If you plot central bank buying in tons per annum it has picked up from roughly 500 tons per annum over 2016 - 2021, to over 1,000 tons per annum since 2022 — and that is even with China’s significant (and self-acknowledged) under-reporting. In a market where annual supply is only 4,700 tons each year, that change is significant. Yes, the recent frothy price action to $4000+ is due to ETFs but there still exists a significant physical underpin from central banks (and Chinese retail investors, who are no longer investing in property) versus years prior. ETFs were in fact net SELLERs of gold over 2021 to H1’2024, and yet the gold price rose from $1,700 to $2,500. That in itself is unusual: since the advent of ETFs in 2008-ish, the two have always been highly correlated with ETFs having a far louder voice in determining the gold price (albeit on smaller volumes) than the physical market. Central banks (and in particular, EM central banks) are a structural underpin in a larger way than they were in the 2010s.

From a 2023 world gold council survey:

• Will the US dollar’s share of global reserves remain unchanged 5 years from now? EM: 20% agree; DM (developed market): 54% agree.

• Will Gold rise as a share of global reserves 5 years from now? EM: 68% agree; DM: 20% agree.