This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

Bitcoin Will Bite the Dust

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.

  • lb

    Hi Kevin,

    I'm not a lunatic fringe type. I own a few dollars worth of bitcoin and that's it. Your points mostly make sense, and apply to most systems where trust is involved in some way.

    I think you're far overstating the case for the certainty of bitcoin failing. Your scenario could certainly play out, but it's not a certainty. What's your time frame on this happening? Would you care to make a long bet (


    Lucas (Computer Science person, not a bitcoin enthusiast).

  • Well done! There is a reason gold/silver have stood the test of millennia and thousands of competing systems. E-money will work only when there is competition between several of them, as in free banking.

  • A fee-based system, rather than diminishing mining returns, would be all that is necessary for a distributed competitive network to be sustainable just as with any other market. But isn't that what one expects to happen once mining returns go to zero?

    Even if GHash mobilizes the fee-less network, once mining returns reach zero, it needs fees to continue to be profitable. But as its fees rise, independent players have more incentive to enter the market, just as one sees with any free market monopoly.

  • barbierir

    I agree that the Proof-of-Work system is fundamentally flawed because it leds to centralization, in contrast to the very reasons Bitcoin was conceived for. Many bitcoiners seem blind to this matter of fact, in the past I had some useless discussions but until now I haven't seen any reverse in this trend.
    Yet there are other systems like many Proof-of-Stake variants and the recent Proof-of-Importance that, IF they prove to work, don't suffer from the same drawbacks. In theory Bitcoin could change the protocol and adopt one of such methods, but any change depends on the miners themselves, so it is very unlikely that they would agree to a change dentrimental to themselves.
    Personally speaking I changed all my Btc to Nxt some months ago for the very reasons expounded above, Nxt is based on a Proof-of-Stake variant and it is more a general-purpose cripto-financial-platform than a simple cripto-currency. The client itself includes an asset exchange with the highest volume and number of assets in the criptoworld. I suggest to have a close look at it:

    • jwputrl

      No: miners depend exclusively on the bitcoin price for their revenues, so they will not risk centralisation as they know this will affect the price. This is exactly what we observe – pools organising at the largest scale which doesn't over-centralise.

  • freakingjesus

    Well, in practice Ghash has taken more then 51% of the network before. When that happened the price of BTC dropped along with Ghash's mining profits so they throttled their miners. Then Ghash was DDoS attacked and people started to look toward altcoins in the system that were designed to discourage a mining monopoly. A new type of decentralized mining pool, called P2Pool, emerged around this time.

    If P2Pool gained 51% of the network the issue of a mining monopoly would be solved.

    P2Pool is one of many solutions to a mining monopoly. The short term problem is that the risk of switching to P2Pool is currently too great for most miners because the payout from the pool is less consistent than with larger mining pools. If a mining monopoly presents itself again miners will again look toward switching to a decentralized mining pool to protect their own mining investments.

    Mining monopolies are different from typical monopolies in that the economies of scale only work until a monopoly takes hold. When the monopoly takes hold people lose faith in the system, as you predict, but miners are incentivised to correct the problem before their mining hardware is worthless from destroying confidence in the system. There are theoretical fixes to this problem, such as P2Pool, that have not yet been implemented but will be implemented long before confidence in the entire system breaks down.

  • keanupromethium

    Your thought experiment regarding the monopolization of mining in Bitcoin is not new and has been debated for a while now in the Bitcoin space.

    If we play the thought experiment out, firstly, the attainment of a monopoly will be very difficult due to the fact that over half (much more than half) of the hash power that is used by these large mining companies is actually owned by the public. If history shows us anything, about 30-40% of the public hashes remove themselves from large mining pools when the pool approaches 50%.

    Secondly, if ever a monopoly is gained, this is not a zero sum game; the monopolist has invested millions of dollars into their own equipment, and collects millions from fees for the people pointing their respective miners towards the monopolist's pool. Ontop of this, they have extremely large monthly obligations (electricity etc) that must be upheld to survive in the space.

    So if a monopolist gets to 50%, unless its motive was non-profitable, would not try and pull off a double spend. The spend would be noticed immediately and the company called out and soon out of business.

    Thought experiments are fun, and they have been done to death regarding this topic. Bitcoin is like the internet and stands on the same pillars. Many have declared that the internet "can never work because how the hell are you gonna find anything" as Andreas Antonopolous so eloquently stated at the Canadian Fed hearing; "This is not a problem, its an opportunity"

    Bitcoin mining presents many opportunities to innovate around the problem that seems impossible today. Wikipedia should never have worked. The internet should never have worked. Internet telephony can never work. Internet video can never work, the internet is too slow. We will never bank on the web its too insecure.

    It did, it did, it did, and as far as Bitcoin; it does work. And it will work.

    As an engineer who has been in banking, oil and gas and currently game development I would like to think that I have a "whos who" list in this domain (IT), and theres one thing i can promise you. The smartest cryptographers, software developers, software architects and other information technology specialists all have vetted this software, and it works. We were all skeptics at first. The informed skeptic is doesnt evolve into an investor, they become a believer.

    And when the smartest people around you are not sitting silent with their shares of Apple but evangelizing their first Bitcoin; then something is happening thhat is very special.

    Great read and I hope to see you donate more to this space!


  • jwputrl

    There is a simple – but devastating – flaw in this argument. Bitcoin miners depend entirely on the value of bitcoin for their revenues. They would do nothing which even looked like an attempt to dominate mining, if it had a negative effect on the price.

    Certainly there are economies of scale. But you would expect rational, profit-maximising mining pools to organise in such a way as to trade off those economies with the imperative of maintaining the integrity of bitcoin. This is exactly what we observe.

  • xdigital

    Can you give us a timeline? 1 year/5 years/10 years/100 years.
    Without a timeline, your prediction is not a valid prediction.
    Even with gold or silver, Everything will bite the dust soon or later.

  • hun8573ec

    Hello, there are two suppositions you make that are incorrect. You have built your entire argument on them but unfortunately the foundation is faulty. Here, you have a unilateral voice, but people usually discuss these things in forums where such suppositions can be pointed out as inaccurate before you masquerade as an authority on the matter.

    The first supposition is the helplessness of the bitcoin network to a mining monopoly. This is false. While miners are incentivized to merge their operations, their influence on the network is immediate and known by all. If a miner gained control of the network, and this was against the other participant's wishes, then the other participants could easily fork the protocol away from that miner using a hashing algorithm that is incompatible with their hardware. Specialized machines have been made to process the SHA256 algorithm, but the bitcoin network could maintain security by an immediate and abrupt switch to X11 or a mixture of algorithms and your personal PC could mine bitcoin again, disrupting the mining monopoly.

    Secondly, you overstate the influence a dominate monopolistic miner has if they did have control over the network. If a dominate miner attempted a double spend attack this would be known by the whole network immediately. There have been no successful double spend attacks in the bitcoin network. None, ever. Miners have had nearly 100% of the network before, and many many times over 51%, although not in recent years. A mining monopoly has the potential to temporarily disrupt transactions as people stop transacting as the network reaches consensus.

    The rest of your argument is based on the idea of a miner attacking the "constitution" of bitcoin. And since this is all false because your suppositions are false, there is no need to address those.

    Now, to McKinney, there is competition between hundreds of "E-money" platforms, that compete directly against bitcoin and use the same concepts as bitcoin.

    Now, VikingVista, mining is unprofitable for many until a bitcoin unit rises in value. Due to the historical probability that this occurs, people are willing to mine at a loss. Mining returns are negative right now. Bitcoin is a 'fee based system', and the reward for mining diminishes with the intent of being replaced by the fees. This thing was thought of, and it works.

    • "the reward for mining diminishes with the intent of being replaced by the fees"

      Makes sense. The notion that some market naturally tends toward monopoly is a fundamental marxian flaw, not a trap that I would expect a free market economist to fall into. Markets tend to find a way. As long as the market is sufficiently free, it is very hard from economic principles, to imagine a market-destroying or particularly abusive monopoly from evolving for any significant length of time. As long as no authority can employ sufficient coercive force against the bitcoin market, the bitcoin market will be sufficiently free.

      That's not to say bitcoin will always have a market. I just don't see its demise coming from malignant monopolization.

    • hun8573ec: "…there is competition between hundreds of "E-money" platforms, that compete directly against bitcoin and use the same concepts as bitcoin."

      That's not the kind of competition that takes place in free banking. It needs competition such that when one producer of money expands the supply the expansion itself hurts the producer and takes away his ability to produce more. The idea in free banking is to limit the supply of new money. I'm not a bitcoin expert but from what I understand its technology that limits the supply of bitcoins, not competition.

      Seems to me that bitcoin is the domain of programming experts, but it will never become widely used because for most people it is a black box.

  • pelagio

    Given you thoughts, you should go and short bitcoin with all your savings. Would you? Why haven't you?

    • Because no one will take the long end of this bet.

  • Chuck Moulton
    • George Selgin

      Chuck, having scrolled through the reddit link you supply, I'm now convinced that Bitcoin is definitely vulnerable to a "51%" threat, the threat being, not what Kevin worries about, but that no currency system can be expected to survive for very long when 51% or more of its most ardent proponents behave like jerks.

  • btcvega

    "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."

    This makes no sense. Expected return is the same as the probability of obtaining a return times the amount of return. Remember that a new block is created every 10 minutes on average, so a large enough pool will always obtain a return with almost complete certainty over a period of a few hours.