Cost and carbon done properly: one boundary, a fair baseline, a stated counterfactual

Audit a fictional bus study with four flaws, fix each one, and watch its conclusion change

Research studio R7 of Energy Systems, Step by Step: audit a fictional study of electric buses with mismatched cost and emissions boundaries, a convenient baseline, missing grid costs and a credit counted twice; fix each flaw, watch the conclusion change in your browser, and write an audit note checked against the course rubric.

Researcher · L4
Research studio
35 min guided · about 90 min practice
After A1 and A11 and M11
Completed

The task

Produce: an audit note on a study that weighs cost against carbon: each flaw, its fix and its effect, then the corrected conclusion with the condition it rests on.
From: a fictional study of 40 electric buses, its tables, and your browser's recomputation of every number after each fix.
Done when: the rubric's bar for Consistent accounting and for Realism: one boundary, the emissions question named, who pays named, inputs' evidence status and a list of what is left out.

About 35 minutes guided here, then about 90 minutes on your own:

pick a study from your field: a report, a paper or a press release;
run the checklist under Self-check on it;
recompute one number it rests on, on one boundary;
write the audit note and rate it against the rubric.

The study (fictional)

Electric buses for Port Alder: cheaper and carbon-free. R. Okonjo and L. Varga, Harbourline Consulting (fictional), report to Port Alder Transit (fictional), 2026.

Replacing Port Alder Transit's 40 oldest diesel buses with battery-electric buses saves $1.15 million a year and cuts the fleet's CO₂ by 100%.

Its inputs (illustrative)

Input	Electric	Diesel
Price per bus	$1,000,000 + $60,000 charger	$560,000
Energy per mile	2.2 kWh, at the depot meter, charging losses included	3.5 mpg (the 2009 buses)
Energy price	$0.09 per kWh, energy rate only	$3.60 per gallon
Repairs per mile	$0.60	$1.20 (the 2009 buses)
Purchase credit	$150,000 per bus, the state's clean-bus credit (fictional)	none

Both fleets: 40 buses, 36,000 miles each a year, a 12-year life, a 3% real discount rate.

From Port Alder Transit's fleet plan (fictional): The 40 buses bought in 2009 retire in 2026; 40 new diesel buses replace them (4.5 mpg, $0.85 a mile in repairs) unless the board funds another option.

Units: t is a metric tonne (1,000 kg); mpg is miles per U.S. gallon; a demand charge is dollars per kW of the month's highest draw (a kW-month), billed each month.

Two numbers worked from these inputs

Charging energy: 40 buses × 36,000 miles × 2.2 kWh per mile = 3,168 MWh a year; at $0.09 per kWh, $285,120 a year.
Diesel fuel, as the study runs it: 1,440,000 miles ÷ 3.5 mpg = 411,429 gallons a year; at $3.60 per gallon, $1,481,143 a year.

Four flaws, each fixed, and the conclusion after each

Fix	What it changes	Saving a year, electric against diesel	Attributed CO₂ cut
The study as published	none	$1.15 million	4,192 t (100%)
One boundary	One boundary: charging 3,168 MWh a year at 0.366 t per MWh is attributed 1,160 t. The CO₂ cut falls from 100% to 72%.	$1.15 million	3,033 t (72%)
A fair baseline	Fair baseline: new diesel buses burn 320,000 gallons a year, not 411,429. Diesel's yearly cost falls by $0.83 million.	$0.32 million	2,101 t (64%)
Grid costs	Grid costs: $288,000 a year in demand charges, and $1.80 million up front for the depot ($180,832 a year as an annual equivalent).	−0.15 million	2,101 t (64%)
The credit once	The credit once: the $6.00 million stays off the price up front; the second count, $500,000 a year, goes.	−0.65 million	2,101 t (64%)

Audited: Electric costs $0.65 million a year more and lowers attributed CO₂ by 2,101 t (64%): about $309 per attributed tonne. That is for the transit agency, with charging at EIA's 2023 U.S. average factor.

With JavaScript on, you find each flaw, fix it and watch the conclusion change; your note stays in this browser.

Fictional study. Its authors, its city and its transit agency are invented; every fleet, price and cost input is illustrative, in 2025 dollars. Real fleets differ.

Two real numbers. Diesel emits 10.19 kg CO₂ per U.S. gallon burned (EIA, 2024). The U.S. grid emitted 1.53 billion tonnes over 4.18 trillion kWh generated in 2023 (EIA): 0.366 t per MWh.

Up front and a year. One-time costs become a yearly equivalent with the capital recovery factor at 3% over 12 years, 0.1005 (A1).

Two notes on the same study, and who pays

Weak note · Developing

Audit: the study looks optimistic [1]. Electric buses still have no emissions [2], but costs are probably higher [3]. More research is needed. [4]

Falls short: No flaw is named, so no one can check or fix it.
Falls short: True at the tailpipe only; the power plants that charge them emit.
Falls short: No number, no baseline, no 'whose cost'.
Falls short: It says nothing a reader could act on.

Strong note · Proficient

Boundary: charging is in the cost but its CO₂ is not; at EIA's 2023 U.S. average factor, the electric fleet is attributed 1,160 t a year. [1] Baseline: without the project the agency buys new diesel buses (4.5 mpg), not the 2009 fleet (3.5 mpg). [2] Grid: add the depot connection ($1.80 million up front) and demand charges ($288,000 a year). [3] Credit: counted once, in the capital line; the second $500,000 a year removed. [4] Evidence: the prices, miles and 2.2 kWh per mile are the study's illustrative assumptions; the replacement plan is the agency's own document (fictional); the CO₂ factors are EIA's published figures. [5] Leaves out: making the buses and batteries, street air and noise, grid losses, a grid that gets cleaner, battery replacement. [6] Conclusion: for the agency, electric costs $0.65 million a year more and lowers attributed CO₂ by 2,101 t (64%), about $309 per tonne, if the grid is at the 2023 U.S. average. [7]

Works: Names the mismatch, the fix and the factor's source.
Works: The counterfactual, stated from the agency's own plan.
Works: Up front and annual kept apart.
Works: Says which count stays and why.
Works: Each key input's evidence status, as the Realism bar asks.
Works: What the audit does not count, said.
Works: Cost and attributed CO₂ on one boundary, whose cost, and the condition that moves it most.

	Weak note	Strong note
Flaws	None named	Four, each with its fix
Numbers	None	Cost, CO₂ and cost per tonne
Whose cost	Unsaid	The agency's, with the credit once
Conditions	None	The grid's CO₂ factor
Evidence and limits	None	Each input's status; what is left out

Who pays, who gains: the audited plan, a year, against new diesel buses

Who	Cost or change	Per attributed tonne of CO₂
Port Alder Transit the agency's budget	+$0.65 million a year	$309 per attributed tonne (about)
The state pays the credit	$6.00 million up front ($0.60 million a year as an annual equivalent)	a transfer to the agency
Agency and state together the credit nets out	+$1.25 million a year	$595 per attributed tonne (about)
Attributed CO₂ operating energy, charging at EIA's 2023 U.S. grid average	−2,101 t a year	64% of the diesel fleet's

Where the audited conclusion holds

It holds when	It leaves out
the inputs are this fictional study's, in 2025 dollars, over a 12-year life at 3%; costs are the agency's, with the state's credit once; CO₂ is each fleet's operating energy, at the tailpipe or the power plant; charging's CO₂ is attributed at the grid's average factor (EIA, 2023): what the grid really emits less needs a consequential analysis (A10); 2.2 kWh per mile is at the depot meter, charging losses included; the agency pays the depot connection; both fleets run the same routes and miles.	making the buses and batteries: embodied CO₂ is outside both fleets' counts (a life-cycle boundary, A11); grid losses between plant and depot (EIA: about 5%), which would raise charging's CO₂ a little; tailpipe air pollution and noise in the streets, which electric buses cut; a grid that gets cleaner over the 12 years, and battery replacement; federal grants, fuel and power price changes.

the inputs are this fictional study's, in 2025 dollars, over a 12-year life at 3%;
costs are the agency's, with the state's credit once; CO₂ is each fleet's operating energy, at the tailpipe or the power plant;
charging's CO₂ is attributed at the grid's average factor (EIA, 2023): what the grid really emits less needs a consequential analysis (A10);
2.2 kWh per mile is at the depot meter, charging losses included;
the agency pays the depot connection; both fleets run the same routes and miles.

making the buses and batteries: embodied CO₂ is outside both fleets' counts (a life-cycle boundary, A11);
grid losses between plant and depot (EIA: about 5%), which would raise charging's CO₂ a little;
tailpipe air pollution and noise in the streets, which electric buses cut;
a grid that gets cleaner over the 12 years, and battery replacement;
federal grants, fuel and power price changes.

How far up the ladder is each number?

The transit board voted the money for 40 electric buses. Funded. Funded: money committed, nothing built yet.
The utility approved the depot's new service connection. Permitted. Permitted: approved, not yet built; its cost estimate is still an estimate.
Ten electric buses have carried passengers since March. Operating. Operating: in service, but a few months do not show a year's energy use.
Over a full year the ten buses used their expected energy per mile, winter included. At design output. At design output: the 2.2 kWh per mile now rests on measured service, not a brochure.
The audit's cost per attributed tonne of CO₂. Not on the ladder: a projection. A projection: a model's output from inputs on several rungs, not a stage of anything built.

The study against the audit, thousands of 2025 dollars a year (rounded)

Line	Study	Audited
Electric: buses and chargers, net of the credit up front $36.40 million	3,657	3,657
Electric: electricity	285	285
Electric: demand charges	0	288
Electric: depot connection up front $1.80 million	0	181
Electric: repairs	864	864
Electric: the credit again, as income	−500	0
Electric, total	4,306	5,275
Diesel: buses up front $22.40 million	2,250	2,250
Diesel: fuel 3.5 then 4.5 miles per gallon	1,481	1,152
Diesel: repairs	1,728	1,224
Diesel, total	5,459	4,626

Go deeper · AnalystWhich grid factor, and which question?

The audit counts charging at the grid's average factor: the buses' share of what the plants emit.
What charging changes depends on which plants respond in the hours the buses charge, now and as the grid is built out: A10's questions 2 and 3.
Name which question your number answers, and show the result at the other factors, as step 6 does.

Go deeperWhat this studio leaves out

A life-cycle boundary: making the buses and their batteries (A11).
Health and noise in the streets, which electric buses improve.
Uncertainty in each input: R6.
Who pays for a depot connection when a utility recovers it from all its customers: the cost moves, it does not vanish.

Check your own work against the rubric

Competence	Developing	Proficient (the bar)	Strong
Consistent accounting Look for: cost and CO₂ for the same service, alternatives and years, the emissions question named (an attributed share or a change in the system), and whose cost each number is.	Costs and emissions on different boundaries.	Cost and emissions on one boundary; emissions question named; who pays and who benefits named.	Distributional and environmental side effects estimated with stated assumptions.
Realism Look for: each key input with its evidence status, and a list of what the audit leaves out.	Assumes announced projects exist.	Evidence status of key inputs; a 'what the analysis leaves out' list.	Realism checked with a practitioner or primary documents.

Cost and carbon, before you trust a result

Same service: Both options deliver the same service: the same routes and miles, the same years.
One boundary: Cost and CO₂ cover the same service, alternatives and years; whose costs count and which emissions stages are in are stated, with what is left out.
A stated counterfactual: One sentence says what happens without the project, from a plan or a rule, not a convenient past.
The emissions question: Average or marginal factor, and why: whose share, or what changes.
Grid and connection costs: Upgrades, demand charges and connection fees are priced, with who pays them.
Credits once: Each grant or credit appears in one line, and is a transfer for the state and the agency together.
Up front and annual apart: One-time amounts and their annual equivalents are labelled, with the rate and the life.
The condition that moves it: The input that changes the conclusion most is named, with the result at its other values.

Myth or fact? “A fleet with no tailpipe emits no CO₂.”

Myth
Fact

Answer

Myth. At the 2023 U.S. average factor, charging Port Alder's buses is attributed about 1,160 t of CO₂ a year, emitted at power plants.

Myth or fact? “A purchase credit makes a project cheaper for everyone.”

Myth
Fact

Answer

Myth. It moves money from the state to the agency: the audited buses cost the agency $0.65 million a year more, and the two together $1.25 million.

Myth or fact? “The fair baseline is the fleet running today.”

Myth
Fact

Answer

Myth. It is what would happen without the project: here the replacement plan's new diesel buses, not the 2009 ones.

Go deeper · ResearcherFrom this studio to your own study

Proficient: cost and CO₂ on one boundary, the emissions question named, who pays named, inputs' evidence status and what is left out.
Strong goes further: side effects for other people and places, estimated with stated assumptions.
To test it: give your note to someone who did not write the study, and ask them to recompute one number.

Sources and what’s next

Numbers to remember

Number	What it means	Source
10.19 kg per gallon	CO₂ from burning a gallon of diesel.	EIA, Carbon Dioxide Emissions Coefficients · released 18 September 2024
0.366 t per MWh	The U.S. grid's average CO₂ per MWh generated in 2023: 1.53 billion tonnes over 4.18 trillion kWh (about 0.81 pounds per kWh).	EIA, Frequently Asked Questions · updated 11 December 2024

R8 · How do you show results so people understand their limits?

In the real world

In the reported scenarios, reductions in attributed emissions were on the order of one to two percent of regional totals or less.
Reductions are computed under average-emission-intensity accounting with perfect foresight; they are not estimates of the consequential (marginal) system response.
Approved excerpt of: Li et al. (2025), Cell Reports Sustainability.

Reading it. Attributed emissions: a user's share of the grid's CO₂, here at its average intensity, as in this studio's audit.

Perfect foresight: the model knows the coming hours in advance. Consequential response: how much the grid's own emissions actually change.

Case file CF6: how much shifting load to cleaner hours can cut emissions, and when.

Sources and further reading

EIA. Carbon Dioxide Emissions Coefficients, 18 September 2024 (diesel and home heating fuel: 10.19 kg CO₂ per gallon).
EIA. How much carbon dioxide is produced per kilowatthour of U.S. electricity generation? Data for 2023; updated 11 December 2024.
EIA. How much electricity is lost in electricity transmission and distribution in the United States? About 5% in 2018 to 2022; updated 7 November 2023.
ISO 14044:2006, Environmental management: life cycle assessment, requirements and guidelines (functional unit, system boundary, allocation), by name.
Li et al. (2025). Real-time greenhouse gas emission intensity informed demand-side load regulation for power grid decarbonization. Cell Reports Sustainability 2, 100367.
Earlier lessons: A1 (cash flows and annual equivalents), A11 (boundaries and life cycles), A4 (demand charges), A10 (which emissions question), A12 (credits and transfers).

Spotted an error, or stuck? Write to hlli@ksu.edu. Corrections are made in place and dated here.

Text licensed CC BY 4.0, code MIT: teachers and students may reuse and adapt it, with credit to Honglin Li.