classification rather than relying on the absence of a rating as a default to the 100% bucket. This means your credit assessment processes and documentation standards need to be robust enough to support a regulatory challenge, and your systems need to capture and store the qualifying criteria at the individual exposure level.
What should banks prioritize first if they are behind on CRR3 readiness with limited time and resources?
If resources are constrained, the highest-priority action is completing a portfolio-level capital impact assessment using the revised standardized approach risk weights, as this quantifies the RWA gap and informs every downstream decision about capital planning, business strategy, and system investment. Alongside this, data quality remediation for the inputs that drive the largest RWA changes — particularly LTV data for real estate and exposure classification for retail and corporate segments — should be treated as urgent, since inaccurate inputs produce unreliable capital numbers regardless of how well-configured your calculation engine is. Once the impact is quantified and data gaps are identified, banks can sequence system and model updates more efficiently based on where the capital materiality is highest.
Related Articles
- What is T0 reconciliation and why does it matter between IFRS 9 and Basel IV?
- What is the difference between regulatory and financial reporting?
- What is a regulatory reporting connector and how does it work?
- What are the benefits of automated regulatory reporting?
- What are conditional scenarios in integrated stress testing?
This content was generated with the help of AI and it may contain mistakes
