Haute Lumière
Commerce · V.05 · MMXXVI · daylight
For the person working inside a gainshare arrangement — where a defined share of verified improvement returns to the people who created it. Skill is the one asset in the scheme that you own outright, that the firm cannot take, and that the ledger almost never counts. This workbook is about getting it counted.
A gainshare has four parts: a baseline, a measure, a share, and a period with a verifier. Chapter V.05 adds a fifth thing that sits underneath all four and is usually invisible.
Every gain your team produces rests on skills somebody paid to build. The question of who paid, how fast those skills wear out, and who captures the return, is not a side issue in a gainshare. It is the arithmetic of the pool.
Three numbers from the chapter, and what each one means from where you sit.
Exercise 1.1 — Read your scheme for the skill clause (90 minutes)
Take your scheme document and find the answers to these. Write them down; do not rely on what you have been told.
Question two is the one that decides whether your scheme quietly punishes teaching. If the hours a senior person spends training a junior are booked as a cost against the pool, and the junior's improved output lands in a later period or a different team, then your scheme pays people not to teach. Nobody designed that. It is simply what falls out when training is a cost and its return is somewhere else.
Exercise 1.2 — Map your team's skill portfolio (one week)
For each person, including yourself, note the main skill they bring and place it on the scale:
delta half-life what it looks like
--------------------------------------------------------------
1.35 %/yr 51.3 yr judgement, diagnosis, the ability to
learn the next thing, handling people
3.04 %/yr 22.8 yr the trade itself, the professional practice
10.41 %/yr 6.7 yr the specific tool, platform or system
Then ask two questions of the map.
Exercise 1.3 — The appreciative conversation (45 minutes)
Run this with your team, and use these words:
"Think of a time somebody here learned something fast. Not the biggest thing — the fastest. What were the conditions? What did we do that we do not normally do?"
Take notes on conditions, not outcomes. Repeatable conditions are what a gainshare can actually pay for.
Exercise 2.1 — Your own payback month (two hours)
Do it for a recent new starter on your team, or for yourself when you joined.
what they cost per month pay + the hours senior people spend
x those people's fully loaded rate
what they produced per month honest estimate: what would you have
paid someone to do exactly that work
Anchors from the Swiss surveys, if you have nothing better: a first-year trainee runs at about 40 per cent of a qualified person's productivity, a third-year at about 80 per cent, with roughly three-quarters of their time on productive work.
Cumulate month by month and find the crossing. That month is a fact about your team and it is currently nobody's number.
Now the question that matters for the pool: does your scheme's measurement period end before or after that month? If a new starter's payback lands in month nineteen and your scheme measures on a twelve-month cycle with a resetting baseline, then every hour your team spends training lands in the cost column and the return lands after the window closes. Your team will rationally stop training, and everyone will call it a culture problem.
Exercise 2.2 — The depreciation claim (90 minutes)
Here is the argument that belongs in your next scheme review, computed.
A rent of €4,000 a year on a skill, held with probability p, decaying at δ, over a horizon T, is worth V × p × (1 − exp(−δT)) / δ. At German retention of 53 per cent over five years:
delta E(5 yrs) E(unbounded)
----------------------------------------
1.35 %/yr 10,250 EUR 157,037 EUR
3.04 %/yr 9,833 EUR 69,657 EUR
10.41 %/yr 8,264 EUR 20,364 EUR
The slow-depreciating skill is worth nearly eight times the fast one over a long horizon and only twenty-four per cent more over five years.
What that gives you to say. If your team maintains fast-depreciating skills — a platform, a stack, a machine that changes every few years — the firm's return on that maintenance arrives quickly and then evaporates. That is precisely the case where the gain should be shared inside the period, because there is no long tail for the firm to collect later. The usual objection to sharing more now is that the firm needs the later years to recover. At δ near ten per cent, there are no later years.
Exercise 2.3 — The baseline-ratchet test (one hour)
Ask: what happens to the baseline when a gain is realised?
If it resets to the improved level each period, you are on a treadmill: every improvement raises the bar you are measured against, and skill investment is worth less each cycle. A well-designed scheme holds the baseline for a stated term — three to five years is typical — or ratchets on a published schedule everyone can see coming.
Now connect it to this chapter. A gain built on a fast-depreciating skill will decay on its own whether or not the baseline ratchets. Ratcheting a baseline against a decaying skill is being paid twice for the same thing, by the firm, out of your pool. That is a specific, arguable, arithmetic point and it is the strongest one in this workbook.
Exercise 3.1 — Propose the teaching credit (one afternoon)
One clause, and it fixes the structural fault in Exercise 1.1.
Hours spent by a scheme member training another member are credited to the pool at the trainer's standard rate, and the trainee's improvement is measured against the trainee's own entry baseline for the following two periods, attributed to the training team.
Three things to be ready for.
Exercise 3.2 — Propose the skill register (one week)
Ask for one thing that costs the firm almost nothing and changes your position permanently: a register of skills held by scheme members, with the date each was certified or last used in anger.
Why it matters to you: asymmetric information is the main reason firm-financed training is affordable at all. The firm knows what you can do; the outside market does not, and so prices you at the average. A register, especially one tied to a recognised standard, makes your capability legible — and legible capability moves pay toward productivity. This is the single most valuable non-cash thing you can ask a gainshare scheme for.
Why it is easy for the firm to say yes: it wants the register for succession planning anyway.
Exercise 3.3 — Propose the pool, if the numbers say pool (one week)
If your firm's P exceeds its T — it trains people who leave before the investment is recovered — the chapter's instrument is a training consortium: an entity that employs apprentices and rotates them among member firms, funded by a hosting fee and a per-employee subscription. Denmark has run the public version since 1977, where every employer pays a flat per-head contribution whether it trains or not and only trainers are reimbursed.
From inside a gainshare, this matters for one reason. In a consortium the training cost stops falling on your pool and starts falling on a subscription paid by the whole membership. The cost line leaves your ledger and the benefit line stays. Raise it in exactly those terms.
Exercise 4.1 — Two lines in the scheme report (one hour)
payback month of our last three joiners (P) ____ months
months they stayed after that (T - P) ____ months
teaching hours credited to the pool this period ____ hours
Three lines in the standing report will outlast every argument you win in a meeting.
Exercise 4.2 — Know what to ask for, and in what order (one afternoon)
Ask in that order. The first two are close to free and establish that you argue from arithmetic, which is what makes the third one winnable.
Exercise 4.3 — Notice the month (ongoing)
There is a week in any real apprenticeship when the person stops checking whether they have done it right and simply knows. It arrives in the middle of something ordinary, and the person teaching them usually sees it first.
Somewhere near that same week, the cumulative line crosses zero. They are the same event, seen from the two sides of a bench. You are in the rare position of being able to see both — and of being in a scheme that could, if you write the clause, pay the person who was watching.
| Yes | No | Nobody knows | |
|---|---|---|---|
| Is training time a cost against the pool? | |||
| Is a trainee's later improvement attributed to whoever trained them? | |||
| Does the baseline reset when a gain is realised? | |||
| Is there a carry-forward for improvements that pay after the period? | |||
| Is there a register of who can do what, with dates? | |||
| Do you know the payback month of your last three joiners? | |||
| Does the firm pay for training that the market can verify — a real certificate? | |||
| Is there a floor in a bad period? |
Every "nobody knows" is a finding, and finding one is worth more this month than winning an argument about the share percentage.
"I want to raise one thing about how the pool treats training, and I have the arithmetic rather than a complaint.
Right now the hours we spend training land in the cost column and the improvement lands in a later period or another team, so the scheme pays us not to teach. Nobody designed that; it just falls out of where the two things sit in time.
Two changes fix it and neither costs the firm anything net. Credit training hours to the pool at standard rate, and attribute a trainee's improvement against their own entry baseline for the next two periods, split with the team that trained them.
And one thing I would like that costs almost nothing: a register of what people here can actually do, with dates. You want it for succession planning. I want it because it makes what we can do visible. Those are the same document."
The sentence that ends these conversations is the problem with training is that people leave. It is said with regret and it sounds like realism.
The measured answer is that Switzerland loses more than sixty per cent of its apprentices within a year of qualifying — the highest rate in Europe — and Swiss firms train more deeply and more willingly than almost anyone, because their payback lands at month 27.7 of a thirty-six-month contract. Losing people is not the problem. A payback that sits behind the departure is the problem, and that is a design choice, not a fact about human loyalty.
You are not in the room to defend loyalty. You are in the room with a date.