Haute Lumière
Commerce · VII.11 · MMXXVI · daylight
For the person with a P&L, a board, a set of auditors and a finance function that has heard enough about the long term. This workbook does not ask you to care more. It asks you to produce one page your competitors cannot produce, at a cost of about four analyst-weeks, and shows you where your own numbers already support it.
You are already running an intergenerational balance sheet. You publish parts of it every year: the fixed asset register, the pension scheme note, the net debt position, the decommissioning provision, the impairment tests. What you do not publish — what nobody publishes — is the consolidated per-unit position across all of them at once, and the one parameter that decides its sign.
The business case is not ethical. It is three commercial facts.
One: your cost of capital already prices this and prices it badly. Lenders and rating agencies are applying an implicit substitutability assumption to your firm because you have not stated an explicit one. An implicit assumption is always conservative, because the person making it carries the downside and none of the upside.
Two: the disclosure regimes are converging on the account and not on the narrative. Everything that has happened in sustainability reporting over the last decade has moved in one direction: from statements towards balance-sheet items with named methods and named verifiers. A firm that has already built a consolidated per-unit account is ahead of a requirement rather than behind one, and being ahead of a requirement is worth real basis points at exactly the moment everyone else is scrambling.
Three: it is cheap. The UK-scale version of this account costs about £0.0047 per person per year to verify. At the scale of a firm, it is four analyst-weeks for the first edition and about one for each one after.
Exercise 1.1 — The five lines, from your own statements (one day)
Pull them from documents you have already published. Do not commission anything.
| Category | Where it already is |
|---|---|
| Debt | Net debt note; maturity ladder; the externally held share |
| Pension promises | Scheme note — the liability, the discount rate, and the movement |
| Produced capital | Fixed asset register, net book value, and remaining useful life |
| Funds set aside | Decommissioning, restoration and rehabilitation provisions |
| Natural position | Scope 1 and 2 emissions; land holdings; water; any restoration liability |
Then divide each by the same denominator. Per employee works for a services firm; per tonne, per hectare or per unit of output works better for anything extractive or industrial; per share works for the equity story. Choose one and hold it, because the whole value of the exercise is that seven numbers become comparable for the first time.
Exercise 1.2 — The consolidation (half a day)
For each liability, ask the two-sided question: who holds the matching asset, and are they inside or outside the boundary? Intercompany debt nets. A funded pension scheme's liability is matched by its assets and the net position is what crosses the boundary. A supplier prepayment is somebody's receivable.
The UK sovereign version of this move is the cleanest teaching case: of £39,941 per head of public debt, £11,183 is external and £28,757 is one Briton's claim on another. Roughly three-quarters of the most-quoted intergenerational liability in British public life is internal. Find the equivalent in your own numbers and you will usually find it is a larger proportion than the board believes.
Exercise 1.3 — The rate-sensitivity audit (half a day)
Take your pension note and find the movement attributable to the discount rate alone. The UK Whole of Government Accounts moved from £2,631 bn to £1,375 bn in twelve months — £1,256 bn, about £18,580 a head — without one pension promise being changed.
Compute the same ratio for your scheme, and put it on one line. That line is the most useful piece of risk disclosure most firms could add this year, and it costs nothing because the number is already in your actuary's report.
Exercise 2.1 — ΔM for the firm (two days)
Take your net operating assets per unit at two points a generation apart — or, if the firm is younger, at founding and now — and deflate. You are asking: is the productive endowment per unit larger than the one this management inherited?
Run the UK national version first as a method check: £174,556 now against £90,276 then in constant prices, a delta of £84,280 per head and a compound real rate of 2.47 percent a year.
Then apply the discipline that makes the number usable: move the earlier figure by ±20 percent and see whether your conclusion survives. The UK's does — ΔM runs from £66,225 to £102,336 and never changes sign. If yours does not survive, you have learned something more valuable than the original number.
Exercise 2.2 — The revaluation switch (one day)
This is the corporate form of the chapter's land switch, and it is where boards most often flatter themselves. Separate, explicitly:
In the UK national accounts the second is 78.2 percent of the headline transfer. In a property-heavy, land-heavy or brand-heavy firm it can be more. Report both. A board that cannot see the split is measuring the market, not the management.
Exercise 2.3 — ΔN and the UNMEASURED register (two days)
Price what you can and register what you cannot.
Cumulative emissions over the period, at three carbon values, not one: the UK figure is 15.64 GtCO₂e, which at £269 a tonne is £62,228 per head, at £120 is £27,760, at £50 is £11,567 and at the government's own 2050 value of £378 is £87,443. Use your own internal carbon price as one of the three and say so — that is a disclosure in itself.
Then the register. Every natural stock the firm draws on that you cannot price: name it, name the unit it would be measured in, and write UNMEASURED. Do not write zero. A zero is a claim; the register is a statement about evidence, and an auditor can tell the difference even when a reader cannot.
Exercise 2.4 — θ* and the board table (half a day)
IGBS(θ) = ΔN + θ · ΔM θ* = −ΔN / ΔM
Compute θ including and excluding revaluation, at each carbon value. Build the four cells. The UK's are +£22,052, −£62,228, −£43,863 and −£62,228; θ is 0.74 with land and 3.39 without; and θ* reaches 1 at £364 a tonne with land and £79 a tonne without.
Then do the thing that makes this a governance instrument rather than a report. Put θ on one slide and ask the board a single question: is our working assumption about substitutability above or below this number?* You will discover that the board has never been asked, that its members disagree, and that they have been voting on capital allocation for years on the strength of an unexamined answer. That meeting is the deliverable.
Exercise 3.1 — Write it. Six pages, no appendix (two days)
Exercise 3.2 — The pre-mortem (90 minutes)
Before it goes: assume it failed and write why. The four standard causes, from the chapter, are that the producer picked θ quietly, that the unmeasured term was set to zero, that the account was used as a score, and that the schedule question was never asked. Each of the four is prevented by disclosure. Say in the paper which of the four you are most at risk of and how you have closed it.
Exercise 3.3 — The instrument (one day)
Take the covenant bond structure from the chapter to your relationship bank's sustainable finance desk. On a $2,000 m ten-year note at 5.00 percent, a 25 bp two-way step is $5.0 m a year, $50.0 m undiscounted, and $38.6 m in present value; a 15 bp step on Uruguay's model is $23.2 m. Verification at $400 k a year is $3.1 m in present value, leaving the covenant net-positive at $35.5 m.
Say the honest negative before the analyst does. That step is 0.0022 percent of the £1,400 bn the UK's own Climate Change Committee puts on net zero to 2050. The instrument finances nothing. What it buys is a priced, dated, externally verified commitment to publish a number that does not currently exist anywhere. Sell it as that and it is cheap and defensible. Sell it as climate finance and the first analyst who divides two numbers will not read page two.
Four analyst-weeks, spread across a quarter, with a named owner and an artifact at every checkpoint. Nothing here needs new capital and nothing needs a mandate you do not already have.
| Day | Action | Owner | Artifact |
|---|---|---|---|
| 1–10 | Pull the five lines from published statements | FP&A analyst | Seven sourced lines, one denominator |
| 11–20 | Consolidate; identify what is external to the boundary | Treasurer | The consolidated ledger |
| 21–30 | Rate-sensitivity audit on the scheme note | Actuary liaison | One line, in cash |
| 31–45 | ΔM in constant prices, with the ±20% sweep | FP&A analyst | ΔM, and whether it survives |
| 46–55 | Split capital formation from revaluation | Controller | The two-column table |
| 56–70 | ΔN at three carbon values; UNMEASURED register | Sustainability lead | A lower bound, labelled |
| 71–80 | θ* and the four cells | FP&A analyst | One slide |
| 81–85 | Pre-mortem; hostile read by somebody who did not build it | Internal audit | A corrected page |
| 86–90 | Board paper; the θ question put and answered | CFO | A minuted working θ |
The one checkpoint that cannot be skipped is day 81. The account must be read by somebody who did not build it and who has an interest in finding the quiet switch. This is not review and it does not block anything — the paper goes either way — but every serious error in this class of work is invisible to its author, because the author is the one person who cannot see the choice they made without noticing they were choosing.
Budget. Four analyst-weeks for the first edition, about one for each subsequent year, plus verification. At the sovereign scale verification runs about $400 k a year and is $3.1 m in present value over a ten-year term; at the scale of a single firm it is a line item in the existing internal audit plan and frequently costs nothing incremental at all.
What you will actually find, based on what everyone finds. Three things, in this order: that a larger share of your headline liabilities is internal to the boundary than the board believes; that a larger share of your asset growth is revaluation than the board believes; and that nobody — including you — has ever stated the substitutability assumption on which the last several years of capital allocation silently rested. The third of those is the finding. The first two are what make the room willing to hear it.
Five places, all of them inside documents you have already signed.
| Now | In two quarters | |
|---|---|---|
| We can state our consolidated intergenerational position per unit | ||
| We separate capital formation from revaluation in what we report | ||
| We know which of our liabilities are external to the boundary | ||
| We disclose what our pension discount rate did, in cash | ||
| We price the natural position at three values, not one | ||
| We keep an UNMEASURED register and never substitute zero | ||
| The board has been asked its working θ and has answered | ||
| θ* is on one slide in the capital allocation pack | ||
| Somebody outside the producer verifies the account | ||
| We have asked when our transfers arrive, and to whom |
Not the account. θ\*.
A single number, on the front page, that says exactly what a person would have to believe for this firm's position to be positive. It ends the category of argument in which everyone agrees on the figures and disagrees completely without being able to say why — and it does it without asking anybody to change their mind, which is why it survives a change of chief executive.
If θ is below your board's working assumption, you have a disclosure that strengthens your credit story and you should publish it. If θ is above 1, no belief anyone in the room holds clears the ledger — and that is not a reason to withhold the account. It is the reason the account exists, and a firm that publishes it first will be treated as the one that understood the problem rather than the one that had it.