Bitcoin will bite the dust.

• Author: Dowd, Kevin
• Position: Essay

Bitcoin is the most radical innovation in the monetary space for a very long time. It is an entirely private monetary system that runs itself and does not depend on trust in any central authority to honor its promises. Instead, it relies on trust in the Bitcoin community or network that verifies transactions and maintains the integrity of the system. This system of distributed trust creates bitcoins and produces an automatic, tamper-proof bitcoin money supply process. (1) As such, it avoids the dangers of discretionary monetary policy--namely, quantitative easing, manipulated interest rates, and the need to rely on wise men or women to withstand political pressure or successfully forecast the future. Indeed, under Bitcoin there is no monetary policy at all. There is just an automatic monetary rule dictated by the Bitcoin protocol designed in 2009 by an anonymous programmer using the alias Satoshi Nakamoto.

Bitcoin has been widely hailed as a success and has won a substantial following. Unfortunately, the underlying economics of Bitcoin mean that it is unsustainable and in all likelihood will be remembered as a failed experiment--at best a pointer to some superior successor.

A first-pass intuition into Bitcoin can be obtained from a comparison with the stone money in Milton Friedman's (1992) case study, "The Island of Stone Money." In this story, the people of the island of Yap in Micronesia used as money large round limestone disks transported from the nearby island of Palau. These were too heavy to conveniently move around, so they were placed in prominent places. When ownership was to be transferred (e.g., as part of a dowry, inheritance, or ransom payment), the current owner would publicly announce the change in ownership but the stone would typically remain where it was and the islanders would maintain a collective memory of the ownership history of the stones. This collective memory ensured that there was no dispute over who owned which stones. Similarly, in Bitcoin, the record of all transactions, the "blockchain," is also public knowledge and is regarded as die definitive record of who owns which bitcoins. Both the stone money and Bitcoin share a critical feature that is highly unusual for a monetary system: both systems operate via a decentralized collective memory.

On February 11, 2009, Nakamoto gave an explanation of the thinking behind Bitcoin in an e-mail announcing its launch: "The root problem with conventional currency is all the trust that is required to make it work. The central bank must be trusted not to debase the currency, but the history of fiat currencies is full of breaches of that trust.... With e-currency based on cryptographic proof, without die need to trust a third-party middleman, money can be secure and transactions complete." Cryptocurrencies, however, face the problem of "double-spending." As Nakamoto notes, "Any owner could try to re-spend an already spent coin by [digitally] signing it again to another owner. The usual solution is for a trusted company with a central database to check for double-spending, but that just gets back to the trust model.... Bitcoin's solution is to use a peer-to-peer network to check for double-spending." Consequently, "the result is a distributed system with no single point of failure." (2)

Kevin Dowd is Professor of Finance and Economics at Durham University in the United Kingdom and a partner at Cobden Partners. Martin Hutchinson is a journalist and author of the Bear's Lair column (www.tbwns.com/category/the-bears-lair). The authors thank Ferdinando Ametrano, Gavin Andresen, Raadhiyah Anees, Steve Baker MP, Roger Brown, Dave Campbell, Akin Fernandez, Dominic Frisby, Jim Harper, Doug Jackson, Gordon Kerr, Jim Rapp, Eric Samieski, Lawrence H. White and Basil Zafiriou for much helpful feedback. We note that several of our readers have expressed serious reservations about our analysis and conclusions. Any remaining mistakes are the authors' own.

The fact that Bitcoin has no single point of failure is highly significant: it means that it cannot be brought down by knocking out any particular individual or organization. (3) It can only be brought down by knocking out the whole network or one of the underlying building blocks on which the network depends. (4) It can and does operate outside of government control: Bitcoin is a dream come true for anarchists, criminals, and proponents of private money.

Despite its success, the Bitcoin system is unsustainable due to a design flaw at the very heart of the system. The problem is that Bitcoin requires competition on the part of "bitcoin miners" who validate transactions blocks, but this competition is unsustainable in the long run because of economies of scale in the mining industry. Indeed, these economies of scale are so large that the bitcoin mining industry is a natural monopoly. Furthermore, there are signs that competition in this industry is already breaking down. Once that happens, the system will no longer be able to function as it hitherto has. Its key attractions (decentralization, absence of a single point of failure, and anonymity) will disappear; there will no longer be any reason for users to stay with it; and the system will collapse.

How Bitcoin Works

Let's start by explaining how Bitcoin works. (5) The first point to appreciate is that the system is based on the use of Public-Key Encryption (PKE) used to digitally authendcate a signature. PKE is the basis of Internet financial security and is widely used to protect sensitive financial information. Each individual user has both a public key, known to everyone, and a private key, known only to it. When Alice transfers a bitcoin to Bob, she adds Bob's public key to the coin and digitally signs the coin using her private key. When Bob receives the bitcoin, he is able to verily that only someone with Alice's private key could have sent it to him: he can then be confident that the bitcoin came from her. (6)

There is, however, a problem: how to prevent Alice double-spending the same bitcoin. So let's look at the general problem of Alice making a payment to Bob: there needs to be a transfer and the transfer needs to be validated without allowing Alice to double-spend. With conventional cash, the transfer is easy: Alice hands over a $1 note to Bob and Bob just needs to check that the note is not a fake. Alice is unable to spend the same dollar note twice: having handed over the note to Bob, she can't then spend it again with Charlie. With a conventional bank check, Alice sends the check to Bob, and the central authority (the bank) verifies the transfer and adjusts Alice's account to prevent her spending the same deposit money twice. Under the Bitcoin system, on the other hand, there is no central authority to validate the transfer and prevent Alice double-spending the same bitcoin. Instead, these tasks are performed by the network itself--that is, by the community of users.

To get the system to perform these tasks we then have to ensure that there is a reliable verification process and we have to incentivize that process. At first sight, the most obvious verification approach would be a voting system: other users could vote on whether the transfer was valid. If the majority agreed, then the transfer would be deemed valid. Unfortunately, this approach does not work. Remember that the users are anonymous and till we really have is user IDs. So Alice can set up a billion different IDs and take over the system: she has one bitcoin, which she then spends with Bob, Charlie, David, and so on, and she uses her billion votes to approve each transaction; she sock-puppets the system by overwhelming it with spam votes, known in the trade as a "Sybil attack." The way round this problem is to make each "vote" costly, and this costliness is achieved by imposing a proof-of-work requirement. In the Bitcoin system, the proof of work is to demonstrate that the user/voter has expended valuable CPU power to solve a difficult mathematical problem. To oversimplify slightly, the veracity of any transaction is then determined by the majority of CPU-adjusted votes. It is now no longer economic for Alice to spam the system.

In addition to solving the double-spending problem, Bitcoin must address the incentive problem--namely, the need to incentivize the network to validate any transaction. Users need to be given some reward for spending their valuable CPU power to validate other people's transactions. Under the Bitcoin system, this reward comes from a combination of newly minted bitcoins and a transaction fee. Leaving aside the fee, the analogy here is with a gold miner mining for gold and occasionally finding it. Similarly, the Bitcoin validator goes searching for bitcoins and occasionally gets lucky. For this reason, the process of validating the bitcoin transactions blocks is usually referred to as "bitcoin mining." In the very early Bitcoin system, a successful bitcoin miner was rewarded with 50 bitcoins for each block validated. However, after every 210,000 validated blocks the rewards halve. This halving has occurred once so far, in November 2012, so the current reward for validating a block is 25 bitcoins. This halving will continue roughly every four years so the production of new bitcoins over time will asymptotically go to zero. It is anticipated that the rate of bitcoin production per block will fall to less than 1 satoshi, or 0.00000001 BTC, by 2141. By that point, the total amount of bitcoin produced will be just short of 21 million. Note, too, that because the time to validate each block is fairly predictable (i.e., about 10 minutes), and even this randomness tends to cancel out over the long run thanks to the law of large numbers, then the rate of bitcoin production, and the hence the total mined by any future time, is highly predictable.

The actual amount of bitcoin available to use, however, is unknown, because of the risk of permanent loss...