Ethereum’s validators don’t really build the blocks they propose anymore. That job has been quietly outsourced to a specialized class of actors called block builders, and the separation between those two roles - proposer-builder separation, or PBS - is now one of the most consequential structural features of the network.
The shift happened gradually. After the Merge in September 2022, MEV-Boost - a piece of middleware built by Flashbots - gave validators a way to auction off their block-building rights to external builders who could extract more value from transaction ordering. Validators adopted it quickly because it was profitable. By early 2024, MEV-Boost was being used in the vast majority of Ethereum blocks. Validators earned more by outsourcing; builders earned from MEV extraction. The incentives were clean.
What emerged from that arrangement is a builder market that is highly concentrated. A small number of builders - often entities with deep connections to trading firms or sophisticated MEV infrastructure - consistently win the auction. This isn’t speculation; block explorer data from platforms like mevboost.pics has tracked builder dominance for years, and the top three or four builders routinely account for the overwhelming majority of Ethereum blocks in any given week.
The Neutrality Problem
Ethereum’s original design assumed validators would be relatively passive, decentralized participants who simply ordered transactions as they arrived. PBS broke that assumption. The actor who decides which transactions appear in a block, and in what order, is now a builder - not the validator signing it.
This creates a censorship surface that didn’t exist before. Builders can choose to exclude transactions, favor certain protocols, or front-run users. OFAC compliance pressure demonstrated this concretely: following U.S. Treasury sanctions on Tornado Cash in August 2022, several builders began filtering sanctioned addresses from their blocks. At peak, a significant share of Ethereum blocks were OFAC-compliant, meaning they excluded certain transactions not because the protocol demanded it, but because builders chose to.
What Enshrined PBS Would Actually Change
Ethereum researchers have been working on enshrinement - building PBS directly into the protocol rather than relying on MEV-Boost as a bolt-on solution. The theoretical benefit is that an enshrined design could include stronger commitments from builders and better protections for validator neutrality.
But enshrinement is hard. It requires changing how the consensus and execution layers interact, and proposals like ePBS have been in research discussion for years without shipping. The builder market has continued to consolidate in the meantime.
The uncomfortable reality is that Ethereum’s block production is already more centralized than its validator set suggests. The validator count is in the hundreds of thousands; the effective builder market is not. Those two facts exist simultaneously, and the protocol hasn’t resolved the tension between them.