At the 32nd Annual Monetary Conference on November 14, 2014, I presented my paper "Bitcoin Will Bite the Dust", which was co-written with Martin Hutchinson. Below is the transcript of my talk and PowerPoint slides. I welcome your comments. A video of the panel is available on Cato's website (my talk starts at minute mark 7:30).
The PowerPoint slides to accompany the talk are here.
[Slide 1]. Good morning everyone. Bitcoin is the most radical innovation in the monetary space for a long time: an entirely private system that runs itself and does not depend on trust in any central authority. Instead, it relies on distributed trust – trust in the network – to maintain the integrity of the system. We can well understand the attractions of such a system – a tamper proof money supply, no monetary discretion, no QE, no central bank.
There is only one small problem: despite its success to date, Bitcoin is not sustainable. This means it will collapse. What cannot go on, will stop.
Let’s go back to basics. As a first pass, compare Bitcoin to the stone money in Milton Friedman’s case study. In this story, the people of the island of Yap in Micronesia used large round limestone disks as money. These were too heavy to move, so when ownership was to be transferred, the owner would publicly announce the change in ownership. The stone would remain where it was and the islanders would maintain a collective memory of the ownership history of the stones.
Similarly, in Bitcoin, the record of all transactions, the blockchain, is also public knowledge. Both the stone money and Bitcoin share a critical feature: both operate via a decentralized collective memory.
[Slide 2]. Bitcoin is a type of e-cash system in which there is no central body to authorize transactions; instead, these tasks are carried out collectively by the network. The network verifies transactions through competition between individual bitcoin ‘miners’ seeking rewards in the form of new bitcoins. It is this competition that maintains the integrity of the system. This takes us to Bitcoin’s value proposition.
[Slide 3]. The first point is that the system does not depend on trust on any one body to keep its promises; instead, the only trust required is distributed trust. The second point is that it has no single point of failure: it cannot be brought down by knocking out any particular entity. The third point is a high degree of anonymity. This has enabled some bitcoiners to operate outside of government control. Bitcoin is a dream come true for anarchists, criminals and proponents of private money. The fourth pillar of the value proposition is security against tampering: this comes from the incentive-compatibility built into the system. Underlying that, security comes from the Bitcoin protocol, the Constitution of the system. These features ensure that players play by the rules and that bitcoins are not over-issued.
[Slide 4]. Unfortunately, there is a fundamental contradiction at the heart of the system. The problem is that it requires atomistic competition on the part of the miners who validate transactions blocks. However, the mining industry is characterized by large economies of scale. In fact, these economies of scale are so large that the industry is a natural monopoly. The problem is that atomistic competition and a natural monopoly are inconsistent: the inbuilt centralization tendencies of the natural monopoly mean that mining firms will become bigger and bigger – and eventually produce an actual monopoly.
There are not one but two reasons to see mining as a natural monopoly. The first is based on risk aversion. If two miners merge their operations, they get the same expected return as if they mined on their own, but they obtain a return with a higher probability. If miners are risk averse, they are better off by pooling and sharing their profits. But if it makes sense for any two individual miners to pool, it makes sense for any two groups of miners to pool. The limiting case is then one big mining pool, a monopoly.
The second reason for a natural monopoly is even stronger: the negative externalities of competitive mining. The expected marginal benefit from mining depends on the amount of hash power expected by an individual miner, but the difficulty of mining depends on the hash power expended across the network. Individual miners do not take into account the negative cost externalities that their own activities impose on other miners. We then get an equilibrium in which excessive resources are devoted to mining activities: there is excessive use of bandwidth, excessive use of energy and excessive investment in computing resources. In the early days, a home PC could produce hundreds of bitcoins a day; now, a state of the art mining machine can expect to mine only a fraction of a bitcoin a day. We estimate that the energy power devoted to bitcoin mining has increased by a factor of at least 10 billion. Most of this is pure waste as the system could be maintained on a single server. A single operator could avoid most of this waste.
The implications of these centralizing tendencies are totally destructive of the Bitcoin system. They destroy every single element of its value proposition: one by one, the dominos fall down.
[Slide 5]. The first casualty is decentralized trust. Once the individual miners coalesce into a dominant player, then that entity has control over the system: it decides which transactions are to be deemed valid, and which are not. We then have to trust that entity not to abuse its position and are back to the trust model that Bitcoin had tried to escape from.
Going back to our island of stone money, imagine if everyone woke up one morning unable to remember who owned which stones. However, one individual still claims that he can remember and helpfully offers to remember for everyone else. One wonders how well that would work!
By this point, the dominant player has taken control over the system: it becomes the monarch – albeit, a constitutional monarch still constrained by the Bitcoin protocol. Once that dominant player emerges, it also becomes a point of failure of the whole system: one can bring down the system by taking him out. One could imagine Uncle Sam being very interested: if he wanted, he could now take Bitcoin down and stop all that money flowing to the bad guys he is after.
The next casualty is anonymity. A dominant player cannot possibly operate in a clandestine manner beyond the knowledge of law enforcement. And if it cannot operate anonymously, then it cannot escape government regulation. Anonymity would then disappear. The likelihood is that the government would destroy anonymity at a stroke by requiring that the dominant player insist that users register themselves by providing photo ID, social security numbers and proof of address.
It would also become clear that the system no longer assures incentive compatibility. In fact, it never did: it’s just that the system’s incentive compatibility weaknesses took time to become clear.
The last domino to fall would be the Bitcoin protocol. The protocol no longer provides any discipline on the system, because the dominant player can rewrite it at will. At some point the temptation to tamper with it would be too much to resist. Just a like a modern central bank, it would start throwing out bits of the protocol it didn’t like – like the bits that constrain over-issue of bitcoins. The Bitcoin monarchy would then become an absolute monarchy - assuming that there was anyone else left in the system by then.
The question crying out for an answer is why users of bitcoin would continue to have any confidence in the system when every single element of its value proposition had been kicked down. The obvious answer: they wouldn’t.
Remember also that the willingness of any individual to accept bitcoin is entirely dependent on his or her confidence that other people will continue to accept it. There is nothing in the system to anchor the value of bitcoins because bitcoins have no alternative use-value. They are not like gold or tulips.
Nor is there any rational reason to trust in the dominant player to behave itself. Trust comes from credible assurances – it comes from credible precommitment, a willingness to post performance bonds and to submit to account. There is simply no way that a shadowy dominant mining pool can provide such assurances. I doubt it would be willing to anyway.
The most prominent mining pool is GHash.IO. Here is a snapshot from its homepage [Slide 6].
The page announces that GHash is the number 1 mining pool, is trusted by 300k users, and dates all the way back to late 2013(!). We don’t know for sure who is behind it or where it is based. What we do know is that it has a logo that looks like the hammer and sickle and it has a bad rep. It pointedly refuses to adhere to the principles of high-level Bitcoin idealism that the other players adhere to. It has also been associated with a double-spend attack on a gambling website last year. Very reassuring.
Now this might just be coincidence: GHash also shares its name with a character in the film Ghostbusters. Here is a picture [Slide 7].
Cute critter, eh? In the film, Ghash is a power-obsessed poltergeist who pulls other ghosts into a massive mouth in his torso. Once swallowed up, they are drained of their powers until there is nothing left. Meanwhile, Ghash gets bigger and more powerful. By the time the ghostbusters encounter him, he had become too powerful to control: he was able to shoot beams from his eyes, pull up floorboards, disarm the ghostbusters and throw them around at will. Perhaps GHash is a spectral entity in more ways than one!
John Pierpont Morgan once said that the essence of banking is character. Someone I do not trust would not get any money from me on all the bonds in Christendom, he said. We don’t see much of that character here! If you really trust such an outfit with your wealth, we have a bridge to sell you. In any case, there is no reason to want to trust such an entity when you can use reputable systems such as PayPal instead.
[Slide 8].Anyway, return to the main storyline: The whole Bitcoin system eventually becomes a house of cards: there is nothing within the system to maintain confidence in the system, and anything – a scandal, a government attack, whatever – could trigger a loss of confidence leading to a run that brings down the entire system. The rational decision is to sell before that happens. If enough individuals think the same way – and why shouldn’t they? – their expectations will become self-fulfilling: there will be a stampede for the exit, the price of bitcoin will drop to its intrinsic value – zero – and the system will collapse. Only question is when. With the specter of GHash hovering over the system, our guess is soon.
Now I dare say our message is a disappointment to Bitcoiners. I share that disappointment: it would have been great if Bitcoin could displace government money. However, Bitcoin is an experiment, most experiments fail – and Bitcoin is another failed experiment. I don’t wish it so, but that is the way it is. We make this prediction before the event: if we are wrong, we will eat humble pie afterwards. But we don’t think so.
There is also the Bitcoiner lunatic fringe. Their response to even the mildest criticism is to foam at the mouth and hurl abuse at wicked Disbelievers. To them we say: OH DO GROW UP! And if you won’t listen to us, take Voltaire’s advice: “To succeed in life it is not enough to be stupid. You must also have good manners.”
In the meantime, Bitcoin is a sell. Thank you.