Haute Lumière

Commerce · VI.01 · MMXXVI · daylight

La Bourse  /  Volume VI  /  Nº VI.01  /  Workbook — the Gainshare employee

Four colleagues walking toward a long table set with dried grasses, the room bright with sun.
Plate VI.01 · Workbook — the Gainshare employeeThe Table Before the Meeting.A decision rule is furniture. It decides who can be in the room, how long they will be there, and who is not in the room at all — and it does this before anybody has said a word.

WORKBOOK — THE LUMINOUS GAINSHARE EMPLOYEE

Chapter VI.01 · Who Decides

For the person working inside a gainshare arrangement — where a defined share of verified improvement returns to the people who created it. This chapter is about decision rights, and decision rights are the part of a gainshare that determines whether the share is real.

WHY THIS CHAPTER IS YOURS

A gainshare has four written parts: a baseline, a measure, a share, and a period with a verifier. Most people inside one focus on the third — the percentage — and that is the wrong place to look first.

The share tells you what you get. The decision rule tells you who can change it. A scheme with a generous share and a unilateral amendment right is worth less than a scheme with a modest share that cannot be amended without your agreement, and the difference is not a matter of trust. It is a threshold, and it is computable.

So read your scheme document twice. The first time for the number. The second time for the five questions in Exercise 1.1, which are about who decides — and which almost nobody asks until the year the number changes.

PART ONE — DISCOVERY

Days 1–30: find the decision rights in your own scheme

Exercise 1.1 — The five questions (2 hours)

Take your scheme document and answer these in writing. Where the document does not say, write "the document does not say" — that is a finding, not a blank.

  1. Who sets the baseline, and under what rule? One person, a committee, a vote?
  2. Who can amend the measure mid-period, and what threshold do they need?
  3. Who verifies, and can the people being measured object to the verification?
  4. Who can suspend or terminate the scheme, and what notice is required?
  5. What is the minimal blocking coalition on an amendment — that is, how many people, and which people, could stop a change you did not want?

Question five is the one this chapter exists for. Under a k-of-n rule the smallest group that can stop something is n − k + 1. If the answer is "nobody on our side," then every other term in your scheme is held at someone else's discretion, and you now know the single most important thing about the arrangement you are in.

Exercise 1.2 — Name the blocking coalition (1 hour)

For each of the five questions, write the actual names of the people who could block. Not the roles. The names.

This is exactly what the chapter does with public institutions. Article V of the United States Constitution needs 38 of 50 states, so 13 can block, and the thirteen least populous hold 14,618,613 people — 4.41 per cent of the country. The Senate's cloture rule needs 60 of 100, so 41 senators block, drawn from at least 21 states holding 11.22 per cent of the population. Naming the coalition is how you find out what a rule really says.

Exercise 1.3 — Find what is already working (2 hours, appreciative)

Ask three colleagues who have been in the scheme longest:

You are looking for the precedent that already exists. Schemes that survive usually survived one well-handled amendment early on, and the way it was handled is the model you should be asking to have written down.

PART TWO — THE ARITHMETIC

Days 31–50: compute what your decision rights are worth

Exercise 2.1 — Price the amendment right (2 hours)

Your scheme's amendment threshold is the external-cost term in this chapter, pointed at you.

Under a rule where the employee side holds k of n seats on the scheme committee, the external cost is decisions × loss × (n − k)/n. Put your own numbers in: how many scheme decisions a year, and what a bad one costs you personally over the period. On the chapter's parameters — nine seats, twelve decisions, £30,000 of loss — a 5-of-9 rule carries £160,000 a year of external cost and a 7-of-9 rule carries £80,000.

You will not get exact figures. You will get an order of magnitude, and an order of magnitude is enough to know whether the threshold in your document is proportionate to the stake.

Exercise 2.2 — Price the delay (1 hour)

Decision rights cut both ways, and this is the exercise that keeps you honest. If the employee side holds a veto, the scheme cannot change without you — and changes you want are also blocked.

At unanimity in a body of nine, with each member accepting a given proposal with probability 0.60, the expected number of rounds is 1 / 0.6⁹ = 99.23. At a 6-of-9 rule it is 2.07. A veto is expensive to everyone including its holder, and a scheme that cannot be improved is not a protection; it is a freeze.

Exercise 2.3 — The consent arithmetic (1 hour)

The instrument that gets you most of the veto for a fraction of the delay is a consent rule: an objection must be argued and paramount — about whether the proposal will work or will harm the aim — and a preference does not qualify.

At an admissible-objection rate of 10 per cent in a body of nine, a proposal carries in 2.58 rounds against unanimity's 99.23 — faster by 38.4 times. But know what you are agreeing to: that rule behaves like 6.23 of 9, 69.2 per cent, and carries £110,918 a year of external cost against £0 under true unanimity. Consent buys speed by billing the people whose objection is a real preference they cannot phrase as a workability argument. Sometimes that is you.

Ask for consent plus one thing: a written objection log, so the grammar can be audited rather than argued.

Exercise 2.4 — The baseline ratchet, as a decision right (1 hour)

The classic way a gainshare dies is baseline ratcheting: each period's improvement becomes next period's baseline, so the same effort yields less each cycle and eventually nothing.

Ratcheting is a decision, and the question is who makes it and under what threshold. Write down which of these your scheme does: baseline fixed for a stated term; ratchets on a published schedule everyone can see coming; or is reset at management discretion. If it is the third, that is the highest-value finding you will produce this quarter, and it is a decision-rights finding, not a pay finding.

PART THREE — DESIGN

Days 51–70: ask for the right thing

Exercise 3.1 — Draft the three classes (3 hours)

Propose a banded schedule for your own scheme, in the same three classes the executive workbook uses, so both sides are reading one instrument:

Compute each one rather than asserting it: band by what a wrong decision in that class costs the employee side over the period, and take the minimum of the two curves.

Exercise 3.2 — The delegation question, answered honestly (1 hour)

You will be offered representation rather than participation, and representation is usually right. It is right because the arithmetic is overwhelming: an assembly of 1,200 deciding directly costs £10,114,446 a year in decision cost for twelve decisions, against £80,562 delegated to a board of nine — a saving of £10,033,884, with a break-even of £3,344,628 per divergent decision.

So do not argue against delegation. Argue about the reserved list. Ask for Class C above the break-even and accept delegation below it, and you are asking for the thing the arithmetic already supports rather than for a concession.

And watch the ratio. The House of Representatives carried 60,449 people per seat in 1789 and 761,952 in 2020 — the principal's stake per agent thinned 12.6 times; each European Parliament seat carries 623,898. If your employee representative speaks for four hundred people, ask what the ratio was when the scheme started.

Exercise 3.3 — Write the ask (2 hours)

One page, four paragraphs, no grievance in it:

  1. What the scheme's current decision rule is, stated accurately.
  2. What class of decision it currently under-protects, with the loss figure.
  3. The proposed three classes, with the computed threshold for each.
  4. The delay cost of the proposal, computed, so that the other side can see you have priced their side of it too. That fourth paragraph is what gets it read.

PART FOUR — DESTINY AND DELIGHT

Days 71–90: make it hold

Exercise 4.1 — The amendment threshold (30 minutes)

Whatever is agreed, the threshold for amending the schedule must sit above every threshold inside it, and must be set before anyone wants to amend it. This single clause is worth more than several points of share percentage.

Exercise 4.2 — The objection log (20 minutes a month)

Date · proposal · objection · workability or preference · resolution. If the share of objections ruled inadmissible is rising, the grammar is tightening and your consent right is quietly shrinking. Nobody will have amended a word. The log is the only thing that sees it.

Exercise 4.3 — The annual recomputation (2 hours a year)

Re-run the bands with the year's actual figures. A scheme reviewed by arithmetic survives changes of personnel; a scheme reviewed by negotiation is re-won every year by whoever is best at negotiating, which over time is not you.

Exercise 4.4 — Notice what changed (30 minutes)

At day ninety, write half a page on the first scheme meeting that ended early because the threshold was met and everyone could see it. That is what a priced rule feels like from the inside: not winning, but being dealt with by something that was built before anybody knew the question.

HOW THE LEDGER WORKS, AND WHAT TO CLAIM

A gainshare ledger has three columns that matter and one that does not. The one that does not is the running total, which is the only one most people look at.

Column one — the baseline, with its date and its signature. A baseline without a date is not a baseline; it is whatever the person holding the spreadsheet believes this quarter. If yours has no signature block, ask for one before you ask for anything else. It costs nobody anything and it is the foundation every other claim rests on.

Column two — the measure, computed the same way twice. Take one period's figure and recompute it yourself from source data. If you get a different answer, the difference is the finding and it belongs in the objection log, phrased as workability: the measure as written does not reproduce. That is an admissible objection under any consent grammar, which is exactly why you phrase it that way.

Column three — the decision trail. Every change to the baseline, the measure or the period, with the date, the threshold it was made under, and who could have blocked it. Most schemes have no such column, and that absence is the whole argument for building one: without it, nobody can tell an improvement from a redefinition.

What to claim. Claim the gain the measure produces, and claim it without apology — it is arithmetic, not a favour. But claim two further things that are usually left on the table. First, a share of the decision cost saved: if the team's self-management removes a layer of approval, that saving is real, countable in hours, and currently appears in nobody's line. Buurtzorg runs roughly 833 teams of at most 12 nurses on a central staff of about 50 — a support ratio of 200 to 1 — and KPMG found 40 per cent fewer care hours per client; those are decision-cost savings showing up as clinical outcomes. Second, the objection log itself as a deliverable, because it is the evidence that the scheme is being run rather than merely held.

KNOW YOUR SCHEME — A CHECKLIST

Tick what you can answer from the document alone. Every unticked line is a decision right held by somebody whose name you should find out.

THE CONVERSATION, SCRIPTED

You. "I have priced our decision rule. Three classes, thresholds computed from what a wrong decision in each class costs. Class C covers the baseline, the share and termination."

Them. "That will slow everything down."

You. "It speeds two of the three classes up. Class A drops to a simple majority, which is most of our volume. Only Class C rises, and Class C is three decisions a year. I have costed the delay: here it is."

Them. "We need to be able to adjust the baseline."

You. "Agreed — on a published schedule everyone can see coming. What we cannot work under is discretionary reset, because then the measure is not a measure. Publish the schedule and I will sign the same page."

Them. "Why does this matter so much to you?"

You. "Because the share tells us what we get and the threshold tells us who can change it. I would rather have a smaller share I can rely on than a larger one I cannot."

APPRECIATIVE QUESTIONS FOR YOUR TEAM

  1. When has a change to this scheme been handled in a way that left everyone — including the people it cost — feeling fairly dealt with? What made that possible?
  2. Which part of our scheme do we understand best, and how did we come to understand it?
  3. If every proposed change arrived with its class and threshold printed on it, what would our scheme meetings be about instead?
  4. What would have to be true for this scheme to still be working, and still trusted, when everyone currently in it has moved on?
  5. Who outside this team pays for something our scheme gains, and how would we find out what it costs them?
  6. What is the smallest change to how scheme decisions are made that would give us the most confidence in the number at the end of the period?