Haute Lumière

Commerce · V.08 · MMXXVI · daylight

La Bourse  /  Volume V  /  Nº V.08  /  Workbook — the executive

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter V.08 · Meaning as an Economic Input

For the person with a P&L, a compensation committee and an auditor. This workbook produces three artifacts: a measured δ with a range, a funding model for the accrual, and a board paper that survives the first analyst who runs the numbers themselves.

---

THE PREMISE, STATED COMMERCIALLY

Your organisation is buying an input it has never priced and selling one it has never invoiced.

The input is meaning: the observable consequence of the work, the judgement you leave with the team, the proportion of a person's week spent on what they are actually for. It has measured returns — 142 and 171 per cent on a fundraising line from five minutes of beneficiary contact, 43 per cent on hours worked from reading what the work prevented, 40 per cent fewer care hours per client where judgement went back to a team of twelve.

The thing you are selling without an invoice is the same input, sold to your own staff and settled in forgone wages. That settlement has a size. At the conservative estimate of 5.5 per cent it is 1,951.48 dollars a year for a worker on the May 2023 median of 33,530 — 1,170,888.89 dollars across six hundred of them, which is a transaction larger than most items on your capital plan and is in none of your papers.

Two commercial facts follow, and they are the whole of this workbook.

The first is that the input is under-supplied, because nobody owns it and its returns land in lines that are not attributed to it. That is a cheap prize.

The second is that the sale is a liability you have not recognised. Not a legal liability — a commercial one. An unpriced transfer from your workforce is a position that reprices violently when it becomes visible, and it becomes visible through the labour market, the press, or a competitor who publishes first.

---

PART ONE — DISCOVERY

Days 1–30: find where meaning is already paying

Exercise 1.1 — The beneficiary-contact map (one week)

For each of your top five business units, answer three questions in writing:

1. Who is the end beneficiary of this unit's work — the person, not the segment?
2. When did a member of this unit last see or speak to one of them?
3. What would it cost, in hours, to put fifteen minutes of that contact on the quarterly rota?

The third answer is nearly always trivial and the second is nearly always never. That pairing is the finding. A return of the size the literature reports, available for a rota change, is the cheapest line on any operations plan you will write this year — and it should be piloted in a unit where the output is counted daily, so the effect clears the noise.

Exercise 1.2 — The 20 per cent audit (two weeks)

Sample forty roles across the organisation. For each, estimate the share of working time spent on the activity that person finds most meaningful. Ask them; do not infer it.

Plot the distribution against the threshold. Below 20 per cent, the physician study found burnout at 53.8 per cent against 29.9 above — 23.9 points. Above the threshold, no further benefit.

Two management conclusions, and they are unusual in being symmetrical. Below the threshold, this is a staffing finding, not a wellbeing one: someone is spending a professional's judgement on work that does not require it, which is a margin question. Above the threshold, stop. Money spent pushing a role from thirty per cent to forty buys nothing measurable and is better spent on pay.

Exercise 1.3 — Retention anomalies (one week)

Find the two teams with the lowest voluntary turnover in the firm, and the two with the highest, controlling roughly for pay band.

Interview four people in each, appreciatively: tell me about a time the work here was unmistakably worth it — what were the conditions? Take notes on conditions, not outcomes.

Then cost the difference. At a replacement cost of 16.1 per cent of salary for low-wage roles and 21.4 per cent for roles under 75,000 dollars, a twenty-point difference in annual turnover across a thousand-person population is a seven- figure line. You are looking for what the good teams have, so you can buy more of it — not for what the bad teams lack.

---

PART TWO — THE ARITHMETIC

Days 31–50: compute δ, with a range

Exercise 2.1 — The skill-matched benchmark (two weeks)

This is the exercise the whole workbook rests on, and it fails in exactly one way: using a sector benchmark.

A sector benchmark in a mission sector has the differential inside it. Bench against your peers and δ comes back near zero, everyone relaxes, and the instrument has certified the thing it was built to find.

Build it properly, per job family:

1. Define the skill bundle: training, licensure, years, hours, physical demand, unsocial hours.
2. Find that bundle's pay in the whole economy — national occupational wage statistics first, a broad commercial survey second, your sector last and only with the contamination stated.
3. Record vintage, geography and the adjustment method. Write down what your comparator does not hold constant. That paragraph is what makes the number defensible.

Exercise 2.2 — Compute δ (three days)

  δ  =  1  −  (median internal pay  /  skill-matched external benchmark)

Per job family. Then the total transfer: δ × W × N.

Worked, at the chapter's conservative end: a benchmark of 35,481.48 against a median of 33,530 gives δ = 5.5 per cent, 1,951.48 per worker, 1,170,888.89 across six hundred. At the aggressive end of 17.0 per cent the benchmark is 40,397.59, the differential 6,867.59 per worker and 4,120,554.22 across the same six hundred.

Carry both. A single point estimate of a compensating differential should never enter a pay decision, because these estimates come back wrong-signed more often than any other coefficient in labour economics — unobserved ability sorts into pleasant jobs and loads onto the amenity.

And watch the denominator. δ of 5.5 per cent is 5.5 per cent of the benchmark, which is 5.82 per cent of your current wage bill. Quote the wrong one in a board paper and you will be corrected in the room.

Exercise 2.3 — The break-even (one day)

The number that decides it is not δ.

  break-even turnover reduction  =  δ / c

At δ = 5.5 per cent and a replacement cost of 16.1, that is 34.16 percentage points of annual turnover — in a sector running at a median of 94 per cent, a fall to 59.84. The firm-level separation elasticities in the monopsony literature run about 1.7 to 4.0, which deliver 8.79 to 20.68 points. Short by 25.37 to 13.48.

At the aggressive δ of 17.0 per cent the break-even is 105.59 points against the 94 that exist: eliminating every separation would not repay it.

So the churn case does not close, and your board paper says so on page one. The residual is found in quality, in client retention, in error and incident rates, in the price your sector's best providers command — or it is presented honestly as a distribution of value the firm has decided to make. Both are defensible. Only one survives an analyst, and it is whichever you said first.

Exercise 2.4 — Clean up the engagement reporting (half a day)

Two corrections, and they will make your pack the most credible in your peer group.

Quartile contrasts. Every headline of the form top-quartile units are 23 per cent more profitable expresses a correlation of about 0.22. Top minus bottom quartile is 2.5422 standard deviations of the predictor; multiplied by r that is 0.559. Report the effect size and the contrast together, in that order.

Same-source correlations. Where engagement and performance both come from your survey, divide the correlation by 1.26 before quoting it: 0.40 becomes 0.317, and the explained variance falls from 0.1600 to 0.1008 — 37.0 per cent of it was method. Better still, pair the survey with an outcome from your own systems. The independent-outcome literature reports 0.22 where the self-report literature reports 0.43, a ratio of 1.95.

---

PART THREE — DESIGN

Days 51–75: the instrument

Exercise 3.1 — Separate the two committees (one week)

Structural, cheap, and it is the move that matters most.

• The compensation paper carries the external skill-matched benchmark on page one, and no mission material.
• The pay-setting model may not contain a mission, purpose or engagement variable. Write that into the terms of reference.
• Any proposal arguing that a role can be paid less because the work is rewarding is returned unread, with that sentence quoted.

Exercise 3.2 — Structure the Mission Differential Account (two weeks)

The forfeiture clause is the one to defend. If unvested balances revert to the firm you have built a retention handcuff and called it restitution, and your people will read it correctly within a quarter.

Exercise 3.3 — Fund it (one week)

Three-year model, three cases: full δ accrued, half accrued, and δ published with a stated affordable accrual.

The third case is not a failure and it is often the right answer. Publish the whole δ; accrue what the covenant and cash position will carry; state the gap. A measured subsidy that is only partly funded is still an enormous improvement on an unmeasured one — and setting the accrual at a level you cannot fund converts an honest measurement into a covenant breach and teaches the organisation never to measure it again.

Exercise 3.4 — Agree the treatment early (two weeks)

Take your auditor the benchmark method, the accrual policy and the vesting terms before you take the board anything. This is a conversation about deferred compensation, which they have every year. Give them a documented, consistently applied method in year one and the treatment is uncontroversial thereafter.

---

PART FOUR — DESTINY AND DELIGHT

Days 76–90: make it hold

Exercise 4.1 — Put δ in the pack (one week)

Owned by finance, not by people. A number owned by human resources drifts toward a narrative; a number owned by finance drifts toward a method, and a method survives the departure of its author.

One page: δ by job family, the range, the benchmark source and vintage, the accrual, and the funded proportion.

Exercise 4.2 — Publish internally (one day)

With the method attached, so several hundred people can check it. That is the most robust governance mechanism available to you and it costs a document.

Expect the method to be attacked in the first year. That is the sign it is measuring something.

Exercise 4.3 — Sequence the supply side (ongoing)

Only now do you improve the supply of meaning: beneficiary contact on the rota, task significance made legible, judgement returned to teams, time protected to the 20 per cent threshold and no further.

The order is not stylistic. An organisation that raises the supply of meaning before it has fixed the pricing question has, in strict arithmetic, increased the amount it can extract — and the mechanism will work perfectly.

---

THE FAILURE MODES, NAMED

So you can see them coming

The sector benchmark. δ comes back near zero, and the instrument certifies the thing it was built to find. Detect it by benchmarking one job family both ways and putting the two numbers side by side.

The unaffordable accrual. An honest measurement becomes a covenant problem and the measurement is abandoned. Publish δ in full; accrue what you can fund; state the gap.

Supply before pricing. Meaning improves, the labour market notices, and applicants accept less. The tell is a falling median offer accepted alongside rising engagement scores. Watch those two lines on the same chart.

The single point estimate. Someone drops the range and δ becomes a target that your current position already satisfies. Keep the range in the pack.

The churn case oversold. You promised the accrual would pay for itself on turnover, it delivered 8.79 to 20.68 points against a 34.16-point break-even, and the programme is now a business case that collapsed rather than a deliberate distribution. Say it on page one, in the first paper.

---

THE NINETY DAYS ON ONE PAGE

---

BOARD PAPER TEMPLATE

Title. The mission wage differential: measurement, treatment and funding.

Page one, four lines.

1. Our median pay in these job families is δ below the skill-matched external benchmark; the range is stated and the method is attached.
2. That differential is worth δ × W × N a year, currently transferred from the workforce to the mission and recorded nowhere.
3. This does not pay for itself on turnover. Break-even requires δ / c points of annual turnover reduction; the elasticities support a fraction of that. The residual case rests on quality, client retention and price — or on a distribution the board chooses to make.
4. We propose to publish δ annually and accrue what is fundable as deferred compensation under IAS 19 — vesting over four years, with forfeitures returning to the pool.

Everything else is an appendix. A board that reads four honest lines will turn the page. A board that is given a case which collapses under its own arithmetic will remember that it did.

---

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

1. Where in this firm is somebody already doing the work of making the purpose of the job visible, without being asked — and what does it cost us?
2. Which of our teams holds the most judgement over its own work, and what does it get right that we have never tried to copy?
3. If our δ were published tomorrow with its method, what would we want it to say in three years, and what is the first step toward that?
4. What would it take for the compensation committee to receive a paper that could not be influenced by how much we love what we do?