Haute Lumière
Commerce · IV.09 · MMXXVI · daylight
For one person, a notebook and a term. Everything here is done on things you can actually reach — your own possessions, your own building, your own town, your own family's records. Nothing requires an organisation, a budget or permission.
Most exercises about the long term ask you to imagine the future. This one asks you to go and find the past, because the evidence for long-horizon design is already standing in your street and almost none of it has been costed.
You are going to do four things across a term. You will learn to read a design life off an object. You will compute, by hand, why a discount rate destroys the case for durable things — and why it destroys the case for replacing them even faster. You will build one endowment, for real, at a scale you can afford. And you will write a succession note for something you will hand on.
By the end you will be able to sit in a meeting in fifteen years' time and say, in one sentence, what a proposal's design life is, what rate is annihilating it, and what instrument would fix that. Very few people in that meeting will be able to do it.
A warning about the chapter's first movement, and take it seriously. You are going to be tempted to quote the seven-generation principle. Before you do, go and read Parker's 1916 rendering of the Great Law and see for yourself that the word seven is not in it, and read what Oren Lyons actually says. A student who learns to check one citation properly has learned more from this chapter than one who memorises every figure in it.
Exercise 1.1 — The object census (one afternoon)
Take twenty objects you own. For each, write three columns: what you paid, how long you expect it to last, and how long the previous one lasted. Most people have never put those three numbers side by side for anything.
Then rank by cost per year of service, not by price. You will find at least two inversions — a cheap thing that is expensive and an expensive thing that is cheap — and those two inversions are the whole chapter in miniature.
Exercise 1.2 — Read a code (two hours)
Find Eurocode EN 1990, Table 2.1, or your jurisdiction's equivalent, and copy out the categories by hand: 10 years temporary; 10–25 replaceable structural parts; 15–30 agricultural; 50 buildings and other common structures; 100 monumental buildings, bridges and civil works. Add AASHTO's 75-year US bridge practice and the UK's 120.
Now walk to the nearest bridge, the nearest school and the nearest block of flats and write down, for each, which category it is in, when it was built, and how many years of its design life remain. Three buildings, twenty minutes, and you have done something most professionals in your future field have never done.
Exercise 1.3 — Find the oldest working thing near you (one week)
Not the oldest thing — the oldest thing still doing its job. A bridge, a wall, a pump, a clock, a tool, a tree in a planted avenue. Find out three facts: who paid for it, who has maintained it since, and where that maintenance money comes from now.
The third question is the one that matters and it is usually the hardest to answer. Write down what you find and what you could not find out. Naming the gap is part of the work.
Exercise 1.4 — Check the citation (one evening)
Read Parker's The Constitution of the Five Nations (1916), the duties of the Confederate Lords. Write out the clause by hand. Then find three commercial or policy documents that quote a "seven generations" rule and note, for each, what they cite. Keep the list. It is your first piece of primary research and it will change how you read everything else this year.
Exercise 2.1 — Build the two cases by hand (one session, then a spreadsheet)
Case A: £40.0m capital, 40-year life, £0.240m a year maintenance, £8.0m refurbishment at year 20 of each cycle, rebuilt at years 40, 80, 120, 160. Case B: £52.0m capital, 200-year life, £0.208m a year, £7.80m at years 50, 100, 150.
Compute the two totals in cash first: A is £288.00m, B is £117.00m. The durable one saves £171.00m and costs 2.46 times less.
Now discount at 3.5 percent: £65.69m against £59.63m. At 7 percent: £48.51m against £55.25m, and the answer flips. Write down, in one sentence of your own words, what physically changed between those two calculations. The answer is nothing, and you should be able to say why.
Exercise 2.2 — Find the indifference rate yourself (one hour)
Do not look it up. Bisect it: try 5 percent, try 4 percent, narrow. You will arrive at 4.510 percent. The exercise is not the number. It is the experience of watching a physical question turn out to have an answer that depends on a parameter nobody in the room chose deliberately.
Exercise 2.3 — The rebuild table (twenty minutes)
Compute the present value of four £40.0m rebuilds at years 40, 80, 120, 160:
| Rate | PV | As % of face |
|---|---|---|
| 0.0% | £160.000m | 100.00% |
| 3.5% | £13.462m | 8.41% |
| 7.0% | £2.862m | 1.79% |
Then write one sentence beginning "A discount rate does not only…" and finish it in your own words. If you can say this out loud without notes, you understand the chapter.
Exercise 2.4 — The seventh generation, priced (thirty minutes)
Compute the present value of £1,000,000 arriving in year 175 at 7.0 percent (£7), 3.5 percent (£2,429), Stern's 1.4 percent (£87,772), and on the Green Book declining schedule (£10,184 — 4.19 times the flat 3.5 percent).
Then compute ln 2 / 175 = 0.396 percent. Sit with it. Above four tenths of one percent a year, the seventh generation is worth less than half of one of us. That is arithmetic, not opinion, and it is why this argument is almost never won in present value.
Exercise 2.5 — Price a foreclosed option (one session)
Take something in your own life that you committed to for a long time — a course, a lease, an instrument, a city. Estimate three numbers: the probability that a materially better option would appear within the commitment period; the gain if it did; the cost of switching. Multiply the first by the difference of the other two and discount it to the day you decided.
That number is what the commitment cost you beyond its price. Most people have never computed it once. Do it honestly, including the cases where the number is small and the commitment was right.
Exercise 3.1 — Endow something, for real (one month)
This is the exercise that changes people, and the scale does not matter.
Pick a small recurring cost you want to survive you: the annual upkeep of a grave, a tree, a bench, a subscription to something a library needs, a tool library's insurance. Establish the annual real cost M. Choose a real spending rate s you can actually defend — 3.0 percent is the number Norway uses for a sovereign fund and it is a reasonable student default.
corpus = M / s
At £50 a year and 3.0 percent, the corpus is £1,667. At £200 a year, £6,667. Write down the number. If you can reach it, reach it — over years if necessary, on the sinking-fund schedule in Exercise 3.3. If you cannot, you have still done the thing that matters: you have converted a wish into a figure, and a figure can be worked toward.
Exercise 3.2 — Compute the identity (fifteen minutes)
For your own endowment, compute the present value of the same flow at a commercial hurdle rate of 7 percent, and divide the corpus by it. You will get 2.333, which is exactly 0.07 / 0.030.
Write this in your notebook and do not forget it: the premium for making a perpetual promise real, over its discounted value, is your hurdle rate divided by the spending rate. It is one of the few exact identities in applied finance and it will be useful to you for the rest of your life.
Exercise 3.3 — Build a sinking fund schedule (one hour)
Pick a replacement you know is coming — a bicycle, a laptop, a boiler, a roof. Target R, years n, real return i:
payment = R x i / ((1 + i)^n - 1)
For £40.0m at 40 years: £1.000m at 0 percent, £0.818m at 1, £0.662m at 2, £0.530m at 3. At 2 percent the fund earns £13.51m of the £40.0m and the payment is 1.66 percent of capital cost per year. Run the same three columns at your own scale, and note how much of the target the returns do. That column is the reason to start early rather than large.
Exercise 3.4 — The indexation test (twenty minutes)
Take any fixed annual amount in your life — a standing order, an allowance, a membership. Compute what it buys in real terms at years 30, 50 and 100 at 2.5 percent inflation: 47.7 percent, 29.1 percent, 8.5 percent. Then find one real document near you — a lease, a trust deed, a charity's constitution — and check whether its money figures are indexed. Most are not.
Exercise 4.1 — Write a succession note (one session)
For one thing you intend to hand on, write a single page naming four items: the drawings — the instructions, in a format that will still open; the materials — what it is made of and where more comes from; the skills — what someone needs to know and who still knows it; the money — the instrument and its number.
Ise Jingu's succession note is a forest planted in 1923 and a rite performed every twenty years. Yours may be a paragraph and a bank account. The form is the same.
Exercise 4.2 — Find the interval (thirty minutes)
For the thing you are handing on, choose the renewal interval deliberately. Shorter than institutional memory means the knowledge travels in people: at Ise, every carpenter has worked alongside someone who did it last time. Longer, and it has to be reinvented each time at full cost. Write down your interval and the reason.
Exercise 4.3 — Write the demolition into the birth certificate (twenty minutes)
For one long-lived commitment you are making this year, write down now: what it would cost to undo, who would pay, and what would have to be true for undoing it to be right. The Elwha dams stood 99 years and cost about USD 325 million to remove — roughly four times per megawatt what building comparable capacity costs. Nobody wrote that down in 1913.
Exercise 4.4 — The delight exercise (one afternoon, no writing)
Go and stand in front of the oldest working thing you found in Exercise 1.3. Stay ten minutes. You are being addressed by someone who knew they would never meet you.
Then go home and make one small, unsupervised, entirely unnoticed improvement to something you own, for the benefit of whoever has it next. Extra cover on the paint. A label. A joint that can be opened. A copy of the instructions in the drawer. This is what the economics feels like from the inside, and nobody will ever thank you for it, and that is the point.
Choose one object, building, tree, instrument or institution near you that is intended to outlive its current custodians. Then produce a single document of eight to twelve pages containing:
The test of the finished project: hand it to someone who has not read this chapter and ask them to tell you, from the document alone, whether the thing should be maintained or replaced, and why. If they can, it is done.
Score each honestly, one to five.
Below 30 — work Parts One and Two again; they are the load-bearing ones. 30 to 40 — you can hold your end of the argument. Do the term project. Above 40 — teach it. Find one person who is about to make a long-lived decision and give them the two columns.
Three habits, and they cost nothing.
Ask for the design life, always. In every meeting, about every asset, in years, with its source. It is a polite question and almost nobody can answer it, which tells you where the room is.
Present cash and present value together, never one alone. The cash column is the one a maintenance budget experiences; the present value column is the one a committee experiences. Both are true. Showing one is advocacy.
Check the citation before you use it. You practised this in week four with the Great Law. Do it with every quotation you intend to repeat in public for the rest of your career, and you will be wrong far less often than the people around you.