Musings on Bitcoin's upper bound price and electricity

[biodieselchris]

If bitcoin is generated 6*25 coins per hour (average) *$315 = $47,100. Plus some fees, maybe, so let's call it $50,000. If people collectively are paying more than $50k per hour in electricity, than miners at the margin begin to fall off. Even at a basal state cost of $0.05 per kWh that's $50,000 per hour / 0.05 $/kWh = 1,000,000 kW or 1,000 MW. This doesn't include the cost of hardware deprecation, and uses a very conservative rate for power.

So how much is 1,000 MW? The world produced about 22,000 TWh last year and there's 8765.76 hours in a typical year, so the average electrical output is 2.5098 TW or 2,509,800 MW. Bitcoin is therefore 1,000 / 2,509,800 = 0.0357% of the total. Again, this may be high if the global average rate for power is higher than $0.05 but to a first approximation it's close enough.

The inevitable arms race to increase hash/W is irrelevant. From a microeconomic standpoint that just serves to distribute the coins to the players, and further from a macroeconomic point of view the amount of money spent on power will almost nearly match the current price, with some latency adjustment period and some caveats for people who voluntarily lose money, steal power, etc (I would assume these are small because they will eventually correct over a reasonable time period).

If this theory holds then does it put an upper limit on the price of bitcoin? If Bitcoin was currently worth 100x at $31,500 per coin right now it would logically mean 3.57% of the grid is being used to service bitcoin mining, and 3.5% would have a significant impact; power starts to become a scarce resource that needs to be allocated efficiently (i.e., cost increases). At $31,500 per coin BTC has a fully diluted market cap of $655B which would make it about the size of Apple. It is not beyond the stretch of imagination that a crypto-protocol, with all of it's payment processing functionality, brings that kind of value to the market (e.g., it's not like $5T or something to high of a percent of global GDP to be a credible theory). The price can go beyond this as it just serves to reward miners with more advanced hashing equipment (higher hash/W) or simply increase the profit margin of miners (which would then collectively go into advancements in technology), so round and around we go.

Basically, I have no point. But it's interesting to think about.

[1715217327]

1715217327

[1715217327]

1715217327

[1715217327]

1715217327

[1715217327]

1715217327

[redHairy]

If bitcoin is generated 6*25 coins per hour (average) *$315 = $47,100. Plus some fees, maybe, so let's call it $50,000. If people collectively are paying more than $50k per hour in electricity, than miners at the margin begin to fall off. Even at a basal state cost of $0.05 per kWh that's $50,000 per hour / 0.05 $/kWh = 1,000,000 kW or 1,000 MW. This doesn't include the cost of hardware deprecation, and uses a very conservative rate for power.

So how much is 1,000 MW? The world produced about 22,000 TWh last year and there's 8765.76 hours in a typical year, so the average electrical output is 2.5098 TW or 2,509,800 MW. Bitcoin is therefore 1,000 / 2,509,800 = 0.0357% of the total. Again, this may be high if the global average rate for power is higher than $0.05 but to a first approximation it's close enough.

The inevitable arms race to increase hash/W is irrelevant. From a microeconomic standpoint that just serves to distribute the coins to the players, and further from a macroeconomic point of view the amount of money spent on power will almost nearly match the current price, with some latency adjustment period and some caveats for people who voluntarily lose money, steal power, etc (I would assume these are small because they will eventually correct over a reasonable time period).

If this theory holds then does it put an upper limit on the price of bitcoin? If Bitcoin was currently worth 100x at $31,500 per coin right now it would logically mean 3.57% of the grid is being used to service bitcoin mining, and 3.5% would have a significant impact; power starts to become a scarce resource that needs to be allocated efficiently (i.e., cost increases). At $31,500 per coin BTC has a fully diluted market cap of $655B which would make it about the size of Apple. It is not beyond the stretch of imagination that a crypto-protocol, with all of it's payment processing functionality, brings that kind of value to the market (e.g., it's not like $5T or something to high of a percent of global GDP to be a credible theory). The price can go beyond this as it just serves to reward miners with more advanced hashing equipment (higher hash/W) or simply increase the profit margin of miners (which would then collectively go into advancements in technology), so round and around we go.

Basically, I have no point. But it's interesting to think about.

thanks for sharing your calculation.
but you forgive a point...
0.05 $/kWh it is not possible in a large amount of nations...
in eu we have 0.25euro /kWh
so electricity cost is 5 times more, without considering change who is EUR/USD --> 1,24!!!
so your calculation is restrictive

[odolvlobo]

I don't think $31k $/BTC is likely anytime soon because adoption would have to be quite high to support a price of $31k. Also, keep in mind that the U.S. monetary base reached $665 billion in 2001. By the time the price reaches $31k, the block reward will be much lower and world power output will be much higher.

Finally, the price of a bitcoin depending on the cost of mining is a common misconception. The cost of mining depends on the price of a bitcoin but not the other way around.

[hazenyc]

Finally, the price of a bitcoin depending on the cost of mining is a common misconception. The cost of mining depends on the price of a bitcoin but not the other way around.

I have looked into this for a working paper actually-
It seems that when looking at Granger Causality, the directionality is hashpower drives price. The logical argument could be made against this that a high price will attract hashpower, but it seems that it is the hashpower itself that buoys the price.

Interestingly, When looking at Granger Causality of the log/log (%change in hashpower vs %change in price), it becomes bi-drectionally significant.

[Q7]

If Bitcoin was currently worth 100x at $31,500 per coin, that means the demand must be so strong to push it to that level and it simply outweighs the selling pressure. When that happens more miners decide to join the group and competition becomes tough and some of them would drop out simply because they can't churn out a profitable amount of bitcoin either it is due to the mining equipment factor or just any other reason. Eventually the pool of miners will still remain about the same. Electricity demand would then ends about the same. What I'm trying to say here, it will eventually reach a stage of equilibrium.