Haute Lumière
Commerce · IV.11 · MMXXVI · daylight
For the person with a P&L, a board, and a sustainability commitment that is starting to collide with a gross margin. This workbook assumes both are real and treats the collision as a sizing problem rather than a values problem, because that is what it is.
Your firm has a set of regenerative substitutions in flight. Some of them will get cheaper. Some of them will not. Right now those two categories are being managed identically, financed identically, and reported identically — and that is the single largest source of write-offs in corporate sustainability programmes.
The commercial argument for this chapter is not ethical. It is that a misclassified cost destroys more value than the cost itself. A permanent premium funded on a payback case defaults, and the default is read internally as evidence that the whole programme was naive — which takes down the box-one and box-two items alongside it, and those were making money.
There is a second commercial argument and it is the better one. A cost that will not come down for you will not come down for an entrant either. Priced correctly and contracted early, a permanent premium is a barrier to entry that your competitors are currently declining to build because they are still waiting for it to fall. The arithmetic is in Part Two, and the answer is that a durable margin worth 40.4 per cent of a temporary one is worth exactly the same money.
Exercise 1.1 — The classification, in one afternoon (half a day, with procurement and finance in the room)
List every regenerative substitution the firm is considering, piloting or committed to. For each, one of three boxes:
The discipline that makes this exercise real: the box-two entry is only valid if it carries the volume at which crossover occurs and the mechanism by which the cost falls. "Scale" is not a mechanism. "Learning in sortation yield, crossing at 7.1× current throughput" is.
Exercise 1.2 — The learning-rate test (1 hour per box-three item)
For each box-three item, decompose the premium into cost terms and ask of each: what makes this fall?
Terms driven by cumulative experience fall. Terms driven by population density, real wages, land area or the price of an hour do not.
Worked, from the chapter: a virgin-to-recycled PET spread of £350/t looks closable — a 12 per cent learning rate against virgin's 3 per cent crosses after 2.84 doublings, about 7.1× volume. Then decompose. Kerbside collection at £1.10 per stop and 0.45 kg per stop is £2,444/t, which is 6.98 times the entire spread, and it is a driver and a vehicle at a door. Deposit return at 2.60 kg per stop is £423/t — 5.78× cheaper. Same polymer, same chemistry, two completely different investment cases, distinguishable only by decomposing the premium.
Exercise 1.3 — Find where you already pay one well (2 hours)
Somewhere in your firm a permanent premium is already being paid and paid well. Find it and interview whoever runs it. Look in: a supplier you have never re-tendered, a specification you have never relaxed, a plant that runs an expensive process because of a customer requirement, a materials choice that survives every cost programme.
Ask what they get for it, who holds the avoided cost, and what would make them stop. You are looking for the internal precedent, because a board approves a second instance of something far more readily than a first.
p* on your own numbersExercise 2.1 — The decision rule (2 hours per material)
p < p* = m + s + τ · λ / (λ + w)
p — the premium per unit, from an actual quote, not a market report.m — the realised revenue premium, from a live price test. A shelf, a basket, real abandonment. A stated-preference survey is not admissible and is the most common way this calculation is corrupted.s — avoided direct cost. Disposal, energy, compliance, warranty, insurance, rework. Get this from the plant, not from a model.τ — the levy per unit you will pay if you do not switch.λ — the hazard rate of that levy arriving. State it. 1/λ is your expected arrival date and everyone in the room will have an opinion, which is the point: it makes a regulatory expectation into a parameter rather than a sentence.w — your WACC.Worked, from the chapter: m = £60/t, s = £40/t, τ = £217.85/t, λ = 0.15 (6.67 years expected), w = 0.09. Then λ/(λ+w) = 0.625, credit £136.16/t, and p = £236.16/t. Against p = £350/t, p/p = 1.48×.
The answer is not "no." Full substitution is a donation of £113.84 per tonne. Partial substitution to the levy's 30 per cent threshold costs £105/t and extinguishes £217.85/t with certainty — a net gain of £112.85/t. Contract that tonnage. Leave the balance on the incumbent. Revisit the day τ rises or λ shortens.
Exercise 2.2 — The instrument's ceiling (15 minutes, and almost nobody does it)
For every levy, tax or incentive with a threshold, compute the maximum premium it can carry:
maximum premium = rate / threshold
UK Plastic Packaging Tax: £217.85 / 0.30 = £726.17 / t
Below the threshold the instrument is powerful. Above it, it pays nothing, because the tax is already avoided. A levy with a threshold buys exactly the threshold — so plan your capacity, your contracts and your public commitments against the threshold, not against the aspiration in the policy's preamble.
Exercise 2.3 — The moat calculation (30 minutes, for the board paper)
This is the number that changes the conversation from defence to strategy.
transient advantage £100 / unit / yr for 6 years
annuity factor at 9 % 4.4859
NPV £448.59 / unit
equivalent perpetuity £40.37 / unit / yr = 40.4 % of the transient
Read it into your own business. Every regenerative advantage that arrives with a crossover is competed away as imitators cross the same line — call it six years. A position built on a premium that will not fall cannot be waited out by an entrant, because they face the same premium. A durable margin needs to be worth only two-fifths as much to be worth the same money, and it does not require you to keep running.
Exercise 2.4 — Write the falsifier (30 minutes per item)
For every item classified permanent, write the observation that would change your mind, with a review date. A volume. A published spread. A wage-to-price ratio. A regulatory rate.
The case that earns this exercise is offshore wind. In 2012 the UK's own cost reduction task force set £100/MWh by 2020 as a stretch target. The 2015 auction cleared at £114.39/MWh. The 2017 auction cleared at £57.50 — 42.5 per cent below the target, 3 years early, and 37.8 per cent below the £92.50/MWh strike granted to Hinkley Point C the previous year. By 2019 it was £39.65, a real decline of 23.3 per cent a year. The firms that had written falsifiers moved first.
The allocation principle, for the board paper.
A permanent premium should be paid by whoever holds the avoided cost.
It converts a fairness argument into a question of fact, which is a far better kind of argument to have in a board meeting.
| Who holds the avoided cost | Instrument | Your move |
|---|---|---|
| The public | Procurement standard, levy | Engage the standard-setter early; supply the arithmetic |
| The buyer, provably | Price | Run the price test; carry the premium on the invoice |
| The incumbent substitute | Levy on the substitute | Support it publicly; you benefit from the level field |
| Nobody | The firm, openly | Book it as a donation, with an owner and a review date |
The instrument: a bilateral contract for difference on the material spread.
£320 × 12,000 = £3.84m/yr, £19.33m discounted at 9 per cent over the term. A contract whose worst case is not on the front page is not a contract.p. At £320/t against a p of £236.16/t the ratio is 1.355×, so the contract is written on the threshold tonnage only.Exercise 3.1 — Draft the term sheet (1 day with treasury and legal)
One page: volume, strike, reference index, settlement frequency, term, credit support, termination, and the worst-case exposure in bold. Then the audit memo on own-use, in parallel, not afterwards.
Exercise 4.1 — Get the classification into the pack (the whole game)
One table in the standing monthly pack: item, premium, box, payer, falsifier, review date. Anything reviewed monthly persists; anything reviewed by exception evaporates.
This table does more work than any presentation you will give, because it makes the next executive's decision legible without requiring them to share your convictions.
Exercise 4.2 — Name the payer, in writing
A premium paid by whoever happens to have budget is paid until the second bad quarter. A premium paid out of a line that would otherwise go on a levy, a landfill charge or a customer discount is paid for as long as the arithmetic holds — by somebody with a self-interested reason to keep paying it.
Exercise 4.3 — Take the donation to the board as a donation
One page, one number, one owner. We pay £113.84 per tonne above p on this material. It buys us the following, which we value for the following reasons. It is a donation, not an investment. It is £X a year. Here is who owns it and when it is reviewed.*
This paper passes far more often than the disguised version, and it does not take the credibility of your genuine investments down with it if it is later stopped.
Comfort classification. A premium called permanent because permanent means nobody has to keep trying. Countermeasure: the falsifier, and a named owner for the review.
All or nothing. The firm cannot justify full substitution so it does none, and forgoes a threshold tonnage that was genuinely profitable. Countermeasure: p* computed at the margin, and a contract written on the fraction that pays.
The survey. m taken from stated preference, p inflated, a donation presented as an investment, a write-off two years later. Countermeasure:* a live price test, or m = 0.
The threshold mistaken for an ambition. Capacity planned against a policy's preamble rather than its arithmetic. Countermeasure: rate divided by threshold, on the first page of every policy response.
The stale classification. The falsifier occurred, the review had become a formality, and the firm paid a premium that had stopped existing. Countermeasure: a date, in the pack, owned by finance rather than by sustainability.
| Day | Action | Artifact |
|---|---|---|
| 1–15 | Classify every substitution into three boxes | The signed classification |
| 16–30 | Decompose each box-three premium; find the term with no learning rate | The decomposition |
| 31–45 | Compute p* per material on your own numbers | The p* sheet |
| 46–55 | Run the live price test for m | The measured m |
| 56–65 | Compute each instrument's ceiling: rate ÷ threshold | The ceiling table |
| 66–75 | Draft the spread CfD on threshold tonnage; open the own-use memo | Term sheet + audit memo |
| 76–85 | Write every falsifier and review date | The falsifier register |
| 86–90 | Board paper: the classification, the donation, the moat | One page, one number |
Recommendation. Contract N tonnes at a strike spread of £X/t for seven years, and continue on the incumbent material for the balance until p* moves.
Why now. p* = £236.16/t on current inputs. The threshold tonnage returns £112.85/t with certainty against a levy of £217.85/t. The balance would be a donation of £113.84/t and is not recommended at this time.
What we are buying that a competitor cannot wait out. The premium does not fall for them either. An advantage that survives imitation is worth the same as a temporary one at 40.4 per cent of the annual margin.
Worst case. £19.33m discounted over the term, if the spread goes to zero.
What would change this. The levy rate rising, the hazard rate shortening, a measured revenue premium above £60/t, or the collection density term falling — reviewed each March, owned by the CFO.