Blockchain Myths #1 — Someone will always pay to run the network
As we’ve introduced Saito and the ideas behind it to blockchain developers, venture capitalists, and attendees at various blockchain gatherings, we’ve noticed some common themes in their reactions. And while most serious bitcoiners end up loving Saito once they understand it, it often takes a while for them to figure out what it is. We believe the stumbling blocks in getting there result from common misconceptions about blockchains: things that people think are true but that are actually false.
This gave us the idea to have a series of posts here correcting the most common “myths” about blockchains. Today we start with the idea that because mining is profitable…
Myth #1: Someone will always pay to run the network
Many bitcoiners argue that miners will pay for everything because they are making money from the network. This sounds reasonable but is wrong: all proof-of-work and proof-of-stake blockchains are characterized by free-rider problems that discourage miners from acting altruistically.
For an example of this, imagine a massive version of bitcoin in which a miner like Bitmain can afford to spend only 20% of its revenue on hashing because it is spending the rest paying for the peer-to-peer network: running servers to collect transactions from users, paying for the extraordinary amounts of bandwidth the network is consuming, laying down fibre-optic cable to provide extremely fast connections fellow miners, and more. In this environment, any competitor that can get away with spending more than 20% of its revenue on hashing will be more profitable than Bitmain. It is theoretically possible in this environment to be up to five times more profitable than the world’s largest miner! And if attackers pour their extra profits into buying more hashpower they will quickly away at Bitmain’s market share and eventually bankrupt the company.
In economics this problem is known as a collective action problem: everyone can agree that the world would be better off if everyone paid their fair share for a public good, but in the absence of a centralized authority forcing people to do exactly that everyone has a strong incentivize to cheat. And the result is that we end up with much less of the public good than is needed. In the case of bitcoin, this means that miners will happily spend money on mining because that makes them money, but when it comes to paying for bandwidth or mining they tend to do header-only mining, or treat bandwidth like a scarce commodity, refusing to upgrade from home connections or throttling bandwidth to support only a limited number of unknown peers. Regardless of whether we might wish it to be otherwise, the economic incentives in bitcoin encourage companies to pour their resources into the activity that is rewarded by the consensus mechanism. Doing anything else is altruistic but unprofitable.
When we point this problem out to hardcore bitcoiners, their first reaction is usually to frown and try to think about why we are wrong. The second reaction is to wonder why we are the first people to be telling them of something that is — in retrospect — perfectly obvious. And we think there are four major reasons that this myth has gained such widespread currency:
The blockchain community is very libertarian. This is a fantastic value to have in a free-speech community, but one unintended consequence is that few people treat markets as things that can fail. As a result, when a prominent member of the community asserts that miners will pay for everything because markets will sort out distribution problems, everyone just takes it at face value. In our experience, less than 5% of the blockchain community is even aware that there is a free-rider problem buried in the economic design.
If people are still not aware that the free-rider problem is a core economic problem, one reason is that we have a bias to believe that what worked in the past will work in the future, and this problem simply wasn’t visible at the beginning of the network when storage requirements were almost nothing and even a mobile phone had the bandwidth to run a full-node. Despite the fact that Bitcoin is nearly ten years old, people’s understanding of whether the technology — can there really be such a foundational problem with a platform that has survived for nearly a decade and clawed its way past an almost inhuman series of challenges?
The cryptocurrency community is focused on building things. Criticism is poorly taken unless it comes paired with an actual solution. This not only prevents people from understanding the nature of the problems that they face, but also encourages people to treat all problems as “technical problems” that can be solved through algorithmic complexity. In reality, algorithmic complexity in blockchains are usually signs of deep-rooted economic problems and attempts to gloss over problems by weakening either the open access or security requirements of the network. In part because of this problem, we estimate that perhaps only 20% of the blockchain community is even aware that the real scaling problems are economic rather than technical.
And what is the solution? We encourage everyone interested in blockchain technology to familiarize themselves with the consensus mechanism that Saito uses: Proof of Transactions (a.k.a proof of routing, proof of relay). The method is not intuitive and takes effort to understand, but it solves this and several other foundational economic problems with existing consensus mechanisms by paying nodes directly and at market rates for the work the network needs to perform.