Haute Lumière

Commerce · VI.11 · MMXXVI · daylight

La Bourse  /  Volume VI  /  Nº VI.11  /  Workbook — the executive

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter VI.11 · The Limits of Self-Governance

For the executive. Applied to a P&L, a board paper and a business unit. The firm's own language, without apology — including the sentence most governance material will not print, which is what the firm is actually for.

---

THE PREMISE, STATED COMMERCIALLY

Your firm already runs self-governing units. It does not call them that.

Safety committees that write their own standing orders. Works councils. A franchise network with a franchisee advisory board. An industry association that sets a code your firm is judged against. An open-source foundation whose roadmap you depend on and do not control. A joint venture board. A supplier consortium. A trading desk with a limits framework it administers itself. An internal platform team that the other teams govern by committee.

Every one of those is a commons with a boundary, a membership, a rule set and a sanction. Almost none of them has ever been sized against the arithmetic, and the cost of that shows up as the same three failures: a group that has grown past the point where peer pressure works, a decision cycle slower than the thing it governs, and a well-run body imposing costs on somebody with no seat.

This workbook costs those three and gives you the instrument that fixes them. It is a ninety-day piece of work and it sits inside one person's discretion.

And the premise underneath it, stated without embarrassment. A firm is a device for making a decision when deliberation is too expensive, and the price of the device is that somebody is subordinate. That is Coase's finding in this chapter's language. You do not have to be uncomfortable about it. You do have to know where the device stops paying, because that is the boundary of your own competence, and it is where delegated governance earns its keep.

---

PART ONE — DISCOVERY

Days 1–30: inventory the self-governing units you already run

Exercise 1.1 — The register (one day)

List every body inside or adjacent to your firm that makes rules affecting your P&L and is not a line-management decision. For each:

Most executives find between four and eleven. The column that will surprise you is the fourth, because for most of these bodies the honest answer is "nothing formal — people would be embarrassed."

Exercise 1.2 — Find the one that is already outperforming (one week)

One of them is doing something the line organisation could not do as cheaply. The usual candidates: near-miss reporting that a formal process would never surface, a franchisee group that polices brand standards more strictly than your field team, a maintenance community that keeps an asset running past its schedule, a security working group that finds things the annual audit does not.

Cost it on the sceptic's metric, which is cost per unit of compliance. What would it cost you to obtain that level of adherence by inspection? The chapter's worked case has a state warden at £1.00 a hectare and a member-monitored charter at £0.55 — a 44.7 percent saving with the oversight cost already included. Find your equivalent ratio. It is usually better than you expect and it has usually never been calculated.

Exercise 1.3 — The shadow roll (half a day)

For your three largest self-governing bodies, list who is materially affected by their decisions and has no seat. Suppliers below a threshold. Contractors. Customers on a legacy tariff. A site that joined after the rules were written. The graduate cohort. A joint-venture partner's employees.

This is your standing exposure, and in a regulated industry it is where the enforcement action comes from.

---

PART TWO — THE ARITHMETIC

Days 31–45: three thresholds, computed for your own bodies

Exercise 2.1 — The reputational ceiling

  n* = D × (S_max / g)                D = 150

g is the gain to an individual from one quiet non-compliance. S_max is the full career cost of being known to have done it in a place where everybody knows everybody.

At the chapter's parameters — a £500 gain against a £1,500 sanction — the ceiling is 450 members. Above it, deterrence requires a detection probability above one, which is not a number: the failure is impossibility, not inefficiency.

Run it on each body in your register. Anything above its own n* needs a formal sanction with a fine or a licence behind it, and it needs it now rather than after the incident.

Exercise 2.2 — The governance clock

  T_g ≤ T_d / 2

For each body, the fastest variable it governs and its own decision cycle. A monthly committee governing a three-day doubling is 20× too slow; a quarterly architecture board governing a weekly release cadence is not governing it.

The board-paper sentence this produces: "The following decisions are reserved to a named individual, acting alone, because the committee cycle is structurally unable to reach them in time. Each such decision expires in N days and is reported to the committee at its next meeting." That sentence is worth more to a risk committee than any amount of governance narrative, because it is the only version that survives an incident review.

Exercise 2.3 — The price of your decision rule

Consensus is a choice with a price and the price is meetings. Thirty members, each assenting with probability 0.9: unanimity passes 4.24 percent of proposals — 23.6 tabled to pass one — and costs 2.65× the Buchanan–Tullock optimum of 29 of 30. The last signature costs more than the whole rest of the decision.

Compute it for your own bodies, then put the number in the terms of reference next to the rule. Groups that see the figure usually change the rule, and the ones that keep it can tell you exactly what they are buying.

Exercise 2.4 — The monitoring crossover

  mutual = c × n(n−1)/2        warden = F + v × n

At the chapter's parameters the crossover is 59 appropriators. Past it, a paid inspector is cheaper than peer monitoring, and the finding is not a judgement about your people — it is that one curve is quadratic and the other is linear.

Exercise 2.5 — The decision inequality for the instrument

      state (or corporate) enforcement cost at the required detection rate
   −  ( member levy + residual oversight + advocate )
  ------------------------------------------------------------------------ > 0

Worked in the chapter: £12,000 of central enforcement against £6,640 of chartered self-enforcement, a saving of £5,360 a year, with a levy of £27.67 per member against a right worth £520 — 5.3 percent.

The soft term is the detection probability, and it is the one that can be gamed to nothing. A charter whose members file monitoring they did not do costs you the levy and deters no one. Everything in Part Three exists to stop that.

---

PART THREE — DESIGN

Days 46–75: the delegated management charter

The instrument, in the form your general counsel will recognise, because it is already running in three industries: FINRA under the Exchange Act, Chile's management areas, Maine's lobster zones, Namibia's gazetted conservancies.

The structure: a delegated management charter with a statutory floor, four reserved powers, a member levy and a reversion trigger.

3.1 The grant. An exclusive right to make and enforce rules over a defined scope, for a defined term — five to ten years. The right is the consideration. Everything the body gives you is paid for with it, which is why the term matters and why an indefinite grant is worth less to both sides than a renewable one.

3.2 The floor. The minima the body may not go below, named individually. A floor described as "applicable law and company policy" is not a floor; it is a sentence. Draft it so that amending it is harder than amending the rules above it — a longer notice period, a higher threshold, a mandatory hearing of the advocate. Maine's V-notch is in statute and the trap limit is in the zone council. That is the right way round and it is the whole design.

3.3 The four reserved powers.

3.4 The levy. The body funds its own monitoring. Set it against Exercise 2.5, and disclose the ratio of levy to the value of the right — below about a tenth it is uncontroversial; above about a third the body will start behaving like a taxed entity rather than a governing one.

3.5 The reversion trigger. Defined, measurable, automatic. A breach of the floor; a failure to file; monitoring below a stated threshold. Chile's management areas revert on failure to submit a management plan, and that is why the plans arrive.

---

PART FOUR — THE BOARD PAPER AND THE ACCOUNTS

Days 76–90

4.1 The balance-sheet treatment, which is where this becomes real.

• The charter is an intangible with a useful life equal to its term, amortised over it, and impaired in full on a reversion trigger. That treatment is the point: it converts the trigger from a sentence in an appendix into a number your controller has to carry.
• The levy is an operating cost, and it must not sit inside the governed body's own discretionary budget — a line the body can cut is not a levy, it is an intention.
• The advocate's budget sits outside the body's control. One to three percent of the scope's annual value is a defensible range. A validation line a sponsor can cut is not independent whatever the charter says, and the same logic applies here exactly.
• The reserved substitution power is a contingent liability and should be disclosed. A counterparty relying on the body's rules is exposed to a power the body does not hold, and your auditors will ask about it eventually. Ask first.

4.2 The board paper, one page.

4.3 The failure modes, named so you see them coming.

Capture. A body funded by its members and staffed from them will discover, with nobody deciding to, that its enforcement is gentle. This is the standing charge against every self-regulator in finance and it is not baseless. Counter: publish the enforcement count annually. A body with no enforcement actions in three years has found nothing, not confirmed everything.

The floor lowered by amendment. The commonest failure, and it never feels like dishonesty from the inside; it feels like clarifying a definition. Counter: the amendment threshold, and the advocate's mandatory hearing.

The emergency power that never sunsets. Each crisis leaves a permanent enlargement, and one year the slow forum wakes up with nothing left to decide. Counter: the expiry is automatic, not discretionary.

Local capture. Bardhan and Mookherjee's finding is that there is no presumption a local body is less captured than a central one; where local inequality is high and accountability weak, devolution moves the rent rather than removing it. Counter: the advocate, and an externally appointed chair.

Scope drift. The body begins deciding things outside its grant because it is the only body meeting. Counter: the approval power, used small and early. A reserved power never exercised has lapsed in practice, and the first exercise after ten years will be litigated as an outrage.

---

WHERE YOUR OWN NUMBERS ALREADY SUPPORT THIS

Three places the case is usually already made inside your accounts and nobody has assembled it.

Assurance cost. Add up what you spend on inspection, audit and compliance verification for one scope. Compare it with the chartered alternative at the same detection probability. The chapter's ratio is 44.7 percent and the oversight cost is already inside it.

Incident response time. Pull the last three incidents in a scope governed by a committee and measure from first signal to decision. Set that against the variable's own T_d. Where the ratio exceeds one, you have been relying on individuals acting outside their authority and getting away with it, which is a finding you would rather have now.

Turnover and adherence in franchised or federated units. Units governed by a body of their peers frequently show better rule adherence and lower churn than units governed by field inspection. If that is true in your estate, it is in your data already, and it is the cheapest evidence you will ever gather.

---

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

1. Which of our self-governing bodies has quietly outperformed the line organisation, and what has it been doing that we have never costed?
2. When has an outside authority — a regulator, a court, a standard-setter — done something for us we could not have done for ourselves?
3. Where have we successfully dealt with a party who had no seat at our table, and what made that work?
4. If every body in our register arrived with its three thresholds computed, what would our governance committee sound like?
5. Imagine our charter is the one our industry copies. What is on its first page?
6. What would we do with the time we got back if every body knew exactly what was not its to decide?
7. What is our floor — what could we not vote to do even if we all wanted to — and would two of us write the list the same way?
8. Who is harmed by a decision we make well, and what would it take for them to have a seat and a budget here?
9. At what size does peer accountability stop working in this firm, and what goes in its place the quarter before we get there?
10. Which decisions arrive too fast for a committee, who should make them, and when does that authority expire?
11. What would be the first sign our floor had been lowered, and who would notice it first?
12. What would have to be true for this charter to be renewed by executives who have never met any of us?