4 ms·

I've always touted the line about comparing the energy consumed to the value produced. And while I still think that you must normalize the numbers, I also star

I've always touted the line about comparing the energy consumed to the value produced. And while I still think that you must normalize the numbers, I also start to grow concerned that its continued growth seems like it might actually be a problem. Oh, wow, -- lightbulb -- I didn't notice this until I started to write this post. I was going to ruminate about what PoW could we possibly do instead (presuming we need PoW for a good cryptocoin). Now I'm thinking that this is the correction that will finally level off speculation? Bitcoin's value is bounded: it can't be worth much more to humanity than the sum of all global financial services (likely worth much less). But its energy consumption is only bounded by the perception of its future value. Hash rate growth is fueled by speculation in continued growth in the bitcoin exchange rate. At some point the energy consumption may grow fast enough to damper excitement about future gains. Then again, with each halving the inflation rate slows. So it's calibrated to keep increasing. I wonder if the next Bitcoin community crisis will be different approaches to difficulty or reward. Those seem much more "untouchable" than block size, so hopefully not.

runeks 9y ago
> But its energy consumption is only bounded by the perception of its future value. Hash rate growth is fueled by speculation in continued growth in the bitcoin exchange rate. How do you figure? Miners mine bitcoins and sell them at the current exchange rate -- not a future one. Mining to speculate on future price increases make no sense. If you believe the price will rise, just buy bitcoins now (no reason to mine them at a loss).
- raiflip 9y ago
  I think he pins it to future price instead of current price because of the time lag between the decision to mine for bitcoin and the time you actually get the coin. If you think bitcoin is going to crash by the time you start making a profit then even if it is worth a lot now you're not going to mine.
- nhaehnle 9y ago
  Are miners actually selling all earned bitcoins immediately? If so, that means there must be at least 20 million USD "entering" Bitcoin every day in net terms. That seems like a lot, certainly much more than could be supported by actual economic activities in Bitcoin (including illegal ones). How many people are there who are willing to use, say, 10k USD of their own money for gambling? How many greater fools? Then again, some of those "greater fools" could be people holding Bitcoin that they collected from ICOs, so who knows. As they say, the really tricky part of attempting to profit from the bursting of bubbles is the timing.
  - killjoywashere 9y ago
    20M USD/d * 365 ~ 7B USD. I suspect people are trying to evacuate north of 1T USD from China alone. This will go on for years.
- jlarocco 9y ago
  A potential bitcoin miner has to decide whether it's worth buying the equipment in the first place, and then later they have to decide whether it's worth it to upgrade. Before they actually start mining they're speculating that the future value will cover the up front equipment cost, on going electricity costs, and all of that. IMO, the interesting number isn't the cost of mining or the value of the bitcoins in dollars, but the value of goods and services being bought directly with bitcoin. If everybody treats it as an investment and never spends it, or immediately cashes out, then isn't it essentially just a big ponzi scheme?
stale2002 9y ago
The sum of all financial transaction services has got to be worth Trillions of dollars. That is a very high "limit". Banks use electricity too, in order to keep the literal lights on at their buildings.
- NTDF9 9y ago
  Do they use as much electricity JUST to validate a transaction? With bitcoin, we have electricity consumption per transaction AANNDDD lights in bldgs to provide services like loans, insurance, trading? Wtf?
  - stale2002 9y ago
    I think you'd be surprised how much money is spent just on fraud prevention, security, storing peoples money, and all the other stuff that goes into "validating transactions" alone. It is not an easy problem. Also, I've never needed to go into a bitcoin bank in order to talk to a bitcoin bank teller, so that the person at the front desk can deposit my Bitcoin check. So I'd argue that there would be a whole lot less literal lights in a building, because you don't need to handle those in person services like that. But this is also ignoring the fact that it is a fundamentally different problem that is being solved. The problem that Bitcoin is trying to solve is a decentralized payment system that can't be censored by a central authority. Let me know when a real life bank can do THAT.
    - NTDF9 9y ago
      But not all of Banking is just validating transactions. Do you get that? The amount of energy that goes into maintaining accounts will be the same with maintaining wallets. The fact that any central authority/gang/thief can seize my wallet by simply stealing my password is a bigger problem than a bank holding on to my transactions...a bank I can trust because of rule of law. Now we'll have to have fraud prevention for bitcoin...which means more additional energy.
erikpukinskis 9y ago
> Bitcoin's value is bounded: it can't be worth much more to humanity than the sum of all global financial services (likely worth much less). What's the difference between "all global financial services" and "the global economy"? If every asset was publicly traded, and those stocks were all denominated in Bitcoin, wouldn't the value of Bitcoin be the entire global economy? Or are you talking more pragmatically, where "financial services" is all assets that it's economically valuable to denominate in monetary terms, and there is some set of other assets that are kept "off the books", like an uncapitalized startup, or an employee, or something of that nature? Like, I would consider "financial services" to be a subset of things that are denominated in the US Dollar, but Bitcoin could potentially be used for that whole set, not just the financial services segment.
crimsonalucard 9y ago
This is worse then gold. Its not even real. Its meaningless data.
- omio 9y ago
  The blockchain contains the data of billions of dollars of transactions. I would say that is very far from meaningless.
JumpCrisscross 9y ago
> [Bitcoin] can't be worth much more to humanity than the sum of all global financial services Finance is about 7.3% of U.S. GDP [1]. The U.S. produces about 1,000 gigawatts of energy per year [2]. A first-order estimation thus yields 73 Gigawatts for finance. So about 22 times Bitcoin’s 3.3 gigawatts [0]. About $1.5 billion of BTC change hands every day [3]. That’s $550 billion a year. U.S. non-cash transaction volume is like $180 trillion a year [4]. So about 330 times the size of the Bitcoin economy. Bitcoin thus appears about 15 times less efficient than the non-cash status quo. Given finance is less energy intensive than most of the U.S. economy, this is likely a lower-bound estimate. Worryingly, I ran these numbers 2 months ago [5]. If the power estimates are accurate, the Bitcoin network is becoming less efficient. (I assume the variance is explained by measurement error.) [0] https://digiconomist.net/bitcoin-energy-consumption#assumptions https://digiconomist.net/bitcoin-energy-consumption#assumpti... ~29 TWh / year [1] https://www.selectusa.gov/financial-services-industry-united-states https://www.selectusa.gov/financial-services-industry-united... [2] https://en.m.wikipedia.org/wiki/Electricity_sector_of_the_United_States https://en.m.wikipedia.org/wiki/Electricity_sector_of_the_Un... [3] https://www.quandl.com/data/BCHAIN/ETRVU-Bitcoin-Estimated-Transaction-Volume-USD https://www.quandl.com/data/BCHAIN/ETRVU-Bitcoin-Estimated-T... [4] https://www.federalreserve.gov/newsevents/press/other/2016-payments-study-20161222.pdf https://www.federalreserve.gov/newsevents/press/other/2016-p... [5] https://news.ycombinator.com/item?id=15387663* https://news.ycombinator.com/item?id=15387663*
- IkmoIkmo 9y ago
  > Bitcoin thus appears about 15 times less efficient than the non-cash status quo. You can make calculations today, but they're somewhat meaningless. The reason is, miners don't mine to secure the system and earn money on transaction fees. They mine to earn from new bitcoins being created. As such, miners total expenses will approach the rewards in bitcoin, which has been >90% of a miner's income. Long-term, bitcoin will either die, or it'll switch to a system where miner's primary income is from transaction fees. At that point you'll see cost per volume come down to something much more sensible. In the short term this transaction cost to volume ratio tends to get worse as the price increase outpaces usage. Bitcoin's price increased over tenfold the past 12 months or so, usage didn't, so you'll get a whole lot more mining expenses (electricity) spread out over not many more transactions. But again, that's only because mining expenses currently are linked to bitcoin rewards for mining which correlate with price increases. Bitcoin is designed to phase that out, so again, this is a temporary situation, whether bitcoin has 100 users or 1 billion, this shouldn't last. disclaimer: not holding any bitcoins.

runeks 9y ago

> But its energy consumption is only bounded by the perception of its future value. Hash rate growth is fueled by speculation in continued growth in the bitcoin exchange rate. How do you figure? Miners mine bitcoins and sell them at the current exchange rate -- not a future one. Mining to speculate on future price increases make no sense. If you believe the price will rise, just buy bitcoins now (no reason to mine them at a loss).

raiflip 9y ago
I think he pins it to future price instead of current price because of the time lag between the decision to mine for bitcoin and the time you actually get the coin. If you think bitcoin is going to crash by the time you start making a profit then even if it is worth a lot now you're not going to mine.
nhaehnle 9y ago
Are miners actually selling all earned bitcoins immediately? If so, that means there must be at least 20 million USD "entering" Bitcoin every day in net terms. That seems like a lot, certainly much more than could be supported by actual economic activities in Bitcoin (including illegal ones). How many people are there who are willing to use, say, 10k USD of their own money for gambling? How many greater fools? Then again, some of those "greater fools" could be people holding Bitcoin that they collected from ICOs, so who knows. As they say, the really tricky part of attempting to profit from the bursting of bubbles is the timing.
- killjoywashere 9y ago
  20M USD/d * 365 ~ 7B USD. I suspect people are trying to evacuate north of 1T USD from China alone. This will go on for years.
jlarocco 9y ago
A potential bitcoin miner has to decide whether it's worth buying the equipment in the first place, and then later they have to decide whether it's worth it to upgrade. Before they actually start mining they're speculating that the future value will cover the up front equipment cost, on going electricity costs, and all of that. IMO, the interesting number isn't the cost of mining or the value of the bitcoins in dollars, but the value of goods and services being bought directly with bitcoin. If everybody treats it as an investment and never spends it, or immediately cashes out, then isn't it essentially just a big ponzi scheme?

stale2002 9y ago

The sum of all financial transaction services has got to be worth Trillions of dollars. That is a very high "limit". Banks use electricity too, in order to keep the literal lights on at their buildings.

NTDF9 9y ago
Do they use as much electricity JUST to validate a transaction? With bitcoin, we have electricity consumption per transaction AANNDDD lights in bldgs to provide services like loans, insurance, trading? Wtf?
- stale2002 9y ago
  I think you'd be surprised how much money is spent just on fraud prevention, security, storing peoples money, and all the other stuff that goes into "validating transactions" alone. It is not an easy problem. Also, I've never needed to go into a bitcoin bank in order to talk to a bitcoin bank teller, so that the person at the front desk can deposit my Bitcoin check. So I'd argue that there would be a whole lot less literal lights in a building, because you don't need to handle those in person services like that. But this is also ignoring the fact that it is a fundamentally different problem that is being solved. The problem that Bitcoin is trying to solve is a decentralized payment system that can't be censored by a central authority. Let me know when a real life bank can do THAT.
  - NTDF9 9y ago
    But not all of Banking is just validating transactions. Do you get that? The amount of energy that goes into maintaining accounts will be the same with maintaining wallets. The fact that any central authority/gang/thief can seize my wallet by simply stealing my password is a bigger problem than a bank holding on to my transactions...a bank I can trust because of rule of law. Now we'll have to have fraud prevention for bitcoin...which means more additional energy.

erikpukinskis 9y ago

> Bitcoin's value is bounded: it can't be worth much more to humanity than the sum of all global financial services (likely worth much less). What's the difference between "all global financial services" and "the global economy"? If every asset was publicly traded, and those stocks were all denominated in Bitcoin, wouldn't the value of Bitcoin be the entire global economy? Or are you talking more pragmatically, where "financial services" is all assets that it's economically valuable to denominate in monetary terms, and there is some set of other assets that are kept "off the books", like an uncapitalized startup, or an employee, or something of that nature? Like, I would consider "financial services" to be a subset of things that are denominated in the US Dollar, but Bitcoin could potentially be used for that whole set, not just the financial services segment.

crimsonalucard 9y ago

This is worse then gold. Its not even real. Its meaningless data.

omio 9y ago
The blockchain contains the data of billions of dollars of transactions. I would say that is very far from meaningless.

JumpCrisscross 9y ago

> [Bitcoin] can't be worth much more to humanity than the sum of all global financial services Finance is about 7.3% of U.S. GDP [1]. The U.S. produces about 1,000 gigawatts of energy per year [2]. A first-order estimation thus yields 73 Gigawatts for finance. So about 22 times Bitcoin’s 3.3 gigawatts [0]. About $1.5 billion of BTC change hands every day [3]. That’s $550 billion a year. U.S. non-cash transaction volume is like $180 trillion a year [4]. So about 330 times the size of the Bitcoin economy. Bitcoin thus appears about 15 times less efficient than the non-cash status quo. Given finance is less energy intensive than most of the U.S. economy, this is likely a lower-bound estimate. Worryingly, I ran these numbers 2 months ago [5]. If the power estimates are accurate, the Bitcoin network is becoming less efficient. (I assume the variance is explained by measurement error.) [0] https://digiconomist.net/bitcoin-energy-consumption#assumptions https://digiconomist.net/bitcoin-energy-consumption#assumpti... ~29 TWh / year [1] https://www.selectusa.gov/financial-services-industry-united-states https://www.selectusa.gov/financial-services-industry-united... [2] https://en.m.wikipedia.org/wiki/Electricity_sector_of_the_United_States https://en.m.wikipedia.org/wiki/Electricity_sector_of_the_Un... [3] https://www.quandl.com/data/BCHAIN/ETRVU-Bitcoin-Estimated-Transaction-Volume-USD https://www.quandl.com/data/BCHAIN/ETRVU-Bitcoin-Estimated-T... [4] https://www.federalreserve.gov/newsevents/press/other/2016-payments-study-20161222.pdf https://www.federalreserve.gov/newsevents/press/other/2016-p... [5] https://news.ycombinator.com/item?id=15387663* https://news.ycombinator.com/item?id=15387663*

IkmoIkmo 9y ago
> Bitcoin thus appears about 15 times less efficient than the non-cash status quo. You can make calculations today, but they're somewhat meaningless. The reason is, miners don't mine to secure the system and earn money on transaction fees. They mine to earn from new bitcoins being created. As such, miners total expenses will approach the rewards in bitcoin, which has been >90% of a miner's income. Long-term, bitcoin will either die, or it'll switch to a system where miner's primary income is from transaction fees. At that point you'll see cost per volume come down to something much more sensible. In the short term this transaction cost to volume ratio tends to get worse as the price increase outpaces usage. Bitcoin's price increased over tenfold the past 12 months or so, usage didn't, so you'll get a whole lot more mining expenses (electricity) spread out over not many more transactions. But again, that's only because mining expenses currently are linked to bitcoin rewards for mining which correlate with price increases. Bitcoin is designed to phase that out, so again, this is a temporary situation, whether bitcoin has 100 users or 1 billion, this shouldn't last. disclaimer: not holding any bitcoins.