Haute Lumière

Commerce · VII.01 · MMXXVI · daylight

La Bourse  /  Volume VII  /  Nº VII.01  /  Workbook — the executive

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter VII.01 · Planetary Boundaries as a Balance Sheet

For the person with a P&L, a signature limit, a board and a disclosure obligation that is arriving whether or not anybody in the building is ready. This workbook uses the language of the firm without apology, because the accounting treatments it needs already exist and are already audited — they have simply never been pointed at this.

---

THE PREMISE, STATED COMMERCIALLY

You are about to be asked to report against nature the way you report against capital, and the frameworks arriving on your desk mostly consolidate four incompatible kinds of evidence onto one page.

Your position is straightforward and it is defensible in front of any auditor: grade before you consolidate, and never sum across grades. Three of the nine planetary boundaries behave like covenants — level tests with step consequences. Two behave like stocks, which are impaired proportionally against specific assets. Two behave like flows, whose damage is local and whose global total carries almost no decision-relevant information. Two are expert judgements with published uncertainty bands wider than their own boundaries, and the correct treatment for those is disclosure, not recognition.

That is not a hedge. It is the difference between a statement an auditor will sign and one they will qualify, and a qualified nature statement is worth less than none at all.

The commercial prize is specific. A segmented statement supports a segmented facility, and a segmented facility prices each risk the way that risk actually behaves — which is the only way the pricing will ever be large enough to change a capital allocation.

---

PART ONE — DISCOVERY

What your firm already has

Exercise 1.1 — The grading sweep (one week, with your controller and your head of risk)

Do not commission a study. Pull the metrics you already report and grade them.

Three tests per metric.

1. Well-mixed. Does one number genuinely describe the whole of the system this metric is about, or is it a sum of places?
2. Directly measured. Is the control variable read by an instrument, or produced by a model whose assumptions you could not recite?
3. Demonstrated threshold. Is there a level at which behaviour changes, shown rather than inferred?

All three: A. A genuine measured stock without a demonstrated global tipping point: B. A measured flow whose damage is local: C. Anything else: D.

Output: every metric you currently report, graded, on one page. Most firms discover in this exercise that they have been managing two or three C-grade totals as though they were covenants, and one genuine covenant as though it were a reporting line.

Exercise 1.2 — Locate (two weeks)

The TNFD's core method begins with L, for Locate, and it is the most valuable instruction in the whole standard. Before evaluating anything, establish where your material assets and dependencies physically are: sites, catchments, airsheds, biomes, supply-chain origins.

This is the expensive part and it is a fixed cost paid once. Everything downstream — the asset register, the flow account, the borrowing base, the assurance scope — reuses it. Budget it as infrastructure, not as a report.

Exercise 1.3 — The appreciative board question (one session)

Before proposing anything, change one question in one meeting. Replace "what is our environmental exposure?" with:

"Where do we already report a genuine level test correctly, and who insisted on it?"

Record what comes back. You are establishing that the firm already knows how to do this in at least one place, which is a materially easier argument than the claim that it must learn.

---

PART TWO — THE ARITHMETIC

Days spent computing, not arguing

Exercise 2.1 — The band-width test (half a day)

For every externally set boundary you are asked to report against, compute the width of its published uncertainty band as a share of its boundary value.

  ozone                     5.3 %          forest cover        28.0 %
  aragonite saturation     12.4 %          nitrogen            32.3 %
  CO2 concentration        28.6 %          radiative forcing   50.0 %
  phosphorus, regional     80.6 %          intactness          66.7 %
  phosphorus, global      809.1 %          extinction rate    900.0 %

The four climate-and-chemistry bands average 24.1 percent of their own boundary, median 20.5 percent. The two biosphere bands average 483.3 percent. That is 20.1 times on the means and 23.6 times on the medians.

Take this number into the room. It is computed from the framework's own published tables, it is not an opinion about the science, and it ends the argument about whether these lines are the same kind of thing in about ninety seconds.

Exercise 2.2 — The misclassification cost (two hours)

Compute this for your own balance sheet, at your own scale. The chapter's worked version:

  position                        $500,000,000      illustrative
  overshoot                                5 %
  ---------------------------------------------------------------
  if it is a COVENANT             $500,000,000      accelerates
  if it is a STOCK                 $25,000,000      impairment
  ratio                                  20.0 x

Present that ratio to your audit committee before you present anything else. It is the argument for the grading, and it is an argument about provisioning rather than about the environment, which is the meeting you want to be in.

Exercise 2.3 — Stop reporting global totals you cannot act on (half a day)

Take every group-level environmental total in your pack and ask one question of each: what decision available to anyone in this building is a decision about this number?

For a global nitrogen or phosphorus figure the answer is almost always none. The chapter's division shows why: 190 Mt N/yr across 1,600 Mha is 118.75 kg N/ha/yr spread evenly and 593.75 kg N/ha/yr on a fifth of it — a factor of 5.0 — and both report the same total. Your group figure behaves identically. Replace it with throughput by catchment or airshed against that place's own allowable input.

Exercise 2.4 — Sensitivities that protect you (two hours)

Run every material claim at your central case and with the two most load-bearing assumptions moved 30 percent against you. Publish all three.

An analysis that only works at the central case is destroyed by the first person who wants it destroyed. The executive who brings all three unprompted is trusted with larger questions afterwards.

---

PART THREE — DESIGN

The four statements and the instrument

Exercise 3.1 — Build the four statements (three weeks, with your controller)

The rule that does the work: never sum across the four. A total that spans a covenant and an expert judgement is not a total.

Exercise 3.2 — The audit conversation (one meeting, early)

Two positions to open, neither of which requires your auditor to accept anything about the environment.

Useful economic life. Where your spend creates or improves a long-lived asset whose productive capacity is demonstrably rising, the depreciation schedule should not assume it is falling. Bring evidence of condition: soil organic carbon trend, yield stability, plant availability, defect rates.

Recognition versus disclosure. Grade D items fail the measurement-reliability test by a wide margin — a published band of 900.0 percent is not an estimate. IAS 37 disclosure is the correct treatment and your auditor will recognise the argument immediately, because it is the argument they make.

You will not close either conversation in one meeting and should not try. You are opening a file, not closing one. The file changes what is arguable next year.

Exercise 3.3 — Draft the segmented facility (two weeks, with treasury)

The market's default is a sustainability-linked loan with a margin ratchet, and it is worth knowing precisely why that will not do this job.

  drawn facility                   $400,000,000     illustrative
  market-norm ratchet                      25 bp
  --------------------------------------------------------------
  annual value of the ratchet        $1,000,000 / yr
  consequence it claims to price   $400,000,000
  ratio                                   400 x
  the ratchet is                         0.25 % of the consequence

A ratchet is not a price on a threshold. It is a rounding error wearing one. That single division explains most of the market's disappointment with sustainability-linked lending and it is the opening line of your term sheet discussion.

The structure that replaces it:

The decision inequality, on the front page:

   avoided impairment  +  margin benefit
   ---------------------------------------  >  1
     verification  +  restoration spend

Worked at mid-cap scale: avoided impairment $2,400,000 and margin benefit $1,000,000 against verification and location data at $340,000 and a restoration programme at $1,250,000 returns 213.8 percent, a net $1,810,000 a year. It clears before a single tonne of anything is restored, because most of what the spend buys is the documented right not to impair assets that were never impaired.

The counterparty. A relationship lender or a development finance institution for the first facility, not a syndicate. Segment A requires a lender willing to negotiate a real breach clause, and that is a two-party conversation. Syndicate on the second facility, once the schedule has a compliance history.

---

PART FOUR — DESTINY AND DELIGHT

Days spent making it hold

Exercise 4.1 — Into the pack (one conversation)

Get the covenant schedule onto the standing board pack — in the borrowings section, beside the financial covenants, not in a sustainability annex. Anything reviewed monthly persists; anything reviewed by exception does not. Where a number sits in a pack determines who argues about it.

Exercise 4.2 — Name the grader (one appointment)

One named person applies the three tests and signs the grades. Publish the tests and publish the dissent. Grades that nobody owns drift, and they drift in both directions: inflation toward A because A gets a meeting, deflation toward D because D requires nothing.

Exercise 4.3 — Refuse the composite (standing)

Somebody will ask for one number for the cover, and they are not being foolish — a cover needs a number. Give them the covenant schedule's worst headroom, which is a real number about a real threshold. Refuse the index. A composite across four evidence grades is the nine-wedge diagram again, with a decimal point.

Exercise 4.4 — Delight, for a firm (ongoing)

Make the covenant schedule one page and beautiful. Make the asset register something a site manager is pleased to see improve. Let the operating unit keep the whole benefit once the facility is repaid.

Every one of those is an adoption decision disguised as an aesthetic one, and the adoption is the mechanism.

---

THE FAILURE MODES, NAMED

So you can see them coming

1. Grade inflation. Everything drifts to A because A gets a meeting. Prevention: published tests, a named grader, and band widths computed from the source.
2. Grade deflation. Everything is pushed to D so nothing has to happen. Prevention: the same arithmetic run the other way — a band of 28.6 percent is not in the same class as one of 900.0 percent, and the grading must be evidence-bound in both directions.
3. Re-aggregation. A composite index reappears, usually from marketing. Prevention: give them the worst headroom instead, in writing, before they ask.
4. A ratchet instead of a covenant. The facility gets signed with a 25 bp adjustment and changes nothing. Prevention: the 400 times division, on page one of the term sheet.
5. Reporting a global total nobody can act on. Prevention: the "what decision is this a decision about?" test, applied to every group-level environmental figure in the pack.
6. Location data treated as a report rather than infrastructure. It gets funded once, used once and rots. Prevention: budget it as a platform with an owner and a refresh cycle.

---

THE FIRST NINETY DAYS ON ONE PAGE

---

BOARD PAPER TEMPLATE

Title. One line, commercial. "Segmented nature facility: 213.8 percent return on verified avoided impairment, self-funding." Not "Nature strategy update."

1. The number. The decision inequality, computed, first paragraph.
2. The grading, and why. Three tests, four grades, one page. Include the band widths — they are the evidence that the grading is not an opinion.
3. The covenant schedule. Three lines. Boundary, reading, headroom, trend, breach clause.
4. The asset register. By site. Gross, impairment, net. Name the two sites that surprised you.
5. The flow account. By catchment or airshed. State plainly that group totals have been removed and why.
6. The watch list. With grades and band widths shown. State that it is not consolidated into anything, and that this is deliberate.
7. The facility. Four segments, four consequences, and the 400 times division explaining why Segment A is a covenant rather than a ratchet.
8. What would make this fail. Three, honestly. This paragraph is why you will be believed.

Length: two pages plus the four statements. If the narrative is longer, you have not decided what matters.

---

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

1. Where do we already report a genuine level test correctly, and who insisted on it?
2. Which of our metrics is a covenant, which is a stock, which is a flow — and where have we been managing one as another?
3. Which group-level total in our pack corresponds to no decision anybody here can take?
4. Which of our assets is demonstrably improving in condition while we depreciate it?
5. If our board pack carried a three-line covenant schedule, what would the first meeting about it actually be about?
6. Where are we summing across places, and what would we see if we stopped?
7. What is the widest uncertainty band we currently publish, and what would it take to print it beside the number?
8. Which site would surprise us most if we reported condition alongside carrying value — and in which direction?
9. Who outside this firm would we want auditing our grades, and what would we have to publish for them to be able to?
10. What would have to be true for this grading to survive the first person who wants one number for the cover?
11. If a lender offered a real covenant instead of a ratchet, what would we attempt that we will not attempt today?
12. What is the first sign we would see that our grades were drifting, and who would notice it first?