Reconstructing realised first-of-a-kind plant performance from public payment records: the NER300 output identity and the first two measured plants

Anyone insuring or lending against a first-of-a-kind (FOAK) industrial plant must price a question with no public statistical base behind it: how much of the promised output will the plant actually deliver? This note sets out a mechanism that makes part of that question answerable from public records, and publishes the first two figures it yields. Article 11(2) of Commission Decision 2010/670/EU pays NER300 renewables demonstration projects for energy produced, at a rate derived from awarded funding and committed output. This yields an identity, not a proxy: realised output equals disbursement times 0.75 times committed output divided by awarded funding. Because awarded funding and committed output are published for all 39 projects in Annex II of European Court of Auditors Special Report 24/2018, a private performance figure becomes readable from a public payment record. Two conditions bound the identity and are stated explicitly: Article 11(3) makes payment conditional on knowledge-sharing compliance, so zero disbursement does not imply zero output; Article 11(5) permits advances not tied to output. Applying this to output figures published by the Joint Research Centre (JRC119497) gives the first two measured plants: BEST (Crescentino, Italy) produced 44.3 GWh in 2017 against 283 GWh committed annually (15.7 per cent), and Verbiostraw (Germany) produced 59.0 GWh in 2018 against 100.4 GWh committed (58.8 per cent). An independent route via nameplate capacity corroborates the BEST figure. The note is explicit about what these two points cannot support: with n=2 this is not a base rate, and the programme-level contrast between bioenergy and wind measures financing failure rather than performance shortfall. The narrower performance contrast documented by the Court covers three plants. The hypothesis that technological novelty predicts output shortfall is raised, not demonstrated. A prediction is registered before the evidence arrives: applying the identity to the published output figures, disbursements should have been approximately EUR 1.17m for BEST in 2017 and EUR 3.45m for Verbiostraw in 2018. Eight access-to-documents requests are pending before the Commission and the EIB, the first falling due on 13 August 2026. Results will be published whether or not they confirm the hypothesis. Spanish and English versions are included in this record.