Haute Lumière
Commerce · IV.01 · MMXXVI · daylight
ASSESSMENT · Chapter IV.01 — Making Things That Add
Three instruments: a ten-point quiz, eight reflection questions, five essay prompts. The quiz checks comprehension rather than recall. The reflections are private and first-person. The essays are arguable from more than one side.
THE QUIZ — ten points
Four on recall.
1. State the four declarations that a net-positive claim must carry, and say in one line what each is for.
Boundary — which processes are counted. Baseline — what the thing is compared against, in the same units, dated and sourced. Counterfactual — what would have happened instead, written as a scenario. Unit — the functional unit and the service life it is held for. One mark for the four, one for noting that a claim missing any of them is not a claim but a sentence.
2. What is a functional unit, as ISO 14040 uses the term, and why is it not the product?
The quantified performance of a product system, used as the reference unit — a square metre of floor covering held for fifteen years, a kilowatt-hour delivered, a wash at 40°C. It is not the product because two different products can only be honestly compared on the function they deliver, not on their mass or their count.
3. Name the EN 15804 module groups and say what each covers.
A1–A3 product stage; A4–A5 transport to site and installation; B1–B7 the use stage, including B2 maintenance; C1–C4 end of life; D benefits and loads beyond the system boundary. Full marks require naming B2 specifically, since that is the module the chapter turns on.
4. What does ISO 14044 require of a comparative assertion disclosed to the public, and what does it forbid?
It requires critical review by a panel of interested parties rather than a single expert. It forbids presenting a weighted single score as an objective comparison. The 2020 amendments make clear that "disclosed to the public" covers advertising, packaging and marketing material, not only academic publication.
Four on application.
5. A supplier's brochure says its product is "carbon negative." You download the declaration and find A1–A3 only. Write the two questions you ask next.
First: what is the whole-life figure — the same product with A4, A5, the B modules over the reference service life, and C1–C4? Second: what end-of-life scenario does the declaration assume for the stored biogenic carbon, and does it hold if the product is incinerated rather than landfilled? Credit any answer that identifies the claim as boundary-dependent rather than false.
6. A colleague proposes reporting the company's avoided emissions net of its own scope 1, 2 and 3 inventory, to show a negative total. Diagnose it.
It is not permitted by the guidance it would be citing: the WBCSD's 2023 guidance treats avoided emissions as a separate ledger and rules out netting against the inventory. The two are computed differently — an inventory is backward-looking against a base year, avoided emissions are a forward-looking comparison against a hypothetical baseline — so they have no common denominator. The stronger answer notes that reporting both, separately, with the counterfactual named, is the credible version of what the colleague wants.
7. Two carpet tiles. Tile A: 2.8 kg CO₂e/m² at the gate, service life eight years. Tile B: 4.0 kg CO₂e/m² at the gate, service life twenty years. Which is better, and what is the trap?
Per square metre-year of service, A costs 0.35 and B costs 0.20 — B is better by nearly half, on the gate stage alone. The trap is comparing gate figures per square metre when the functional units differ in duration. The stronger answer adds that neither comparison is complete until the maintenance module is annualised across each product's own service life, which may reverse it again.
8. Why does the Verified Product Carbon Covenant exclude avoided-emissions credit from the price by covenant, rather than simply not using it?
Because an excluded term is auditable and an unused one is not. Writing the exclusion into the contract makes the price a function only of measured, verified inventory movement, which is what allows a verifier to attest it and a buyer's auditor to test it. It also aligns the instrument with the separate-ledger requirement rather than leaving the alignment to practice.
Two that require the arithmetic to be done.
9. A declaration gives A1–A3 = 2.760, A4 = 0.323, A5 = 0.238, B2 = 0.403 per year, C2 = 0.00710 and C4 = 0.104 kg CO₂e/m², and a reference service life of fifteen years, with B2 declared for one year. Compute the whole-life total and the share of it that is maintenance. Show your working.
B2 over the life:
0.403 × 15 = 6.045. Total:2.76 + 0.323 + 0.238 + 6.045 + 0.00710 + 0.104 = 9.477 kg CO₂e/m². Maintenance share:6.045 / 9.477 = 63.8 %; product stage2.76 / 9.477 = 29.1 %. The whole-life figure is 3.43× the cradle-to-gate figure. The point of the question is that the multiplier was printed on the same page as the rate, and the answer changes which department the problem belongs to.
10. A developer generates 10.0 TWh a year from offshore wind at a lifecycle intensity of 12 g CO₂e/kWh. Compute the avoided emissions against a coal baseline of 820 g/kWh and against an onshore-wind baseline of 11 g/kWh. State what the pair of answers shows.
Against coal:
(820 − 12) × 10.0 TWh = 808 g/kWh × 10¹³ Wh = 8.080 Mt CO₂e/yr. Against onshore wind:(11 − 12) × 10.0 TWh = −0.010 Mt CO₂e/yr. A spread of 8.090 Mt on the same physical asset. Full marks require stating the conclusion: the number is not a measurement of the wind farm, it is a measurement of the counterfactual chosen by the claimant — and the low case is below zero, which is the answer rather than an arithmetic failure.
REFLECTION — eight questions, for one person and a pen
These are not for a room. Write the answers by hand if you can; the slowness is the point.
- Think of a claim you have made about your own work whose boundary you chose without noticing you were choosing. What was outside it, and what would have changed if it had been inside?
- Where in your work is the equivalent of the maintenance module — the recurring, unglamorous, per-period cost that is much larger than the one-off you keep optimising?
- When did you last read a document all the way through that you were only supposed to skim? What did you find, and what did you do with it?
- Whose numbers do you accept without checking, and is that trust earned by their record or by the difficulty of checking?
- Recall a time you changed a definition rather than a result, and it worked. Were you honest with yourself at the time about which one you had changed?
- What is the counterfactual you tell yourself about your own career — the version of your life that did not happen? What work is that story doing?
- Something you make or specify is already better than it is credited for. What is it, and what would it take to measure the thing that makes it better rather than the thing that is easy to measure?
- Where have you avoided computing an implied cost per unit of benefit because you suspected you would not like the answer? Compute one of them now.
ESSAY PROMPTS — five
Each is arguable from more than one side. Each requires at least one source the chapter cites and at least one it does not.
1. Is boundary choice a methodological problem or a governance problem? The chapter shows the same square metre moving from −0.30 to +11.325 kg CO₂e/m² across six defensible boundaries. Argue either that this is a defect in the standards, curable by tightening ISO 14044 and EN 15804 into a single mandatory boundary, or that variable boundaries are a necessary feature and the cure is governance — signature, freezing and restatement, as the chapter proposes. Engage ISO 14044 directly, and one source on standard-setting or accounting policy change that the chapter does not cite.
2. The offset withdrawal. Interface discontinued its Carbon Neutral Floors programme in 2024, retiring a claim it had made since January 2019. Write the case that this was a loss — that a funded, verified offset portfolio delivered real mitigation and the withdrawal removed it — and then the case that it was a gain. Use West et al. (2023) on forest-conservation offset baselines, and at least one source arguing the value of the voluntary carbon market that the chapter does not cite.
3. Divestment as abatement. Ørsted reports a 98 percent reduction in generation intensity, from about 500 to 10 g CO₂e/kWh, and separately divested an upstream business producing 100,000 barrels of oil equivalent a day — on the order of 10.95 to 15.70 Mt CO₂ a year of eventual combustion, depending on the oil-and-gas split, which is not disclosed. Argue whether a company that sells a carbon-intensive business has mitigated, transferred, or done both — and what a disclosure regime should require of it. Use the Ørsted disclosures, and one source on carbon leakage or divestment effects that the chapter does not cite.
4. The vacuum cleaner problem. Held over its own declared service life, the largest impact of the worked product is maintenance at 63.8 percent, not manufacture at 29.1 percent. Argue either that product-level declarations should require use-stage modules annualised across the reference service life as a condition of publication, or that use-stage impacts are properly the building operator's and belong in an operational assessment rather than a product one. Use EN 15804 and the declaration itself, and one source on building operational carbon that the chapter does not cite.
5. When the cheapest tonne is the wrong tonne. The chapter computes an implied abatement cost of $717/t CO₂e for the lower-carbon specification, against internal carbon prices of $50 to $250, and notes that a maintenance change delivers 0.74× as much at no capital cost. Argue the case that procurement should rank purely by implied $/t across the whole boundary — and then the case that it should not, because early premiums fund the learning curve that makes later tonnes cheap, and because supplier capability is not purchasable at the moment it is needed. Use the chapter's Interface material, and one source on experience curves, learning rates or green premiums that the chapter does not cite.