Worked structure three · Structural Options Analysis
A class of investors the vehicle cannot hold.
The vehicle is constituted and its tier is fixed. A class of investors regulated under one regime cannot hold it, because their own rules restrict the legal form they may hold, the disclosure they must receive, the oversight the vehicle must carry, or the leverage it may run. The restriction attaches to the regulatory regime those investors are supervised under, and it applies to every investor under that regime.
Seven constraints, three routes, four dependencies and one question that decides between them. Written as a type: no manager, investor, asset, value, vintage or date appears in it.
Worked structures · fifteen rooms
- 01An existing platform
- 02An ineligible asset
- 03An investor class
- 04A narrow exit
- 05A repeatable structure
- 06A co-investment
- 07A continuation
- 08Unavailable security
- 09The seed terms
- 10A change of control
- 11Drawn elsewhere
- 12An in-kind distribution
- 13A change of domicile
- 14An investor's perimeter
- 15A strategy, no vehicle
01 · The transaction as it arrives
Worked structuresThe vehicle is not broken. It is finished, and the second population arrived after it was.
A fund that has closed is a set of settled facts. Its tier was elected, its constitution was signed, its holders subscribed on terms they read, and every one of those facts was correct for the investors it had. Then a second population arrives, with capital, with appetite, and with a rule of its own that says it cannot hold what the first population holds.
Five facts describe the transaction as it reaches the desk. Only one of them is new.
The vehicle is constituted, its tier was elected at formation, and a first closing has happened. Capital is committed and some of it is usually invested.
A class of investors supervised under a regime that is not the vehicle's. They want the portfolio and agree on the economics. Their own rules do not permit them to hold the thing the portfolio sits inside.
The assets, the strategy and the merits. Those belong to the manager and were settled before anybody arrived here. The problem is about the container.
Not the investors. The restriction sits in the regime that supervises them and applies to every institution under it. There is nobody in the room to persuade.
Between closings, on a raise that already has a date. The answer has to be describable to an investment committee weeks before it can be formed.
The manager asks whether the new population can be let in. Stated that way the question has no answer, because letting them in is the one thing the vehicle cannot do. Stated structurally it becomes two questions that do have answers: what can be built beside the vehicle that the constrained population may hold, and whether it can be built without asking the investors already in for anything.
02 · The structural problem
Vehicles and legal formAdmitting a class of investors whose own regime will not permit the vehicle as drawn.
What makes this structural rather than commercial is that one of the two populations has had the question decided for it, in a rulebook it did not write, by a supervisor who will not attend the meeting.
The manager wants one portfolio and two populations. The populations do not disagree about the assets; they disagree about the wrapper, and they disagree because somebody else decided for them. An investor supervised under a regime that prescribes what it may hold does not have the option of being persuaded.
The difficulty is that the vehicle's tier was elected once, at formation, and the election fixed the disclosure it must carry, the oversight it must run and who may subscribe to it. That election cannot be taken again on different facts. So the question is never how to change the vehicle. It is what to build beside it, and what the second thing costs to run for the life of the fund.
Cannot hold is four prohibitions wearing one phrase, and they do not fail in the same way. Separating them is the whole of the early work.
What kind of thing the institution may hold. Answered only by giving them a different thing to hold. No class term and no side letter reaches it.
What it must receive before it subscribes and while it holds. Answered by a document, and a second document is a running cost rather than a structural obstacle.
What the vehicle itself must carry. The tier fixed that at formation, and no subscriber adds a function to a fund by agreeing to bear its cost.
What the vehicle may borrow. About the portfolio rather than the wrapper, and it removes every single-vehicle answer: two populations that may not run the same borrowing are two portfolios wearing one name.
Not a prohibition of the arriving regime at all. The terms the first population agreed. Any answer that changes them requires their consent, and consent has a price not paid in cash.
The first four decide which structure. The fifth decides which structures are available at all, and it is settled by reading the constitution the manager already holds.
03 · What binds
Marketing and placementSeven constraints from three directions, and six of them are already written down somewhere in the building.
Sorting constraints by source is the first useful thing anybody does. One written into the fund's own constitution can be amended by a vote at a price. One written into a rulebook cannot be amended by anybody in the transaction.
Seven, and they arrive from three directions.
From the investors' own regime: the legal form they may hold, the disclosure they must receive, the oversight the vehicle must carry, and any limit on the leverage it may run. From the vehicle's own centre: the tier, elected at formation, with the holder conditions and any cap on holder numbers the tier carries, and the classification of each subscriber against the standard the centre applies. From every place an approach is made: the promotion perimeter, which is engaged the first time somebody communicates into a territory and is engaged separately in each one.
The perimeter constraint is the one most often met late, because it is the only one of the seven that is engaged by an act rather than by a document. Nothing about it is repaired afterwards by drafting.
Written years before this fund existed. Surfaces when their counsel reads the constitution, which is after the manager has described the vehicle to them.
Same source. Surfaces in the subscription pack, as a request for a document that does not exist yet.
Same source, and the constraint that most often ends the single-vehicle answer outright. A class cannot add a function to a vehicle, and a side letter cannot promise one.
Same source. Surfaces at the first facility, or at the first asset that was to be bought with one.
Taken by the manager once, before the first closing, and not taken again on different facts. Surfaces when somebody asks whether the vehicle can be reclassified.
Applied by the centre the vehicle sits in, to each subscriber. An administrative item until the day one of them does not meet it.
Created separately by every territory an approach is made into. Surfaces when somebody asks who made the first call, and the answer is a person rather than a document.
Six of the seven can be read in an afternoon, from papers already in the building. The seventh was engaged by a telephone call, and nothing drawn afterwards repairs it.
04 · The architectures considered
Master and feederThree routes, and not one of them changes what the vehicle already is.
Three, and the comparison is set out below across the three dimensions that decide it: what the investor actually holds, what has to be run twice, and where the arrangement breaks.
A feeder that subscribes into the existing vehicle and issues its own interests to the constrained population. A parallel vehicle that invests alongside the existing one, asset by asset, on a written allocation rule. A separate class inside the vehicle as drawn, with the differences carried by the class terms and by side letters.
A fourth thing managers propose is deliberately not on the list. Re-elect the tier, or amend the constitution until the vehicle becomes the kind of thing the constrained population may hold. It fails twice over. The election was taken once, on the facts as they stood at formation. And an amendment reaches into the document the first population signed, converting a structural question into a consent sought from holders entitled to ask what they receive for it.
Three routes read on three dimensions. The last column is the one that decides.
| The route · the dimension | What the investor holds | What has to be run twice | Where it breaks |
|---|---|---|---|
| A feeder into the existing vehicle | Interests in a fund in its own right, which happens to hold a participating interest in the vehicle rather than the assets. It is a fund, it is formed as one, and everything that attaches to a fund attaches to it at formation. | A second constitution, a second offer document, a second regulator relationship and a second set of holder reporting. One portfolio, two documents. | At the first difference between the populations that is not about the wrapper. If they need different assets, different borrowing, different timing or different economics, the feeder is pretending, and the pretence surfaces at an allocation nobody can explain. |
| A parallel vehicle investing alongside | Interests in a fund that holds its own share of each asset directly, on a written allocation rule applied asset by asset. | Everything. Two constitutions, two sets of books, two valuation cycles, two boards, and an allocation decision on every transaction for the life of both funds. | At the asset that cannot be divided, and at the exit that cannot be taken by both vehicles on the same day. The allocation rule is tested by the transactions it did not anticipate. |
| A separate class in the vehicle as drawn | Interests in the same vehicle, on class terms, with the differences carried by the class and by side letters rather than by a second structure. | Nothing structural. One constitution, one set of books, one administrator, one regulator relationship. | Immediately, if the constraint is about legal form or oversight. A class is a term of a document. It cannot change what the vehicle is, which tier it was formed at, or what oversight it carries. |
05 · The critical dependencies
The one-way doorsFour dependencies, four owners, and not one of the four is ours.
Each is stated against the party who can make it true and the week it has to be true by. A dependency with no named owner is not a dependency.
Four, and the first two are answered before anything is built.
- 01 That the wrapper may be held at all That the constrained population may hold the proposed wrapper at all. Owned by counsel qualified in the investors' own regime, and answered before the wrapper is drawn rather than after it is offered.
- 02 That the approach may lawfully be made That the person who makes the approach in each territory may lawfully make it, under an instrument that permits them. Owned by the manager, mapped territory by territory, and settled before anybody speaks.
- 03 That the allocation rule exists before it is needed That the allocation rule between two vehicles is written, adopted and capable of being applied to an asset that cannot be divided. Owned by the manager and visible to both advisory committees.
- 04 That two of everything can actually be run That the administrator and the auditor can run two sets of books, two valuation cycles and two sets of holder reporting on one timetable. Owned by the administrator, and the item that decides whether the answer is affordable rather than whether it is possible.
The first two sit ahead of the drawing for the same reason. Either can end the project, and both are cheap to ask. A wrapper drawn before the holding question is answered may have to be thrown away, and an approach made before the perimeter is mapped cannot be withdrawn.
The third and the fourth decide what the structure costs to live inside rather than whether it stands. An allocation rule adopted after the first asset has been identified is a rule written by people who already know which vehicle they want to win it, and it is read that way afterwards by both advisory committees. Where the administrator cannot carry two of everything on one timetable, the parallel vehicle is not unavailable. It is unaffordable, which is a different sentence, and it sends the design back to the feeder while there is still time to change it.
A fifth item looks like a dependency and is not one. The consent of the investors already in. It is absent because the design exists to avoid needing it, and where it appears on a route's list it is the finding rather than the dependency.
06 · The architecture that survives
Parallel vehiclesOne question settles the design, and it is asked before anything is drawn.
Is the constraint about the wrapper, or about the portfolio. The three routes are not ranked against one another. Each is correct for one of the two answers, and the answer is knowable in the first week from documents already in the building.
Where the populations differ only in the wrapper they may hold, and agree about the assets, the borrowing, the timing and the economics, the feeder is the cheapest honest answer and the master and feeder topology is what it produces. Where they differ about any of those four, the feeder is a pretence and the pretence surfaces at the first allocation nobody can explain. The honest structure there is a parallel vehicle, run as two funds and sold as two funds.
The class route survives only where the constraint is about economics or reporting rather than about legal form or oversight, because a class is a term of a document and cannot change what the vehicle is. A regime that prescribes the form of the holding is not answered by a class, and it is not answered by a side letter either.
The design is therefore settled by one question asked early: is the constraint about the wrapper, or about the portfolio. The two answers lead to different structures and the wrong one is expensive for a decade.
Stated as a structure rather than as a preference, the surviving architecture is short. Five lines, each a fact about an entity rather than an intention about a transaction.
The existing vehicle stays as it is, at the tier it elected. The second is formed wherever the constrained population's own regime permits them to hold, which their counsel answers.
In the feeder, a participating interest in the existing vehicle, held by a fund whose only asset it is. In the parallel, a direct interest in each asset.
Two constitutions, two boards, two advisory committees, with the allocation rule written into both constitutions rather than into an agreement between them. A rule living in one document and not the other has a side to it.
Distributions travel to each vehicle's holders on its own waterfall. In the feeder one waterfall runs and the second passes through. In the parallel two run, and the difference is visible every quarter.
The existing vehicle touches its own centre. The second touches its place of formation and the regime of the investors holding it. Every approach touches the perimeter of the territory it is made into.
The property that matters most in both surviving routes is the easiest to overlook. Neither opens the existing constitution, so the first population is never asked to vote or to waive anything. That is the reason for building beside rather than a side benefit of it.
A second vehicle is expensive and visible, and its cost is stated at the outset. A consent sought from investors who did not have to give it is priced by them, at a moment of their choosing, and it is never given back.
07 · The implementation framework
Execution-readinessFive items in one order, and the second vehicle is formed last.
Written as dependency statements rather than as tasks. The sequence is easy to get wrong and expensive to correct, and every item on it blocks the one after it.
Map the perimeter before the first approach, territory by territory, naming the person who may lawfully make it and the instrument that permits them. Take the holding question to counsel in the investors' own regime. Adopt the allocation rule, in writing, before an asset exists to test it. Confirm what the administrator can run. Then form the second vehicle, and not before.
Prepared on counsel's reading, territory by territory. It blocks the first approach, which means it blocks the raise, and it is the only item that must be finished before anybody speaks to anybody.
By counsel qualified in the investors' own regime. It blocks the drawing. A wrapper drawn before this answer arrives is drawn on a hypothesis, and the hypothesis is somebody else's rulebook.
In writing, and visible to both advisory committees. It blocks the second closing rather than the formation, which is why it is deferred most often. No asset yet exists to make it urgent, and that is the one condition under which it can be written honestly.
Together with the auditor. It blocks nothing structural and decides almost everything about cost. Asked early, because the answer changes the route rather than the date.
By counsel in the place of formation, after the four above are settled. Forming the second vehicle first is the most common way this structure gets built twice.
The rest sits with the manager's own parties, and it is the part that has to be signed. Counsel in each relevant regime confirms what the constrained population may hold and how it may be approached. The manager appoints whatever locally licensed intermediary the perimeter map requires, which is a lead-time item and belongs on the critical path from week one. The administrator and the auditor take on the second set of books. We draw the two topologies, state what each one does to the portfolio, and mark the point at which one of them stops being honest.
One item there runs on a clock nobody in the transaction controls. Appointing a locally licensed intermediary is an onboarding process run by somebody else, on their timetable. It decides whether a raise reaches its date, and it is started last.
08 · What this case generalises to
Where structures breakRead one level up, this is not a problem about investors. It is a problem about a decision that could only be taken once.
The particular facts are a class of investors and a tier elected at formation. The shape underneath recurs wherever a binding constraint belongs to somebody else's regime and a document has already been signed by people entitled to be left alone.
Where the binding rule belongs to a counterparty's own regime rather than the manager's, the only variable is what gets built beside it. Persuasion, relationship and portfolio quality are not inputs. Sorting these out in the first week stops a manager spending a quarter on a conversation with no counterparty.
Tier at formation, domicile, legal form, the register a title sits on. Each is a one-way door, walked through before the problem arrived, and each sets the shape of every later accommodation.
Any route needing a consent from the investors already in converts a design question into a negotiation, held with people who now know the manager needs something. The cost of a second structure is stated at the outset; the cost of a consent is set by the other side.
Six of the seven constraints here live in paper and are answered on paper. The seventh is engaged by an act, in a place, by a named person. Finding it is a question about what people did rather than what was signed.
A second vehicle is a second constitution, a second set of books, a second valuation cycle, a second board and a second reporting obligation, carried for a decade. The comparison between routes is a comparison of running costs.
What transfers to another transaction is the ordering rather than the facts. Sort the constraints by who wrote them and when they can be changed. Ask whether the binding one is about the wrapper or the portfolio. Count what each answer runs twice. Then check whether the preferred route needs a signature from investors already in.
Our part ends where the drawings do. We map the constraints, draw the topologies, compare them and mark the point at which one stops being honest. Counsel in each relevant regime confirms what may be held and how it may be approached, the administrator confirms what it can carry, and the manager decides which route it adopts.
Written as a type · stated as at August 2026
The question is not which structure is better. It is whether the constraint is about the wrapper or about the portfolio, and that question has one answer.
Write to us.
Complex transactions fail at the interfaces between otherwise workable components. We resolve the structural complexity between investment intent and transaction execution.
Contact us