In the past few weeks I’ve written about that blockchains are a kind of distributed ledger, and how they can be useful. But in these posts I left with a few unanswered questions that I’ll try to answer here.
So far everything about blockchains seem wonderful. Blockchains will allow software, people and organisations to transact with each other in more transparent and democratic ways where authenticity can be verified in a decentralised manner. If more software was built like this we would no longer need to pay with freedom, privacy and surveillance. But what do we have to pay with instead? There ain’t no such thing as a free lunch.
What is the cost?
Running a world-wide blockchain does indeed cost something. In order to understand the costs, we must first look into what it takes to run a blockchain such as Bitcoin.
Anyone in the world can use their computer to participate in appending Bitcoin’s distributed ledger with transactions between people and organisations. This is done by running a heavy cryptographic program which verifies the authenticity of transactions in the ledger. The more computers that participate, the stronger the authenticity of the ledger becomes. This model is called “proof-of-work” (PoW) and was designed to discourage people who want to maliciously append the ledger by having you prove with costly work that you are serving the network’s interest and not your own.
The aforementioned costs are variable costs such as electricity and degrading computer hardware and fixed costs such as renting space and internet access.
Why would you do costly work?
The first incentive to run these costly programs is of course to simply uphold the blockchain as a beneficial system for everyone. But humans are selfish by nature, so that incentive is not enough. That’s why everyone who does a transaction on the ledger has to pay a small fee to those who run the aforementioned programs. This fee comes in the form of digital currency (called BTC on Bitcoin’s blockchain) that is used for any transaction that writes to the ledger, such as buying goods, registering for an event or voting in an organisation. But this fee alone won’t cover the costs, which is why further reward is given from a fixed supply of digital currency that’s being “injected” to the economical system in a programmed way without causing inflation (this system deserves it’s own blog post for more details).
Not only do you earn a monetary reward, the blockchain is also programmed in such a way that any changes in the technical protocol must be agreed on by at least 51% of all work being done (notice how that differs to everyone who does work).
So to summarise — you only need to run a computer program in order to make a net-gain of digital currency and by doing so you will also have influence over protocol changes in this digital economy. Sounds fantastic and very democratic, doesn’t it?
So what’s the problem?
Firstly, with this proof-of-work model, there are economies of scale that lead to inequality between people. You can optimise your costs by using more computers in environments where electricity costs are lower. This gives an unfair advantage to people in certain geographical areas or people that are rich and can buy many computers.
For Bitcoin this has effectively lead to a centralisation of Bitcoin’s governance in China where production of computer hardware and electricity is very low. It’s a very real possibility that someone can create computers that do at least 51% of all proof-of-work and take control of the blockchain. This is counter productive to a system that is meant to be decentralised!
Secondly, with proof-of-work you essentially must burn a lot of electricity to prove your good intention in writing to the distributed ledger. There are various estimates for how bad this really is for the environment. Someone estimated that Bitcoin’s proof-of-work model could consume as much electricity as Denmark by 2020.
Thirdly, with proof-of-work one can enter the blockchain economy solely for egoistic purpose by earning fees and rewards and selling the digital currency for fiat money. Other than computers and electricity costs (that can be cheap) there’s very little at stake. This is counter productive for an economical system that is supposed to be of benefit for everyone.
How to minimize the costs and make it fair?
Because of these three big problems, the Ethereum blockchain have decided to switch from proof-of-work to something called “proof-of-stake” (PoS). In this model much less computational power is used for creation of cryptographic proof, and instead more work is spent on verification, which is much cheaper. The transaction fees, and the reward from the fixed supply of currency, is given to those who do verification instead.
People who want to make malicious verifications are discouraged by the fact that you have to prove with monetary stake that you are serving the network’s interest, and not your own. What this means is that you must lock-in a deposit of digital currency in order to run verifications, uphold the blockchain and earn reward.
The good thing with this model, soon to be used by Ethereum, is that fees and rewards are paid to everyone who does proof-of-stake in proportion to their stake without any economies of scale. So it does not proportionally benefit rich people.
The proof-of-stake model will have a much smaller impact on the environment since a lot less energy is needed. Further, this model also encourage participation of people who actually have some kind of stake in the system itself, i.e. not there only for egoistic reasons.
Is it all unicorns then?
While the Ethereum blockchain is a big improvement to the Bitcoin blockchain, there are still socio-economical challenges to overcome. In economic systems where you can pay for entry with privacy, freedom and transparency you at least have some means to participate even if you don’t have any money at all. While in the blockchain economy there are both financial and technical barriers to entry. And even if these barriers are small they are still not insignificant.
For the blockchain economy to really take off, it will be important to focus on these barriers to entry. This should include, but not be limited to, clever technical and financial solutions, private incentives and usability improvements.