Haute Lumière
Commerce · V.01 · MMXXVI · daylight
For the person with a P&L, a signature limit, an insurance programme and a risk committee. This workbook uses the language of the firm without apology, because the firm's own numbers already support most of what follows — they have simply never been arranged so that anyone could see them at once.
Your organisation already puts a price on a human life. It does it several times a quarter, in capital deferrals, specification changes, staffing levels, maintenance intervals and insurance retentions. The only open question is whether that price is chosen and written down, or left behind by a budget and discovered afterwards by somebody with subpoena powers.
That is the commercial case, and it has three limbs.
It is cheaper. The implied price left behind by an unexamined budget is reliably far below what the same organisation would defend in daylight. The cladding substitution at Grenfell Tower saved a recorded 293,368 GBP against seventy-two deaths: 4,075 GBP a life, 5,175 USD, against a transport-department figure of 13,200,000 USD — a factor of 2,551 x. Nobody chose it. The exposure created by not choosing it is not insurable retrospectively.
It is faster. Most safety capital is delayed not because it fails a test but because there is no test to fail. A published price turns a three-month discussion into one line: a guard-rail programme costing 2,400,000 USD that avoids 6.0 statistical deaths implies 400,000 USD per statistical life, 33.00 x inside policy, approved.
It has a cash flow. Insurance experience rating pays for it, and that is Part Four.
Exercise 1.1 — The five-room sweep (one week, with your controller and your broker)
Go looking for the places your organisation already prices a person, and bring them onto one page.
1. The wage band. Pull one role family and place three numbers beside each other: what the market clears at, your best estimate of the role's marginal revenue product, and what you pay. If the third is far below the second, that is not a secret advantage — it is a measured elasticity, and it will revert when the frictions holding it in place move. At an elasticity of 2.50 the wage sits at 71.4 per cent of product: a 62.00 USD product predicts 44.29 USD, a 17.71 USD hourly gap, or 33,657 USD a year per head.
2. The insurance programme. Ask your broker for the manual premium, the experience modifier and its five-year history. This is the file that turns everything in this workbook into a financeable proposition, and most executives have never seen it.
3. The deferral log. Every capital item deferred in the last three years that had a safety justification. You are not looking for wrongdoing. You are looking for the arithmetic that was left behind.
4. The turnover line. Voluntary exits by team. Cost one properly: a loaded salary of 58,000 USD with a replacement cost of 0.75 of salary is 43,500 USD an exit, and twelve avoidable exits a year is 522,000 USD. That is a human-capital number and it belongs in the same conversation as the training budget it is usually kept away from.
5. The claims file. What your organisation has actually paid in settlements, and on what basis. This is your own revealed price, and it is the one number in the sweep that a court has already seen.
Output: five numbers on one page, roughly costed, with their sources.
Exercise 1.2 — The appreciative question, asked out loud (one meeting)
Ask your operations leaders: "When did we spend money on safety here and it was clearly the right call — what made that decision possible?"
Take notes on the conditions rather than the outcome. You are looking for the sponsorship pattern that already works in your organisation, because you are going to reuse it rather than invent one.
Exercise 2.1 — Compute three implied prices (one morning)
Take three decisions already taken and recover the price each one implied.
implied price = money saved / statistical deaths accepted
A worked case to calibrate against: a site declining a retrofit costing 1,800,000 USD that would have avoided 0.45 statistical deaths a year over twenty years — 9.0 deaths — implies 200,000 USD a life, which a published figure of 13,200,000 USD exceeds by 66.00 x.
Expect the estimates of avoided deaths to be poor. Use your engineers' ranges, compute the implied price at both ends, and state the range rather than resolving it silently. A range that spans policy and does not is still a finding, and it is the one that gets attention.
Exercise 2.2 — Set the floor (45 minutes)
Establish what your regulator uses and adopt it as a floor. If you operate in multiple jurisdictions, take the highest, for one commercial reason: a company that can show it applied its highest standard everywhere has a defence, and one that applied the cheapest local number has a disclosure.
Note the range that exists inside a single government. The EPA's central value of 7,400,000 USD in 2006 money is 11.5 million in 2024 money; the transport figure stands 1.15 x above it. Different statutes, different risks. Say which one you are adopting and why, in one line.
Exercise 2.3 — Separate the five questions in your own language (one hour)
Rewrite the chapter's five-row table using the names your organisation actually uses — your appraisal template, your insurance programme, your reward framework, your legal reserve. Then circulate it to three people and see whether anyone argues. The argument is the value; consensus on this table in your own vocabulary is worth more than any policy document.
Exercise 3.1 — The life-price resolution (one page)
One page to the board, containing exactly five things.
Exercise 3.2 — The template line (one afternoon)
Add one field to the capital appraisal template: cost per statistical life saved. Not a conclusion about safety — a computed figure with its denominator stated.
This is the single highest-return administrative act in the whole workbook. A line in a template outlives every champion, and it causes the arithmetic to be done in perpetuity by people who have never read this chapter.
Exercise 3.3 — Decide what you will disclose (one hour with the auditors)
Nothing here asks you to misstate a financial statement. IAS 38 forbids capitalising an internally generated workforce, and it will go on forbidding it — which is why the World Bank can find human capital at 64.0 per cent of global wealth while no balance sheet in your industry carries any of it.
What you can do is a memorandum note: the published life price, and the cost per statistical life saved of the programmes funded this year. It changes no recognised number, it costs a paragraph, and it makes the policy expensive to abandon quietly. That last property is the reason to do it.
Exercise 4.1 — Size the facility (one week, with treasury and the broker)
Workers' compensation and employers' liability premium is manual premium times an experience modifier that moves with claims history. The modifier is audited by a third party, which means the repayment mechanism for a safety facility already exists and is already verified.
insured payroll 40,000,000 USD
manual rate 0.025 of payroll
manual premium 1,000,000 USD / year
experience modifier 1.05 -> 0.80
annual premium saving 250,000 USD
facility size 900,000 USD
simple payback 3.60 years
return on facility 27.8 %
Run yours. Then present it in the register a treasurer recognises: at 27.8 per cent this is not a safety proposal competing for scarce capital, it is one of the better-returning uses of capital in the business, and the fact that it also prevents injuries is the second sentence, not the first.
Exercise 4.2 — The two inequalities on the front page
annual premium saving cost of the programme
------------------------ > WACC and --------------------- < published life price
facility size statistical lives saved
The first says the money is worth having. The second says the decision is worth taking. A proposal that clears the second and fails the first is still right — you fund it another way and say so out loud, because that is the sentence that proves the resolution means something.
Exercise 4.3 — The counterparty sequence
Internal first: treasury lends to the operating unit, documented in a week, and you build a track record on your own paper. After two completed cycles, take the structure to the employers' liability carrier or a captive cell. A carrier shown your implied life price and your cost per statistical life saved is being handed better underwriting information than it normally receives, and it prices accordingly.
The life price is the easier half of this chapter for a firm. The harder half is pay, and it is harder because the arithmetic says something uncomfortable and the uncomfortable thing is also commercially useful.
Exercise 5.1 — Name the markdown in your own numbers (one afternoon)
For one role family, write the three figures side by side — market clearing, estimated marginal revenue product, actual pay — and compute the gap. The chapter's worked case: a 62.00 USD product at an elasticity of 2.50 predicts 44.29 USD, a gap of 17.71 USD an hour, 33,657 USD across a 1,900-hour year.
Then ask the only question that matters commercially: what is holding the elasticity down, and how durable is it? Geography, a non-compete, a firm-specific system, a visa, a benefit that does not travel. Every one of those is a friction, every friction has a shelf life, and a wage that depends on a friction reprices abruptly when the friction goes. Firms are routinely surprised by this and call it a labour shortage. It is a repricing.
Exercise 5.2 — Cost the alternative properly (one hour)
Put the turnover arithmetic beside it: 43,500 USD an exit at a loaded salary of 58,000 USD and a replacement cost of 0.75, twelve avoidable exits a year, 522,000 USD. Set that against what closing part of the gap would cost. In a surprising number of role families the retention arithmetic already favours paying more, and nobody has put the two numbers on one page because they live in two different budgets.
Exercise 5.3 — The honest negative, for your own board
Say this part plainly when you present, because somebody will find it otherwise. The pass-through evidence is modest: a rent-sharing elasticity of 0.05 to 0.15 means a 10 per cent rise in value added per worker historically moves pay by 0.5 to 1.5 per cent. Naming the markdown does not by itself raise wages, and a firm that names it and does nothing has merely documented a gap. What it does is make the level a decision the firm owns and can defend, rather than a fact of nature it shelters behind — and decisions, unlike facts of nature, can be improved deliberately and claimed credit for.
| Day | Action | Artifact |
|---|---|---|
| 1–15 | Five-room sweep with controller and broker | Five numbers, one page |
| 16–30 | Appreciative question in one operations meeting | The sponsorship pattern |
| 31–45 | Compute three implied prices from decisions already taken | Three implied prices |
| 46–55 | Draft the resolution; set the floor and the index | The one-page resolution |
| 56–65 | Add the template line; brief the appraisal community | The amended template |
| 66–80 | Size the facility; secure the single signature | Facility memo |
| 81–90 | Deploy on the clearest case; book the modifier review | The disclosure note |
This company prices human life several times a quarter and has never written the price down. The three decisions at appendix A implied prices between X and Y, against a published regulatory figure of 13,200,000 USD. The resolution at appendix B sets our price, its index, its scope and the constraint that it may never be applied to a named individual. The facility at appendix C funds the first programme at a 27.8 per cent return, repaid out of experience-rating credit, requiring one signature. The recommendation is to approve all three today, and to publish the memorandum note with the year-end accounts.
That paragraph is the product of this workbook. Everything before it is how you earn the right to write it in those words.
| 1 | 3 | 5 | |
|---|---|---|---|
| I know what price my decisions have implied | Never computed one | Computed three | Computed and published |
| My appraisal template asks for it | No field | Field added | Field completed as routine |
| My price is indexed and dated | No price | Price set | Price set, indexed, reviewed |
| The constraint is written beside the number | Not discussed | Drafted | Board-adopted and understood |
| The programme pays for itself | Funded from goodwill | Facility sized | Facility deployed and repaying |
| I can say which instrument I am using, in the sentence | Rarely | Usually | Always, and I notice when others do not |
The last row is again the one that matters, and in a firm it has a commercial edge as well as an intellectual one. An executive who names the instrument in the sentence is very hard to argue into a category error, and category errors in this territory are where the expensive mistakes live.