Transacted PPA prices routinely sit away from published indices. Shape, curtailment risk allocation and offtaker credit move the headline number more than technology does — the index is a useful direction and a poor input.
What is a PPA priced on?
Nominally on energy delivered, and actually on a bundle: the shape of that delivery against demand, who carries curtailment risk, the tenor, and the credit of the party promising to pay.
Two agreements at the same headline price per megawatt hour can therefore carry materially different economics, and the headline is the part that gets published — the same limitation that sits inside a published cost curve.
Tenor deserves separating out because it does two things at once. A longer agreement lowers the risk premium the financing carries, which supports a lower headline price, and it lengthens the period over which any mispricing compounds. A short high-priced agreement and a long low-priced one can produce the same project return and very different exposure to what happens afterward.
Why do published indices lag?
Because they aggregate reported deals with a delay and normalize away the terms that differ. By the time a transaction appears in an index, the conditions that priced it have usually moved.
That is a design constraint rather than a flaw. An index has to be comparable across deals, and comparability requires discarding exactly the terms that made each deal what it was.
The lag also means an index is least reliable exactly when it is most consulted. Prices are checked hardest when the market has moved, which is when the reported set is furthest behind, so the moment of maximum interest is the moment of maximum error. Anyone using an index in a fast-moving market should treat it as a description of the recent past.
Which terms move the headline price?
Shape first — whether generation matches the offtaker's demand profile or has to be firmed. Then curtailment: an agreement where the generator carries curtailment risk prices differently from one where the offtaker does.
Offtaker credit is the third and it is frequently underweighted. A corporate offtaker without an investment-grade rating changes the financing cost, which flows back into the price the generator can accept.
The three terms also interact rather than adding up. A generator carrying curtailment risk will accept a lower price only if the offtaker's credit supports cheaper financing, and an offtaker with weak credit will find that shape concessions cost more than they otherwise would. Pricing each term separately and summing the effects misses that, and the sum is usually optimistic.
Who can tell you what actually cleared?
People who negotiated on either side: developers, offtaker energy managers, and the advisers who papered the deal. On an expert call they will rarely quote a number and will readily describe where a deal sat relative to the market and why.
That relative positioning is usually enough. A model needs to know whether an assumption is at, above or below market and what drove the difference.
Advisers are frequently the most efficient single source here, for a specific reason: they have seen both sides of several deals in the same period, which neither a developer nor an offtaker has. They will not describe any single transaction, and they can characterize where the market has been clearing and what has been moving, which is what the model needs.
What does this change in underwriting?
The contracted revenue assumption and, more importantly, the risk allocation behind it. A model that takes an index price and assumes standard terms is pricing a contract nobody signed, which is the sort of thing commercial diligence exists to catch.
It also changes the sensitivity analysis. If curtailment sits with the generator, the downside case has to flex volume as well as price, and many models flex only the latter — often before the connection date is even settled.