Who else should own this?

Propose an ownership and governance structure for a new build hospital in this market, involving external partners, that answers the grounds in the board paper without destroying the commercial proposition.

A complete answer names the partners by type rather than by name, states what each contributes and what each receives, sets out who decides what, and explains how the commitments in the ethics paper survive a year in which the group would prefer to relax them. It must also say how the thing ends, because a structure that cannot be unwound has not been designed.

Two questions are being asked at once here and they are routinely confused, so the answer has to separate them. One is who controls the hospital, which is what the comparator set below is about. The other is who provides the capital, which is a different axis running through all of them: institutional investors, Nigerians abroad, the clinicians themselves, or the group’s own balance sheet. A structure proposal that settles control and is silent on capital is half a proposal, and one that settles capital and calls it a partnership has answered the easier question.

What the answer may not do is treat this as a financing exercise wearing a governance costume. The grounds in the board paper are ethical, legal and operational, and equity placed with investors who bring only money answers none of them. If the proposal would work identically with every partner replaced by a bank, it has missed the instruction. That is not an argument against private investors, who are in the comparator set and may well be the right source of capital. It is a requirement that the control question is answered on its own terms.

It also may not assume the answer is partnership. The comparator set below includes the wholly owned company the board paper argued against, deliberately, because a comparison that excludes the incumbent option is advocacy with a table attached.

Four findings are carried into this instruction as settled, in the sense that the answer may build on them and should not spend its length relitigating them.

The extraction charge is the decisive ground and it is the one a structure can remove rather than repay. Joint clinical appointments across two institutions mean the consultant is shared instead of taken, and there is then nothing to repay. That is the strongest single idea available here and it is the reason a hospital partner comes before a financial one.

The exclusion charge cannot be fully answered by any structure, and can be answered in part by one. A share of capacity given away is charity, which a board can quietly reduce in a difficult year. Contracted volumes from a public or insured payer are an obligation to a counterparty who notices, which is a different and more durable thing.

The enforcement problem has no internal solution. Every commitment the group has made is measured, audited and reported by arrangements the group chooses, on behalf of two groups with no exit and no voice. Only an owner or a trustee with standing and no interest in relaxation changes that, which makes it a structural question rather than a governance one.

And the operating standard is the product. The group sells demonstrated, transferable performance, so any structure that allows a partner to dilute the standard is not a compromise on control, it is the destruction of the asset. That is why control of the standard carries a heavy weight below while control of everything else does not.

One further point belongs here because it cuts against the direction of everything else. Bringing in private investors raises capital and lowers enforceability, because each one is a party with a financial interest in relaxing exactly the commitments this instruction exists to protect. That is not a reason to exclude them. It is the reason the structure has to hold the commitments somewhere an investor cannot reach, and a proposal that raises equity without doing that has made the original problem worse while appearing to address it.

Six structures, named before anything is proposed, so that the eventual recommendation is scored against a set it did not choose.

Two are excluded and the exclusions are stated rather than left as absences. Full public ownership is not available to this group on any realistic terms and would not return the capital. Full charitable ownership is a different business rather than a different structure for this one, and it is dealt with in the board paper as an alternative use of the money rather than as a way of doing this.

Anybody who wants to argue with the answer that eventually arrives should start with those two exclusions, or with a seventh structure nobody here thought of, rather than with the scoring inside the set.

Seven criteria, weighted, fixed before any structure has been drawn. The weights are the argument, and somebody disagreeing with the recommendation should be able to point at the weight they would change.

The first carries more than a quarter on its own, for the same reason the equivalent criterion dominated the first instruction: it is the only one where a structure can change the sign of the outcome rather than its size. Every other criterion makes the project better or worse. This one decides whether the project takes something from people who cannot object.

The third deserves a note because it runs against the direction of the rest. Six of these criteria reward giving things away, and one rewards keeping something. That is not an inconsistency. The operating standard is not a lever of control in the ordinary sense, it is the asset, and a structure that trades it has sold the thing it was built to protect.

The threshold is a matter for the board rather than for this page. What the instruction does require is that any structure scoring below the wholly owned option on this set is reported as such, plainly, including if that is the recommendation.

Five, and three of them are not the group’s to relax.

The commitments in the ethics paper stand in whatever structure emerges. Emergency treatment regardless of means, no detention for debt, a published share of capacity at no charge, the clinician count from a pre-opening baseline, a stated crossover date and a price fixed before consent. A structure that requires any of them to be softened has answered a different question.

No structure may require or tolerate a breach of the emergency treatment provision, which binds every establishment in this market whatever its ownership. A partner who expects otherwise is not a partner.

Any vehicle that will hold land must have the Governor’s consent position contemplated at formation, including on a future change of control. A structure that is sound at incorporation and void at the first transfer is worse than no structure, because it will be discovered at the moment of maximum value.

The academy is contracted before land is committed, which was a condition in the board paper and survives into this one unchanged.

And the operating standard is not delegable. Whatever the equity, the group retains the ability to set, measure and hold the clinical and operating standard, in the constitution rather than in a side agreement.

Four risks that belong to this decision rather than to hospitals in general. An answer is not complete unless it says what the proposed structure does about each.

The fourth is the one most often skipped, and it is worth stating why. Multi party vehicles are negotiated by people whose whole attention is on getting started, and the provisions that matter most are the ones that operate fifteen years later when everybody in the room has gone. Model the exit before signing the entry. If the structure cannot be shown to unwind, it has not been designed, it has been begun.

Six things, and the fourth is the one that decides whether this exercise was worth doing.

Everything else in a structure proposal is recoverable. Get the equity split wrong and it can be renegotiated. Get the reserved matters wrong and they can be amended. But the entire reason for bringing partners in, on the analysis that produced this instruction, is that the commitments need somebody outside the group standing behind them. A proposal that brings in partners and leaves the group still grading itself has added complexity and changed nothing.

This instruction sits ahead of the first gate rather than inside it, which is unusual and is the point.

A multi party vehicle has to exist before it can option land, because the entity taking the option is the entity that will hold the title, and in this jurisdiction the consent position on any later transfer depends on who holds it. Optioning land in the group’s own name and contributing it to a vehicle later is not a neutral sequencing choice: it is a transfer, with a consent requirement and a cost attached.

It also comes first because the structure changes the brief. A vehicle with a public partner may be offered land rather than buying it, which reopens the parcel question entirely. A clinical partnership may put the hospital next to the partner rather than where the catchment analysis would put it. Both of those are decisions that instruction 01 assumed it was making alone.

So the honest position is that this instruction partially reopens the first one, and the answer to it may change the answer to that. That is uncomfortable and it is better than the alternative, which is committing to a parcel and then discovering that the structure needed a different one.