This post runs quite close to the previous one, but I think it’s important to flesh out the argument that the marginal cost of gold, in particular, is not a support to its price.
In recent days large falls in the price of gold have led people to ask whether and what level a floor might be found. Some have responded by taking out rulers and drawing trend lines and contracting triangles. Those with a bit more real-world grounding have argued that producers will cease supplying if the price falls below marginal cost, and therefore the price should find a support there.
This logic works quite well in almost all commodity markets. However, there are a couple of important characteristics of the gold market that lead to perverse effects.
Firstly, gold is not consumed, and annual production only accounts for a small proportion of the total stock. Even if there were no new bars minted, trade in the secondary market would still set a price. And this price would be entirely a function of fear and greed, not marginal cost. Importantly, gold is very different from most other commodities which derive value the role they play in a production chain.
Secondly, marginal cost is a function of price, not the other way around. The marginal cost of production increases with the amount mined. When the gold price is high, producers mine lower graded ore and increase through-put. This means cash costs go up, but the higher volume means more revenue. Producers alter the production profile to match what they see in the end market. Following the recent drop in the gold price, miners will cut production and switch to higher grade ore. There may be a lag of a few months before this is evident.
The case for buying gold rests on the self-fulfilling legacy that in times when many other assets are losing their value it does well. But the cost of bearing this insurance during more benign times may outweigh the ostensible merits, especially if there is no theoretical limit to how close to zero the price can go.