Risk-weighted assets are a bank’s holdings adjusted by regulatory percentages that reflect how likely each asset is to lose value. To calculate the figure, a bank multiplies the dollar value of each exposure by the risk weight assigned to that exposure’s category, then adds every weighted result together. The total sits in the denominator of the capital ratios regulators use to decide whether the bank holds enough of a cushion to survive losses. A U.S. Treasury bond carries a 0% weight because default is essentially inconceivable; a standard corporate loan carries 100% because the borrower might fail. Everything else falls somewhere in between, or, in a few cases, well above.
How the Calculation Works
The arithmetic itself is simple. Take the face value of an asset, multiply by the weight assigned to its category, and the product is the risk-weighted amount. A $10 million corporate loan at 100% contributes $10 million to RWA. A $10 million holding of U.S. Treasuries at 0% contributes nothing. A $10 million first-lien residential mortgage at 50% contributes $5 million. Do that for every exposure on the balance sheet, add the results, and you have total RWA.
The design has a deliberate effect on bank behavior. The higher the total, the more capital the bank must hold. Loading up on risky assets forces the bank to set aside more capital, which means less money to lend or return to shareholders. Shifting toward safer assets lets the bank operate with a thinner cushion. That tension between risk-taking and capital efficiency drives much of how banks manage their portfolios.
Total RWA captures three broad types of risk, each with its own methodology. Credit risk, the chance a borrower or counterparty fails to pay, accounts for the largest share at most banks. Market risk covers potential losses from changes in interest rates, equity prices, or exchange rates in the trading book. Operational risk covers losses from internal failures, fraud, legal liability, or events like cyberattacks. The results of the three calculations are summed to produce the reported figure.
For market risk, the current international framework is the Fundamental Review of the Trading Book, which uses an Expected Shortfall measure and lets banks choose between a standardized approach built on risk factor sensitivities or, with supervisory approval, an internal models approach applied desk by desk. For operational risk, the Basel III standardized measurement approach calculates a Business Indicator from three components of the bank’s financial statements, averages it over three years, and scales the result by coefficients that increase with the bank’s size.1Bank for International Settlements. Basel Framework2Bank for International Settlements. OPE25 – Standardised Approach
Standardized Risk Weights
Under the standardized approach, regulators set the weights and banks slot each exposure into a category based on the counterparty’s identity and any collateral. The percentages below are those used in the United States and illustrate the general pattern worldwide, though exact figures can differ by jurisdiction.3eCFR. 12 CFR Part 3 Subpart D – Risk-Weighted Assets—Standardized Approach
- 0%: Cash held in a bank’s own vaults, U.S. government obligations, and exposures to certain international institutions like the IMF, the Bank for International Settlements, and multilateral development banks.
- 20%: Exposures to U.S. government-sponsored enterprises such as Fannie Mae and Freddie Mac, U.S. depository institutions and credit unions, general obligation bonds from U.S. public sector entities, claims conditionally guaranteed by the FDIC or a U.S. government agency, and exposures to foreign banks from highly rated countries.4eCFR. 12 CFR 217.32 – General Risk Weights
- 50%: First-lien residential mortgages that are owner-occupied or rented, underwritten prudently, and not delinquent or restructured. The Basel international standard uses 35% for similar loans, which is why both figures appear in different sources.
- 100%: Standard corporate loans. This is the default category: any exposure to a business that does not qualify for a lower weight lands here.
- 150%: The unsecured portion of past-due exposures that lack guarantees.
- 250%: Certain mortgage servicing assets and deferred tax assets arising from temporary differences, to the extent they are not deducted from capital.
- 300%: Publicly traded equity exposures not otherwise categorized.
- 1,250%: Securitization exposures where the bank cannot demonstrate it fully understands the position, or where standard calculation methods do not apply. This is the regulatory ceiling and effectively requires the bank to hold capital equal to the entire exposure value.5eCFR. 12 CFR Part 3 Subpart D – Risk-Weighted Assets for Securitization Exposures
Off-Balance Sheet Items and the Credit Conversion Factor
Some exposures do not appear as assets on the balance sheet but still put the bank at risk. Undrawn credit lines, letters of credit, and guarantees are the common examples. Before these items get a risk weight, they run through a credit conversion factor, or CCF, that estimates what portion is likely to become an actual on-balance-sheet exposure.
- 0% CCF: Commitments the bank can cancel unconditionally at any time, such as a credit card line the bank can revoke without notice.
- 20% CCF: Short-term commitments with an original maturity of one year or less that the bank cannot cancel unconditionally.
- 50% CCF: Longer-term commitments with an original maturity over one year that the bank cannot cancel unconditionally.
- 100% CCF: Guarantees, where the bank has committed to cover the full obligation if the primary party defaults.
The converted figure is then risk-weighted the same way as an on-balance-sheet asset. A $100 million revolving credit facility with a 50% CCF and a 100% corporate risk weight contributes $50 million to RWA.6eCFR. 12 CFR 217.33 – Off-Balance Sheet Exposures
Internal Models: The IRB Approach
Larger and more sophisticated banks may use their own statistical models instead of the fixed weights in the standardized approach. This is the Internal Ratings-Based approach, and its appeal is straightforward. A bank with granular data on its borrowers can, in theory, produce more accurate risk estimates than a one-size-fits-all regulatory table.
The IRB approach revolves around three inputs for each exposure:
- Probability of Default (PD): How likely the borrower is to fail to pay within a given time horizon.
- Loss Given Default (LGD): The percentage of the exposure the bank expects to lose if default occurs, after accounting for collateral and recoveries.
- Exposure at Default (EAD): The total amount the bank expects to be owed at the moment of default.
These parameters feed into regulatory formulas that produce a risk weight for each exposure. Two versions exist. Under Foundation IRB, the bank estimates only PD, and the regulator supplies LGD and EAD. Under Advanced IRB, the bank estimates all three. Advanced gives more control over the capital calculation but demands more robust data infrastructure and model validation, and regulators must approve the models before a bank can use them.7Bank for International Settlements. CRE20 – IRB Approach Overview
A bank with strong data on middle-market corporate lending, for instance, might produce PD estimates that are lower than what the standardized approach assumes, which lowers the risk weight and reduces required capital. That is the upside. The downside is model risk: if the estimates are wrong, the cushion may be thinner than the true risk warrants.
Why RWA Matters: Capital Ratios
The RWA figure is meaningful because it sits in the denominator of the ratios regulators use to judge a bank’s resilience. Divide eligible regulatory capital by total RWA and you get the capital adequacy ratio.
Regulatory capital is organized in tiers based on loss-absorbing quality. Common Equity Tier 1 (CET1) is the strongest: common stock, retained earnings, and accumulated other comprehensive income. Additional Tier 1 includes instruments like certain preferred stock that can absorb losses but are less permanent than common equity. Tier 2 includes subordinated debt and qualifying loan loss provisions.8Bank for International Settlements. Definition of Capital in Basel III – Executive Summary
Under both Basel III and U.S. federal banking regulations, banks must maintain at least a 4.5% CET1 ratio, a 6.0% Tier 1 ratio, and an 8.0% total capital ratio, all expressed as a percentage of total RWA.9eCFR. 12 CFR 217.10 – Minimum Capital Requirements
On top of the minimums, banks must hold a capital conservation buffer of at least 2.5% of RWA in CET1, which effectively raises the working CET1 floor to 7.0%. Dipping into the buffer is not an immediate violation of minimum ratios, but it triggers escalating restrictions on dividends, share buybacks, and discretionary bonuses. The deeper the cut, the tighter the restrictions, down to a complete ban on distributions if the buffer falls below 0.625% of RWA.10eCFR. 12 CFR 217.11 – Capital Conservation Buffer
Banks designated as global systemically important face an additional surcharge on top of the conservation buffer. In the United States, this surcharge starts at 1.0% and increases based on a scoring methodology covering size, interconnectedness, cross-jurisdictional activity, substitutability, and complexity. The largest U.S. banks currently face surcharges ranging from 1.0% to 4.5% or more. A bank with a 2.0% G-SIB surcharge and the 2.5% conservation buffer must maintain CET1 of at least 9.0% of RWA before any distribution restrictions apply.11Federal Register. Regulatory Capital Rule (Regulation Q) – Risk-Based Capital Surcharges for Global Systemically Important Bank Holding Companies
Where the Framework Is Headed
The Basel Committee finalized the last set of Basel III reforms in 2017, and they are being phased in through 2028. One central piece is the output floor, which prevents a bank using internal models from reporting RWA below a specified percentage of what the standardized approach would produce. Under the international timeline, the floor moves from 65% in 2026 to 70% in 2027 to 72.5% in 2028, at which point internal-model RWA can never be more than 27.5% below standardized RWA.12Bank for International Settlements. Basel III Transitional Arrangements, 2017-2028
The U.S. path is different. In March 2026, the Federal Reserve and other banking agencies released a re-proposal that would apply primarily to the largest, most internationally active banks (Category I and II institutions). Rather than layering an output floor on top of the current approaches, the proposal would replace both the standardized and advanced approaches with a single “expanded risk-based approach” that is itself largely standardized. Because the new requirements would already be built on standardized methodologies, the output floor was excluded from the proposal on the view that it would be unlikely to bind. Comments were due by June 18, 2026, and the final rule could differ from what was proposed.13Federal Reserve System. Agencies Request Comment on Proposals
Whether through an explicit floor or a rebuilt standardized framework, regulators are converging on the same principle: model-based flexibility should have limits, and the gap between a bank’s internal calculation and what the regulatory table would produce cannot grow without bound.