The Marginal Effects of Ethereum Network MEV Transaction Re-Ordering

Bruce Mizrach Nathaniel Yoshida · 2025

Read The Marginal Effects of Ethereum Network MEV Transaction Re-Ordering by Bruce Mizrach Nathaniel Yoshida (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-04003.txt.

Introduction

Bruce Mizrach and Nathaniel Yoshida measured the costs of transaction reordering on the Ethereum blockchain in 2025. Two block builders now produce the large majority of Ethereum blocks, and builders can rearrange pending transactions within a block for profit, a practice known as maximal extractable value. That power can harm ordinary participants whose trades get jumped or sandwiched.

Using sandwich attack data, the authors estimate that participants in aggregate would pay millions of dollars per month to guarantee a place in the first quarter of a block. Sandwich attacks are frequent, occurring on average more than once every two blocks, and the gas fees on those transactions fund a notable share of the payments flowing to validators. The authors suggest reforms such as gas fee priority rules or private transaction pools could help.

Catalog record

CREATORBruce Mizrach Nathaniel Yoshida
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-04003

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The Marginal Effects of Ethereum Network MEV Transaction Re-Ordering Correspondence: Department of Economics, Rutgers University, 75 Hamilton Street, New Brunswick, NJ 08901 USA. email: [email protected]. We would like to thank EigenPhi for providing the sandwich data. Bruce Mizrach Nathaniel Yoshida Abstract Two MEV builders now produce nearly 80% of Ethereum blocks. Block builders have the ability to reorder transactions on the blockchain in a way that can be harmful to participants. We estimate they would pay in the aggregate nearly $14 million per month to ensure that they remained in the first quartile of the block. Sandwich attacks, in which a transaction is front-run, are frequent, averaging more than one per block. Gas fees on these transactions pay for nearly 15% of the MEV payments to the validator. These attacks have especially large marginal effects and skew the distribution. Reforms such as gas fee priority or private transaction pools might be helpful. Keywords: Ethereum; maximum extractable value. JEL Codes: G12; G23. 1 Introduction The Ethereum blockchain has evolved considerably since it was created in July 2015 by Vitalik Buterin. 1 1 1 https://etherscan.io/block/1 Most tokens utilize the ERC-20 standard which was adopted in 2017. Since September 2022, blocks are validated by a system called proof-of-stake which replaced the energy intensive mining in proof-of-work. Validators who stake Ether, the network’s utility token, are eligible to approve blocks. Block creators receive rewards in the form of gas fees, payments denominated in Ether to defer the processing and storage costs of adding new blocks to the chain. Buterin raised concerns about the centralization of the network under proof-of-stake. Validators have twelve seconds to approve a block with hundreds of transactions, which can occur at any time in the day. These high fixed costs and technical sophistication might lead to a small group of staking pools validating most of the blocks. In the first month of the merge [ 14 ] found that the Ethereum network Herfindahl index rose slightly compared to the mining pools that were validating blocks under proof-of-work. [ 11 ] advanced the idea of proposer builder separation as a potential solution. In this framework, the validator can choose from a set of proposed builder blocks. The block builder seeks to maximize extractable value (MEV) in the block by re-ordering transactions or inserting transactions from searchers. 2 2 2 Searchers are participants who scan the public mempool for profitable atribtrages This extractable value was previously obtained by the miner under proof-of-work. From the perspective of the Ethereum developers this proposer-builder separation is a way for validators to share in MEV revenue and enable less sophisticated validators to participate in the blockchain. EigenPhi estimates 3 3 3 https://eigenphi.substack.com/p/30m-72-of-searchers-mev-revenue-went that in January-February 2023 $48.3 million was extracted from blockchain users. This was split between: searchers - $7.3 million (17.4%); builders - $4.4 million. (10.5%); and validators - $30.3 million (72.1%). Toni Wahrstätter, a researcher at the Ethereum Foundation, has created a dashboard 4 4 4 MEVBoost.pics for the number of blocks built by maximum-extractable value (MEV) algorithms. On September 23, 2022, the day after the Beacon merge, MEV builders completed 34.2% of blocks. By December 2022, the MEV share had grown to over 90%. While validators are still chosen randomly, many use open source software from Flashbots called MEV-Boost 5 5 5 https://docs.flashbots.net/flashbots-mev-boost/introduction to access a selection of proposed blocks. They generally select the block offering the highest payment to the validator. We typically observe the payment of ETH to the validator as the last transaction in the block. Information on the builders and relays is from Mevboost.pics. From these transactions, sourced from Google Big Query, 6 6 6 bigquery-public-data.crypto_ethereum we can compute the market share of the MEV builders. By August 2024, as shown in Figure 1 , just two builders have come to dominate the market. Figure 1: MEV Block Formation Note: The MEV builder addresses are from MEV Boost.pics, and we detect payments in Mainnet blocks from them to the validators. The sample is from January 23, 2024 to April 16, 2025. Some MEV arises because of efficiencies in gas utilization. The Ethereum foundation notes 7 7 7 These practices are known as gas golfing , https://ethereum.org/en/developers/docs/mev/ two examples: “[1] using addresses that start with a long string of zeroes …since they take less space (and hence gas) to store; [2] leaving small ERC-20 token balances in contracts, since it costs more gas to initialize a storage slot… than to update a storage slot.” The other sources of MEV are less benign. They include various types of profitable manipulation of the transactions in the block. [ 13 ] describes several categories of front running behavior. They include displacement in which a builder simply substitutes their own address for a potentially profitable transaction. A second group involves insertion of a transaction, generally sandwiched in between two transactions drawn from the mempool. A third category is suppression of a transaction to a later sequence in the block or even to a subsequent block. [ 12 ] estimate that of the 6.6 mllion blocks with transactions on Uniswap V2 decentralized exchange between May 15, 2020 and January 13, 2024, more than 90% were at risk of front running, and more than one million blocks actually experienced front running attacks. We show in Figure 2 that MEV Boost revenues are substantial. They comprise gas fees net of base fees that are burned, and payments from MEV algorithms. Figure 2: Validator Revenue Note: The MEV payments data are from the last transaction in the block and range from January 23, 2024 to April 16, 2025. The net gas revenue calculations utilize Ethereum Mainnet blocks sourced from Google Big Query. We exclude four surges in MEV payments: August 5, 2024 that coincides with a 30% price decline in ETH that appears to have multiple causes 8 8 8 https://medium.com/@bikingex/crypto-market-black-monday-on-august-5th-2024-9a2c0dbf7430 ; August 26, 2024 when the Ethereum Foundation transferred a large supply of ETH to Kraken. 9 9 9 https://blockcast.cc/market-roundup-bitcoin-and-ethereum-26-august-2024/ February 3, 2025 in which Ethereum suffered a 27% flash crash 10 10 10 https://www.soliduslabs.com/post/ether-feb3-flash-crash-a-stark-reminder-of-crypto-market-vulnerabilities ; and April 7, 2025 is the day that U.S. President Trump announced new global tariffs. Median total revenues are 635 ETH per day ($1.59 million per day at an ETH price of $2,500), with MEV payments making up 41% of that total. There is an active literature within the Ethereum community to address protocol changes that might be the fairest for validators and/or reduce the harmful effects of front running. Several proposals have been offered to smooth out MEV payments across validators, including full sharing of MEV payments 11 11 11 https://ethresear.ch/t/committee-driven-mev-smoothing/10408 , burning MEV 12 12 12 https://ethresear.ch/t/mev-burn-a-simple-design/15590 , offering revert protection to failed transactions 13 13 13 [ 18 ] , and more complex auction designs. 14 14 14 [ 15 ] To our knowledge, the community has not proposed a system of absolute time priority. This is particularly surprising since Ethereum developers acknowledge that there is an incentive to bid as late as possible in an MEV auction. 15 15 15 https://ethresear.ch/t/reducing-latency-games-by-levelling-the-playing-field-on-block-size-for-pbs/19356 We estimate the harm to participants by first estimating a probit model for the probability that a transaction will be placed in the first quartile of

Read the complete text