Long-lead equipment procurement finance

In an ordinary development sequence, capital arrives in an order everyone recognises: control the land, establish the power position, obtain entitlements, complete design, close the financing, then buy the equipment. Procurement sits near the end because that is where it belongs — you buy what you have decided to build, once someone has agreed to pay for it.

Electrical balance of plant has broken that sequence. Lead times on large power transformers, generator step-up units and the associated switchgear and breakers are now measured in years rather than months, and the queue for a slot is longer than the development cycle that would otherwise justify joining it. A developer that waits for entitlement before ordering has, by waiting, added the whole of that lead time to its energization date — and on a project whose entire commercial premise is [speed-to-power](/learn/speed-to-power), that is usually the difference between a viable schedule and no schedule at all.

So the order moves to the front. It is placed during site control, frequently before entitlement is certain, generally before final design is frozen, and almost always before any lender has approved anything. The consequence is a cash profile that no development model was built to describe:

• Money leaves early and in size. A slot is reserved against a deposit, and the payment schedule runs through manufacture rather than on delivery. A material share of the equipment cost is spent while the project is still a set of options and applications.
• It leaves before anyone has underwritten it. The payments are made against a plan, not against an approved facility, so the developer is funding them from its own resources or from whatever early-stage capital it has available.
• It leaves against a project that may not proceed. Entitlement can fail, the power position can prove weaker than represented — the whole subject of [verifying a power claim](/sites/verifying-a-power-claim) — and the design the equipment was specified against can change.

This is a financing problem before it is a procurement problem, and it is usually treated as the second thing. The developer's own working capital absorbs an exposure that is larger, earlier and less reversible than anything else at that stage of the project.

The difficulty is not that the equipment is bad collateral. It is that, for most of the period in question, there is no equipment.

Four characteristics compound, and each one on its own is manageable.

There is no asset. What exists is a contractual right to receive equipment at a future date. Until manufacture progresses far enough for goods to be identified, there is nothing to attach security to in the ordinary way — no serial number, no marked unit, nothing in a yard. Security in this period is taken over rights and money, not over things.

There is no approved project. Permits are outstanding, design is not frozen, and the site's own position is still being established. A financier looking at the order is looking through it at a project that does not yet clear its own gates, which is the analysis set out in [what makes a site financeable](/sites/what-makes-a-site-financeable).

There is often no borrower a financier would recognise. The order is placed by a development entity with thin capitalisation, whose main asset is the option package it is spending money to convert. The obligation to keep paying the manufacturer runs for years and outlasts most of the assumptions behind it.

The payments are front-weighted and non-refundable in practice. Deposit and progress payments are structured to fund the manufacturer's own procurement and production, and the refund terms available to a buyer in a seller's market are narrower than a buyer would like. Cancellation is not a costless reversal; it is a loss crystallised at whatever point the buyer stops.

The honest summary is that this exposure is financed in the market less often than it needs to be. The technical and scheduling literature on long-lead electrical equipment is extensive and good. The question of who carries the deposit-to-delivery position, on what security, and on what terms is answered almost nowhere — which is why it is generally answered by default, on the developer's balance sheet, rather than by design.

The practical work is to recognise that the security position is not static. It changes at identifiable points between the deposit and energization, and both the exposure and the available protection change with it.

The table below reads as a progression. What matters when reviewing a structure is not whether it has security in the abstract, but whether it has the security that is actually available at the stage the exposure sits in — and whether the documents step it up automatically as the position matures, or require someone to remember.

The order does not stay unfinanceable. It becomes financeable as the project underneath it catches up, and structuring this well means arranging in advance for each of those steps to convert into terms rather than negotiating them from the beginning each time.

Four things change the picture, roughly in the order they arrive.

Entitlement resolves. Once permits are in hand and the site's power position is documented, the equipment is no longer speculative — it is committed to a project that can be built. The same order that was an unsecured bet becomes a critical-path component of an approved development.

The order is novated or assigned to the project entity. Equipment ordered by a development company has to end up held by the entity that will own the plant, and the mechanics of moving it matter. Manufacturer consent to assignment is usually required, is easier to obtain when asked at the outset than when needed, and is one of the most commonly overlooked provisions in a supply contract. Where a financier is expected to take security over the contract, its consent regime is part of what is being underwritten.

Credit support substitutes for the missing collateral. This is the mechanism that does most of the work in the interval, and it runs in both directions. A manufacturer taking an order from a thinly capitalised buyer wants assurance the remaining payments will be made; a buyer paying substantial sums in advance wants assurance it is not simply an unsecured creditor of its supplier. Letters of credit, advance payment guarantees, performance bonds and parent support are the ordinary answers, and the way they are structured — validity periods, documentary conditions, what actually triggers a drawing — is set out in [the LC-backed equipment facility](lc-backed-equipment-facility). Two points are specific to procurement at this stage: the support has to run for a period measured against a delivery date that may move, and cover posted for it is capital the developer cannot spend on the project.

The equipment becomes conventional collateral. Once units are identified, delivered and installed, the position converts into ordinary equipment security and folds into the project's own package. At that point the analysis is the familiar one — identification, access, priority — and the procurement structure has done its job and should be designed to retire cleanly into whatever takes it out.

The practical structuring question is which of those milestones a facility is sized against, and whether the terms step down as each one is reached. A structure priced for the pre-entitlement position that never reprices as the project matures has charged for a risk that stopped existing, and the developer will notice.

Every long-lead order is ultimately measured against one date: when the site can take load. That date drives the offtake commencement, the hosting commitments made to customers, the conversion test on the construction facility, and the point at which the project starts producing anything.

What makes electrical balance of plant distinctive is that it is unforgiving in both directions.

Late delivery is not a delay, it is a stop. A site with generation, a building and no energized connection produces nothing. Unlike many components on a construction schedule, this equipment has no realistic substitute and no expedited path — a replacement order rejoins the same queue at the back. A slip of months therefore propagates in full to the energization date, and from there into every contract that referenced it. This is the exposure [bridge to energization](bridge-to-energization) is written against, and long-lead equipment is one of the most common reasons the bridge period extends.

Early delivery is a cost, not a win. Equipment that arrives ahead of civil works sits in storage, insured under policies that may not contemplate long-term outdoor storage of high-value items, exposed to site conditions and to whatever the site's own security arrangements are. It also means the money went out earlier than it needed to. A schedule that pulls the order forward for safety should be explicit about what that safety costs.

Because of that, the delivery date is a credit term rather than a logistics term, and it deserves the same reading a financial covenant would get. The questions that settle it: what the contractual delivery window actually is, as opposed to the indicative date in the correspondence; what the manufacturer's remedy for delay is and whether it is capped at a level that means anything against the project's real loss; whether the schedule is re-forecast on a defined cadence or reported when asked; and what the project's contingency is if the units are late — because on this equipment, in most cases, there is not one.

The most useful discipline is to hold the delivery date and the downstream commitments on the same page. Offtake commencement, customer capacity dates and facility conversion tests are frequently negotiated by different teams against different assumptions, and the equipment schedule is the only one of them that a third party controls.

How procurement positions on this equipment go wrong:

• The deposit was never secured at all. Substantial payments made against a supply contract with no assignment, no refund arrangement, no vesting provision and no credit support from the manufacturer's side. This is the single most common and largest unmitigated exposure in the whole sequence, and it is usually discovered when someone tries to finance the project and asks what the payments bought.
• The order cannot be moved. Equipment ordered by a development entity that the supply contract does not permit to assign or novate without consent the manufacturer has no incentive to give quickly. The project entity that is supposed to own the plant cannot be given the order, and the financing waits on a consent nobody sought.
• The specification changed and the order did not. Units released to manufacture against a design that was subsequently revised. Change orders on long-lead equipment carry both cost and schedule, and a change made late can return the buyer to the queue for the affected items.
• Credit support expires before delivery does. Support instruments issued for validity periods set against an original delivery date that has since moved. The equipment arrives after the protection has lapsed, which is the tenor problem described on [the LC-backed equipment facility](lc-backed-equipment-facility) in its most avoidable form.
• The project failed and the slot is orphaned. Entitlement lost, power position not confirmed, or the sponsor unable to continue paying. What the order is then worth depends entirely on whether the slot or the units can be transferred to another project — which is a question about the contract's assignment terms and about demand, not about the equipment.
• The equipment arrived and the interconnection did not. Energization requires both the equipment and the utility position, and they are held by different parties on different schedules. A project that solved its procurement problem and not its interconnection problem has bought expensive equipment it cannot use, which is why [who holds the interconnection position](who-holds-the-interconnection-position) is asked at the same time.
• Procurement was financed on the project's terms rather than its own. The deposit-to-delivery exposure is a distinct position with a distinct tenor and a distinct risk profile. Absorbed into a construction facility or into the developer's general working capital, it is funded by capital that never priced it and is not visible to anyone as a position — the same layering argument made in [the data-center capital stack](/learn/the-data-center-capital-stack), applied to the earliest money in the project.

Continuum structures and arranges transactions around long-lead equipment procurement and coordinates the parties in them. It does not procure, own, supply, store or take title to equipment, does not issue or confirm credit support instruments, and is not a bank, a broker-dealer or a direct lender; it does not hold client funds. Equipment specification, ratings and selection are covered by Pantheon, not here.