Haute Lumière

Commerce · IV.03 · MMXXVI · daylight

La Bourse  /  Volume IV  /  Nº IV.03  /  Workbook — the Gainshare employee

WORKBOOK — THE LUMINOUS GAINSHARE EMPLOYEE

Chapter IV.03 · Agriculture as the Model Case

For the person working inside a gainshare arrangement where the gain is measured in soil — an agronomist, a farm manager, a field officer, a co-operative technician, a grower inside a shared-upside scheme. How the mechanism reads from the inside, what to measure, what to claim, how the ledger works, and what to ask for.

---

WHY THIS CHAPTER IS DIFFERENT FROM THE INSIDE

In most gainshare schemes the thing being measured is a cost, a throughput or a defect rate, and it is measured by somebody else's system that already exists.

Here the thing being measured is a stock in the ground that changes by about 0.6 percent a year, is spatially variable at 15 percent, and will not show up at all unless somebody designs the measurement properly. That is a completely different position to be in, and it has one enormous implication:

In a soil gainshare, the person who designs the sampling decides who gets paid. Not the person who does the work. Not the person who signs the contract. The person who chooses the number of cores, the depth, the pairing and the strata.

If that person is not you, make it your business to understand what they chose. If it can be you, take it. It is the highest-leverage role in the whole scheme and it is usually vacant.

---

THE FOUR PARTS, AND WHERE A SOIL SCHEME BREAKS EACH ONE

A gainshare is a written promise with four parts. In a soil scheme each part has a specific failure mode, and they are all avoidable.

Every one of those is a question you can ask in a meeting this week, and every one of them is worth money.

---

PART ONE — DISCOVERY

Days 1–30: find where the gain actually comes from

Exercise 1.1 — Read your own scheme against the four parts (2 hours)

Take the scheme document and answer in writing:

1. What is the baseline, to what depth, in what year, and was bulk density measured or assumed? An assumed bulk density can move a stock figure by ten percent on its own.
2. Is the measure the stock or the difference against a control? If there is no control, what is the counterfactual and who chose it?
3. What is the share, and is the verification cost inside or outside it?
4. What is the period, and does it reset?
5. What happens to the baseline when a gain is realised?

Question five decides whether the scheme is worth being in. If the baseline resets to the improved level each period, you are on a treadmill: every gain raises the bar, and in soil — where the rate already decays by saturation — a ratcheting baseline means your pay falls twice for the same good work.

Exercise 1.2 — Compute your scheme's saturation profile (90 minutes)

This is the exercise nobody else in your scheme has done and it is the one that will get you taken seriously.

  C(t) = C∞ − (C∞ − C₀) e^(−kt)      rate(t) = k (C∞ − C(t))

Broadbalk, fitted: k = 0.01399/yr, and the rate falls from 0.867 t C/ha/yr at the start to 0.656 at year 20, 0.431 at year 50, 0.214 at year 100 and 0.070 at year 180.

Now ask the question that follows: if the gain rate falls every year, and the share is a percentage of the gain, my pay falls every year for identical work.

That is not a complaint. It is a design finding, and it has three fixes you can propose in the same breath:

1. Stage new ground. A scheme whose members enrol new blocks on a rolling cycle has a flat aggregate rate even though every field's rate is falling.
2. Pay on practice as well as stock. Hectares under four-year rotation is cheap to verify, moves monthly, and does not saturate.
3. Lengthen the period. A four-year measurement cycle matched to a four-year contract lets a slow gain be paid for at all.

Exercise 1.3 — The stratification walk (one day)

Walk the ground with the soil map. Count the strata. Then count what the scheme counts.

If the scheme is sampling on field boundaries rather than soil boundaries, it is paying for geology and calling it management, and the variance it is fighting is variance it created. Say this once, with the map in your hand, and then let the map do the work.

Exercise 1.4 — The appreciative team conversation (45 minutes)

Run this with the people who actually do the fieldwork:

"Think of a block here that has done better than its soil class says it should. Not the best block — the surprising one. What has been done to it, and for how long, and who did it?"

Take notes on practice and duration, not on yield. What you are assembling is a list of things the scheme could be paying for and currently is not.

---

PART TWO — THE ARITHMETIC

Days 31–45: compute what you are owed

Exercise 2.1 — Your share, computed from the ground up (2 hours)

  verified gain  =  (stock now − baseline)  −  (control now − control baseline)
  pool           =  verified gain × share %
  your line      =  pool × your allocation basis

Worked, from Rodale's twenty-two years:

  organic-animal    0.634 t C/ha/yr
  conventional      0.195 t C/ha/yr
  verified gain     0.439 t C/ha/yr      ← the only figure a contract may pay for

If your scheme is quoting 0.634 you are in a scheme that will be repriced, and the repricing will be presented as a cut to you. Raise it before it is discovered, and raise it as the thing that makes the scheme survivable.

Exercise 2.2 — What the carbon is worth per hectare (60 minutes)

  0.30 t C/ha/yr × 4 yr = 1.20 t C/ha
  × 3.6667              = 4.40 t CO₂e/ha
  × $30/t               = $132.00/ha gross over four years
  less MRV at co-op scale  $4.17/ha
  less a 22 % buffer
  net                      $22.45/t CO₂e  =  $24.70/ha/yr

Now the sentence to carry into any meeting about this: the whole carbon stream is about twenty-five dollars a hectare a year. A scheme sharing 20 percent of it is offering five dollars a hectare. That is not an insult and it is not a windfall — it is a fact, and knowing it stops you negotiating hard over the small stream and softly over the large one.

The large one is the input bill. Marsden: synthetic nitrogen down 86 percent, herbicide down 88 percent, at higher yield. On most arable operations that saving is an order of magnitude larger than the carbon. Ask whether it is in the gainshare measure. In most schemes it is not, and putting it in is the single most valuable change you can propose.

Exercise 2.3 — Check the measurement can see you (90 minutes)

This is the exercise that protects your pay, because a scheme that cannot detect a gain will pay you nothing and report it as no gain.

  n_unpaired = 2 (1.960 + 0.8416)² σ² / Δ²   →  614 cores per group
  n_paired   =   (1.960 + 0.8416)² σ_d² / Δ² →   50 points re-sampled

with σ = 7.50 (15 percent of 50 t C/ha), σ_d = 3.00 (6 percent), Δ = 1.20.

Ask the scheme for its n. Then ask which design. If the answer is an unpaired design with fewer than several hundred cores per stratum, the scheme is very likely unable to resolve the gain it is promising to pay for.

And bring the fix in the same sentence, because this is the moment the conversation either becomes collaborative or becomes adversarial:

"On a paired design we need about fifty fixed points per stratum instead of six hundred cores, at the same confidence. That is $2,500 a stratum instead of $30,700 — twelve times cheaper — and it is the difference between this scheme being able to pay and not."

Exercise 2.4 — Your own crossover (45 minutes)

Many improvements cost effort now and pay later. Under a short measurement period, an improvement that takes four years to show is one you are paid nothing for during four years.

The farm-level version of the same problem is computable: three transition years at $350, $270 and $190 a hectare of shortfall — $810 in total — recovered from year eight at a 32 percent premium, or year four at a 100 percent premium.

Does your scheme's period let you be paid for a year-eight improvement? If the measurement period is one year with a resetting baseline, the honest answer is no, and the rational response of every person in the scheme is to make only improvements that show inside twelve months. That is not cynicism. It is the mechanism working exactly as designed, and it is the most important thing you can put in front of whoever designed it.

---

PART THREE — DESIGN

Days 46–70: make the uncounted countable

Exercise 3.1 — Propose the paired grid (one week)

Write one page. It should contain: the strata count, fifty georeferenced points per stratum, coring to 30 cm, bulk density measured rather than assumed, duplicate splits archived, and the cost — $2,500 a stratum against $30,700.

Then the line that gets it approved: the archive means a change in protocol does not destroy the baseline. Every scheme manager has either been burned by that or knows somebody who has.

Exercise 3.2 — Propose the practice metric alongside the stock metric (3 days)

Hectares under four-year-or-longer rotation, as a share of managed area. Cheap, unambiguous, monthly, and it does not saturate.

Argue it on the scheme's own interest rather than on yours: a stock-only scheme pays nothing for three years and then argues about the result. A scheme with a practice metric pays something every month while the stock metric matures, which is what keeps members enrolled long enough for the stock metric to work at all.

Exercise 3.3 — Propose the buffer instead of the clawback (half a day)

If your scheme has a personal clawback for reversal, it has an instruction to till in a bad year written into it.

  1 % annual reversal hazard  →  9.6 % over a ten-year vintage
  buffer held                    22 %

A buffer at 22 percent against a 9.6 percent ten-year probability carries real margin, converts a reversal from a dispute into an accounting entry, and costs the scheme a known percentage instead of an unknown lawsuit. Bring the two numbers and the argument makes itself.

Exercise 3.4 — Write down what the scheme does not measure (2 hours)

Every scheme has a list. Water infiltration. Worm counts. Time to field capacity after rain. Machinery hours. Rejection rate at the gate. Nitrogen purchased.

Pick the one where you can already see a gain nobody is counting, and start counting it yourself, monthly, in a notebook with dates. An uncounted gain is never shared. A counted one is, eventually, always. Twelve months of your own dated series is the strongest negotiating position available to anybody in a gainshare, and it costs nothing but the habit.

---

PART FOUR — DESTINY AND DELIGHT

Days 71–90: make it hold

Exercise 4.1 — Get one number into the standing pack (one week)

Whatever gets reviewed monthly survives. Hectares under long rotation is the one to fight for, because it is cheap, it moves, and it is the leading indicator of everything the scheme is actually trying to buy.

Exercise 4.2 — Recruit the second owner (3 days)

One person is a hobby; two is a practice. Give them the credit for the first finding — genuinely, in writing, to their manager. It is the cheapest thing you will ever buy and it is what keeps the scheme alive when you move.

Exercise 4.3 — Count the worms (30 minutes, twice a year)

Five spade samples, same places, same season, written down with the date.

It is not sentimental. It is a time series that nobody else is keeping, it costs half an hour, and in three years it will be the only continuous biological record the scheme has. The person who kept the series is the person the scheme consults, and that has a value that shows up nowhere in the share percentage.

---

KNOW YOUR SCHEME — A CHECKLIST

Ten questions. Take them into the meeting on one page.

1. What is the baseline, to what depth, in which year?
2. Was bulk density measured or assumed?
3. Are we paid on stock or on the difference against a control?
4. What is the counterfactual, and who chose it?
5. What is n, and is the design paired or unpaired?
6. How many strata, and are they soil boundaries or field boundaries?
7. Is the verification cost inside or outside the share?
8. Does the baseline ratchet, and on what published schedule?
9. Is there a clawback, and does it fall on a person or on a buffer?
10. Does the measurement period match the time the improvement takes to appear?

Any scheme that can answer all ten is a good scheme. Any scheme that cannot answer five of them is not yet a gainshare — and the act of asking is itself the highest-value work available to you this quarter, because nobody else in the room has the list.

---

THE CONVERSATION, SCRIPTED

When you raise the 0.439 against the 0.634, or the 614 cores against the 50, you are telling somebody their scheme has a flaw. Three moves keep it collaborative.

Open with what holds. "The baseline is properly archived and the strata are soil-series rather than field boundaries, which is better than most schemes manage." Say something true and specific before anything else.

Name the finding once, with its number. "We are quoting 0.634 where the contract can only support 0.439 — the difference is what the control was building anyway." Once. Do not repeat it and do not check that it landed.

Arrive with three ways out. Reprice the contract to the difference now while it is a correction rather than a discovery. Add the practice metric so the scheme still pays while the stock metric matures. Move to paired sampling and spend the savings on a second stratum. A finding with three routes out of it is a contribution. The same finding without them is a complaint, and it will be remembered as one.

---

APPRECIATIVE QUESTIONS FOR YOUR TEAM

1. Which block here has surprised us, and what has been done to it that nobody wrote down?
2. When has this scheme paid somebody for something they genuinely did? What made that measurement work when others have not?
3. What are we already improving that nobody is counting — and what would it take to start counting it this month?