--- title: "Wales 20 mph Casualty Attribution Audit: A Constructive Non-Identification Result" date: 2026-08-11 version: "1.0.0" doi: 10.5281/zenodo.21894035 pdf: repository: https://github.com/ipitchford/wales-20mph-policy-analysis archive: https://zenodo.org/records/21894035 license: CC0-1.0 status: unrefereed (internally replayed; not peer reviewed, not independently reproduced, not formally verified) --- # Wales 20 mph Casualty Attribution Audit: A Constructive Non-Identification Result ## Plain-English summary Wales changed the default speed limit on restricted roads from 30 mph to 20 mph on 17 September 2023. A provisional Welsh Government release later reported that casualties on roads recorded as 20 or 30 mph were 28% lower in the first four complete quarters after implementation than in the preceding four-quarter window. That decline is reproducible as a description of one published data vintage. The provisional totals are 2,402 casualties before and 1,725 after, an exact change of about -28.18%. But the comparison does not identify the policy's causal effect on the roads whose speed limit actually changed. The reason is simple but consequential. The published 20/30 mph series is a changing mixture of affected roads, pre-existing 20 mph roads, retained 30 mph exceptions and later reversions. The frozen evidence does not supply a versioned historical map showing which road segments were treated, and it does not reveal what casualties would have occurred on those affected roads without the policy. This candidate makes the limit exact. It constructs three hypothetical worlds with the same observed aggregate totals but affected-road causal effects of -50%, 0% and +100%. None of those worlds is asserted to be true. Together they show that, within the registered aggregate evidence and admitted rival class, neither the point value nor even the sign of the effect is identified. The result therefore does not say that the policy worked, failed or had no effect. It says that the positive 28% attribution remains unsupported by this evidence and that a defensible causal estimate requires different measurement or independently justified restrictions. ## The central claim or finding The registered target is the percentage causal effect of the default-speed- limit change on the total number of police-recorded casualties on Welsh road segments whose restricted-road limit actually changed from 30 mph to 20 mph, for 2023 Q4 through 2024 Q3, relative to the same segments' no-policy casualty count over that period. The candidate finding is narrower than a policy-effect estimate: > For the frozen aggregate official corpus and declared rival set, the > affected-road causal estimand is not point-identified, its sign is not > identified, and no finite identified interval is claimed. This is a corpus-specific constructive non-identification result. It is not a claim that no future design could identify the effect. A historical treatment map, linked collision locations, a defensible counterfactual design or externally justified bounds could change the evidence state. The descriptive and causal quantities must remain separate: | Quantity | Population | Contrast | What it can support | |---|---|---|---| | Provisional 28% figure | All casualties on roads recorded as 20 or 30 mph | Four complete post-policy quarters versus the preceding four-quarter bin | A vintage-specific before/after description | | Registered estimand | Casualties on road segments whose limit actually changed | Policy outcome versus the same segments' no-policy outcome | A causal effect, if treatment membership and the counterfactual are identified | The baseline bin for the descriptive comparison is also not wholly untreated: 2023 Q3 contains the 14 calendar days from implementation on 17 September to quarter end. ## How the result works The audit first preserves the source-vintage ladder instead of selecting the number most favourable to one interpretation: | Frozen route | Baseline | Post | Approximate change | Evidential role | |---|---:|---:|---:|---| | Provisional Welsh Government graphic | 2,402 | 1,725 | -28.18% | Reproduces the reported first-year headline | | Current StatsWales extract | 2,402 | 1,775 | -26.10% | Shows revision across official vintages | | Current DfT collision-row route | 2,395 | 1,816 | -24.18% | Unreconciled diagnostic, not independent replication | Those figures describe aggregate changes. They do not reveal affected-road membership or the missing no-policy outcome. The constructive witness then fixes one hypothetical observed decomposition in all three worlds. The aggregate baseline is 2,000 casualties on affected roads plus 402 elsewhere, giving 2,402. The observed policy-period aggregate is 1,000 on affected roads plus 725 elsewhere, giving 1,725. Only the unobserved affected-road no-policy outcome changes: | Hypothetical world | Observed affected-road policy outcome | Unobserved no-policy outcome | Implied causal effect | |---|---:|---:|---:| | Negative-effect world | 1,000 | 2,000 | -50% | | Zero-effect world | 1,000 | 1,000 | 0% | | Positive-effect world | 1,000 | 500 | +100% | Every world is compatible with the same hypothetical observed aggregate decomposition. The observation map is therefore non-injective: different causal signs map to identical observables. The three values are counterexamples, not estimates, confidence limits or endpoints of an identified interval. An England comparison is retained only as adversarial pressure. In the current DfT route, the Wales-versus-England descriptive differential changes sign between the immediately preceding diagnostic window and the first-year window. That does not rule out every conditional donor design, but it gives no basis for assuming national parallel trends without additional work. ## What is classical, and what is offered here The general principles are established: a before/after comparison is not a causal effect; counterfactual outcomes are not observed; treatment misclassification can break an observation map; and observationally equivalent worlds can establish non-identification. This candidate does not present those ideas as a new theorem. Earlier working papers already offered Wales/England causal evaluations of the default 20 mph policy, and both the posted-speed reclassification problem and the value of combining 20 and 30 mph categories had already been discussed. Their existence establishes prior art, not the correctness of their estimates or assumptions. Within the bounded search completed on 11 August 2026, the residual contribution is a case-specific bridge: 1. exact reconstruction of the provisional first-year graphic and reconciliation with later frozen source vintages; 2. separation of the combined posted-speed proxy from the affected-road causal estimand; and 3. an explicit point-and-sign non-identification witness tied to the registered Wales claim, corpus and rival set. The novelty assessment is `PARTLY_NOVEL`, not an exhaustive priority claim. No exact Wales application of the registered witness was found within the recorded searches, but subscription indexes, Welsh-language-only material, some grey literature and a complete citation network were outside the boundary. ## Evidence and audit trail The research repository registers 25 sources and three frozen search records. It binds all 25 sources through 125 source references, checks 29 narrow frozen- text anchors and replays 17 exact calculations. The non-circular scientific review target contains 45 entries and has SHA-256 `8f4434763bd9c2970b140ac222dfec388a3f29b93a3f6eb5894b4130023cf135`. The source set includes Welsh Government statistical releases, StatsWales extracts, Department for Transport collision data, legislation and Office for Statistics Regulation guidance. Source records retain version, retrieval, coverage, transformation, licence, redistribution and limitation fields. The case also preserves official-data measurement cautions, including unreported incidents, speed-limit coding corrections, historical inconsistencies, reporting-system changes and the absence of traffic exposure by speed-limit class. Successive read-only external-model adversarial reviews and their exact transport receipts are retained rather than collapsed into one badge. They are model-assisted continuity review inside a producer-coordinated workflow. They are not unaffiliated empirical reproduction, reviewer-panel corroboration, external human specialist review or editorial peer review. The release gate additionally requires exact manifest coverage, ordinary and optimized execution, hostile controls, clean-checkout replay, a retained attestation and detached package verification. The authoritative public receipt must state which of those checks passed for the released bytes. Hashes establish byte identity and producer replay establishes declared transformations; neither establishes source validity, causal truth, policy merit or real-world impact. ## What the result does not establish - It does not establish that the policy caused the reported casualty decline. - It does not establish that the effect was negative, zero, positive, 28% or any other magnitude. - It does not turn the illustrative -50%, 0% and +100% worlds into estimates or bounds. - It does not show that every possible future design is non-identified. - It does not validate or refute earlier causal evaluations merely by citing them as prior art. - It does not identify risk per journey, displacement or realised-speed effects; affected-road traffic exposure is absent. - It does not establish welfare effects, distributional consequences, implementation quality or whether the policy should be retained, changed or reversed. - It does not supply independent rerun, independent reimplementation, formal verification, human specialist review or editorial peer review. - It does not show that Evidence Press makes policy assessment faster or better. That remains a separate comparative hypothesis requiring prospective measurement. ## Relationship to earlier work This release reuses the identification-before-estimation discipline illustrated by [*Stocks Are Not Flows*](https://evidencepress.org/releases/stocks-are-not-flows/): write the estimand, specify the observation map and admitted rivals, and stop a point attribution when the evidence cannot distinguish it. Reusing that method does not make this case independent confirmation of the earlier release. Three APEP working papers located in the bounded novelty search had already presented Wales/England difference-in-differences evaluations: *Slower Streets, Safer Streets?*, *Does the Default Kill?*, and *The Speed Penalty*. This audit does not adopt their estimates as true. It asks the separate question whether the registered affected-road causal estimand is recovered from the frozen aggregate evidence and unrestricted counterfactual class used here. The [Office for Statistics Regulation's 20 mph guidance](https://osr.statisticsauthority.gov.uk/guidance/2026-what-to-look-out-for-20mph/) is a close conceptual antecedent: it highlights reclassification, exposure, reporting and causal-language cautions. The present candidate turns those measurement concerns into a source-bound estimand distinction and constructive non-identification witness; it does not convert regulatory guidance into an empirical effect estimate. ## Who should care, and why | Audience | Potential use | Required caution | |---|---|---| | Welsh and local-government policymakers | Distinguish a reported trend from the evidence needed to attribute it, and identify the missing measurements that could change the decision state. | Non-identification is not evidence of zero effect and does not settle the policy's merits. | | Road-safety statisticians and data custodians | Prioritise a versioned historical treatment, exception and reversion map linked to collision locations and exposure. | More aggregate quarters alone may leave the same observation-map failure intact. | | Causal-inference and policy-evaluation researchers | Stress-test donor designs, treatment coding, interference and counterfactual restrictions before estimating a point effect. | The witness is conditional on the registered corpus and rival class, not a universal impossibility theorem. | | Journalists and public commentators | Report the provisional 28% as a descriptive, vintage-specific comparison rather than a measured causal contribution. | Later source vintages differ, and none of the three aggregate routes identifies the affected-road effect. | | Research agents and assurance engineers | Reuse the machine-readable estimand, rival, source, citation, calculation and review objects or challenge them with a new counterexample. | Structural conformance and model review do not substitute for human semantic judgment or unaffiliated reproduction. | ## How to reproduce the recorded checks From the exact public research release, verify the published checksum ledger, extract into a fresh directory and use the pinned offline environment. The recorded lifecycle entry point is: ```sh uv run --no-editable --offline --frozen --no-sync pio lifecycle \ cases/wales-20mph-casualty-attribution \ --schema-root schemas \ --output OUTPUT_DIRECTORY ``` The expected lifecycle outputs include `MANIFEST.sha256`, `lifecycle-summary.json` and `validation-receipt.json`. The package also exposes the focused analysis check: ```sh uv run --no-editable --offline --frozen --no-sync \ python tools/build_wales_20mph_tables.py check ``` Use the release receipt, not this prose, for the exact commit, dependency lock, command vector, output hashes and clean-checkout result. A successful rerun is producer-workflow evidence unless performed and reported by an unaffiliated party. ## The most valuable next projects 1. Freeze a versioned historical segment-level map of the roads that changed, remained exceptions or later reverted. 2. Link collision and casualty locations to that map under explicit coding and linkage-error rules. 3. Specify a defensible comparison design with multiple pre-period diagnostics, placebo dates and explicit interference and anticipation assumptions. 4. Measure affected-segment traffic, realised speeds, enforcement, weather, reporting changes and concurrent interventions where they bear on the target. 5. Register source-backed restrictions prospectively and compute an empirical identified set only if those restrictions support one; retain sign ambiguity or `NOT_BOUNDED` when they do not. 6. Independently reconstruct the source-vintage ladder and reimplement the observation-map witness without using the producer code or intermediates. 7. Compare this workflow with a matched conventional expert assessment at the same assurance endpoint, recording elapsed time, human effort, compute, rework and missed valid effects rather than inferring acceleration from publication speed. ## What is in the public package The complete public package should contain the case records, frozen official sources authorised for redistribution, source and citation manifests, exact estimand, observations, assumptions, rivals, missingness, calculation and identification records, derived tables, adversarial-review history, ordinary and optimized test evidence, hostile controls, clean-checkout attestation, final receipt, provenance, licences, environment lock, replay instructions and detached verifier. The GitHub prerelease and Zenodo record must distribute the same declared release assets, with public readback hashes matching the local release ledger. The Evidence Press page, art and any audio or video are communication surfaces, not additional policy evidence. ## Open directions for follow-up research - Freeze a versioned historical segment-level map of the roads that changed, remained 30 mph exceptions or later reverted, then link collision locations under explicit coding and linkage-error rules. - Develop and independently review a comparison design with multiple pre-period diagnostics, placebo dates and explicit interference, anticipation and concurrent-intervention assumptions. - Measure affected-segment traffic exposure, realised speeds, enforcement, weather and reporting changes where they bear on a risk or displacement estimand. - Register source-backed restrictions prospectively and compute an empirical identified set only if the restrictions support one; preserve sign ambiguity or NOT_BOUNDED otherwise. - Independently reconstruct the official source-vintage ladder and reimplement the observation-map witness without producer code or intermediate files. - Compare this workflow with a matched conventional expert assessment at the same assurance endpoint, measuring elapsed time, human effort, compute, rework and missed valid effects. ## Research process and reusable methods This is prospective process metadata. It records the intended handoff and claim boundary; it is not evidence that the method accelerated this work. - Work ID: ep-work:wales-20mph-casualty-attribution - Attempt receipts: ep-attempt:wales-20mph-casualty-attribution: published / negative; measured-partial; active human minutes missing; compute minutes missing; rework minutes missing; assurance endpoint measured-partial. Work ledger: https://evidencepress.org/api/work-ledger.json - Intended aims: policy - Artifact roles: research-output, evidence-assessment, method-demonstration - Decision object: counterexample — Three hypothetical affected-road worlds preserve the same aggregate observed baseline and policy-period totals while implying negative, zero and positive causal effects. Scope: The first-four-complete-quarters affected-road casualty estimand, frozen official aggregate corpus and admitted rival class registered in the Wales 20 mph case. The worlds are counterexamples, not estimates or bounds. - Reusable methods: Adversarial scientific controls (adversarial-controls); Productive failure and stop receipts (productive-failure); Identification before estimation (identification-gate); Counterexample- and proxy-first analysis (counterexample-proxy-first); Assurance as a vector (assurance-vector); Agent-readable research objects (agent-readable-research-object). Registry: https://evidencepress.org/api/method-registry.json - Targeted clocks: discovery, assurance - Semantic bridge: partial — The release explicitly separates the combined 20/30 mph descriptive proxy from the affected-segment causal estimand and machine-checks the constructive observation-map witness. The empirical bridge cannot be completed because the frozen corpus lacks a versioned historical treatment map and affected-road no-policy outcome. Remaining risks: The non-identification result is conditional on the registered aggregate corpus, observation map and admitted rival class.; Speed-limit coding corrections, source-vintage changes, under-reporting and reporting-system changes limit empirical interpretation.; No human road-safety, official-statistics or causal-inference specialist has independently assessed the semantic bridge.; Future linked segment data or independently justified restrictions could change the identification state.. - Human judgement gates: Assess whether the affected-road estimand and admitted rivals adequately represent the substantive policy question.; Judge the credibility, legality and ethical acceptability of any future linked treatment, collision and exposure data.; Evaluate policy merits, welfare, distributional effects, rights and implementation choices; the identification audit does not decide them.; Authorize authorship, archival publication and any public characterization of external-model review. - Next assurance action: An unaffiliated multidisciplinary team should reconstruct the official source-vintage ladder, independently assess the affected-road estimand and rival class, and reimplement the observation-map witness without producer code or intermediates; a later design should add the missing historical treatment map before estimating an effect. - Claim ceiling: The released aggregate corpus and declared rival class recover neither the point value nor the sign of the affected-road causal effect. This does not establish the policy's true effect, a universal impossibility result, exhaustive novelty, independent reproduction, peer review, policy merit or an acceleration impact. - Aim-scoped impact evidence: - policy: NO_IMPACT_EVIDENCE — Faster and more decision-adequate assessment of causal policy claims with fewer unsupported point attributions in Evidence assessment of the Wales default 20 mph first-year casualty attribution claim; design none; comparator No matched conventional expert assessment was conducted.; estimand No comparative assessment-time, decision-quality or unsupported-attribution effect was estimated.; no real-world effect evidence asserted - Parent handoffs: reuses-method stocks-are-not-flows; inherited claim: The project reuses the identification-before-estimation workflow: distinguish the estimand from available proxies, register admitted rivals, test the observation map and publish a justified stopping boundary when point attribution fails.; inherited ceiling: Method reuse is not independent confirmation of the housing release or of this Wales result. Both remain conditional on their declared observation maps and neither establishes a real-world policy or acceleration effect. - Defeasible explanations: Identification gates, rival-mechanism audits and bounded decision maps produce decision-adequate policy assessments faster and with fewer unsupported causal claims. (policy-assessment-improvement); Published blockers, counterexamples and stop receipts prevent duplicated dead ends and redirect effort toward discriminating data or tractable subproblems. (negative-result-option-value). IBE ledger: https://evidencepress.org/api/ibe-ledger.json ## Verification status Unrefereed candidate. The claim is limited to corpus-specific non-identification of the affected-road causal point and sign. Producer-side source, citation and exact-calculation checks exist, together with model-assisted adversarial review inside the coordinated workflow. No unaffiliated rerun, independent reimplementation, formal verification, external human specialist review, editorial peer review, policy-effect estimate or real-world impact evaluation has been established. The positive 28% attribution remains unsupported. ## Sources and related work - Welsh Government (2025). Police recorded road collisions: July to September 2024 (provisional). - Office for Statistics Regulation (2026). 20mph speed limit. - Slower Streets, Safer Streets? The Causal Effect of Wales's 20 mph Default Speed Limit on Road Casualties and Property Values. APEP 0513 v1 (2026). - Does the Default Kill? Speed Limit Reclassification and Road Casualties in Wales. APEP 0734 v1 (2026). - The Speed Penalty: Causal Evidence from Wales's Default 20mph Limit. APEP 0744 v1 (2026). - Edinburgh 20 mph evaluation inspected as a changed-road/stayed-road analogue in the bounded novelty audit. - Anonymous (2026). Stocks Are Not Flows: An Identification Audit of Investor Pressure and Tenure Change in England. Evidence Press.