In previous articles, Vitalik has discussed the “Merge” and “Surge” upgrades for Ethereum, and mentioned possible improvements to the Ethereum staking system, while setting an ambitious goal of achieving the ability to process 100,000 transactions per second on both the first and second layer networks of the blockchain.
In the latest article, Vitalik emphasizes that the centralization risk in the proof-of-stake system is “one of the biggest risks Ethereum L1 faces” due to economic pressure, and outlines various methods to mitigate this risk in the “Scourge” upgrade. Buterin points out that these risks primarily manifest in two aspects: block construction and staking capital provision.
In the first part of the article, Vitalik discusses the issue of maximum extractable value (MEV) and points out that currently “two actors (proposers and block constructors) decide about 88% of Ethereum block content”, which increases the risk of censorship and can lead to transaction delays of several minutes, especially in time-sensitive settlements or token exchanges.
Vitalik believes that one “key” element in the solution may be an encrypted transaction pool, which would make it more difficult for block proposers to censor specific transactions. He also acknowledges that designing a system that is “both robust and relatively simple, and reasonably feasible” still requires further effort.
Vitalik points out that a “central challenge” lies in countering MEV with different methods: “any substantial power retained in the hands of stakers could become power related to MEV”, which means there must be a balance between stakers’ ability to choose transactions and their potential to extract value from the blockchain.
He proposes two schemes with different trade-offs: one is the use of inclusion lists, where stakers propose a list of transactions that block constructors must include in the next block; the other is the use of multiple simultaneous proposers, distributing the block production process among multiple actors.
Vitalik concludes by stating:
“A conservative effective strategy is to ‘wait and see’, where we can start with a scheme that restricts stakers’ power and auctions off most of the power, and gradually increase stakers’ authority as our understanding of the MEV market grows.”
Vitalik points out that currently about 30% of Ethereum’s supply is staked, which “is already enough to protect Ethereum from 51% attacks”. However, if this ratio approaches 100%, it could bring some risks, including weakening the effectiveness of slashing, unnecessary issuance of approximately 1 million Ether per year, and the potential replacement of Ethereum’s “currency” network effect by a single liquid staking token (LST).
He presents two major approaches: setting a limit on the amount of Ether that users can stake, and implementing a two-layer staking system that divides staked Ether into slashable and non-slashable portions. In the article, Vitalik writes:
“The main remaining task here is either to accept the risk of almost all Ether entering LST, or to finalize the details of one of the proposals and achieve consensus on its parameters.”
Additionally, Vitalik also proposes application-layer solutions, including developing and promoting dedicated staking hardware, rewarding individual stakers through airdrops, and reducing MEV through more sophisticated application design.
Reference:
