Hedge Agreement

A hedge is not an offtake — nobody buys anything

Under a tolling agreement or an offtake agreement, someone buys what the battery does — capacity, energy, ancillary services — and pays the project for it. Under a hedge agreement, nobody buys anything. The counterparty is a bank or a commodity trading desk; the project keeps dispatching into the market and earning whatever its revenue stacking produces; and the two parties simply exchange cash at settlement.

If the reference price or the project's merchant revenue lands below the agreed level, the counterparty pays the difference. In a swap or a collar, payments can run the other way too. The instrument is insurance-shaped rather than sale-shaped — which means the project keeps operational control, and also keeps every risk the hedge does not name.

Basis risk: paid at the hub, earning at the node

Most hedges settle against a liquid reference — a trading hub or a published index — because that is what the counterparty can price and lay off. The plant earns at its own node, behind whatever congestion the local network produces. The gap between the two is basis risk, and it stays with the project: the hedge can settle exactly as written while the node underperforms the hub.

This is a US framing — nodal pricing in ERCOT, CAISO, and the other ISOs — and it is one of the first things a lender's market consultant probes: how far, and how often, has the project's node diverged from the reference the hedge settles on. Revenue puts that settle on the project's own merchant performance sidestep hub basis, which is part of why lenders like them.

The tenor gap — and the naked tail

Hedges commonly run shorter than the debt. A counterparty will price storage revenue risk for a limited window starting at the commercial operation date; the financing agreement behind the project usually amortizes over a longer horizon. That mismatch is the standing caution with hedged merchant cases: a debt sizing propped up on expiring hedges leaves the tail years naked, and the tail years are precisely when the cells have degraded, augmentation spending arrives, and the original price forecast is at its least defensible.

Lenders respond by sizing debt primarily against the hedged window, sculpting amortization so most of it repays inside the hedge term, or requiring a re-hedging covenant. What no credit committee will do is treat the hedge as if it covered years it does not.

What certainty costs — and the shapes lenders accept

The counterparty prices the risk it absorbs and adds margin, so a hedge converts upside into certainty — it never creates value. Expected revenue goes down; bankability goes up. That trade is the entire point, and it is worth being honest about: if merchant prices come in strong, the project will have paid away real money for protection it never used.

The honest answer to "should we hedge?" is therefore a financing question, not a market-view question — you hedge because the debt you want requires it. Some structures also demand collateral posting, which turns a paper obligation into a live liquidity call at exactly the moments prices move.

For storage specifically, the shapes lenders have accepted are the revenue put — the project pays a premium and the counterparty tops merchant revenue up to a floor — and the related revenue floor and collar structures. These are US structures, developed largely around fully merchant ERCOT batteries, where a project with no offtaker needed something a credit committee could underwrite. Classic fixed-for-floating swaps of the kind wind and solar projects use map poorly onto storage, because a battery's revenue depends on dispatch decisions rather than a weather-driven production profile.