If you believe Bitcoin has the potential to replace traditional global financial systems, a new economic analysis is here to rain on your parade.
The discussion of digital money thus far has been dominated by libertarians and computer geeks, but the massive popularity of crypto-tokens has gotten the attention of academics like the University of Chicago’s Eric Budish. In a new paper (PDF) Budish concludes, based on a model of Bitcoin’s incentive system, that there are “intrinsic economic limits to how economically important it can become.”
Some “Bitcoin maximalists” say that the currency is a lot like gold—it works as a store of value, even if it’s not very efficient as a true currency. But if Bitcoin got anywhere close to gold’s value, Burdish argues, people would attack its network for profit.
Before we dive into the argument, a little context: Bitcoin’s market capitalization over the last year or so has oscillated between $100 and $200 billion. Gold stock is worth about $7.5 trilllion. So yeah, in these terms Bitcoin is nowhere close to being “economically important.”
And according to Budish, it never will be. That’s because, if it ever gets too large, the genius of Bitcoin’s design would be its undoing.
Bitcoin’s security arises from a competition between members of the network called “miners.” Each miner is in pursuit of chances to add new transactions to the blockchain and earn bitcoins in return. To play, they use large amounts of computing power in a race to solve a complicated math problem. An attacker couldn’t defeat this system unless they coordinated so much computing power that they could overwhelm the network and manipulate the record of transactions in such a way that they could spend the same bitcoins repeatedly. This is called a “majority attack” and it’s Bitcoin’s biggest weakness, but at the moment mining coins is more profitable than it is trying to overthrow the network would be—so that’s what people do, and the network stays safe. (See “How secure is a blockchain really?”)
But this protection is very expensive, writes Budish (the Bitcoin network uses about as much power as Ireland to run). And although Bitcoin’s value could theoretically increase almost without end, the blockchain’s security can only increase linearly, as more mining power is added to the network. That’s unlike other forms of security, like cryptography used in the traditional financial system, which, like adding a lock to a door, increases security much more than the cost of deploying it.
The cost of running the Bitcoin blockchain today is on the order of $100,000 per 10 minutes, whereas the cost of attacking the system is in the neighborhood of $1.5 to $2 billion, according to Budish’s calculations. A big reason it’s so expensive is that Bitcoin mining is currently dominated by chips that are purpose-built for mining and can’t be redeployed to perform other tasks. An attack could also drastically lower the value of Bitcoin—and in turn, the attacker’s own holdings—but this wouldn’t deter someone who was simply looking to sabotage or destroy Bitcoin.
Though the new paper has gotten a fair amount of praise from other economists, some cryptocurrency enthusiasts have been dismissive. The conclusion that Bitcoin may not be sustainable because it might become too expensive to deter sabotage “may be true,” says Ari Paul, co-founder of BlockTower Capital, but it has long been a topic of debate in popular online forums. The paper “adds no new data or logic to the debate,” he says.
Joshua Gans, an economist at the University of Toronto, argues that those online discussions lacked scientific rigor. Economists are just beginning to discuss these issues, and the research community will benefit from Budish’s “rigorous work of putting this all together,” he says. “It is that kind of approach that leads to better science.” | 
