Haute Lumière

Commerce · I.11 · MMXXVI · daylight

La Bourse  /  Volume I  /  Nº I.11  /  Ten concept briefs

TEN CONCEPT BRIEFS · Chapter I.11 — Handing It On

One page each. A reader who reads only these ten pages has the chapter.

---

BRIEF 1 — The Separation of Economic and Control Rights

The idea. Ownership is not one right. It is at least two — the right to the money, and the right to decide — and nothing requires them to be held by the same person. Almost every enterprise that has survived its founder by a century separated them deliberately.

Worked example. Robert Bosch GmbH, as filed:

The foundation holds the money and cannot outvote anybody, so it cannot sell. The industrial trust holds the votes and owns nothing, so it cannot enrich itself by selling. Each is the other's constraint, and no shareholder alive has a personal financial reason to break the company up.

Why it matters. Most succession planning tries to find a trustworthy owner. That is a search for a person, and people are mortal and changeable. Separating the rights is a search for a structure, and a structure can be written down. It converts trust from a hope into an arrangement.

You already know this because you have seen a board where the person with the loudest financial interest was the worst judge of the long-term question, and you have wished, silently, that those two things could be held apart.

---

BRIEF 2 — The Perpetual Purpose Trust, and the Twenty-One Year Problem

The idea. A trust normally exists for beneficiaries — people who can go to court and enforce it. A purpose trust exists for a stated purpose instead, with an enforcer appointed to police it. That is how a company can be owned by an idea rather than by a person.

The trap. Under Uniform Trust Code § 409, a non-charitable purpose trust is valid for twenty-one years and no longer. A working generation is about twenty-five. The default American purpose trust therefore expires before the first successor retires — 0.84 of a generation.

The escape. Delaware, 12 Del. C. § 3556 removes the duration limit entirely. Wyoming, New Hampshire and South Dakota have comparable provisions. This is why American steward-ownership conversions are structured as Delaware trusts holding operating companies chartered anywhere.

Worked example. The Patagonia Purpose Trust holds the two percent of shares carrying all voting power. It is perpetual by design; if it had been settled under the UTC default, control would revert in 2043.

Why it matters. The filing state is not paperwork. It is the instrument. A perfectly drafted deed in the wrong jurisdiction is a twenty-one year lease on your own company.

You already know this because you have watched a good agreement fail on a technicality of where it was signed, and thought afterwards that somebody should have checked first.

---

BRIEF 3 — The Sustainable Growth Ceiling

The idea. An enterprise grows from three sources: retained profit, new debt, new equity. Lock the ownership and you remove the third one permanently. What is left has a computable maximum, from Robert Higgins in 1977:

        g*  =  ROE  x  b

  ROE  return on equity (on beginning-of-period equity)
  b    retention ratio — profit not distributed

Worked example. A firm earning 12 percent on equity and retaining all of it can grow at 12.0 percent a year and not one point faster. Retain 70 percent and the ceiling falls to 8.4 percent. At 8 percent ROE and 70 percent retention it is 5.6 percent.

The test. Put the ceiling beside the growth the business must hold to keep its market position:

        ROE x b   >=   required growth

At a 12 percent ceiling, mature retail at 3 percent is comfortable; automotive electronics at 15 percent is short by three points a year; semiconductor fabrication at 22 percent is short by ten.

Why it matters. This is the chapter's central claim in one line: succession is a financing decision before it is a governance decision, and the number can be computed in four minutes, before anything is filed.

You already know this because you have watched a business that could not raise money slowly lose a market it used to lead, and nobody called it an ownership problem.

---

BRIEF 4 — Steward-Ownership, in Two Principles

The idea. Steward-ownership is not a legal form. It is a pair of constraints that several legal forms can carry:

1. Self-governance. Control stays with people actively involved in the enterprise, and cannot be bought by people who are not.
2. Profits serve purpose. Profit is a means, not an end. Returns to capital are real, fair and capped; the residual belongs to the mission.

Worked example. The Carl-Zeiss-Stiftung statute, settled by Ernst Abbe in 1889 and substantially complete by 1896, did both a century before the term existed: no individual may own the enterprise, profit shares go to employees, a ratio caps the spread between highest and lowest pay.

What it is not. It is not a B-Corp certification, which is a standard you meet and can stop meeting. It is not a mission statement. It is not a promise by the current owner. It is a constraint on the current owner, enforceable by somebody else.

Why it matters. The distinction is between intending the right thing and being unable to do the wrong one, and only the second survives a good offer on a bad day.

You already know this because you have made a rule for yourself and then given someone else the power to hold you to it, precisely because you knew your future self would negotiate.

---

BRIEF 5 — The Employee Ownership Trust, and What It Really Costs

The idea. A UK structure from the Finance Act 2014 (TCGA 1992 s.236H onward). Sell a controlling interest to a trust holding shares for all employees and the seller pays no capital gains tax at all, against a main rate of 24 percent. The company may also pay each employee up to £3,600 a year free of income tax — worth £720 to a basic-rate taxpayer, £1,440 to a higher-rate one. National Insurance is still due.

The catch, and it decides the deal. An EOT almost never pays cash on completion. It pays out of future profits, over years, unsecured, at a discount to a trade sale because there are no synergies and no competitive tension.

Worked example. £10m business, 20 percent discount, seller discounting at 8 percent:

The break-even discount is 11.5 percent over three years and 1.4 percent over six.

Why it matters. The tax relief is not what decides an EOT. The deferral is. Shorten the payout or shrink the discount — both are negotiable and both are routinely left un-negotiated because everyone is being polite about purpose.

You already know this because you have been offered more money later and taken less money now, correctly.

---

BRIEF 6 — The ESOP and the S-Corporation Exemption

The idea. The American employee ownership vehicle, created under ERISA 1974 out of Louis Kelso's work: a qualified retirement plan that invests primarily in the stock of the sponsoring employer, usually funded by debt the company itself services.

The two tax mechanisms that do the work.

• IRC § 1042. A C-corporation seller who sells at least 30 percent to an ESOP may defer capital gain by rolling proceeds into qualified replacement property.
• The S-corporation exemption. An ESOP trust is tax-exempt. Where the plan owns 100 percent of an S corporation, the entire passthrough income is untaxed.

Worked example. $10m of pretax income at an effective passthrough rate of about 29.6 percent leaves $2.96m as tax under ordinary ownership. Under a 100 percent ESOP-owned S corporation, that 29.6 percent stays in the business every year — and it is available to service the very debt that bought the company.

Scale. On NCEO's count of Form 5500 filings, roughly 6,500 plans covering about 14 million participants.

Why it matters. It is the most heavily used employee-ownership structure in the world and it is self-financing by design: the company buys itself with money it would otherwise have paid in tax.

You already know this because you understand that a mortgage paid by the tenant is a different proposition from a mortgage paid by the landlord.

---

BRIEF 7 — The Operating Memory

The idea. Across a long tenure, roughly nine to ten thousand non-routine judgements are made, of which perhaps 85 percent are never written down — call it eight and a half thousand decisions living in one head. You cannot transfer them and should not try. You transfer four things.

The four.

1. Twenty standing decisions. Each with: what was decided, what the alternative was, and what would have to change for the answer to change.
2. Ten relationships that an introduction does not transfer, handed over in person, one at a time, over a year.
3. Five numbers you actually read to know the business is well. Not the pack — the five.
4. One page on what it is for, in your own sentences, short enough to sit in the trust deed as a recital.

The arithmetic. About 120 hours of work. Across an eighteen-year tenure that is 6.7 hours a year — the cheapest item in the whole succession.

The clause that carries it. What would have to change for this answer to change is what converts a rule into a judgement a successor can exercise. A rule without it produces obedience; with it, it produces a colleague.

Why it matters. Half the entries will surprise you and two will turn out to have been wrong for years, because writing the change-condition is a test the answer can fail.

You already know this because you have explained something out loud to a new person and discovered, mid-sentence, that you no longer believe it.

---

BRIEF 8 — Reserved Matters: the Constraint, Not the Wish

The idea. A succession structure protects the enterprise by naming a very short list of things the board may not do without a specific consent — and then leaving everything else genuinely free.

The three that recur. Disposal of the enterprise or a controlling interest. Alteration of the stated purpose. Alteration of this list.

Why three and not eight. Every additional reserved matter converts a successor into an applicant. A structure that permits nothing is broken within a decade by people who need to run a business; a structure that reserves three things can be honoured indefinitely, because honouring it costs almost nothing day to day. The century-old structures reserve almost nothing.

The drafting failure. A purpose written as "to act in the long-term interests of the company" constrains nothing, and any competent buyer's counsel can satisfy it. The purpose must be specific enough that a breach is visible to an ordinary reader.

The governance failure. A single trustee is a single point of capture. Three to five, staggered terms, at least one with no prior relationship to the founder — and a stated method for appointing trustees themselves, which is the omission that has voided more of these structures than any other.

You already know this because you have worked under a policy so detailed that everyone routed around it, and under a short rule everyone kept.

---

BRIEF 9 — The Deferral Discount

The idea. Money promised is not money held. Any succession paid out of the company's future profits — an EOT, a leveraged ESOP, a management buyout, vendor finance — hands the seller an unsecured asset of uncertain timing, and it must be discounted at the seller's own rate, not at zero.

The formula.

  PV  =  (price / n)  x  sum over t=1..n of 1 / (1 + i)^t

Worked example. £8m over six years at 8 percent: the annuity factor is 4.623, so 8.0/6 × 4.623 = £6.16m — not £8m. Against a taxed trade sale netting £7.60m today, the purpose-led route is £1.44m worse, and the relief has been entirely consumed by time.

The useful inversion. Solve for the discount at which the two are equal:

Why it matters. This single table turns a values conversation into a negotiation with two live variables — term and discount — and it is the honest negative of the whole chapter. Discounting at zero is the most common error in these transactions and it is always made in the seller's own favour, right up until it is not.

You already know this because you have been paid in instalments and noticed that the last one felt smaller than the first.

---

BRIEF 10 — Financing One Level Down

The idea. When the lock is right but the ceiling is too low, the answer is not a looser lock. It is to raise capital one level down: against a named asset, in a named subsidiary, with a named lender — never against the purpose.

Worked example, and it is the documented one. The Carl-Zeiss-Stiftung has owned its enterprises since 1889. By the 1990s the capital requirement had outrun what the statute could finance. Zeiss did two things: in 2000 it floated Carl Zeiss Meditec on the Frankfurt exchange, and in 2004 the statute was reformed so that Carl Zeiss and Schott became joint-stock companies wholly owned by the foundation, expressly so they could reach capital markets. Control of the parent never moved.

The counter-example. John Lewis Partnership: £1bn of stated investment need, £56m of profit before tax in the year to January 2024 after a £234m loss the year before — 17.9 years of profit to fund a five-year plan, with no equity to issue and no financeable subsidiary to float. The partnership bonus, paid every year from 1953, was not paid in 2022, 2023 or 2024.

The rule. Design the lock and the capital route as one decision. Taken separately, the lock always goes first — it is the emotionally satisfying one — and the capital question arrives four years later as a crisis.

You already know this because you have remortgaged one property rather than sell the house you actually live in.