Who are the block proposers?
A collective and open-membership group who provide the network with a transaction ordering service
Early e-cash protocols relied on the broker to issue new coins and permit redemption of coins. Finding an agent to take on the role of a broker was non-trivial and plagued the adoption of e-cash. There are many reasons why financial institutions avoided taking on the broker role including:
Regulatory red tape,
Financial and reputation risk,
Lack of understanding or appeal (“why not credit cards?”).
Satoshi Nakamoto’s core innovation was to replace the broker’s role for issuing new coins with a deterministic monetary policy. The policy is enshrined in the protocol specification and executed by all participants. It side-stepped the issue of finding an appropriate broker and the only remaining meatspace problem was to decide who will collect and order user-generated transactions.
Transaction ordering service
Collectively, the block proposer offers a single service:
Transaction ordering service. Decide the final order of transactions for execution.
All transactions carry a fee (or tip) to help prioritise the ordering of transactions. It is expected that block proposers will pick transactions with the highest fee first as this maximises their profit. Pending transactions are received from the peer to peer network and/or users can send transactions directly to the block proposers.
To avoid any misconceptions, block proposers DO NOT in ANY WAY run or control the network:
Block proposers are NOT responsible for enforcing consensus rules.
Block proposers CANNOT change the consensus rules.
It is a very simple, narrow and well-defined role. They are financially rewarded for providing a good quality service for ordering transactions.
Why are the block proposer’s honest?
If they propose a block that breaks the rules, then the peer to peer network (economic majority) will reject the block and the block proposer forfeits the reward. In fact, there is only a single decision a block proposer can make:
Decide to extend a block. If a block proposer receives a new block, should they extend it or continue to mine a competing block?
Given this single decision, there are two ways for a block proposer to deviate from the honest behaviour:
Filter transactions. A bock proposer will not include certain transactions in their blocks.
Filter blocks. A block proposer will ignore blocks if it contains a censored transaction.
We cover the topic of censorship and how both actions can be leveraged by an adversarial block proposer in more-depth here. It is a growing issue due to the rise of OFAC-enforcing Validators on Ethereum’s proof of stake protocol.
There are two reasons why block proposer’s are honest:
Profitability. It should be more profitable, in the long term, for block proposers to follow the rules than to break them.
Trust, but verify. The peer to peer network can decide to accept/reject the outcome of their actions in real-time.
It is interesting that “honest behaviour” is a social concept and what it means to be honest has changed over time based on who is participating. For example, the idea of sending a transaction directly to the block proposers and bypassing the peer to peer network was considered “dishonest” for a long time. Today, direct-transactions is now a popular service run by Flashbots as it is the only way to avoid the dark forest.
How do we decide who can become a block proposers?
A leader election protocol has two components:
Registering participants. Deciding who is allowed to participate in the leader election protocol,
Picking the leader. All participants agree on the new leader for a single round.
In terms of registering participants, there is no trusted admin to curate a potential list of candidates nor is there a strong notion of identity. As we will see, it is possible to build an open-process that links a weak notion of identity by using the ownership of a financially expensive and scarce resource called Proof of X.
Once a list of candidates is available, then it is fairly straight-forward to decide who can be the next leader. For example, a round-robin style protocol can be implemented for each participate to take turn in becoming a leader or a random beacon can be used select participants at random. AWS has a nice writeup on the pros and cons of a leader election protocol for classical distributed systems.
Interestingly, in proof of work, a participant can register for the election without publicly signalling their intent to participate.
No strong notion of identity
The leader election protocol’s goal is not to simply widen access for anyone to participate in the process, but to rate-limit membership only to parties who are financially motivated to remain online and actively participate. It is hoped this will ensure block proposers are financially aligned to honestly follow the protocol (and not deviate by censoring transactions).
In practice, it is difficult to rate-limit membership as there is no strong notion of identity on the protocol. It is very easy for an individual to create one, two, or several identities. While there are some fun protocols for Proof of Personhood, there is no realistic way to restrict participation based on real-world identities.
Proof of X.
The solution for rate-limiting who can participate is to pick block proposers based on their ownership of a scarce and financially expensive resource to acquire.
We call this Proof of X as the participant must prove they own the asset X before they can participate as a block proposer. It leads to a weak notion of identity as it is linked to the resource and the participant can remain pseudonymous.
Two prominent examples include:
Proof of Work. Prove computational power under your control by producing the solution to a difficult puzzle (ownership of electricity and hardware efficiency).
Proof of Stake. Prove ownership of a set of coins.
In both cases, there is a limited supply of the resource and it is financially expensive to acquire it. The side-effect of Proof of X is that it can help align the financial incentives of all participants to protect the long-term health of the system. If they deviate from the honest protocol, it can be detrimental their investment.
Misconception: Proof of X has nothing to do with transaction throughput. It only serves as a way to identify who is participating in the leader election for ordering transactions.
The financial cost to become a block proposer
We consider the financial cost for someone to self-appoint themselves as a block proposer. This includes:
Capital cost. The upfront investment before a participant is eligible to become a block producer.
Operational cost. The on-going cost for the participant to continue as a block producer.
Put another way, the capital cost relates to the registration process and the operational cost relates to competing in the leader election.
Proof of work. It is a meat-space process. The capital cost is incurred when the operator acquires hardware and a warehouse to run the mining farm. In 2019, this article discusses how Bitmain invested $80m to deploy their own hardware to mine bitcoins. The only way to recoup the capital cost is to make a profit or to sell the physical setup to another miner on the secondary market.
The dominant operational cost to run a mining farm is electricity. This leads to a financial incentive for operators to set up a mining facility in locations that offer cheap (or even free) electricity. This is the main argument by the Bitcoin community on why mining incentives renewable energy as it can be the bidder of last resort for the energy suppliers.
Proof of stake. The capital cost is locking coins into a staking contract (in 32 ETH segments). There is no expenditure and the staked coins can be returned in a timely manner by withdrawing from the staking contract (not yet enabled). Put another way, block proposers must demonstrate that in-band coins are locked up before they are permitted to participate as a block proposer.
The operational cost is insignificant. There is no mining facility or proof of work. The participation involves a light-weight cryptographic protocol to help pick leaders at random and proportional to their capital lockup. A staker simply needs a machine that can process blocks and cast a vote for new blocks in real-time.
Simple mental models. Proof of work is a professional and commoditised business model. Long term, participants fight for minimal returns on investment due to the high operational costs and the reducing network reward. On the other hand, proof of stake is more like a savings account that earns interest as the operational costs are virtually non-existent. The network only needs to offer a minimal viable income as opposed to subsidising the block proposer’s participation.
Reward function
The final aspect we need to consider is how a block proposer gets paid for their participation and working together as a collective to order transactions:
Reward function. The algorithm for checking a block proposer’s participation before computing their reward alongside penalties.
The actions that are checked by the reward function differs based on proof of work or proof of stake. Before we consider the actions in more detail, there are two sources of funds that can be used to pay them:
Block subsidy. The issuance of new coins by the blockchain protocol,
Fees & tips. A payment from the user for including their transaction in a block.
At present, the majority of the reward for most networks is the block subsidy as new coins are issued to the block proposers. Long term, both Bitcoin and Ethereum take vastly differently approaches.
Bitcoin’s fee market. The monetary policy is enshrined to solely rely on the fee market as the block subsidy will eventually disappear when there are approximately 21 million coins issued. It is worth noting that Bitcoin’s fee market has yet to surpass the bock subsidy in any meaningful manner.
Ethereum’s minimal issuance. The community is pursuing a concept called minimal viable issuance which issues new coins as a guaranteed minimum income for the block proposers. With EIP-1559, it is expected the new issued coins will be offset by burning most transaction fees paid on the network. This is the current state of the Ethereum network and it is being battle-tested in real-time.
Frequency of payments and pooled operations
Regardless, assuming there is a reward function and we can compute the number of coins to send a block proposer, we need to consider the frequency of payments which differs for proof of work and proof of stake.
Only paid upon mining a block (proof of work). In Bitcoin’s proof of work, a miner is rewarded if they successfully mine a block and it is eventually accepted as part of the canonical blockchain. The reward is only transferable after a block has achieved a depth of 100 confirmations. This frequency of payment is problematic for small miners as it may take years before they can mine a block and collect the reward. It has led to the proliferation of mining pools as miners reduce the variance of payouts by collaborating as a collective to solve blocks and share the reward amongst themselves.
Continuous payout (proof of stake). In Ethereum’s proof of stake, the block proposers are continuously rewarded for their involvement in the protocol and there is no significant sunk cost for participating. A block proposer is rewarded when they vote for proposed blocks (once per epoch) or when they propose their own block. Thanks to the small block times (12 seconds), a solo staker is likely to propose a block within ~90 days. Thus, there is no motivation to participate in a staking pool to reduce the variance of payoffs. However, with the rise of MEV and the significant profits distributed by MEV-boost, many stakers sign up to Relay services as they may receive a ‘large bonus payment’ when proposing their block.
Pooling remains popular. Pooling resources together was useful in proof of work to reduce the variance of payments, but it remains popular in proof of stake as it can significantly increase the block proposer’s overall profit. The rise of Flashbots and Relays has led to an interesting insight that block proposers will ignore pending transactions from the peer to peer network if they can maximise their profit elsewhere (i.e., OFAC-enforcing relays).
Joining a pool impacts decentralization as the act of block building (“picking transactions”) is no longer the sum of all block proposers, but the set of pool masters who submit block templates to the block proposers. In proof of work, this involves less than <10 agents. It appears to be slightly better in proof of stake / MEV as block building is competitive based on a first-priced auction, but the set of block builders is expected to centralise overtime and there are only a few trusted Relays (an intermediary between the block builder and block proposer) in operation as of 7/11/2022.
Wrapping up.
In a future block post, we will cover the long-term sustainability of Bitcoin and Ethereum as this is critical for attracting the “right type of players.” If you want to know more now, you can watch Lecture 4 or the wonderful Q&A session on Fee Market Sustainability.
As you can imagine, the design by Satoshi Nakamoto was radically different to any leader election protocol proposed for the past 30 years. It led to a new frontier of research for a topic that was typically considered dull and mostly complete.
p.s., Distributed systems literature calls the process a leader election, but in reality it is a competition to become the leader. We stick with historical terminology.