Who holds the interconnection position

An interconnection position is a bundle of rights and obligations owed by and to a named party. Somebody applied, somebody paid the deposits, somebody signed the study agreements, and somebody will be liable for the network upgrades. That party is the customer of record, and the position is theirs, not the site's.

This is the point most often missed. A queue position is not an attribute of a parcel the way a zoning classification is. It is a contractual relationship attached to an applicant, studied against a specified parcel and a specified load. Land and position can be separated — by accident more often than by design — and when they are, neither half is worth what the pair was.

The consequences run through everything above it. The shell cannot be financed on a longer horizon than the power under it. A hosting agreement cannot commit capacity the facility's own supply arrangements do not durably provide. A compute contract that runs past the power position behind it commits capacity that may not exist for the whole term. Each of those is a seam, and each of them is only as good as the entity holding the bottom of it.

So the question is not merely administrative. Before the position can be verified as real — which is a separate exercise, worked through in [how to verify a power claim](/sites/verifying-a-power-claim) — somebody has to decide where it will live. That decision determines what can be sold, what can be secured, what needs consent, and what happens on enforcement. It is cheap to make deliberately at the outset and expensive to unwind afterwards, because unwinding it usually means asking a utility for a consent it has no obligation to give.

There are only a few realistic answers, and they trade against each other. None is correct in the abstract; the right one follows from what has to be financed, sold or separated later, and from which party the utility will accept as its counterparty.

Two criteria decide most cases. The first is whether the holder will still be there. A position held by an entity that is going to be sold, wound up or restructured has a transfer in its future, and every transfer is a consent event. The second is whether the holder can meet the obligations that come with it. Utilities generally require deposits, study payments and, where network upgrades are involved, commitments that can be substantial. An entity holding valuable rights and no means of funding the obligations attached to them is a structure that will need support from somewhere, and the shape of that support determines how separated the layers really are.

A third consideration applies where the project intends to raise capital at more than one layer: the position should sit at, or below, the level where the capital that depends on it is being raised. A position held above a project entity is outside the reach of anyone financing that project, however comfortable the group relationship looks from inside.

Whatever the answer, it has to survive the transactions the project is going to need, and interconnection paper is generally unfriendly to those.

The recurring restrictions are ordinary but consequential. Assignment usually requires the utility's or operator's consent, and consent is a process rather than a formality. Some positions are non-transferable outright. Others can move but only alongside the site, or only to a party meeting stated creditworthiness criteria, or only with a re-study — and a re-study is not an administrative step but a reset of the thing that made the site valuable. Changes to the studied parcel, the point of interconnection, or the requested load can all reopen the analysis.

A change of control is a transfer in substance even where the named holder never changes. Structures that put the position in an SPV precisely so it can be sold by selling the SPV work only where the documents do not treat a change in the SPV's ownership as an assignment. That is a drafting question with a large commercial consequence, and it is answered by reading the agreements rather than by assuming the structure does what it was designed to do.

The practical discipline is to establish, in writing and early: what consent a transfer requires, who gives it, what tests they apply, how long they take, and what a change of control does. Those four answers determine whether the position is an asset that can be moved into a financeable structure, or one that is stuck where it was first filed.

Most of the stack can be secured directly. Land takes a mortgage or charge, equipment takes a security interest, contracts are assigned. The interconnection position frequently fits none of those cleanly, and pretending otherwise produces a package that reads better than it performs.

The usual answer is indirect. Rather than taking security over the position, capital takes security over the entity that holds it — a pledge of the SPV's equity — so that enforcement delivers the company with the position inside it, rather than an attempt to assign a right that cannot be assigned. This is why the SPV answer is common: it makes the unpledgeable pledgeable by wrapping it.

That structure still has to be tested against the same restrictions. If the documents treat a change in the SPV's ownership as an assignment requiring consent, enforcement runs straight into the utility. The customary response is to ask for acknowledgement in advance: a direct agreement under which the counterparty accepts a specified financing party, agrees to notify it of default, and permits a transfer on enforcement to a party meeting stated criteria. Whether such an arrangement is available differs by market and by counterparty. Establishing whether it is available at all is worth doing before the structure depends on it.

The other half is the obligations. Deposits already paid, study costs, and network-upgrade commitments sit with the holder. A structure that puts the rights in one entity and the funding obligations in another creates an entity that owes money it has no means of paying — and utilities generally have remedies, including forfeiture of the position, when those obligations are not met.

The ways this goes wrong are consistent enough to list.

• The position and the land drift apart. A position filed by an early developer, over a parcel later sold, is a live problem for both halves. The buyer holds land whose power belongs to somebody else; the original applicant holds a position over land it does not control.
• The studied parcel is not the parcel. Boundaries move during assembly, options lapse over part of a site, and a position studied against a different footprint is not obviously portable to the new one.
• Site control expires before energization. An option or lease that runs out before power arrives leaves the holder with a position over land it no longer controls — the countdown problem set out in the [site financeability gates](/sites/what-makes-a-site-financeable).
• The holder is the wrong credit. Utilities frequently require credit support from the customer of record. A thinly capitalised SPV may be unable to provide it, which forces a parent guarantee and quietly re-couples the layers the structure was separating.
• Enforcement delivers half an asset. A financing party that can take the land but not the position is left with a parcel whose value was the position. The pledge structure exists to prevent exactly this, and it prevents it only if it was put in place before it was needed.
• Nobody decided. The most common failure is not a wrong answer but no answer: the application was filed by whichever entity existed at the time, and every subsequent structure inherits that choice.

Continuum advises on how these positions are held and structures the entity arrangements around them. It is not a utility, not a bank, not a broker-dealer and not a direct lender; it does not hold client funds.