ESG Case Interviews 2026: Cost Per Metric Ton Avoided

An ESG case objective is a quantity, and it carries a unit

Take a prompt in the form these arrive in. A manufacturer has committed to cutting 30,000 metric tons of carbon dioxide a year by 2030, and the chief financial officer wants to know what it costs and whether it is reachable.

Three things in that sentence do work. The quantity, 30,000 metric tons. The unit, metric tons of carbon dioxide a year. And the implicit second objective, which is money.

Write the objective function before anything else. You are minimizing dollars spent per metric ton avoided, subject to the sum of tons avoided reaching 30,000 a year. That formulation hands you two questions per lever rather than one: what does a ton cost here, and how many tons can this lever physically deliver.

The second question is the one that gets skipped, and it is usually the binding one. A lever with a low cost per ton and a low ceiling cannot close a target no matter how attractive its unit economics look. That asymmetry is the reason an ESG structure needs a capacity branch, and our page on structured thinking covers why branches should fall out of the equation rather than out of a memorized shape.

Money enters twice, which is worth naming. It is a constraint, because the budget is finite. It is also the tie breaker among levers that clear the tonnage. Keeping those two roles apart in your structure prevents the answer that recommends the cheapest lever and misses the target.

Turning a kilowatt hour into a metric ton

Everything on this page runs on one conversion, and the Environmental Protection Agency publishes it with its full arithmetic at epa.gov, checked 30 September 2026.

The published equation reads: "1,405.3 lbs CO2/MWh × 1 metric ton/2,204.6 lbs × 1/(1-0.051) MWh delivered/MWh generated × 1 MWh/1,000 kWh = 6.72 × 10-4 metric tons CO2/kWh". The rate is the national weighted average marginal rate from eGRID 2022 data, and the 5.1 percent term grosses up for line losses.

Redo it by hand, because a conversion you have not checked is a conversion you will misuse. Divide 1,405.3 by 2,204.6 and you get 0.637533 metric tons per MWh. Divide by 0.949 and you get 0.671794. Divide by 1,000 and you get 0.000671794 metric tons per kilowatt hour, which the agency rounds to 6.72 times ten to the minus four. The rounding is upward and the agency flags it: "Due to rounding, performing the calculations given in the equations below may not return the exact results shown."

Two rates live on that page and they are used for different questions. The marginal rate of 1,405.3 pounds per MWh applies to electricity avoided, because efficiency displaces the plants running at the margin. The baseload rate of 823.1 pounds per MWh applies to electricity used. Picking the wrong one moves an answer by 71 percent, since 1,405.3 divided by 823.1 is 1.71.

The figures below use EPA's published rounded factor for avoided power and the baseload rate for power consumed, which is the agency's own convention.

Three levers in an ESG case, priced per metric ton

Prices come from the Energy Information Administration. July 2026 national averages were 9.77 cents per kilowatt hour industrial and 14.53 cents commercial, at eia.gov, checked 30 September 2026 and labeled preliminary. Regular gasoline averaged 4.465 dollars a gallon for the week of 28 September 2026, at eia.gov.

Every operating assumption below is ours, declared as such, and none of it is discounted.

Lever one, an efficiency retrofit. It cuts 12 million kilowatt hours a year, costs 9 million dollars, and lasts ten years. Tons avoided: 12,000,000 times 0.000672 is 8,064 metric tons a year. Energy saved: 12,000,000 times 0.0977 dollars is 1,172,400 dollars a year. Annualized capital is 9,000,000 over ten years, so 900,000 dollars a year. Net annual cash is 1,172,400 minus 900,000, a saving of 272,400 dollars. Cost per ton is minus 272,400 divided by 8,064, so the retrofit earns 33.78 dollars per metric ton avoided. Simple payback is 9,000,000 over 1,172,400, which is 7.68 years.

Lever two, electrifying the light fleet. The firm runs 400 vehicles. EPA puts an average gasoline car at 4.29 metric tons of CO2 equivalent a year, on 10,917 miles at 22.8 miles per gallon. An electric one covers the same 10,917 miles at 3.60 miles per kilowatt hour, so 3,032.5 kilowatt hours a year, which is 3.0325 MWh. At 823.1 pounds per MWh that is 2,496.0 pounds, and dividing by 2,204.6 gives 1.132 metric tons. Avoided per vehicle: 4.29 minus 1.13, so 3.16 metric tons a year. Fuel: 10,917 over 22.8 is 478.8 gallons, times 4.465 dollars is 2,137.8 dollars. Power: 3,032.5 times 0.1453 dollars is 440.6 dollars. Operating saving is 1,697.2 dollars a year. Assume an 8,000 dollar purchase premium over an eight year life, so 1,000 dollars a year, and net annual cash is a saving of 697.2 dollars. Cost per ton is minus 697.2 over 3.16, so the swap earns 220.63 dollars per metric ton.

Lever three, a cleaner power contract. Assume a 1.5 cent per kilowatt hour premium on 30 million kilowatt hours of purchased power. Cost is 450,000 dollars a year. Tons avoided: 30,000,000 times 0.000672 is 20,160 metric tons a year. Cost per ton is 450,000 over 20,160, so 22.32 dollars.

Ranked by cost, ranked by capacity, and the two orders invert

Put the three levers in one table and the finding is immediate.

The fleet figure is 400 times 3.16, which is 1,264 metric tons, and its cash is 400 times 697.2, which is 278,880 dollars.

Now read the two rightmost columns against the tonnage column. By cost per ton the order is fleet, retrofit, contract. By tons available the order is contract, retrofit, fleet. The ranking is exactly reversed, and the reversal is the reason a merit order alone does not answer the prompt.

The fleet lever makes the point sharpest. It is the cheapest per ton by a factor of six and a half against the retrofit, and it delivers 1,264 tons against a 30,000 ton target, which is 4.2 percent of it. Reaching the target on fleet alone would take 30,000 divided by 3.16, so 9,494 vehicles, against a fleet of 400.

A candidate who ranks levers and stops has answered a question about efficiency. The prompt asked about a quantity.

A portfolio that pays for itself and misses the target

Add the three levers at full capacity. Tons: 1,264 plus 8,064 plus 20,160, which is 29,488 metric tons a year. Against a 30,000 ton commitment that is 98.3 percent, leaving a shortfall of 512 tons.

Now the cash. Savings of 278,880 and 272,400 dollars against a spend of 450,000 dollars leaves a net saving of 101,280 dollars a year. Divide by 29,488 tons and the whole program earns 3.43 dollars per metric ton avoided.

So the portfolio pays for itself, reaches 98.3 percent of the commitment, and fails it. Money was never the binding constraint, and any answer framed as a budget question would have missed that. The recommendation has to be one of three things: find a fourth lever worth at least 512 tons a year, expand the ceiling on an existing one, or renegotiate the target with the evidence attached.

Expanding a ceiling is usually the cheapest of the three and it is the one candidates forget to price. The contract lever scales with purchased power, so covering the 512 ton gap needs 512 divided by 0.000672, which is 761,905 additional kilowatt hours a year, at 1.5 cents, so 11,429 dollars. The gap that broke the target costs eleven thousand dollars to close, on a program already saving a hundred thousand.

That last line is what a chief financial officer will remember, and it exists only because the tonnage and the cash were tracked in separate columns from the start.

Regulation, cost and demand give three ESG case structures

The arithmetic above assumes a voluntary target. The status of the objective changes the structure, and there are three statuses worth separating before you draw.

1. A binding rule. The quantity is set outside the company and missing it carries a penalty. Feasibility comes first and cost per ton is a tie breaker among compliant plans. The infeasible portfolio above would be a reportable problem rather than a trade-off.
2. A cost to minimize. The company has chosen the target. Now cost per ton leads, capacity constrains, and the target itself is negotiable evidence, which is what the 512 ton shortfall becomes.
3. A source of demand. Customers pay more for the lower-carbon version, so the emissions work sits in the revenue line and the case becomes a pricing case with a physical constraint attached.

Oliver Wyman publishes a case in that third shape. Its digital therapy exercise has a client whose product "reduces the need for in-person therapy, allowing children to use it at home and expanding access for those who currently cannot access traditional therapy", and the published answer defends a lower price on the reasoning that "A subscription model reduces the initial financial burden, supporting the client's goal of making the product affordable", at oliverwyman.com, checked 30 September 2026. Access and margin sit in the same model, and neither one is decoration.

Ask which of the three you are in before you branch. The question takes one sentence and it changes every number after it. Our page on energy cases works the sector version of the same distinction.

What the scale rewards on an ESG prompt

The weights below are ours, read from the grading code in src/lib/grading. No consulting firm publishes a marking scheme, on ESG or anything else, which holds across every research file we keep.

Drilling the tonnage arithmetic an ESG case interview needs

The conversions are the drillable part, because each has an inverse.

1. Go backwards from tons to kilowatt hours. A 5,000 ton target on avoided power needs 5,000 divided by 0.000672, which is 7,440,476 kilowatt hours a year. At 9.77 cents that power costs 726,934 dollars, which is the saving available if the target is met by efficiency.
2. Flip the rate on purpose. Redo the fleet calculation using the marginal rate of 1,405.3 pounds per MWh for the electric vehicle instead of baseload. It becomes 3.0325 times 1,405.3, which is 4,261.6 pounds, so 1.933 metric tons, and avoided tons per vehicle fall from 3.16 to 2.36. Then say out loud which rate the question deserves.
3. Price the premium that breaks a lever. The contract lever earns nothing above the point where its premium exceeds the value of the tons. At a 22.32 dollar per ton internal carbon price it breaks even, since that is where 450,000 dollars and 20,160 tons meet.

Run these against a clock in the math drill, which returns Answer, Working, Units and Interpretation separately with the first correction named. It is free and needs no account, it knows nothing about emission factors, and it will not tell you which rate the prompt wants.

For prompts where the beneficiary rather than the atmosphere is the objective, see non-profit cases and public sector cases.