Benchmark governance

Methodology change & consultation

What counts as a material change, the 14-day consultation window, and why changes are prospective.

Applies to: the benchmark class (WETGRI, WETFED, and any benchmark added later).

Standing disclosure — this process was SUSPENDED for the 2026-08-05 rebuild

The process below describes how a rule change reaches a published number in normal operation: a 14-day consultation, a backtest published with the proposal, prospective effect only. It was not followed for the change that produced the current series, and that is stated here rather than left for a reader to discover by checking.

An adversarial review found that several published rules were not implemented by the code — the orientation of de-escalation contracts, divisor continuity across departures, the weight cap's enforcement space, and the settlement-window claim. Correcting those changed every value. The administrator directed that correctness come before process, on the grounds that the series was three days old and had never been quoted, cited or licensed. The v1 series was therefore retired rather than restated — it is preserved verbatim at content/indices/benchmarks/retired-v1-2026-08-05/, including the copies as they were actually served in production — and the family re-inceptioned at 100 on 2026-08-05 under methodology v2.0.

What that costs, stated plainly. No consultation window was opened, no impact analysis was published in advance, and the change was not prospective. A reader who relied on this page during that window was owed a process they did not get. The version stamp does distinguish the two series — every v1 row stamps v1.0, every row from the re-inception stamps v2.0 — so no print is ambiguous about which rules produced it.

This suspension is not standing. It covered the rebuild and ends with it. Every subsequent change goes through the process below, including the two changes currently drafted as consultations in consultations/.

What counts as a material change

Anything that could alter a published value or its interpretation:

  • the severity rubric or any multiplier in it (lib/indices/severity.ts)
  • the classification rules (lib/indices/classify.ts)
  • eligibility criteria, the liquidity floor, or the resolution-source tiering
  • the weighting scheme, the weight cap, or the settlement window
  • the consolidation weights, guards, or outlier treatment (lib/indices/consolidated.ts)
  • adding or removing a venue from the consolidated set

Adding a constituent that satisfies the existing published rules is not a methodology change — that is reconstitution, and it is covered by the index's own methodology document.

The process

  1. Proposal. Written up with the rule as it stands, the rule as proposed, and the reason.
  2. Impact analysis. The change is backtested against the full ledger history and the effect on past values is published with the proposal. A change whose impact is not shown has not been proposed.
  3. Consultation — 14 days. Posted publicly before it can take effect. Feedback to the address in complaints.md; all responses are logged, including the ones rejected.
  4. Decision. Logged with reasons.
  5. Effect. Prospective, from the next settlement after the window closes.

Version discipline

Every benchmark carries a methodologyVersion stamped into each settlement row. A material change bumps it. Because the version is in the row, any historical print states the rules that produced it — you never have to trust that today's methodology page describes yesterday's number.

The one thing that is never allowed

A rule change may not be applied retroactively to restate published values. If a rule was wrong, the values it produced still stand as what was published under the rules of the day, and the correction is prospective. The only exception is a data error, which is a different document: data-errors.md.

The rest of the set
Data licence & rights
What the free tier grants, what attribution it requires, and what commercial use needs.
The loop architecture
The nine scheduled agents that cultivate indices, which one may publish, and the six mechanisms that assure accuracy.
The confirm gate
What human confirmation covers — merges only — and how to read the provisional and single-venue counts.
Restatement policy
The three narrow conditions under which a published value may be corrected — and the many under which it may not.
Cessation & fallback
What happens when a benchmark cannot be computed, and why the history stays published forever.
Conflicts & the conflict wall
The wall between index operations and the trading circuit, plus standing disclosures.
Data-error playbook
Detection, triage, quarantine — and the commitment to disclose errors nobody noticed.
Complaints & challenges
How to challenge a value, a rule, or a constituent. No account required.
Settlement-reference licensing
What licensing a benchmark to settle against requires — and why publication alone does not grant it. Reading a value is free; referencing it in something that settles is not the same act.
Determination, disruption & cessation
Written for contracts rather than charts: who determines a final value, the disruption fallback ladder, and why a refusal is not a cessation.
The independent committee seat
The seven-part independence test, published before a candidate exists; the power to block rather than comment; and what the administrator may do while the seat is vacant.
Incident process
Who is paged, what they do, and what gets published when the machinery fails — as distinct from a wrong value, which is the data-error playbook.
Finding — the WETX coverage breach, August 2026
Why WETX refused for five consecutive days, the change that caused it, and what is still not established. Published because an availability figure without the incidents behind it is a number nobody can check.