Briefing
The core research problem is the fundamental trade-off in designing a transaction fee mechanism (TFM) that is simultaneously safe for users, profitable for block producers, and resistant to collusion. This paper establishes a powerful impossibility result by proving that the three core desiderata → Dominant Strategy Incentive Compatibility (DSIC) for users, Myopic Miner Incentive Compatibility (MMIC) for miners, and Off-Chain Agreement-proofness (OCA-proofness) against collusion → cannot be satisfied by any non-trivial deterministic TFM. The single most important implication is that all practical, revenue-generating blockchain auction designs must inherently compromise on at least one of these three properties, forcing a systemic re-evaluation of the trade-offs between user truthfulness, protocol revenue, and economic security.
Context
Before this research, a major open question in cryptoeconomics, posed by Roughgarden, asked whether a “good” Transaction Fee Mechanism could satisfy all desired properties → user truthfulness (DSIC), miner incentive compatibility (MMIC), and collusion resistance (OCA-proofness). Prevailing theoretical limitations centered on the difficulty of designing an auction that could resist the formation of profitable off-chain agreements (OCAs) between users and block producers, a key vector for Maximal Extractable Value (MEV) extraction and market instability. Prior work had shown impossibility results under the stronger notion of Side-Channel Proofness (SCP), leaving the precise boundary for the more practical OCA-proofness undefined.
Analysis
The paper’s core mechanism is a precise characterization of the intersection of three key properties in auction theory applied to blockchain transaction ordering. It formally defines the three desiderata → DSIC (users bidding truthfully is a dominant strategy), MMIC (miners maximizing their immediate revenue), and OCA-proofness (no coalition of users and the miner can jointly deviate for profit). The breakthrough is the mathematical proof, specifically Theorem 4.7, which demonstrates that the only deterministic mechanism satisfying all three is the “trivial TFM,” one that allocates no transactions and yields zero revenue.
This result is achieved by first characterizing all DSIC+OCA-proof mechanisms and all MMIC+OCA-proof mechanisms, then showing that their intersection is only the trivial case. For randomized mechanisms, the paper shows a necessary trade-off between collusion resistance and efficiency, proving a strict upper bound on achievable social welfare.
Parameters
- Impossibility Result Condition → DSIC + MMIC + 1-OCA-proofness (The three simultaneous properties that only the trivial mechanism satisfies.)
- Maximum Efficiency Bound → 0.842 (The worst-case upper bound on efficiency approximation for any non-trivial randomized DSIC+MMIC+OCA-proof mechanism.)
- Trivial Mechanism Revenue → 0 (The miner revenue of the only deterministic mechanism that satisfies all three properties.)
Outlook
This impossibility result immediately shifts the focus of mechanism design research from seeking a perfect, fully-incentivized TFM to exploring mechanisms that strategically relax one of the three properties, or operate in a Bayesian-Nash equilibrium rather than a Dominant Strategy one. Future research will concentrate on practical, non-deterministic mechanisms that can approximate the maximum efficiency bound of 0.842, or on novel protocol architectures, such as Proposer-Builder Separation (PBS), that fundamentally decouple the roles to bypass the impossibility by changing the game’s structure. This theoretical boundary will guide all future design of fair and stable decentralized economic systems.
Verdict
This research establishes a foundational, non-negotiable economic constraint, proving that perfect incentive alignment, revenue generation, and collusion resistance are mutually exclusive in deterministic blockchain transaction fee mechanisms.
