Revision to GAFTA sampling rules
Since I started in this business in the 1990s, much of my work has been with grains and animal feedstuffs. Huge quantities of grain and feeds are traded by sea, and it is no exaggeration to say that this trade is necessary to feed the world. It is a widely quoted statistic that 80% of the world trade in grain and animal feeds is by means of GAFTA contracts.
GAFTA is an acronym for the Grain and Feed Trade Association. I’m pleased to be able to say that Sheard Scientific is a member of GAFTA.
Trade based on GAFTA contracts usually adopts the concept that the definitive and “final” check on the quality of the goods being traded occurs at the time of loading. This enables a degree of certainty to those participating in the trade, but the concept does require reliable sampling and analysis. That sampling is required to be carried out according to GAFTA 124 sampling rules.
I may write in more detail about sampling as a general topic in a later post (time allowing!), but suffice it to say here that representative samples by definition are to represent the average properties of the parcel of cargo being sampled. Since those samples have to be much smaller in size than the cargo they are drawn from, they have to be drawn in a manner which gives equal statistical chance for any given particle in the cargo being sampled to be in the working/drawn sample.
The purpose of GAFTA 124 is to set a standard for how that is to be carried out, how the integrity of the samples should be preserved, and what should be done with them. Whilst the official scope of GAFTA 124 is in reference to the GAFTA sales and purchase contracts, to ensure quality parameters are properly and reliably assessed when a cargo is traded, it is in my experience so commonplace as to be universal that they are used as a benchmark “gold standard” for sampling of cargoes where there is a dispute – in circumstances where there has been transit damage, for instance. The GAFTA sampling rules don’t necessarily suit all such circumstances, but they are always where surveyors and experts start when considering how to sample average parameters of bulk grain or feeds.
Bearing that in mind, a change to the GAFTA 124 rules is potentially big news, and so I read with interest when I received notification of the changes. There have been substantial changes to the GAFTA 124 rules over the years I have been working in this industry – when I first started, in all cases sublots were 500MT in size – meaning dozens of sublots for a moderate sized vessel. That has long-since been changed so that the selection of sublot sizing scales with the overall consignment quantity.
One change being introduced in the new version of the rules is that there is greater clarity regarding the numbers of individual bags from which samples must be taken when bagged goods are being sampled.
Paragraph 9.1.8 now contains the following text, seemingly to ensure that all are made aware of the status of “certificate final” provisions and that sets of samples reserved for arbitration are not to be used to overturn compliant load port certification.
Arbitration samples drawn for goods traded on contracts stating ‘Quality (Certificate) Final’ at loading or at discharge cannot be used to challenge the quality specifications which are included in the ‘Quality (Certificate) Final’ certificate, unless otherwise determined by arbitrators or Board of Appeal.
But to my eyes, the biggest change is in the selection of laboratories. Again, back when I started this work, in the case of any disputes or disagreements, specific laboratories were to be used as “referee” laboratories. In recent years there have been four designated as referee laboratories and their identities were written into GAFTA 124. The four reference laboratories were undeniably “household names” to all in the grain trade, and will no doubt continue to be used as reliable grain testing facilities. However, it is now the case that any GAFTA-approved laboratory can serve this purpose as all approved labs are considered equivalent.
