Cessation & fallback
What happens when a benchmark cannot be computed, and why the history stays published forever.
A benchmark can stop being computable — a venue delists a whole category, a resolution source disappears, liquidity falls below the floor and stays there. Users need to know in advance what happens, because the alternative is discovering it from a gap in a chart.
Insufficient data (temporary)
If a benchmark cannot meet its published minimum constituent count on a given day:
- the settlement row is written with
status: insufficient_dataand no value; - yesterday's value is not carried forward. Filling a hole to keep a chart continuous is a fabricated print, and it is the single most tempting corrosive act available to an index;
- the pages show the gap and say why.
Fallback
Where a benchmark has a defined fallback in its methodology — for example WETFED reverting to the nearest available meeting cycle when the constant-maturity bracket is incomplete — the fallback is published in that methodology document and its use is flagged on the settlement row. An unflagged fallback is indistinguishable from an ordinary print, which defeats the purpose.
Cessation
If a benchmark cannot be computed under its methodology for 20 consecutive settlement days, or if its underlying market structure has changed such that the methodology no longer measures what it claims:
- A cessation notice is published, with the reason and the last valid settlement.
- A 30-day notice period runs before the benchmark stops publishing, so anyone referencing it can transition.
- The full historical ledger remains published permanently, with its checksum chain intact. A ceased benchmark's history is still a record; deleting it would break every citation ever made and destroy the audit trail the record exists to be.
- Where a successor exists, the notice names it and states explicitly that the series are not continuous across the change.