So Bitcoin’s lack of a “fundamental use value” floor is not a serious disadvantage of Bitcoin against gold.
Theoretically, no, practically, it still is. Humans are humans and the whole history and, um, aura of gold makes a “simple equilibrium flip” not very likely in the near future.
Measuring Bitcoin against “gold market cap” is dangerous because gold has uses which Bitcoin cannot replace. For example, traditionally most of the wealth of Indian families (those who have wealth, of course) have been kept as gold, specifically golden jewelry. Bitcoin will not replace that use. Another big advantage of gold is anonymity which Bitcoin will not be able to replicate either.
The probabilities of the various events required for BTC to receive this status are roughly within an order of magnitude of the 5% mark.
Why do you think so?
And what exactly is the reference class that you put Bitcoin into?
my expected-value estimation (10% of gold market cap = $700b = $34000 per BTC * 0.05 chance = $1700 per BTC EV
Don’t think in terms of point estimates, think it terms of full distributions (which here will not be symmetric).
in those worlds where someone actually can torture 3^^^3 people for fun they will likely be able and willing to do it again, so my cooperation may end up leading to the torture of more than 3^^^3 people
We are entering nonsense territory here, as the set up implies that someone who can torture ^^^3^^^ people for fun still need something from you and, moreover, needs your free and willing consent. Recall that the original Pascal Mugger is Omega and I doubt that you can incentivize Omega by cooperating or not.
And what exactly is the reference class that you put Bitcoin into?
Roughly speaking, assets which have received a major amount of public attention and whose store-of-value functionality is the dominant factor in their price. Gold, silver, land and internet domain names are the only others I can think of that vaguely fit there.
So your argument is that gold has a very very large network effect? Reasonable I suppose, but technology has disrupted similarly entrenched things over the past two decades, so you have to add a lot of fundamental uncertainty.
Another big advantage of gold is anonymity which Bitcoin will not be able to replicate either.
It’s quite possible to be anonymous with BTC; the $400m MtGox theft at least is pretty good evidence of that. It just takes a lot of skill. Also, with the growth of meatspace surveillance, I would not be surprised if gold became easier to track than BTC over the next few years.
Recall that the original Pascal Mugger is Omega and I doubt that you can incentivize Omega by cooperating or not.
So for every world where the mugger is Omega, there is a world where the mugger is someone incentivizeable. And I have no idea how to weigh the relative probability of those worlds.
We are entering nonsense territory here, as the set up implies that someone who can torture ^^^3^^^ people for fun still need something from you and, moreover, needs your free and willing consent.
He clearly very likely wants something from me: the subjective experience of seeing my submissive reaction. Bully psychology 101.
Don’t think in terms of point estimates, think it terms of full distributions (which here will not be symmetric).
I fully agree. However, the full distribution is at least the size of the rectangle defined by my point estimate, so that’s actually a point in my favor.
whose store-of-value functionality is the dominant factor in their price. Gold, silver, land and internet domain names are the only others I can think of that vaguely fit there.
Don’t think land (which is productive and “consumable”—in the sense that you can live on it) fits in here. I am also not sure that silver has much store-of-value role nowadays.
In different times in different societies the store-of-value function was fulfilled by different things. For example, right now empty housing is a major store-of-value in China. US dollar banknotes are a notable store-of-value around the world, e.g. in Russia. Government bonds, especially of reputable governments, are often used as store-of-value.
All in all it’s fairly complicated and context-dependant :-) In the West the predominant store-of-value right now is financial securities (stocks and bonds).
But I wonder why are you focusing solely on the store-of-value function, you don’t think Bitcoin will be valuable as a medium of exchange?
It’s quite possible to be anonymous with BTC
Depends on the resources brought against you, threat model still matters.
the full distribution is at least the size of the rectangle defined by my point estimate
Huh? I don’t understand that. And, by the way, what is your point estimate? A mean? A median? Mode, maybe? :-)
In the West the predominant store-of-value right now is financial securities (stocks and bonds).
Mostly, yes. But gold still has a $7t market cap. But this does open up an interesting argument: that gold is on the whole dying as a store of value and it’s being propped up almost entirely by tradition; in this case central bank gold holdings will probably decline 90%+ over the next century. In this scenario, BTC has no chance to replace gold because there’s no new interest in gold anyway.
However, I would still argue that there is diversification value in BTC if it simultaneously manages to (i) have value increase proportionately to economic growth in the long term (that basically requires maintaining constant salience in a growing society), and (ii) be countercyclical to stocks; if that pattern repeats over two or three business cycles then I could see investment specialists advocating it in place of gold as part of an “all seasons” portfolio (essentially replacing gold here: http://mebfaber.com/2014/10/24/the-all-seasons-portfolio-aka-the-tony-robbins-portfolio/ ).
But I wonder why are you focusing solely on the store-of-value function, you don’t think Bitcoin will be valuable as a medium of exchange?
A couple of reasons:
MoE usage brings much lower valuation prospects. You can see a currency getting to $10 billion simply from people trading it, but the serious $1t+ valuations come from people actually holding it in huge quantities. MV = PQ (where currency value is P^-1); M is constant, so the use cases that push P^-1 high are the ones where V is very low.
With MoE you can make a credible case that it will simply be continually replaced by superior technologies that improve on block time, scalability, anonymity, transaction cost, functionality, etc. Blockstream’s sidechains project potentially allows protocol upgrades to come together with continued BTC use, but that itself is a bet. Also, for MoE people desire price stability much more, so my bets as far as cryptocurrency goes have been on stablecoins ( https://blog.ethereum.org/2014/11/11/search-stable-cryptocurrency/ ). So because of that last point particularly I would place MoE dominance probability at under 5%.
Depends on the resources brought against you, threat model still matters.
Agree. Anonymity depends pretty much completely on threat model in our high-info-inequality society.
Huh? I don’t understand that. And, by the way, what is your point estimate? A mean? A median? Mode, maybe? :-)
$34000 is my “5% chance it will be above this” target (though my discussions here and in that debate have revised me a bit down to 2.5-4%). My mean is around $5000 I suppose, though I include only the 34000 * 0.05 rectangle in my final answer of $1700 as a sort of way of giving myself a safety margin. Median is under $200, mode is $0 :-)
that gold is on the whole dying as a store of value and it’s being propped up almost entirely by tradition
I think this is a reasonable assumption to work under.
there is diversification value in BTC
Maybe—that entirely depends on whether BTC actually has value and that hinges on, as you put it, “maintaining constant salience” which is the real issue.
$34000 is my “5% chance it will be above this” target
Ah, so $34K is your 5% (or 3%) quantile for the distribution, right?
Maybe—that entirely depends on whether BTC actually has value and that hinges on, as you put it, “maintaining constant salience” which is the real issue.
Okay, this helps. So “will BTC be able to (i) maintain constant salience, and (ii) be countercyclical in the long term, and if it does what value will it have?” seems like the object level issue; definitely makes things clearer than some abstract notion of replacing gold.
Ah I think I’ve been misunderstanding you. I was thinking in terms of the probability distribution over BTC’s long-term value, but I think you’re referring to about my probability distribution over the probability that BTC will get to the $34k (or more precisely, my probability distribution over what probability estimate I would have on the topic if I knew more and was wiser). Is that closer to correct?
Well, most any kind of asset has some diversification value. I see no particular reason for BTC to be countercyclical (not to mention that the traditional business cycle of the late XX century seems to be dead at the moment, or at least much transformed) and in any case there’s too little data to tell.
And if you think BTC has some extra special value because it will be {un|low|negatively} correlated to the S&P then you need to compare it to a different reference class.
With respect to the probability distribution, no, you were right the first time—we’re both talking about the probability distribution of the value of 1BTC at some long-term point (and you really should define what does “long-term” mean here, in years, for obviousreasons). I’m not talking about hyper- or meta- distribution of your credence.
Also, Bitcoin has an advantage over gold in terms of security. Guarding a lot of gold in a vault is expensive, but keeping a Brainwallet is not.
Which brings up another hilarious Bitcoin Pascal’s Mugging scenario: “What is the probability that you die, are cryopreserved, and are resurrected at some point in the future. But this future world is a dystopia, and cryoresurrected people are indentured servants of the company that revived them until they are able to pay back their debt. Since most jobs have been automated it is extremely difficult for you to get the money to ever earn your freedom. However, if you had created a Bitcoin brain wallet, you would be able to access it (because the cryoresurrection process is very good), and you would have something of value with which to pay off your debt and live comfortably in the new world”.
Good luck trying to come up with numbers for the probability and value of that scenario. :)
Also, Bitcoin has an advantage over gold in terms of security. Guarding a lot of gold in a vault is expensive, but keeping a Brainwallet is not.
Really? There is a reason why bank valuts have timed emergency locks.
Which brings up another hilarious Bitcoin Pascal’s Mugging scenario: “What is the probability that you die, are cryopreserved, and are resurrected at some point in the future. But this future world is a dystopia, and cryoresurrected people are indentured servants of the company that revived them until they are able to pay back their debt. Since most jobs have been automated it is extremely difficult for you to get the money to ever earn your freedom. However, if you had created a Bitcoin brain wallet, you would be able to access it (because the cryoresurrection process is very good), and you would have something of value with which to pay off your debt and live comfortably in the new world”.
They’ll just extract the brainwallet from your memories using their advanced brain scan technology. Or a rubber hose.
Also, Bitcoin has an advantage over gold in terms of security. Guarding a lot of gold in a vault is expensive, but keeping a Brainwallet is not.
Bitcoin security is much cheaper to implement, but the drawback is that it’s a lot more susceptible to fraud and large-scale theft. Such thefts have occurred in the last few years.
Which brings up another hilarious Bitcoin Pascal’s Mugging scenario: “What is the probability that you die, are cryopreserved, and are resurrected at some point in the future. But this future world is a dystopia, and cryoresurrected people are indentured servants of the company that revived them until they are able to pay back their debt. Since most jobs have been automated it is extremely difficult for you to get the money to ever earn your freedom. However, if you had created a Bitcoin brain wallet, you would be able to access it (because the cryoresurrection process is very good), and you would have something of value with which to pay off your debt and live comfortably in the new world”.
This assumes the future world still values Bitcoin. I’m reminded of a Twilight Zone episode (The Rip Van Winkle Caper) where a group of thieves steal a large amount of gold, then use a form of suspended animation to sleep for 100 years, intending to wake up in a future where no one remembers their crime and they can live in wealth.
Measuring Bitcoin against “gold market cap” is dangerous because gold has uses which Bitcoin cannot replace. For example, traditionally most of the wealth of Indian families (those who have wealth, of course) have been kept as gold, specifically golden jewelry. Bitcoin will not replace that use. Another big advantage of gold is anonymity which Bitcoin will not be able to replicate either.
I don’t think its unreasonable. Bitcoin competes for best-store-of-value status with Gold. Indian families and many others store wealth in Gold, which indicates that Gold has a strong network effect: a large network of people who highly value it as a store of value. For Bitcoin to capture X% of the market of Gold, it would mean that it captured some fraction of that network size.
Bitcoin has one massive advantage over Gold, which is its capacity to be transported quickly anywhere in the world, which makes it possible to use as a convenient means of payment.
Anonymity is debatable, and its also debatable whether it is a positive or negative, but it is quite possible that other blockchain technologies will develop or have already developed better anonymity features, which means that potentially these could be incorporated into Bitcoin.
which indicates that Gold has a strong network effect
Not just that. Being only a store-of-value is a poor functionality set. The Indian gold jewelry doesn’t just sit in a vault—it is worn on big occasions and serves a major status display.
Not just that. Being only a store-of-value is a poor functionality set. The Indian gold jewelry doesn’t just sit in a vault—it is worn on big occasions and serves a major status display.
I would say the property of Bitcoin to be both a store of value and easily transferable anywhere in the world extremely quickly far exceeds the value of Gold to “look pretty when you wear it”.
Also, if by “Bitcoin”, we mean “Bitcoin and/or any future blockchain technology that replaces it” (such as Ethereum or others), then features can be developed using the blockchain technology which would have immense value, such as prediction markets, assets (stocks, etc) tradeable in the blockchain, voting, website naming, smart contracts, escrow, and many others. While also being a store of value, these features would have immensely more value than Gold’s “looks pretty when you wear it”.
You still have to account for the probability of Bitcoin holders seeing the change coming and deciding to modify the Bitcoin codebase to adapt the new desirable features, but still use the Bitcoin ledger (aka current ownership of Bitcoins).
I don’t know how to evaluate the probabilities of these various outcomes happening, however it only costs about 10-20% more to go from ‘buy X bitcoins’ to ‘buy X bitcoins, and also diversify by buying an equivalent percentage stake in all other promising blockchain technologies’.
If you do that you can change the equation from Bitcoin winning and continuing to have value, versus the blockchain technology succeeding and some instance of it continuing to have value.
I’m referring to bitcoin specifically, as I was specifically trying to determine whether or not it’s a good idea to hold BTC right now. I’m obviously more bullish than 5% on “future blockchain technology that replaces it” (such as Ethereum or others)”; if I wasn’t I would not be a full-time member of the industry :)
If you do that you can change the equation from Bitcoin winning and continuing to have value, versus the blockchain technology succeeding and some instance of it continuing to have value.
But then, the question becomes: if you’re bullish in blockchain tech, but not bitcoin, then why not invest exclusively in the other blockchain tech and not bitcoin?
Some random comments:
Theoretically, no, practically, it still is. Humans are humans and the whole history and, um, aura of gold makes a “simple equilibrium flip” not very likely in the near future.
Measuring Bitcoin against “gold market cap” is dangerous because gold has uses which Bitcoin cannot replace. For example, traditionally most of the wealth of Indian families (those who have wealth, of course) have been kept as gold, specifically golden jewelry. Bitcoin will not replace that use. Another big advantage of gold is anonymity which Bitcoin will not be able to replicate either.
Why do you think so?
And what exactly is the reference class that you put Bitcoin into?
Don’t think in terms of point estimates, think it terms of full distributions (which here will not be symmetric).
We are entering nonsense territory here, as the set up implies that someone who can torture ^^^3^^^ people for fun still need something from you and, moreover, needs your free and willing consent. Recall that the original Pascal Mugger is Omega and I doubt that you can incentivize Omega by cooperating or not.
Roughly speaking, assets which have received a major amount of public attention and whose store-of-value functionality is the dominant factor in their price. Gold, silver, land and internet domain names are the only others I can think of that vaguely fit there.
So your argument is that gold has a very very large network effect? Reasonable I suppose, but technology has disrupted similarly entrenched things over the past two decades, so you have to add a lot of fundamental uncertainty.
It’s quite possible to be anonymous with BTC; the $400m MtGox theft at least is pretty good evidence of that. It just takes a lot of skill. Also, with the growth of meatspace surveillance, I would not be surprised if gold became easier to track than BTC over the next few years.
So for every world where the mugger is Omega, there is a world where the mugger is someone incentivizeable. And I have no idea how to weigh the relative probability of those worlds.
He clearly very likely wants something from me: the subjective experience of seeing my submissive reaction. Bully psychology 101.
I fully agree. However, the full distribution is at least the size of the rectangle defined by my point estimate, so that’s actually a point in my favor.
Don’t think land (which is productive and “consumable”—in the sense that you can live on it) fits in here. I am also not sure that silver has much store-of-value role nowadays.
In different times in different societies the store-of-value function was fulfilled by different things. For example, right now empty housing is a major store-of-value in China. US dollar banknotes are a notable store-of-value around the world, e.g. in Russia. Government bonds, especially of reputable governments, are often used as store-of-value.
All in all it’s fairly complicated and context-dependant :-) In the West the predominant store-of-value right now is financial securities (stocks and bonds).
But I wonder why are you focusing solely on the store-of-value function, you don’t think Bitcoin will be valuable as a medium of exchange?
Depends on the resources brought against you, threat model still matters.
Huh? I don’t understand that. And, by the way, what is your point estimate? A mean? A median? Mode, maybe? :-)
Mostly, yes. But gold still has a $7t market cap. But this does open up an interesting argument: that gold is on the whole dying as a store of value and it’s being propped up almost entirely by tradition; in this case central bank gold holdings will probably decline 90%+ over the next century. In this scenario, BTC has no chance to replace gold because there’s no new interest in gold anyway.
However, I would still argue that there is diversification value in BTC if it simultaneously manages to (i) have value increase proportionately to economic growth in the long term (that basically requires maintaining constant salience in a growing society), and (ii) be countercyclical to stocks; if that pattern repeats over two or three business cycles then I could see investment specialists advocating it in place of gold as part of an “all seasons” portfolio (essentially replacing gold here: http://mebfaber.com/2014/10/24/the-all-seasons-portfolio-aka-the-tony-robbins-portfolio/ ).
A couple of reasons:
MoE usage brings much lower valuation prospects. You can see a currency getting to $10 billion simply from people trading it, but the serious $1t+ valuations come from people actually holding it in huge quantities. MV = PQ (where currency value is P^-1); M is constant, so the use cases that push P^-1 high are the ones where V is very low.
With MoE you can make a credible case that it will simply be continually replaced by superior technologies that improve on block time, scalability, anonymity, transaction cost, functionality, etc. Blockstream’s sidechains project potentially allows protocol upgrades to come together with continued BTC use, but that itself is a bet. Also, for MoE people desire price stability much more, so my bets as far as cryptocurrency goes have been on stablecoins ( https://blog.ethereum.org/2014/11/11/search-stable-cryptocurrency/ ). So because of that last point particularly I would place MoE dominance probability at under 5%.
Agree. Anonymity depends pretty much completely on threat model in our high-info-inequality society.
$34000 is my “5% chance it will be above this” target (though my discussions here and in that debate have revised me a bit down to 2.5-4%). My mean is around $5000 I suppose, though I include only the 34000 * 0.05 rectangle in my final answer of $1700 as a sort of way of giving myself a safety margin. Median is under $200, mode is $0 :-)
I think this is a reasonable assumption to work under.
Maybe—that entirely depends on whether BTC actually has value and that hinges on, as you put it, “maintaining constant salience” which is the real issue.
Ah, so $34K is your 5% (or 3%) quantile for the distribution, right?
Okay, this helps. So “will BTC be able to (i) maintain constant salience, and (ii) be countercyclical in the long term, and if it does what value will it have?” seems like the object level issue; definitely makes things clearer than some abstract notion of replacing gold.
Ah I think I’ve been misunderstanding you. I was thinking in terms of the probability distribution over BTC’s long-term value, but I think you’re referring to about my probability distribution over the probability that BTC will get to the $34k (or more precisely, my probability distribution over what probability estimate I would have on the topic if I knew more and was wiser). Is that closer to correct?
Well, most any kind of asset has some diversification value. I see no particular reason for BTC to be countercyclical (not to mention that the traditional business cycle of the late XX century seems to be dead at the moment, or at least much transformed) and in any case there’s too little data to tell.
And if you think BTC has some extra special value because it will be {un|low|negatively} correlated to the S&P then you need to compare it to a different reference class.
With respect to the probability distribution, no, you were right the first time—we’re both talking about the probability distribution of the value of 1BTC at some long-term point (and you really should define what does “long-term” mean here, in years, for obvious reasons). I’m not talking about hyper- or meta- distribution of your credence.
Also, Bitcoin has an advantage over gold in terms of security. Guarding a lot of gold in a vault is expensive, but keeping a Brainwallet is not.
Which brings up another hilarious Bitcoin Pascal’s Mugging scenario: “What is the probability that you die, are cryopreserved, and are resurrected at some point in the future. But this future world is a dystopia, and cryoresurrected people are indentured servants of the company that revived them until they are able to pay back their debt. Since most jobs have been automated it is extremely difficult for you to get the money to ever earn your freedom. However, if you had created a Bitcoin brain wallet, you would be able to access it (because the cryoresurrection process is very good), and you would have something of value with which to pay off your debt and live comfortably in the new world”.
Good luck trying to come up with numbers for the probability and value of that scenario. :)
Really?
There is a reason why bank valuts have timed emergency locks.
They’ll just extract the brainwallet from your memories using their advanced brain scan technology. Or a rubber hose.
The odds for that scenario are approximately equal to the odds that quantum computers are impossible.
Good answer. :)
Bitcoin security is much cheaper to implement, but the drawback is that it’s a lot more susceptible to fraud and large-scale theft. Such thefts have occurred in the last few years.
This assumes the future world still values Bitcoin. I’m reminded of a Twilight Zone episode (The Rip Van Winkle Caper) where a group of thieves steal a large amount of gold, then use a form of suspended animation to sleep for 100 years, intending to wake up in a future where no one remembers their crime and they can live in wealth.
Given how often Bitcoins get stolen I don’t think the practical security of bitcoin is that high.
A brainwallet has the problem that you risk forgetting it.
I don’t think its unreasonable. Bitcoin competes for best-store-of-value status with Gold. Indian families and many others store wealth in Gold, which indicates that Gold has a strong network effect: a large network of people who highly value it as a store of value. For Bitcoin to capture X% of the market of Gold, it would mean that it captured some fraction of that network size.
Bitcoin has one massive advantage over Gold, which is its capacity to be transported quickly anywhere in the world, which makes it possible to use as a convenient means of payment.
Anonymity is debatable, and its also debatable whether it is a positive or negative, but it is quite possible that other blockchain technologies will develop or have already developed better anonymity features, which means that potentially these could be incorporated into Bitcoin.
Not just that. Being only a store-of-value is a poor functionality set. The Indian gold jewelry doesn’t just sit in a vault—it is worn on big occasions and serves a major status display.
I would say the property of Bitcoin to be both a store of value and easily transferable anywhere in the world extremely quickly far exceeds the value of Gold to “look pretty when you wear it”.
Also, if by “Bitcoin”, we mean “Bitcoin and/or any future blockchain technology that replaces it” (such as Ethereum or others), then features can be developed using the blockchain technology which would have immense value, such as prediction markets, assets (stocks, etc) tradeable in the blockchain, voting, website naming, smart contracts, escrow, and many others. While also being a store of value, these features would have immensely more value than Gold’s “looks pretty when you wear it”.
To you, maybe, to an Indian family, not likely.
No, we do not, because at issue is the future value of the current investment in Bitcoin.
You still have to account for the probability of Bitcoin holders seeing the change coming and deciding to modify the Bitcoin codebase to adapt the new desirable features, but still use the Bitcoin ledger (aka current ownership of Bitcoins).
I don’t know how to evaluate the probabilities of these various outcomes happening, however it only costs about 10-20% more to go from ‘buy X bitcoins’ to ‘buy X bitcoins, and also diversify by buying an equivalent percentage stake in all other promising blockchain technologies’.
If you do that you can change the equation from Bitcoin winning and continuing to have value, versus the blockchain technology succeeding and some instance of it continuing to have value.
I’m referring to bitcoin specifically, as I was specifically trying to determine whether or not it’s a good idea to hold BTC right now. I’m obviously more bullish than 5% on “future blockchain technology that replaces it” (such as Ethereum or others)”; if I wasn’t I would not be a full-time member of the industry :)
But then, the question becomes: if you’re bullish in blockchain tech, but not bitcoin, then why not invest exclusively in the other blockchain tech and not bitcoin?