Brent, WTI and Refinery Energy Costs: A Margin Scenario Bridge

AI-generated editorial illustration; it does not depict a verified project, actual prices or chart data.

1. Economic decision

A rising Brent–WTI spread does not automatically increase refinery profitability. The spread compares two crude benchmarks with different delivery locations and specifications; a refinery must purchase compatible feedstock at its own gate and sell a particular product slate. Energy costs matter through gas, electricity, steam and hydrogen, while inventory timing can create gains or losses unrelated to steady-state processing. The useful economic question is the incremental contribution of a barrel processed under a defined crude and product price window, followed by a separate fixed-cost and capital boundary.

2. Dated market evidence

The World Bank October Pink Sheet gives August→September 2026 Brent 90.9→116.8 and WTI 82.7→96.9 USD/barrel. Within those series, the benchmark spread widens from 8.2 to 19.9 USD/barrel, a change of 11.7. It is not a quoted freight-adjusted feedstock saving. EIA's 6 October outlook forecasts 4Q2026 Brent at 105 USD/barrel; that forecast is separate from the World Bank September observation. EIA's refining explanation describes processing into products, supporting the need for a yield-weighted basket.

3. Cost boundary and calculation

All worked costs are illustrative nominal-currency scenarios as of 7 October 2026, not vendor quotations or observed October averages. Recoverable VAT, finance, penalties and carbon costs are excluded unless expressly stated. Sources retain their own observation period and forecast classification.

Assume a hypothetical yield-weighted product value of 125 USD per feed barrel and delivered crude of 100 USD/barrel. Use external gas 0.15 MMBtu/barrel at 15 USD/MMBtu, electricity 0.015 MWh/barrel at 100 USD/MWh and other variable cost 1 USD/barrel. Gas costs 2.25 and electricity 1.5 USD/barrel. Contribution = 125 − 100 − 2.25 − 1.5 − 1 = 20.25 USD/barrel. At 100000 barrels/day, that is 2025000 USD/day before fixed costs, capital, tax, inventory effects and downtime. This is a budgeting example, not a refinery quotation or profit disclosure.

Throughput changes the daily result but should not be used to assume the same unit margin at every load. At the illustrative 20.25 USD/barrel contribution, processing 80000 rather than 100000 barrels/day would yield 1620000 USD/day if unit quantities stayed constant. Actual lower-load gas intensity may worsen, fixed utilities may remain, and product yield may shift. Use an operating model or plant records for those effects; a linear arithmetic check is not a refinery performance curve.

4. Sensitivity and cost table

Illustrative refinery: product basket 125, delivered crude 100, electricity 1.5 and other variable costs 1 USD/barrel. Gas use 0.15 MMBtu/barrel at 5/15/25 USD/MMBtu gives margins 21.75/20.25/18.75 USD/barrel, before fixed costs, capital and tax.

A product basket needs its own scenario, independent of the gas chart. If the basket falls from 125 to 115 USD per feed barrel while crude remains 100 and base utilities remain unchanged, contribution falls to 10.25 USD/barrel. If crude simultaneously rises to 110, contribution becomes 0.25 USD/barrel. These are constructed stresses, not a predicted crack spread. The crude and product movements can be linked in a real market; the worksheet must explain which prices reset together and which are fixed, rather than assert independence.

5. Operational interpretation

Holding crude, products, throughput and power unchanged, gas 5/15/25 USD/MMBtu produces contribution 21.75/20.25/18.75 USD/barrel. A 10 USD/MMBtu gas increase reduces contribution by 0.15 × 10 = 1.5 USD/barrel or 150000 USD/day at stated throughput. Separately, power moving from 100 to 150 USD/MWh reduces contribution by 0.015 × 50 = 0.75 USD/barrel. A combined stress adds those two independently defined utility exposures only if neither gas quantity already includes purchased power generation. The chart varies gas alone.

Inventory can obscure the current operating economics. A cargo purchased in an earlier month may enter processing at its booked cost while product sales settle at later prices. Reconciliation should show physical operating contribution at consistent replacement prices and inventory accounting separately. Do not count an inventory revaluation as a persistent energy-efficiency saving. A utility upgrade that avoids 0.005 MWh/barrel at 100 USD/MWh saves 0.5 USD/barrel, or 50000 USD/day at 100000 barrels/day, only if saleable output and purchased energy actually match the assumed change.

6. Contract evidence and decision trigger

A refinery cannot assume WTI replaces Brent at equal yield and equal processing duty. Specify crude assay, sulphur, density, delivery premium, voyage cost and equipment limits. Build product value as saleable yield multiplied by each net realisation, including by-products and disposal burdens; do not treat one diesel quotation as the whole barrel. Align crude and product valuation windows and identify inventory accounting separately. The operating trigger is a verified marginal processing contribution after utility and logistics costs, checked against constraints and shutdown risk, rather than a benchmark spread headline.

A shutdown decision includes restart expense, minimum safe throughput, contractual product commitments and equipment condition. Positive contribution does not establish that every production increase is feasible, and a low contribution does not establish an immediate safe stop. Energy purchasing should identify whether gas is fixed, indexed or interruptible and whether hydrogen is bought separately. If purchased hydrogen embeds energy already modelled elsewhere, remove that overlap. Require site-approved utility balances and crude/product settlement terms before using the daily figures to commit production; the example is an economic decomposition, not an operating instruction.

7. Frequently asked questions

Is contribution equal to net profit? No: fixed expenses, capital and tax remain outside. Can a benchmark spread be counted as guaranteed feedstock saving? No: delivery, quality, yield and constraints must match. Should internally generated fuel be free? No: define its opportunity value and avoid counting the same gas both as sold product and consumed utility.

8. References and related engineering

• World Bank October 2026 Pink Sheet
• EIA October STEO
• EIA refining crude oil

• Refinery hydrogen and utility cost
• Pipeline pumping energy scenarios