Five questions to ask of any risk number

A risk figure is a summary of many decisions: which positions were included, which prices were used, which method turned them into a number. When the figure travels — into a report, a meeting, a message — those decisions usually stay behind. Five questions bring them back.

1. Where did it come from?

Name the sources: the position records, the market data, the reference data. If a figure combines several sources, each one should be identifiable. A number that cannot be traced is an opinion with decimal places.

2. How fresh is it?

Every input has an age. Prices from this morning and holdings from last night can sit in the same calculation, and the result is only as current as its oldest input. Ask for timestamps, and look for inputs marked as stale or missing rather than silently filled in.

3. How was it calculated?

The method should be describable in a sentence a non-specialist can follow, with the technical detail one step away. If the method changed recently, that change belongs next to the number, not in a release note.

4. What does it assume?

Scenarios assume shocks; models assume relationships; groupings assume rules. Assumptions are not weaknesses — hidden assumptions are. A useful presentation puts the assumption beside the result, not in an appendix.

5. Who has reviewed it?

A figure that has been checked is different from one that has merely been produced. Knowing who reviewed it, when and with what comment tells the reader how much weight it can bear.

None of these questions needs new mathematics. They need the context to travel with the number. That is a design problem — and a solvable one.