Typical accounting depreciation is very long and makes some assumptions about stuff breaking by chance and getting outdated and what not. Many assets outlive their depreciation schedules.

I suspect crypto mining rigs are a very different beast, actually burning out after a relatively short period of overuse. Turning it off may well extend the actual life of the assets. And if they’re regularly buying more for the rig then that’s worth considering too.

Disclaimer: just thinking aloud with no evidence

I think I worded it badly, I meant that if depreciation is high then you wouldn't want to turn it off because you'd want to mine as much as possible while it was still worth something.

So I meant the depreciation would need to be low enough to make turning them off a worthwhile strategy.

Edit: I see what you mean about sunk costs, I had worded it badly but you were still arguing against my intended meaning.

I don’t think I misunderstood. How fast equipment is “depreciating” is entirely irrelevant, unless you are using the term in a manner I am misunderstanding.

Understanding marginal economics is difficult because it is counter-intuitive. I am currently on a personal binge trying to help understand the economic theory because it helps me challenge whether I have understood and whether I can pass that understanding on to others.

Unfortunately marginal economics is poorly explained - the link I referenced was the only one of five I quickly skimmed that seemed to be free of gross errors and had no misleading content.

Edit: latchkey seems pretty froody and made a couple of relevant comments:

* Bitcoin OPEX (marginal cost) estimation of $7k to $9k per Bitcoin: https://news.ycombinator.com/item?id=31797201

* CAPEX vs OPEX: https://news.ycombinator.com/item?id=31796669

Capex is spending on your crap, ie buying it and maintenance. Opex is what you spend to run your crap.

Doesn't really matter. If the price of Bitcoin is below the cost of electricity and labor required to mine it, then it's irrelevant that the card won't be as productive (lose money at a greater rate) in future. You're still losing money either way.

The only valid reason to continue mining would be that you expected that the price of Bitcoin would exceed the mining costs in the future. Then you would stockpile them and take advantage of the reduced hashrate as other market participants exited.