Gold

As outlined in the Introduction, the differences between Metals Focus and LBMA data largely reflect the relatively small amount of ASM treated by GDLs, the fact that Metals Focus’ coverage extends to the market as a whole and not simply GDL refiners, and the exclusion of Russian activity from the 2022 COO survey. Even allowing for the above points, it is also revealing to see the extent to which the growth rates can also vary between the two datasets, some of which are highlighted in the two main tables below. In the text which follows, we also highlight some of the main differences at the country level, for both mine supply and recycling.

Mine Production

Looking first at mine supply, the COO data for Burkina Faso captures the above point about mine supply. That said, the 45t difference exceeds Metals Focus’ estimate of the country’s ASM production, which suggests that some LSM is refined at non-GDLs. Furthermore, the 25% drop shown in the COO data versus Metals Focus’ 7% fall, at a time of apparent little change in ASM, suggest that GDLs have lost market share of Burkina Faso’s LSM gold production.

The same theme also broadly applies to Côte d’Ivoire, with respective year-on-year changes of 0% and 15% for LBMA and Metals Focus in 2022. The latter mostly reflected the continued ramp-up at Perseus Mining / Cote d’Ivoire government’s Yaouré mine.

In contrast, the reported COO Ghanaian volume of mine supply for 2022, that was treated by GDLs, was up 16% year-on-year, compared with a more modest 10% gain estimated by Metals Focus. This reflected firmer LSM, with little change reported for ASM. The former was due to the ramp-up of AngloGold Ashanti’s Obuasi mine, improved throughput, a mill recovery and higher grades at Newmont’s Ahafo operation and the start- up of the Bibiani mine, which is owned by Asante Gold and the Ghanaian government, Interestingly, the 14 operating LSMs in Ghana, monitored by Metals Focus (in their Gold and Silver Doré Flows Service), produced a combined total of 93t in 2022, which comes very close to the 95t of LSM reported in the COO survey. This may suggest that GDL’s gained market share in 2022 in refining Ghanaian material, at the expense of non-GDLs.

Turning to Mali and Tanzania, with similar growth rates for their mine output estimated by both organisations in 2022, the differences reported of 42t and 17t respectively owed much to Metals Focus’ ASM estimates. With regards to Mali, Metals Focus estimated 58t of LSM across eight mines, all of which therefore appear to be treated by GDLs (given that the COO shows no ASM from Mali being treated by GDLs). The same conclusion also applies to Tanzania, with the combined output of the country’s five LSM operations totalling 34t, which also matches that reported in the COO survey for 2022.

In contrast, as the table shows the data reported for Indonesia varies considerably between the two organisations for 2022; the COO shows a 35% year-on-year decline to 36t, against a 20% rise to 130t estimated by Metals Focus. In terms of the absolute totals, the differences reflect two areas, ASM and base-metal concentrates. Metals Focus’ estimate for the former is around 30t, none of which is refined by a GDL (according to the COO survey). The same is also true for the gold-containing copper concentrate, much of which is also treated by a non-GDL refiner.

Source: Metals Focus, LBMA;
*LSM plus ASM

Turning to gold recycling, the extra granularity in the COO data sheds some additional light on the GDL activities in this area and, to some extent, helps with a comparison with the Metals Focus data on scrap supply. Stripping out “legacy stocks” and “industrial” recycling from the COO survey offers a useful starting point as, respectively, the melt of 400oz bars and manufacturing process (or production) scrap are both excluded from Metals Focus’ recycling series. However, as the recycling table shows, there are still some big differences, which reflect two issues. First, the “unprocessed recyclable” category will contain a mixture of old jewellery scrap, investment bars, post-2011 400oz bars (in other words, excluding legacy stocks) and some production scrap. Metals Focus includes the first of these in its recycling series. Investment bars can also be included, but only when unsold bars are melted, while those sold back by investors are defined as disinvestment. As touched on above, Metals Focus also excludes the melt of 400oz bars and production scrap.

If we look at the COO data of gold that was of Swiss origin, for example, this shows that 463t of “unprocessed” metal was refined in 2022. Given the strength of retail investment at that time, any bars that were melted were therefore more likely to be 400oz GDL bars. The other significant portion of this would have been imported scrap jewellery, given the price strength at the time. As such, the amount of old scrap of Swiss origin that would have been treated would have been trivial, as indicated by Metals Focus’ estimate of this flow at being just over 2t per annum. The treatment of 400oz GDL bars also explains the 271t of “unprocessed” and “melted” gold of UK origin that was processed in 2022, which compared with Metals Focus’ estimate of 14t of gold recycling for that year.

Continuing this theme, the Chinese COO total of 563t (for processed, 449t, and melted gold, 115t) far exceeds Metals Focus’ recycling figure of 154t. This largely reflects the melt of imported 49s kilobars, but will also include melt of products sold back by consumers and the treatment of production scrap from the supply chain. The Turkish figure of 133t also comfortably outstrips the 54t estimated by Metals Focus. This owes much to the nature of the Turkish market, where each year a large quantity of old jewellery is exchanged for a new piece, with the consumer paying just the making charge. This exchange is popular across much of the Middle East, South Asia and parts of East Asia, and will therefore appear in LBMA’s “unprocessed” and “melted recycling” segments. However, this exchange is also excluded from Metals Focus’ recycling series, but helps explain why the Turkish total is so much higher in the COO survey.

Conversely, the COO gold recycling total for India of 20t is eclipsed by Metals Focus’ estimate of 98t. The scrap is processed entirely in-country and the disparity is due to there being just one gold GDL, with the bulk of gold recycling therefore treated by India’s 40 or so non-GDL gold refiner.

Source: Metals Focus, LBMA