Ethereum ePBS: Builders Face Capital Lock or Trusted Payment Trade-off
In brief
- Jason Vranek identifies three ePBS costs for builders: ETH reserve maintenance, high-value block coverage, and payment failure exposure.
- Trusted payment arrangements reduce collateral costs but shift risk to builder reputation rather than protocol security.
- Ethereum.org expects ePBS testnet (Glamsterdam) on mainnet in Q4 2026.
The collateral question
Jason Vranek, a Commit-Boost contributor, outlined the challenge in a Sept. 8-11 Lido discussion. Builders funding protocol-backed payments face costs from idle Ethereum, failed delivery, and offers they wanted to cancel. The proposed EIP-7732 design accommodates two payment forms: collateral-backed and trusted.
Collateral-backed payment draws on Ethereum the builder has deposited into the protocol. Trusted payment depends on the builder honoring a promise through another payment route. Vranek identified three potential costs for the collateral model: maintaining ETH reserves inside the protocol, covering unusually valuable blocks, and exposure to payment without successful delivery. Capital held in reserves to secure payments cannot simultaneously serve another use—a hidden tax on builders' liquidity.
Where trusted payments fit
A trusted arrangement could reduce those costs and leave more room to pay the proposer. Builders with existing trust relationships or access to off-protocol liquidity could gain an edge, though the protocol design may include safeguards to mitigate this advantage. Meanwhile, Titan Builder said it expects validators to continue reaching it through relays that organize auctions and handle publication, suggesting auction-based mechanisms may still help smaller builders compete.
Timeline and broader context
As of Sept. 13, Ethereum.org lists Glamsterdam as testing on devnets, with mainnet expected in the fourth quarter of 2026. The Ethereum Foundation's priorities also identify fork-choice enforced inclusion lists (FOCIL) as a planned mechanism. FOCIL would let validators impose inclusion requirements on builders' blocks, and separating consensus work from execution processing gives validators more time for heavy work. Both moves aim to reduce builder centralization, though the payment trade-off Vranek highlighted remains unresolved.
The choice between locked capital and trusted relationships reflects a deeper design tension: how to secure builder commitments without recreating the intermediaries ePBS was meant to disintermediate.
Frequently asked questions
What is enshrined proposer-builder separation (ePBS)?
ePBS formalizes the exchange between a validator proposing a block and the builder assembling its transactions. In EIP-7732, the proposer includes a builder's signed commitment in its consensus block, and the execution payload containing the transactions follows separately. This design aims to reduce builder centralization by separating the roles.
Why does collateral-backed payment cost builders money?
Builders must maintain ETH reserves inside the protocol to secure payments, cover unusually valuable blocks, and expose themselves to payment without successful delivery. Capital locked as collateral cannot be used elsewhere, creating an opportunity cost that reduces the amount builders can pay proposers.
How could trusted payments change builder economics?
A trusted arrangement could reduce costs by avoiding the need to lock capital inside the protocol, leaving more room to pay proposers. However, this shifts risk from the protocol to the builder's reputation and depends on off-protocol liquidity or existing relationships.
