Haute Lumière

Commerce · IV.09 · MMXXVI · daylight

La Bourse  /  Volume IV  /  Nº IV.09  /  Workbook — the executive

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter IV.09 · Design for the Seventh Generation

For the person with a P&L, a signature limit, a board and a quarter. This workbook uses the language of the firm without apology, because the firm's own asset register already supports most of what follows — it has simply never been arranged to show it.

---

THE PREMISE, STATED COMMERCIALLY

You are being asked to do one thing: stop letting a hurdle rate silently choose the design life of your fixed assets, and start funding maintenance as a corpus rather than as an intention.

That is not a values proposition. It is a capital allocation observation with a treasury consequence, and it has two numbers behind it.

The first: over a two-hundred-year study period, a structure designed for forty years costs £288.00m in cash and one designed for two hundred costs £117.00m — a saving of £171.00m and a factor of 2.46. The second: at a 7 percent hurdle those same two cost £48.51m and £55.25m, and the durable one loses by £6.74m. Nothing physical changed. The reversal is worth £177.74m and it happened in a spreadsheet cell.

The indifference rate is 4.510 percent. Your hurdle rate is almost certainly above it, which means your organisation has a standing, unexamined policy of buying the short-lived version of everything. Nobody decided that. It is what the template does.

Ninety days is enough to put the design life on the register, adopt a declining schedule beyond year thirty, get one spending rate published by the board, and endow one asset properly — at a corpus of 13.3 percent of that asset's capital cost. None of it competes for capital with anything else, and all of it is precedented.

---

PART ONE — DISCOVERY

Days 1–30

Exercise 1.1 — The design life sweep (one week, with your engineering lead and your controller)

Do not commission a study. Pull the fixed asset register and add one column: design life, in years, with its source.

Sources, in order of preference: the code category (Eurocode EN 1990 Table 2.1 — 50 years for buildings and common structures, 100 for monumental buildings, bridges and civil works; AASHTO's 75 for US bridges; UK practice at 120); the manufacturer's stated life; an engineer's written judgement with a name attached. Any of the three. What is not acceptable is a blank, and most of your register will be blank.

Then add a second column: the depreciation life you are using. Sort by the difference.

Every large gap is a story, and the two directions mean opposite things. An asset physically good for sixty years written off over fifteen is a systematically understated asset base and a systematically overstated cost of capital employed. An asset written off over thirty that will physically fail at eighteen is an unprovisioned liability with a date on it.

Output: the register with two new columns, and the ten largest gaps in both directions on one page.

Exercise 1.2 — Find where you already do this (two hours)

Your organisation almost certainly already holds a corpus against a future obligation and does not think of it as an instrument. Look in four places:

1. Decommissioning and restoration provisions — mining, energy, telecoms and utilities carry them by law. Somebody in your building already knows how to size an obligation over a fifty-year horizon.
2. Pension scheme funding — the actuarial machinery for a sixty-year promise is in-house, and it is the same machinery.
3. Concession handback reserves, if you hold any PPP or DBFO contract. The residual-life condition test at transfer is exactly the trigger this chapter recommends.
4. Warranty and service reserves.

You are not introducing a new financial concept. You are moving one you already use from liabilities you were compelled to provision to assets you were not. That is a materially easier argument and it should be the first slide.

Exercise 1.3 — The appreciative board conversation (one session)

Before you propose anything, change one question in one meeting. Replace "what is our maintenance backlog?" with:

"What do we own that has already outlived the people who specified it — and what did they do that made that possible?"

Record what comes back. You are building the evidence that the organisation already knows how to do this.

---

PART TWO — THE ARITHMETIC

Days 31–45

Exercise 2.1 — Run your own two cases (one week)

Take one real asset facing a renewal decision in the next three years. Build two options at the same service level: the conventional specification, and a specification at the next design-life category up. Get the capital premium from your engineers — the chapter's worked case uses 30 percent, and yours will differ; use yours.

Produce four columns, and all four go in the paper:

1. Cash, undiscounted, over a study period of four or five conventional lives.
2. Present value at your hurdle rate.
3. Present value at your national public appraisal rate.
4. Present value on the Green Book declining schedule — 3.5 percent to year 30, 3.0 to 75, 2.5 to 125, 2.0 to 200.

The chapter's case gives £288.00m / £48.51m / £65.69m / £68.97m for the short-lived option against £117.00m / £55.25m / £59.63m / £60.48m for the durable one. Two of your four columns will say one thing and two will say the other, and that is the finding. A paper reporting one column has reported one cell of its own answer.

Exercise 2.2 — Compute your indifference rate (one afternoon)

Bisect until the two present values are equal. The chapter's case gives 4.510 percent. Put your number on the front page beside your hurdle rate and let the committee look at both.

This is the exercise that changes the meeting. You stop being the person arguing for quality and become the person who has found that a £177.74m swing turns on a parameter nobody in the room set deliberately. Those two people get very different hearings.

Exercise 2.3 — The replacement liability, stated properly (two hours)

Compute the present value of your short-lived option's replacements: at 0 percent £160.000m, at 3.5 percent £13.462m, at 7 percent £2.862m — 1.79 percent of face.

Put this in the paper with one sentence: at our hurdle rate, buying this asset four more times costs less in present value than painting the durable alternative. It is true, it is arithmetic, and it is the moment the discussion stops being about sustainability.

Exercise 2.4 — Price the option you would be foreclosing (one session)

Honest work, and it must be done or the paper is advocacy. State three numbers: the probability a materially better answer exists by the mid-point of the durable asset's life; the service-value gain if it does; the switching cost. The chapter's declared assumptions — 40 percent, 35 percent of capital, 25 percent of capital — give a net £4.00m and an option value of £0.316m on the declining schedule, against a durable-case advantage of £8.48m. Net £8.17m, and durability survives.

Then run it the other way: the foreclosed option would have to be worth £107.5m — 2.69 times the asset's own capital cost — to overturn the case. That is your falsification test, and the answer is not always no. Elwha Dam stood 99 years and its removal cost about USD 11.61m per MW of the 28 MW it carried, roughly four times what building comparable capacity costs. Any asset whose design life exceeds the confidence interval on its own purpose should be designed for removal, with the removal costed in the same paper.

---

PART THREE — DESIGN

Days 46–60: the instrument

Exercise 3.1 — Get one number published by the board (one session)

The board publishes a real spending rate for maintenance corpora, once, as a policy parameter, the way it sets an insurance retention. Norway's fiscal rule is the citation: transfers capped at the expected real return, cut from 4 percent to 3 in 2017.

3.0 percent is a defensible default and the argument takes four minutes. What matters is not the level but that it is set by the board rather than by whoever is writing the paper that spends it.

Exercise 3.2 — Size and raise one corpus (two weeks)

  corpus = annual real maintenance cost / real spending rate

The chapter's durable case: £0.208m / 0.030 = £6.933m, which is 13.3 percent of the asset's capital cost, for a total year-zero commitment of £58.933m against £288.00m of cash over the same two centuries — a ratio of 4.89 times.

Raise the corpus in the same approval as the capital, as one line. Not a second paper, not a later paper. The Chorley Formula has run this way at the National Trust since 1946: a property is accepted only with an endowment sufficient to maintain it in perpetuity.

Have the identity ready, because your CFO will ask why the corpus costs more than the discounted value of the flow:

  corpus / PV  =  hurdle rate / spending rate
     7.0% hurdle, 3.0% spend   ->  2.333x

That is the entire answer. It is the price of making the promise real rather than intended, and it is two published numbers and nothing else.

Exercise 3.3 — Draft the covenant and the trigger (one week)

Three clauses and the deal is done.

• The indexed obligation. The operator's annual maintenance duty, with the index named in the document. Unindexed, it buys 29.1 percent of its original work by year fifty at 2.5 percent inflation, and 8.5 percent by year one hundred. One sentence.
• The condition survey. A named independent engineer, a fixed cycle, a defined residual-life measure. Florida's post-Surfside regime is the working precedent: milestone inspections at thirty years — twenty-five within three miles of the coast — and structural integrity reserve studies with mandatory funding from 31 December 2024, the previous right to waive reserves removed.
• The handback test. Where a third party operates, tie a final payment to the measured residual life at transfer. That converts a future condition into a present cash flow for the party who controls it, which is the only mechanism in this chapter that works on somebody else's incentives.

Exercise 3.4 — Where the liability has a date, use a sinking fund instead (one afternoon)

  payment = target x i / ((1 + i)^n - 1)

£40.0m at year 40: £1.000m/yr at 0 percent real, £0.818m at 1, £0.662m at 2, £0.530m at 3. At 2 percent the fund earns £13.51m of the target and the payment is 1.66 percent of capital cost per year — an operating line, not a capital event.

---

PART FOUR — DESTINY AND DELIGHT

Days 61–90

Exercise 4.1 — Put the four numbers on one page of asset policy (one week)

Design life basis. Discount schedule beyond year thirty. Real spending rate. Indexation basis. Four numbers, one page, reviewed annually, signed.

They must not live in a spreadsheet template, because a number that lives in a template is owned by whoever last edited it, and within two cycles it is a negotiating position rather than a parameter.

Exercise 4.2 — Write one succession note (one session)

For the asset you have just endowed: drawings in a format that will still open; materials and where they will come from; skills and who will still have them; money with its instrument named. One page. Lodge it with the trust deed.

Ise Jingu's succession note is a two-hundred-year forestry plan begun in 1923 to grow hinoki cypress for rebuildings from about 2130 — now 103 years old with 97 still to run, for a shrine rebuilt entirely every twenty years. The institution is 1,336 years old and no timber in it is more than twenty. Procurement with the lead time stated correctly.

Exercise 4.3 — Tell one true story to one person (one week)

One page to whoever controls the next allocation: the asset, its design life, the four columns, the indifference rate, the corpus, and the ratio. Not the board, not the all-hands. One person, one page, one number — and the number is the last line of the decision inequality below.

---

THE FAILURE MODES, NAMED

So you can see them coming

The spending rate drifts up. A corpus is only perpetual at a rate below its real return. Moving from 3.0 to 4.0 percent lets the same £6.933m fund £0.277m a year, which feels like a gift and is a slow liquidation. Defence: the rate is a board parameter with a date, reviewed against realised real return, never set in the paper that spends it.

The covenant is not indexed. By year fifty it is buying under a third of its work. Defence: one sentence, never omitted twice.

Durability is used to avoid a decision. The Elwha dams were excellent engineering aimed at a question that stopped being the right question within one lifetime. Defence: write the demolition into the birth certificate — costed, provisioned, in the same paper.

The register column goes stale. Design life is entered once and never revisited after a change of use. Defence: it is refreshed at every condition survey, by the same engineer who signs the survey.

---

THE NINETY DAYS ON ONE PAGE

---

BOARD PAPER TEMPLATE

Title. Design life and maintenance funding: [asset].

The decision. Approve capital of £X and a maintenance corpus of £Y, held in trust at a real spending rate of Z percent, with an indexed maintenance covenant and a condition survey on an N-year cycle.

Front page, five lines.

1. Design life, in years, with source. Study period, in years.
2. Whole-life cost, in cash, both options.
3. Present value at our hurdle rate, at the public appraisal rate, and on the declining schedule. Three cells, not one.
4. The indifference rate, and our hurdle rate, side by side.
5. The decision number:

   capital  +  (annual maintenance / real spending rate)
   ---------------------------------------------------------   <  1
      undiscounted whole-life cost of the replacement cycle

For the worked case: (52.000 + 6.933) / 288.00 = 0.205.

The honest negative, in the paper. The option foreclosed, priced with declared assumptions, and the value at which the case reverses. Plus the removal cost, if this asset ever stops being the right answer, and who holds it.

What we are not asking. Nobody is being asked to believe in two hundred years. They are being asked for one sum today, against a schedule of sums that are already in somebody's budget line, four times over.

---

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

1. What do we own that has already outlived the people who specified it, and what did they do?
2. Where do we already fund a future obligation as a corpus rather than an intention — and who built that?
3. Which of our assets could be handed to a stranger tomorrow with everything they would need?
4. If every capital paper carried design life beside the rate, what would we approve next year that we would not approve this year?
5. If the board published one spending rate, what would become arguable that is currently only assertable?
6. Which of the things we are building now might stop being the right answer within fifty years, and what would it cost to take it out?