5 ms·

Only tangentially related: How do bitcoin network costs scale with market price? We can expect the marginal cost to mine a coin to scale with price. So if btc

by graeme 6y ago

Only tangentially related: How do bitcoin network costs scale with market price? We can expect the marginal cost to mine a coin to scale with price. So if btc worth $52,000 then people will invest on average $52,000 in equipment and electricity. There are 6.5 coins mined every 10 min. So daily network cost approximates to 6.5 * 6 * 24 * $52,000 = $48,000,000 This is not strictly accurate as past equipment costs less and people may over invest expecting appreciation. But it should be close enough. So to run the network at a price of $52,000 you need $17.76 billion dollars to enter as fresh capital to be burned as electricity and equipment, correct? And if the price hit $1,000,000 as boosters claim, mining difficulty would rise to adjust and the new annual maintenance cost would be: 6.5 * 6 * 24 * * 365 * $1,000,000 = $341,640,000,000 And then this would decrease as halving happened. In May 2020 and then 2024. So the annual cost to run a $1,000,000 btc network post the may 2020 halving would be about $170 billion per year. That much new capital needs to enter the market and be burned up simply to maintain the price at $1,000,000. And to be useful to the world a $1,000,000 bitcoin would need to generate $170,000,000,000 in value annually. Have I got this right or did I make a grievous error somewhere?

martinko 6y ago
You've got it right. That being said, I dont think anyone realistically expects 1m / btc this decade.
- graeme 6y ago
  Right just thinking through the claims of the boosters to their logical end. If it did happen in nine years the next halving is 2028 so the cost would be ~$85 billion per year. And that would be less than it is now due to inflation. Ok that’s actually not as bad as I thought, though still pretty expensive. You need $85 billion from new entrants in burnt capital just to maintain the value of existing bitcoins at that price, and people need some reason for putting in capital other than price appreciation assuming appreciation stops.
  - usehackernews 6y ago
    You seem to have a good grasp of creating high level models. I always see people referring to the energy consumption of bitcoin as though it’s strictly additive. What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies. Second, what people tend to ignore is that innovation doesn’t come in a vacuum - there are second, and third degree improvements triggered from it. In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin. Third, and this isn’t directed at you, just a general statement based on arguments I see on HN - cryptocurrency itself does not have high energy consumption. Proof-of-work mining does. There are methods to support cryptocurrency that do not result in high energy consumption (Proof of Stake, DAGs). It’s wrong to generalize energy consumption issues to all cryptocurrencies.
    - graeme 6y ago
      > What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies. Very hard to compare. One thing I’m unsure of for bitcoin: do costs go up if there are more transactions or does mining difficulty adjust so that costs don’t rise no matter how many transactions bitcoin does? I realize this is a major gap in my understanding. The finance and currency system does much that bitcoin doesn’t currently do. So comparing as a whole isn’t adequate unless bitcoin network costs don’t scale up with numbers of transactions because mining gets easier. Actually how do miners make money once there are no more coins to mine? > In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin. That actually doesn’t follow. All that follows is that the world has decided to move $170 billion in capital into bitcoin per year. The world is presently set to move about $20 billion of capital into bitcoin per year, and there’s no real consensus above usefulness. The world could be putting $170 billion in because there is value, or it could be a massive waste of resources. The world has certainly misallocated resources before. And yeah I was restricting my own comments to bitcoin. Ethereum et al I know much less about.
- hfsp 6y ago
  Have fun staying poor.
csomar 6y ago
You are not wrong. However 1= $1m/bitcoin is x20 from here. And here is already a pretty high price. 2= $170bn is a lot of money. But on an international/global scale, it's a drop in the bucket.
- graeme 6y ago
  Yeah I just want to think through the claims of the maximalists. For example I often hear them say “it’s digital gold, so it has 10x to grow”. That would be $85 billion for mining, whereas gold mining annually is about $1.3 billion. ~$1000 per ounce mining cost * 130 million ounces mined. Of course it’s not an exact comparison as you prob ought to add in storage costs for gold but then maybe you’d need to add crypto online wallets and it gets messy. Mining to mining is clear and simple. It is a drop in the bucket globally, but it also means you need that much capital flowing in to maintain the market price. Just to maintain. So you would want the network to be producing that much value at least.
  - csomar 6y ago
    Think about it this way: There is 7bn people on the planet. $170bn works out an average of $24/person. Or $2/month. Even poor people in poor countries can afford that. It might actually be cheaper for them, many poor countries (think Africa, Argentina, Brazil, etc...) are going through hyperinflation that it is safer and less volatile to hold Bitcoin.
    - graeme 6y ago
      That calculus doesn’t work as a store of value. For the substantial minority of people earnings ~$2 per day, $24 per year is a sizeable fraction of wealth. As far as currency goes, it bitcoin transactions rise, does that have no impact on price? I realized my price thinking makes no mention of them and I don’t know how they affect things.
      - csomar 6y ago
        $24 is not a sizeable fraction of wealth by any measure. And here we are talking averages. The average per capita GDP is $18k. $24 is a very small number to that.
soared 6y ago
> We can expect the marginal cost to mine a coin to scale with price. This feels correct but I don’t think it is. For example gold has decreased by $200/oz recently - it definitely didn’t just get cheaper to mine. I know mining is different and it’s not a great comparison, but costs don’t scale immediately, miners don’t make decisions entirely based on the price of btc now, etc. Economic principles like supply and demand meeting on an agreed price are very muddy.
- graeme 6y ago
  No no bitcoin mining and gold mining don’t have the same economics. Gold mining average costs are determined by the actual physical costs of mining. Marginal costs will rise in marginal mines if the gold price goes up. But that won’t affect average gold mining costs in existing mines. But in bitcoin average costs have to work out to marginal costs. When more mining power enters mining gets harder to compensate and mining costs rise.
  - pilord8 6y ago
    That makes a lot of sense. On the other hand, that also suggests that if the price of bitcoin were to rise rapidly, mining costs would take some time to catch up, no? The connection between mining costs and price is an economic one (that can be in temporary disequilibrium), right?
    - graeme 6y ago
      Depends how rapidly. You can buy the hardware as a commodity, and connect to the power grid. Existing miners prob have expansion plans too they can accelerate. So the market may move pretty fast. Though I read an analysis that there were surplus profits in the 2017 bubble.
- samvher 6y ago
  You say "it definitely didn't just get cheaper to mine", which sounds obviously true, but in a way I'm not sure it is. If the price of gold decreases by $200/oz, that probably makes a bunch of mines unprofitable, leading them to close down (or at least pause) their operations. Which in a way would mean that (on average) gold got "cheaper to mine". And it seems to me that the Bitcoin mining market works basically the same in that sense.
Cullinet 6y ago
stock exchanges separate (and occasionally reacquire for commercial reasons) trading from custodian services. I'm not at all up to date but I long ago presumed bitcoin needed to move to a custody transfer market instead of processing on the block chain and the expense involved or beholden attachment that creates to a miner. if Citi (eg) will lend me to trade against collateral of my existing holdings finding risk less (cp counterparty and ops*) liquidity from stock borrowers (typically short players) and happily create a new income that's limited by the value at risk of my trading up to my collateral underlying assets and therefore greater than the principal, surely so can Coinbase. I assumed that was why the London Stock Exchange invested in them. edit I forgot *: see Herstatt Risk inherent in closing currency cash dealing causing the eponymous bank's failure in 1974. possibly more interesting to today was a year earlier in the Cedar Bank crisis in '73 that although little known was the most existential threat to the markets certainly pre LTCM and I'd say still far more dangerous. relevant today because Cedar was a pseudo official shadow bank living by profits from high risk secondary mortgage loans. Cedar was only a bank under a statutory loophole the infamous (to my career at least) Section 123 of the English banking act. interestingly none of the UK so called challenger banks are clearing banks the only new one in 400 years is a white label specialist running on Azure called Clear.bank at that url. a notable lot of UK transactions are happening on Azure with Clear.
TheTaytay 6y ago
I think you’re on track. I found this article to be a great analysis of the electric consumption as it relates to the price of BTC. https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-coins-are-an-esg-nightmare/ https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-c...
t-vi 6y ago
The maths seem right, but I don't think the 6.5 btc/10 minutes is a given for the future, so you're extrapolation contains the rather large assumption that it stays the same.

by graeme 6y ago

martinko 6y ago

You've got it right. That being said, I dont think anyone realistically expects 1m / btc this decade.

graeme 6y ago
Right just thinking through the claims of the boosters to their logical end. If it did happen in nine years the next halving is 2028 so the cost would be ~$85 billion per year. And that would be less than it is now due to inflation. Ok that’s actually not as bad as I thought, though still pretty expensive. You need $85 billion from new entrants in burnt capital just to maintain the value of existing bitcoins at that price, and people need some reason for putting in capital other than price appreciation assuming appreciation stops.
- usehackernews 6y ago
  You seem to have a good grasp of creating high level models. I always see people referring to the energy consumption of bitcoin as though it’s strictly additive. What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies. Second, what people tend to ignore is that innovation doesn’t come in a vacuum - there are second, and third degree improvements triggered from it. In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin. Third, and this isn’t directed at you, just a general statement based on arguments I see on HN - cryptocurrency itself does not have high energy consumption. Proof-of-work mining does. There are methods to support cryptocurrency that do not result in high energy consumption (Proof of Stake, DAGs). It’s wrong to generalize energy consumption issues to all cryptocurrencies.
  - graeme 6y ago
    > What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies. Very hard to compare. One thing I’m unsure of for bitcoin: do costs go up if there are more transactions or does mining difficulty adjust so that costs don’t rise no matter how many transactions bitcoin does? I realize this is a major gap in my understanding. The finance and currency system does much that bitcoin doesn’t currently do. So comparing as a whole isn’t adequate unless bitcoin network costs don’t scale up with numbers of transactions because mining gets easier. Actually how do miners make money once there are no more coins to mine? > In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin. That actually doesn’t follow. All that follows is that the world has decided to move $170 billion in capital into bitcoin per year. The world is presently set to move about $20 billion of capital into bitcoin per year, and there’s no real consensus above usefulness. The world could be putting $170 billion in because there is value, or it could be a massive waste of resources. The world has certainly misallocated resources before. And yeah I was restricting my own comments to bitcoin. Ethereum et al I know much less about.
hfsp 6y ago
Have fun staying poor.

csomar 6y ago

You are not wrong. However 1= $1m/bitcoin is x20 from here. And here is already a pretty high price. 2= $170bn is a lot of money. But on an international/global scale, it's a drop in the bucket.

graeme 6y ago
Yeah I just want to think through the claims of the maximalists. For example I often hear them say “it’s digital gold, so it has 10x to grow”. That would be $85 billion for mining, whereas gold mining annually is about $1.3 billion. ~$1000 per ounce mining cost * 130 million ounces mined. Of course it’s not an exact comparison as you prob ought to add in storage costs for gold but then maybe you’d need to add crypto online wallets and it gets messy. Mining to mining is clear and simple. It is a drop in the bucket globally, but it also means you need that much capital flowing in to maintain the market price. Just to maintain. So you would want the network to be producing that much value at least.
- csomar 6y ago
  Think about it this way: There is 7bn people on the planet. $170bn works out an average of $24/person. Or $2/month. Even poor people in poor countries can afford that. It might actually be cheaper for them, many poor countries (think Africa, Argentina, Brazil, etc...) are going through hyperinflation that it is safer and less volatile to hold Bitcoin.
  - graeme 6y ago
    That calculus doesn’t work as a store of value. For the substantial minority of people earnings ~$2 per day, $24 per year is a sizeable fraction of wealth. As far as currency goes, it bitcoin transactions rise, does that have no impact on price? I realized my price thinking makes no mention of them and I don’t know how they affect things.
    - csomar 6y ago
      $24 is not a sizeable fraction of wealth by any measure. And here we are talking averages. The average per capita GDP is $18k. $24 is a very small number to that.

soared 6y ago

> We can expect the marginal cost to mine a coin to scale with price. This feels correct but I don’t think it is. For example gold has decreased by $200/oz recently - it definitely didn’t just get cheaper to mine. I know mining is different and it’s not a great comparison, but costs don’t scale immediately, miners don’t make decisions entirely based on the price of btc now, etc. Economic principles like supply and demand meeting on an agreed price are very muddy.

graeme 6y ago
No no bitcoin mining and gold mining don’t have the same economics. Gold mining average costs are determined by the actual physical costs of mining. Marginal costs will rise in marginal mines if the gold price goes up. But that won’t affect average gold mining costs in existing mines. But in bitcoin average costs have to work out to marginal costs. When more mining power enters mining gets harder to compensate and mining costs rise.
- pilord8 6y ago
  That makes a lot of sense. On the other hand, that also suggests that if the price of bitcoin were to rise rapidly, mining costs would take some time to catch up, no? The connection between mining costs and price is an economic one (that can be in temporary disequilibrium), right?
  - graeme 6y ago
    Depends how rapidly. You can buy the hardware as a commodity, and connect to the power grid. Existing miners prob have expansion plans too they can accelerate. So the market may move pretty fast. Though I read an analysis that there were surplus profits in the 2017 bubble.
samvher 6y ago
You say "it definitely didn't just get cheaper to mine", which sounds obviously true, but in a way I'm not sure it is. If the price of gold decreases by $200/oz, that probably makes a bunch of mines unprofitable, leading them to close down (or at least pause) their operations. Which in a way would mean that (on average) gold got "cheaper to mine". And it seems to me that the Bitcoin mining market works basically the same in that sense.

Cullinet 6y ago

stock exchanges separate (and occasionally reacquire for commercial reasons) trading from custodian services. I'm not at all up to date but I long ago presumed bitcoin needed to move to a custody transfer market instead of processing on the block chain and the expense involved or beholden attachment that creates to a miner. if Citi (eg) will lend me to trade against collateral of my existing holdings finding risk less (cp counterparty and ops*) liquidity from stock borrowers (typically short players) and happily create a new income that's limited by the value at risk of my trading up to my collateral underlying assets and therefore greater than the principal, surely so can Coinbase. I assumed that was why the London Stock Exchange invested in them. edit I forgot *: see Herstatt Risk inherent in closing currency cash dealing causing the eponymous bank's failure in 1974. possibly more interesting to today was a year earlier in the Cedar Bank crisis in '73 that although little known was the most existential threat to the markets certainly pre LTCM and I'd say still far more dangerous. relevant today because Cedar was a pseudo official shadow bank living by profits from high risk secondary mortgage loans. Cedar was only a bank under a statutory loophole the infamous (to my career at least) Section 123 of the English banking act. interestingly none of the UK so called challenger banks are clearing banks the only new one in 400 years is a white label specialist running on Azure called Clear.bank at that url. a notable lot of UK transactions are happening on Azure with Clear.

TheTaytay 6y ago

I think you’re on track. I found this article to be a great analysis of the electric consumption as it relates to the price of BTC. https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-coins-are-an-esg-nightmare/ https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-c...

t-vi 6y ago

The maths seem right, but I don't think the 6.5 btc/10 minutes is a given for the future, so you're extrapolation contains the rather large assumption that it stays the same.