The Economic Limits of Bitcoin and the Blockchain

From Eric Budish at the Booth School of Business at Chicago:

The amount of computational power devoted to anonymous, decentralized blockchains such as Bitcoin’s must simultaneously satisfy two conditions in equilibrium: (1) a zero-profit condition among miners, who engage in a rent-seeking competition for the prize associated with adding the next block to the chain; and (2) an incentive compatibility condition on the system’s vulnerability to a “majority attack”, namely that the computational costs of such an attack must exceed the benefits. Together, these two equations imply that (3) the recurring, “flow”, payments to miners for running the blockchain must be large relative to the one-off, “stock”, benefits of attacking it. This is very expensive! The constraint is softer (i.e., stock versus stock) if both (i) the mining technology used to run the blockchain is both scarce and non-repurposable, and (ii) any majority attack is a “sabotage” in that it causes a collapse in the economic value of the blockchain; however, reliance on non-repurposable technology for security and vulnerability to sabotage each raise their own concerns, and point to specific collapse scenarios. In particular, the model suggests that Bitcoin would be majority attacked if it became sufficiently economically important — e.g., if it became a “store of value” akin to gold — which suggests that there are intrinsic economic limits to how economically important it can become in the first place.

I like the framework of this paper, though I wonder if there shouldn’t be more on the coordination costs of mounting a “double spending” attack, namely how exactly the returns from the attack should be divided.  Perhaps the most positive scenario for Bitcoin is if those coordination costs rise with the returns to the attack itself, in which case a much higher market value for Bitcoin still might be stable.


What would Rene Girard do? He is more important than Adam Smith in understanding capitalism. Facebook is mimetic desire, squared. Peter Thiel recognized Facebook for what it is, and he is a very wealthy man as a result. No, it didn't take spread sheets and economic theory to make Thiel a billionaire, it took an appreciation of what motivates human behavior; Thiel has Girard to thank for his good fortune. That Thiel has invested in Bitcoin mining and blockchain should give one pause before viewing blockchain through the lens of economic theory and the rational investor. What might give owners of capital pause before making any investment is the possibility (likelihood?) that, with an ever present media, mimetic desire will accelerate, causing today's winning investment to be tomorrow's losing investment. Short-termism may take on an entirely new meaning.

Suppose a conspiracy of miners succeed at a history revision that assigns a significant fraction of the token base to their accounts. As soon as this is known, there's a significant chance that the value of the tokens as exchanged for fiat currencies approaches zero. To succeed, the conspiracy must revise history while simultaneously retaining a broader community's trust that the token base is a worthwhile investment asset.

The conspiracy can borrow tokens to buy real assets before doing the history revision.

You are right that a significant coalition of miners could double spend transactions in the latest block, but you overstate how easy it is to revise historical transactions.

The miners not only have to re-do the proof of work for each block going back, they simultaneously have to do this while also outcompeting any miners that are working on just the newest block.

This means if 90% of the mining network wishes to change a historical transaction, they would *maybe* be able to revise the history of the past 8 7-8 blocks while still on a probabilistic basis winning the race for the latest block.
If any other miner gets the latest block while the attacking coalition are using their hashing power on re-doing the work on the previous blocks, all the coalition's work is wasted and they have to start all over again.

As a user, this means even if we are transacting in a hostile miner environment, the solution is just to wait for more block confirmations before feeling secure your transaction is irreversible.

Compare this to how current finances work.

The coalition of miners has a gazillion dollars invested in Bitcoin mining hardware. Once they run their history revision attack and steal a bunch of Bitcoin, the result is that everyone stops trusting the Bitcoin blockchain. I suspect Bitcoin would die, other PoW blockchains where the Bitcoin mining hardware could be repurposed to attack them would collapse in value, and the miners would lose all their existing investment in mining hardware.

Coalitions of miners could do other bad things (censoring transactions) without necessarily destroying their existing investment, but revising history to steal money by double-spending probably isn't workable unless you've decided that your Bitcoin mining hardware is about to lose all its value.

So, here's a fun game-theoretic thought: It's 2020, and the world seems to be moving onto ethereum or something, and Bitcoin is seen as less and less an attractive place to invest your money into mining hardware. The existing miners see the end coming, and have an incentive to consider trying to do a mass theft to get $X right away, rather than getting their dwindling stream of ongoing mining revenue that gives a present value of $Y. Once $X>Y, a coalition can form that will rationally want to do the double-spending attack and wreck the network.

The establishment of a BTC futures market with settlement in dollars greatly reduces the coordination issue and largely addresses "and then what" question raised by a couple commenters concerning the resulting value of coins held by attackers after their attack causes a general loss in confidence in the coin. I have always had a sneaking a suspicion this was one of the possible motivations for the consortium that forked Bitcoin Cash from Bitcoin.

Blockchain is the latest free lunch economics shiny object the free lunch economists are chasing in the efforts to eliminate costs.

Costs are almost always labor costs.

The only other cost is profits from rent seeking and monopoly restriction of supply of production and capital. But free lunch economists increasing profit to 100% of revenue to be virtue, eliminating all labor costs.

Until economists go back treating money and it's substitutes as proxy for labor, for work done by people, economists are chasing the perpetual motion machine of engineering, the free lunch.


Costs equal benefits. If you want high benefits, you must incur high costs. Zero sum.

Blockchain is less of an innovation than double entry bookkeeping.

We now know that decentralised mining is not economically feasible; blockchains either have to be abandoned or we have to figure out how to create the economic incentives, for those with the financial backing to build something that can control 51% of mining resources, to 'do the right thing'.

Not sure if that link works as intended:

The latest proposal from one of the developers is a mining protocol that keeps the advantage of mining pools with shared payouts but allows the individual miner to create their own block template / they can choose which transactions they include in their proposed block (previously this was done by the central mining pool operator)

Since you can now short bitcoin on the futures market, the incentive to do something whose sole effect is to reduce its value is very high. A few commentators note that some kind of attacks are unlikely because the miners will lose the value of their own bitcoin but this consideration is not present if you can short.

this is 2018, no one cares about bitcoin anymore

Comments for this post are closed