Basel IV is not a single rule change. It is a comprehensive overhaul of how banks calculate, report, and manage capital requirements. If you are responsible for risk-weighted assets (RWA), capital planning, or Basel IV compliance at your institution, understanding exactly which changes affect your numbers is the most practical place to start. Here are the seven Basel IV changes most likely to move your RWA figures, and what each one means for your day-to-day risk management.

How Basel IV reshapes capital requirements for banks

Basel IV, which is being implemented across jurisdictions through CRR3 in Europe and equivalent frameworks in Asia Pacific, represents the most significant revision to the Basel framework since Basel III. The core objective is straightforward: reduce excessive variability in RWA calculations across institutions and make capital requirements more comparable and credible. In practice, this means tighter constraints on internal models, revised standardised approaches with higher granularity, and new floors that limit how far internal models can reduce your capital requirements relative to standardised methods.

For many banks, the combined effect of these changes will increase overall RWA. The scale of that increase depends heavily on your current portfolio composition, how much you rely on internal models, and how well your data infrastructure supports the new calculation requirements. The seven changes below are the ones most likely to have a direct, measurable impact on your numbers.

1: The output floor limits internal model advantages

The output floor is arguably the single most consequential change in the entire Basel IV package. It sets a minimum level of RWA calculated using internal models at 72.5% of the RWA that would result from applying the standardised approaches. In other words, no matter how sophisticated your internal models are, your total RWA cannot fall below 72.5% of the standardised floor.

For banks that have historically relied on advanced internal ratings-based (IRB) or internal market risk models to achieve lower capital requirements, this floor directly limits that advantage. The practical implication is that you now need to run both internal model calculations and full standardised calculations in parallel, compare them, and apply the floor where it binds. This doubles the calculation workload and increases the importance of having a platform that handles both approaches within a single environment.

The output floor is phased in, but by 2030 it will be fully effective. Banks that start modelling its impact now will be better positioned to manage capital allocation decisions before the floor becomes binding.

2: Revised standardised approach for credit risk

The revised standardised approach (RSA) for credit risk introduces more risk-sensitive treatment across a wider range of exposure classes. Where the previous standardised approach applied relatively flat risk weights to broad categories, the revised version differentiates more granularly based on factors such as loan-to-value ratios for real estate, external ratings where available, and specific exposure characteristics for retail and corporate portfolios.

For banks using the standardised approach as their primary method, this means recalibrating risk weights across the entire credit portfolio. For IRB banks, it matters because the output floor requires a credible parallel standardised calculation. Either way, the revised approach demands more detailed data at the contract level, including collateral values, borrower characteristics, and exposure classifications that may not have been systematically captured under the old framework.

The revised standardised approach also introduces a more structured treatment for off-balance-sheet items and unrated corporates, which can affect RWA for banks with significant trade finance or SME lending portfolios.

3: Tighter constraints on IRB models

Basel IV places significant restrictions on which exposure classes can use the advanced IRB (A-IRB) approach and introduces input floors that limit how low your probability of default (PD), loss given default (LGD), and exposure at default (EAD) estimates can go, regardless of what your models produce.

Specifically, large corporate, bank, and financial institution exposures are restricted to the foundation IRB (F-IRB) approach, removing the ability to apply internal LGD and EAD estimates for these segments. Equity exposures are excluded from IRB entirely under the revised framework. For retail and SME portfolios where A-IRB is still permitted, input floors apply: a minimum PD of 0.05% for most exposures, minimum LGD values depending on whether exposures are secured or unsecured, and minimum EAD conversion factors for off-balance-sheet items.

These floors can materially increase RWA for portfolios where your current model estimates fall below the regulatory minimums. Banks with strong historical performance data that supported very low PD estimates in certain segments may see the largest increases here.

4: New credit valuation adjustment framework

The revised credit valuation adjustment (CVA) framework changes how banks calculate capital requirements for the risk of mark-to-market losses on derivative counterparties. The basic approach from Basel III has been significantly revised, and the new framework introduces a standardised approach based on sensitivities that is more risk-sensitive but also more complex to implement.

Under the revised framework, the basic approach is only available to banks with limited derivatives activity. Most institutions will need to implement the standardised CVA approach, which requires calculating sensitivities to credit spreads and interest rates for each counterparty and netting set. This increases both the data requirements and the computational load compared to the previous method.

For banks with significant OTC derivatives books, CVA capital requirements under Basel IV can increase meaningfully. The interaction between CVA capital and counterparty credit risk (CCR) calculations also needs careful management, since both frameworks apply to the same underlying exposures.

5: Fundamental review of the trading book

The Fundamental Review of the Trading Book (FRTB) redraws the boundary between the banking book and the trading book, introduces stricter criteria for internal model approval, and establishes the sensitivity-based method (SBM) as the new standardised approach for market risk capital.

The clearer boundary between banking and trading books reduces the scope for regulatory arbitrage but also means some instruments previously classified in one book may need to move to the other, with capital implications in either direction. The internal models approach under FRTB requires desk-level model approval, which is a significantly higher bar than the portfolio-level approval under the previous framework. Banks that lose model approval for specific desks must apply the standardised approach for those desks, potentially at higher capital cost.

The SBM requires calculating delta, vega, and curvature sensitivities across a defined set of risk factors, which demands robust sensitivity calculation infrastructure. For many banks, FRTB represents one of the most technically demanding implementation challenges within the Basel IV package.

6: Revised operational risk standardised approach

Basel IV replaces the previous menu of operational risk approaches, including the basic indicator approach, the standardised approach, and the advanced measurement approaches, with a single standardised approach. This new approach calculates operational risk RWA based on a business indicator component, which reflects the size and nature of a bank’s business activities, adjusted by an internal loss multiplier for larger institutions.

The removal of the advanced measurement approaches (AMA) is significant for banks that had invested in sophisticated internal operational risk models. Under Basel IV, those models no longer reduce your regulatory capital requirement directly. The internal loss multiplier does allow historical loss experience to influence capital for larger banks, but the relationship is constrained and the overall framework is less model-dependent than before.

For smaller and mid-sized banks, the new standardised approach may actually simplify operational risk capital calculation. For larger institutions with good historical loss records, the removal of AMA could increase operational risk RWA.

7: Real estate exposure risk weights increase

Real estate exposures receive more differentiated and, in many cases, higher risk weights under the revised standardised approach. The key driver is the loan-to-value ratio. Where the previous framework applied relatively uniform treatment to residential mortgages, Basel IV introduces a granular LTV-based risk weight table that increases capital requirements as the LTV ratio rises.

Income-producing real estate exposures, where repayment depends on cash flows generated by the property rather than the borrower’s general capacity to repay, attract particularly high risk weights under the new framework. This affects banks with significant commercial real estate lending, property investment financing, or buy-to-let mortgage portfolios.

The practical implication is that banks need accurate, up-to-date LTV data at the contract level to calculate risk weights correctly. Where property valuations are stale or LTV data is incomplete, the framework requires applying conservative fallback risk weights, which can further increase RWA.

Preparing your RWA strategy for Basel IV compliance

The seven changes above do not operate in isolation. The output floor connects internal model results to standardised calculations. The revised credit risk standardised approach feeds into the floor calculation. FRTB affects trading book RWA, which flows into the overall capital ratio alongside credit and operational risk. Managing your RWA strategy for Basel IV compliance means understanding how these moving parts interact across your specific portfolio.

The most useful starting point is a comprehensive impact assessment that runs your current portfolio through the new standardised approaches and compares the results to your internal model outputs. This tells you where the output floor is likely to bind, which exposure classes will see the largest RWA increases, and where your data gaps are most likely to create compliance risk.

Data quality is not a secondary concern here. The revised standardised approach for credit risk requires granular contract-level data, including LTV ratios, collateral types, and borrower classifications. The CVA framework requires sensitivity data at the counterparty level. FRTB requires desk-level sensitivity calculations. If your data infrastructure cannot support these requirements reliably and at speed, your Basel IV reporting will be constrained regardless of how good your models are.

Stress testing and what-if analysis are also more important under Basel IV than they were under previous frameworks. The output floor means that changes to your standardised RWA directly affect your capital requirements, even if your internal model results do not change. Running scenarios that show how portfolio composition changes, new business, or market movements affect both standardised and internal model RWA gives you the visibility to make informed capital allocation decisions.

We built our Basel.NXT credit risk solution to handle exactly this kind of parallel calculation, connecting standardised and IRB results within a single environment so you can manage the output floor, run stress tests, and drill down from top-level capital ratios to individual contract contributions without switching between systems. If you are working through your Basel IV compliance roadmap, that kind of integrated view makes the process significantly more manageable.

Related Articles

• Which risk platforms can replace both Moody's and SAS?
• What is CSRBB and how does it differ from IRRBB?
• How can banks ensure audit trail when submitting regulatory reports?
• How does systemic risk factor into bank stress tests?
• How do you design realistic stress test scenarios?

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