Prices and unused volume in Britain's demand flexibility auctions
Citable version that has not completed AIRR screening. Not peer reviewed. Screening record
We reconstruct 620 live half-hour windows of Britain's Demand Flexibility Service in July–September 2026. Among 518 demand-reduction windows, 206 across 40 events had unfilled published requirements and both accepted and rejected positive-volume bids. Every such window contained a whole rejected bid that fit the residual national and zonal volume limits. Yet in every window the highest accepted price was strictly below the lowest rejected price, even before filtering for volume. The smallest price gap was GBP 0.20/MWh. Thus prices alone can reproduce the observed binary decisions within each mixed window, without identifying the operator's actual threshold or the value of additional flexibility. A separate settlement analysis preserves missing and negative values.
Preview PDF on this page v1
Uses your browser's PDF viewer. Open the PDF directly if a preview is unavailable.
Cite this paper
Lluis Eriksson (2026). Prices and unused volume in Britain's demand flexibility auctions. AIRR.SCIENCE, ARR-2026-30XJ0Y39DG886RFP, v1. https://airr.science/papers/ARR-2026-30XJ0Y39DG886RFP/versions/v1/ Research preprint; ARR-2026-30XJ0Y39DG886RFP v1. Not peer reviewed by AIRR. Working paper; not admitted to the AIRR accepted collection.
Use this version link to cite the manuscript you read. A Working paper label means the version has not completed AIRR screening. AIRR screening is not journal peer review. Import BibTeX, RIS or CSL JSON into your reference manager to apply a citation style.
Version history
The paper identifier remains stable. Each version has its own immutable release, timestamp and version identifier.
- v1 · source snapshot available · viewing
Verification record
- Frontier-model screening
- Not assessed
- Source integrity
- Pass
- Bibliographic integrity
- Partial
- Reproducibility
- Partial
- Lean 4
- Not applicable
Recorded under ARR-VERIFY-1.0. AIRR verification and screening are not peer review.
AI assistance statement
OpenAI Codex participated in analysis, writing and the first review of the exact PDF. Its shared-context assessment is 3.65/10. Report SHA-256 54364d95638234753e2038420151be0a11ef373a13e53d8e7e42e5c4721ab965. The author administers AIRR. Independent peer review has not been documented.
Frontier-model screening
Status: not_assessed. Any listed reports correspond to this exact version under ARR-SCREEN-1.0; no absent assessment is represented as a pass.
Model assessments and review context
- Distinct models
- 1
- Range
- 3.65–3.65
- Method
- bounded capability-weighted mean · exact version only
Limited evidence · 0/1 benchmark matches. This is an evidence description, not a probability of correctness.
Only one distinct model; One provider; shared blind spots possible; Fresh blinded review context not documented for every model; Mathematical benchmark not applied to this subject; Adverse reports or unresolved objections require editorial disposition
OpenAI · OpenAI Codex, GPT-6 family (exact serving revision unavailable) · involved in manuscript · review context: shared3.65
Recommendation: Minor Revision · Material objections: 0
The manuscript examines a census of 206 mixed underfilled Downwards windows in 40 events. Volume caps alone leave a whole rejected offer feasible in each window; across all zones and before any cap filtering, the accepted and rejected prices are strictly separated in each window. This descriptive conclusion survives the separate exact-decimal reproduction. The introductory event and the smallest observed price gap have traceable identifiers. Two-window transfer, welfare and operator intent remain outside the result.
- Correctness confidence
- All 620 window joins and accepted MW reconciled, with matching results from the offline reconstruction. The half-open price interval and signed settlement identity follow directly from the definitions.
- Rigor
- The sequence of cohort selection, the whole-bid volume restriction, and the separate all-bid threshold are explicit. Some diagnostics were designed after the basic cap result was observed.
- Novelty
- Ranking valid bids by price is a published procurement rule. The exact joint enumeration and narrowest observed separation add evidence for one quarter. Broader priority has not been established.
- Significance
- Decision-relevant for interpreting underfilled requests and rejected offers in this service. It gives neither the marginal value of capacity nor a validated forecasting rule.
- Reproducibility
- Three source digests, frozen operator CSVs and a runnable offline script reproduce the figures. Two clean renders yielded the same PDF, and all three pages were inspected.
Strengths
- The all-zone test is stronger than the pairwise comparison and correctly includes the one window with no same-zone comparator.
- The construction that attains the divisible volume bound is valid for the stated relaxation. Single-bid totals are kept distinct from feasible multi-period dispatch.
- Missing and negative settlements are exposed rather than imputed. The sensitivity threshold retains the half-hour conversion.
Weaknesses
- A reader could use a flow table for all excluded window types and the distribution of price gaps by event rather than only extreme values.
- The paper does not test changes in the price interval across time or control for repeated bids within events. It limits its claims to its observed cohort.
- The reference set covers operational sources. It is insufficient for a general priority claim in empirical auction economics.
Potential errors
- The word unused in the title could suggest available electricity. Only the published requirement and two nominal volume caps are tested.
- Closely spaced prices in the narrowest window raise sensitivity to rounding of the posted prices. The result concerns the public-price precision.
- The physical meaning of negative settled volumes is not established by their signed appearance in the ledger. Their role here is confined to the accounting identity.
Strong novelty candidates
- Window-level census jointly demonstrating residual national and zonal capacity and complete all-zone strict price separation in this post-April-2026 cohort.
Unresolved material objections
None reported.
arr:assessment:f5156fd2-6396-402e-843f-6dffa785453d · prompt ARR-ASSESS-1.0 · response SHA-256 3dede6b1c61c316fc5f8ba2473f2e411342751fc9fc83449d4840cdc25964728 · canonical PDF SHA-256 98d711c32e8f7a65dda4411a99ac8a314ff9403f5995c882ee25fea6d62f32c5
A model assessment is not peer review or a correctness certificate. AIRR preserves disagreement, exact-version provenance and later reassessments.
Editorial disclosure
Working-paper deposit under exact-hash permission. Collection screening and independent peer review remain outstanding.