What an assessed price is, and what it is not

For minerals no exchange quotes, the number in a market report is a reporting agency's judgment anchored to a written specification — and the specification is doing most of the work.

Pending review

A filled woven bulk bag on a pallet with its top folded open, pale sandy material heaped just proud of the rim and dusted down one side. — Illustrative artwork: bulk mineral concentrate of the kind traded against published specifications. Not a facility, equipment or material connected to this project. · Illustration · Osmond Hub

The short version

No number appears on this page, and none will. What a reader of a market report for these minerals actually needs is not the figure but the thing behind it: a published specification defining the product being described, a set of data an agency was told about, and an explicit exercise of judgment over that data. Understanding those three is most of what makes a market report legible.

The specification comes first

An assessment begins with a definition of the product. Fastmarkets states its own objective as the prevailing level at which "a commodity of stated specification has, or could be expected to have transacted over a defined period of time"(opens in a new tab) — and the load-bearing words there are of stated specification.

What that means in practice is visible in the published specification fields. The zircon premium-grade assessment is defined on quality as premium grade, "66.5% ZrO2 min"(opens in a new tab); the standard-grade assessment as standard grade "min 65.5% ZrO2"(opens in a new tab). Both are further defined by quantity — a minimum of 1,000 tonnes — by location, cif China, by timing, spot, and by payment terms of within 30 days.

One percentage point of zirconia separates two assessments that are, commercially, describing different products. Any comparison of "the zircon price" between two documents is really a comparison of two specifications, and if those differ the numbers were never measuring the same thing.

Then the data, which arrives by being told

An exchange observes trades because trades happen on it. An agency has to go and ask. Fastmarkets describes reporters as aiming to collect data "from a broad sample of market participants specifically involved in the buying and selling" of the commodity, "with a good representation of both sides of the market, including producers and consumers, as well as traders and intermediaries"(opens in a new tab).

Three published rules govern what happens to that data, and each is worth knowing before reading any assessment.

Data can be adjusted. Where necessary in certain assessments, data falling within the stated specification ranges "may be normalized to determine the equivalent price for the respective base specification if one exists"(opens in a new tab) — so a reported transaction in a slightly different product can inform an assessment of the base product, after adjustment.

Data can be excluded. Reporters may "apply expert judgment to exclude data deemed unrepresentative, questionable or unreliable prior to consideration in the final assessment"(opens in a new tab).

And there is no floor on how much data there has to be. Since commodity markets differ in liquidity at different periods, the methodology "does not set any minimum number, or threshold, of transactions to be gathered on which to base the assessment"(opens in a new tab). Where there is little or no data of sufficient quality — the methodology's own qualifier — the published fallback is caution rather than movement: in such sessions "extra caution will be applied and reporters may exercise their judgment to keep a price assessment unchanged as a fallback"(opens in a new tab). No retrieved extract of the methodology carries any continuation of that sentence. A longer ending — "…until activity can be confirmed with greater certainty" — would turn an open-ended fallback into one with a stated release condition, and is not quoted here.

An unchanged assessment does not mean nothing moved. It can mean nobody could see whether anything moved.

Which is a different instrument from a settlement

IOSCO's work on price reporting agencies, written for oil rather than minerals, describes the category in terms that transfer: PRAs are publishers and information providers "who report prices transacted in physical and some derivatives markets"(opens in a new tab), and give "an informed assessment of price levels at distinct points in time"(opens in a new tab). On method it is equally direct: some methodologies "involve actual concluded transactions while others may take into account bids, offers and other relevant market information"(opens in a new tab).

IOSCO also recorded, at the time of that report, that "PRAs and their activities relating to price assessments in oil markets remain subject to no form of regulatory scrutiny or accountability"(opens in a new tab) beyond that expected of ordinary commercial enterprises — which is why the discipline the report proposes is transparency rather than supervision. Its first principle is that a PRA should "formalize, document, and make public any methodology that it uses for a price assessment"(opens in a new tab).

The comparison with an exchange settlement is ours, not IOSCO's: the report does not draw it. But the two instruments plainly answer different questions, and reading an assessment as though it were a cleared settlement imports a precision the published method never claimed.

Where this turns up in official documents

It is not only trade press that runs on assessments. The Mineral Commodity Summaries attach source lines to their price footnotes, and those lines name the agencies: "Source: Fastmarkets IM."(opens in a new tab) against one series and "Source: Argus Media group, Argus Non-Ferrous Markets, annual average."(opens in a new tab) against another, on the same page of the zirconium and hafnium chapter.

So when a government publication, a company presentation and a news article appear to agree on a market, they may be agreeing because they all took the same assessment from the same agency, built on the same specification. That is not a criticism of any of them. It is simply the structure of information in a market with no public settlement, and it is the reason this hub explains the mechanism and publishes no number of its own.

Related