With equity risk complete, we move to the fifth risk class: commodity risk. This is the risk class with the widest range of underlyings in the whole framework — everything from crude oil to livestock to rare earths — organised into 11 buckets grouped by shared physical and market characteristics.
The Risk Factor Structure
Commodity delta risk factors are, at their core, all commodity spot prices — with one practical adjustment: for certain commodities such as electricity, which falls within bucket 3, the relevant risk factor can be either the spot price or the forward price, because transactions in electricity happen far more frequently on the forward price than the spot price.
Delta commodity risk factors are then defined along two dimensions: the legal delivery location terms of the commodity, and the time to maturity of the traded instrument — mapped to eleven tenors, notably more than the five- or ten-tenor structures we have seen in other risk classes.
| The Eleven Commodity Delta Tenors |
| 0 years |
| 0.25 years |
| 0.5 years |
| 1 year |
| 2 years |
| 3 years |
| 5 years |
| 10 years |
| 15 years |
| 20 years |
| 30 years |
| What “Delivery Location” Actually Means The standard’s own footnote example: a contract that can be delivered in five specific ports can only be considered to share the same delivery location as another contract if that other contract can also be delivered in the same five ports. A contract deliverable in only four (or fewer) of those five ports does not count as sharing the same delivery location — the match must be exact, not merely overlapping. |
| Risk Measure | Dimensions | Structure |
| Delta | 2 | Delivery location × 11 maturity tenors. |
| Vega | 1 | Option maturity only, mapped to one or several of five tenors: 0.5, 1, 3, 5 and 10 years. The underlying risk factor is the implied volatility of options referencing commodity spot prices, with no differentiation required by underlying maturity or delivery location. |
| Curvature | 1 | The constructed curve per commodity spot price only — no term structure decomposition. All delta tenors are shifted in parallel for the curvature calculation. |
Sourced from MAR21.13, including all three sub-points and footnote [6].
The Sensitivity Formula: A Quick Recap
The delta commodity sensitivity follows the same relative-shift pattern as equity spot and FX, which we covered in Article 8C: shift the commodity spot price by 1 percentage point, and divide the resulting change in market value by 0.01.
Sourced from MAR21.23.
The 11 Commodity Buckets and Their Risk Weights
For delta commodity risk, 11 buckets group commodities by shared characteristics. Each bucket carries its own flat risk weight, applied identically across all tenors within that bucket.
| Bucket | Commodity Type | Example Commodities | Risk Weight |
| 1 | Energy — solid combustibles | Coal, charcoal, wood pellets, uranium | 30% |
| 2 | Energy — liquid combustibles | Crude oil (light-sweet, heavy, WTI, Brent); biofuels; petrochemicals; refined fuels | 35% |
| 3 | Energy — electricity and carbon trading | Spot/day-ahead/peak/off-peak electricity; carbon trading emissions | 60% |
| 4 | Freight | Dry-bulk routes (Capesize, Panamax); liquid-bulk/gas shipping routes | 80% |
| 5 | Metals — non-precious | Base metals (aluminium, copper, lead, nickel, tin, zinc); steel raw materials; minor metals | 40% |
| 6 | Gaseous combustibles | Natural gas, liquefied natural gas | 45% |
| 7 | Precious metals (including gold) | Gold, silver, platinum, palladium | 20% |
| 8 | Grains and oilseed | Corn, wheat, soybean, oats, palm oil, canola, rapeseed, rice, and related products | 35% |
| 9 | Livestock and dairy | Cattle, hog, poultry, lamb, fish, shrimp, milk, whey, eggs, butter, cheese | 25% |
| 10 | Softs and other agriculturals | Cocoa, coffee, tea, citrus juice, sugar, cotton, wool, lumber, pulp, rubber | 35% |
| 11 | Other commodity | Potash, fertilizer, phosphate rocks, rare earths, terephthalic acid, flat glass | 50% |
The example commodities listed are explicitly non-exhaustive — they illustrate the kind of underlyings each bucket is meant to capture, not a complete or closed list.
Sourced from MAR21.81–MAR21.82 (Table 11).
https://www.bis.org/basel_framework/chapter/MAR/21.htm?inforce=20230101&published=20260323
Correlation Within a Bucket
For aggregating commodity risk positions within a bucket, the correlation parameter ρkl is built from three multiplicative factors — a structure that will feel familiar from the credit spread risk classes, though the first factor works differently here.
| Factor | Value When Matching | Value Otherwise |
| Commodity (same commodity?) | 100% where the two sensitivities relate to identical commodities | The specific intra-bucket correlation from Table 12 (below) |
| Tenor (same tenor?) | 100% if the two tenors are identical | 99.00% otherwise |
| Basis (same delivery location?) | 100% if identical in delivery location | 99.90% otherwise |
Two commodities count as distinct whenever the market itself would treat two contracts as distinct purely because of a difference in the underlying commodity to be delivered — WTI and Brent crude, both sitting in bucket 2, are the standard’s own example of commodities that would typically be treated as distinct from each other, and therefore fall back to the Table 12 correlation value rather than 100%.
| Bucket | Commodity Type | Intra-Bucket Correlation (Table 12) |
| 1 | Energy — Solid combustibles | 55% |
| 2 | Energy — Liquid combustibles | 95% |
| 3 | Energy — Electricity and carbon trading | 40% |
| 4 | Freight | 80% |
| 5 | Metals — non-precious | 60% |
| 6 | Gaseous combustibles | 65% |
| 7 | Precious metals (including gold) | 55% |
| 8 | Grains and oilseed | 45% |
| 9 | Livestock and dairy | 15% |
| 10 | Softs and other agriculturals | 40% |
| 11 | Other commodity | 15% |
| The Source Document’s Own Worked Example The correlation between a sensitivity to Brent crude, one-year tenor, deliverable in Le Havre, and a sensitivity to WTI crude, five-year tenor, deliverable in Oklahoma — both bucket 2 — is 95% × 99.00% × 99.90% = 93.96%. Different commodity (Brent vs WTI, so the Table 12 value of 95% applies), different tenor (99.00%), different delivery location (99.90%) — all three factors multiply together. |
Sourced from MAR21.83, including footnote [21] with the worked example.
https://www.bis.org/basel_framework/chapter/MAR/21.htm?inforce=20230101&published=20260323
Further Guidance on What Counts as “Distinct”
MAR21.84 gives non-exhaustive further examples of how to determine distinct commodities within two specific buckets where the question is genuinely less obvious than “WTI versus Brent”:
- Bucket 3 (energy — electricity and carbon trading): each time interval at which electricity can be delivered, as specified in a financial market contract, counts as a distinct electricity commodity (for example, peak versus off-peak). Electricity produced in a specific region — the standard’s own examples are “Electricity NE,” “Electricity SE” and “Electricity North” — also counts as a distinct electricity commodity.
- Bucket 4 (freight): each combination of freight type and route counts as a distinct commodity, and each week in which a good must be delivered also counts as a distinct commodity.
Sourced from MAR21.84, including both sub-points.
Correlation Across Buckets
For aggregating delta commodity risk positions across different buckets, the cross-bucket correlation parameter γbc follows just two rules — simpler than either the credit spread risk or equity risk cross-bucket structures.
| Bucket Pairing | Cross-Bucket Correlation (γbc) |
| Both bucket b and bucket c fall within buckets 1 to 10 | 20% |
| Either bucket b or bucket c is bucket 11 (other commodity) | 0% |
Sourced from MAR21.85, including both sub-points.
| A Structural Difference From CSR and Equity Risk, Worth Noting Explicitly Readers coming from Articles 10, 12B and 13 might expect bucket 11 (“other commodity”) to follow the same pattern as the “other sector” buckets in those risk classes — a special no-diversification override formula, summing absolute sensitivities rather than using a correlation parameter at all. Commodity risk does not work that way. There is no equivalent special aggregation formula for bucket 11 anywhere in MAR21.81 through MAR21.85. Bucket 11 is simply assigned its own intra-bucket correlation of 15% in Table 12, like every other bucket, and its cross-bucket correlation against every other bucket is set at 0% under MAR21.85(2) — achieving a broadly similar economic effect (bucket 11 does not diversify against anything else) through the ordinary correlation mechanism, rather than through a distinct formula. |
Looking Ahead
Commodity risk is now complete. The next article covers the sixth and final individual risk class in the delta framework: foreign exchange risk — genuinely the shortest risk class in the entire Sensitivities-Based Method, before we turn to vega and curvature buckets across all classes together in Articles 16 and 17.
Frequently Asked Questions
How many tenors does delta commodity risk use?
Eleven: 0, 0.25, 0.5, 1, 2, 3, 5, 10, 15, 20 and 30 years — more than the five- or ten-tenor structures used elsewhere in the Sensitivities-Based Method.
Does the “other commodity” bucket get special no-diversification treatment like the other-sector buckets in CSR and equity risk?
No. Unlike CSR’s bucket 16/25 and equity’s bucket 11, commodity risk’s bucket 11 has no special override formula — it simply uses the ordinary Table 12 intra-bucket correlation (15%) and a 0% cross-bucket correlation against every other bucket.
What is the risk weight for precious metals under FRTB?
20% — the lowest risk weight of any commodity bucket, reflecting the deep liquidity of gold, silver, platinum and palladium markets relative to other commodity types.
Link to previous article
Pingback: FRTB Series 15 - MAR21: FX Risk, Buckets, RW and Correlation