Since 1980 it’s had a negative real return, so for the last 28 years it hasn’t looked so rosy – but of course that’s data mining in that I looked for the peak.

I think the discrepancy is that the “global gold supply” is different from “all the gold ever mined”. Wikipedia indicates that at the end of 2006 there was 158,000 tonnes of gold ever mined in history (that they know of). This is confirmed by The World Gold Council as well.

It is also interesting to note that the same article estimates that the gold supply will last another 45 years at current consumption rates. Perhaps with a limited supply and seemingly increasing demand there is some hope for gold as a long term investment/speculation? Again, I’m happy to watch from the sidelines.

However, an additional consideration is that the transaction costs, but more importantly the storage and insurance costs, will eat into the returns as well. So if gold continues at 3.4% real gross of costs, then the real returns after costs will be less than half of that I would guess.

The stat about all the gold ever mined would fit in a cube only 19 meters on a side is interesting. According to Wikipedia, “in 2005 the World Gold Council estimated total global gold supply to be 3,859 tonnes.” At 19.3 tonnes per cubic meter, this is a cube only 6 meters on a side. This is a factor of 30 times smaller than your cube. Perhaps the discrepancy comes from most of the gold ever mined having been used for various purposes and lost?