Same's true of gold, silver, or cash-settled fresh bacon index futures. Consequently such commodities tend to have long-run zero returns, or returns that just equal the per-capita GDP growth rate, depending on how you look at it. I've watched people lose fortunes at that poker table.
The fresh bacon index is maybe less exposed to this kind of thing because, unless the world mass-converts to Judaism or something, someone will pay to eat bacon, so there's a floor on how low the value can go. But gold? Industrial use of gold is minuscule compared to speculative gold trading. Today GC trades at US$1773 per troy ounce, which is 10% down from a few months ago, late 02020, and 100% up from 02008. In 02001 it was barely above US$200. It could drop to US$200 again, and everyone who bought today and held will have lost 80% of what they invested.
That can happen with fresh bacon, too. What's different with gold is that, if enough people decide to sell, it could drop to US$20. It could drop to US$2. Or it could rise to US$20000. We have more history about gold: it's been a precious metal for many millennia and a widespread currency for the last three. So it's a lot less likely for it to lose 99% or 99.9% of its value like that, or go up 100× (though, as I said, it's gone up very close to 10× in a mere score of years.)
People — and, especially, central banks and governments — invest in gold because they dont think it's likely for that to happen, and because it doesn't have the secular inflationary tendency that fiat currencies do. It may bounce up and down by a factor of 10 in a couple of decades, but in 01687 it was probably also within that same factor-of-10 band.
They're not looking for an expectation of profit when they seek a "store of value".
They're just looking to reduce the risk of indigency.