{"paper_id":"2507.08302","title":"Arbitrage on Decentralized Exchanges","authors":["Xue Dong He","Chen Yang","Yutian Zhou"],"paper_date":"2025-07-11","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.08302v1","page_url":"https://thequant.space/flowcharts/arbitrage-on-decentralized-exchanges/","scores":{"empirical_rigor":8,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes the first equilibrium model of gas fee competition between two arbitrageurs across three transaction reversion settings.","quote":"We develop the first equilibrium model of gas fee competition between two arbitrageurs under three transaction reversion settings: no-revert, auto-revert, and selectable-revert.","summary":"Decentralized exchanges (DEXs) are alternative venues to centralized exchanges (CEXs) for trading cryptocurrencies and have become increasingly popular. An arbitrage opportunity arises when the exchange rate of two cryptocurrencies in a DEX differs from that in a CEX. Arbitrageurs can then trade on…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Price discrepancies between DEX and CEX are public information.","type":"market","stated":"explicit","quote":"since price discrepancies are public information, multiple arbitrageurs are likely to compete for the same arbitrage opportunity."},{"assumption":"Arbitrageurs face inventory risk.","type":"model","stated":"explicit","quote":"confirm that arbitrageurs face positive inventory risk."},{"assumption":"In the auto-revert setting, reverted transactions incur a small cost (e.g., gas fee).","type":"execution","stated":"explicit","quote":"in the auto-revert setting we assume a small cost (e.g., gas fee) for transactions that are submitted but reverted."},{"assumption":"Traders can choose the transaction reversion setting (no‑revert, auto‑revert, selectable‑revert).","type":"model","stated":"explicit","quote":"Our goal is to analyze the arbitrageur’s strategy in the gas fee competition game in each of these settings."},{"assumption":"Trading on a DEX incurs a proportional fee f = 0.3%.","type":"execution","stated":"explicit","quote":"Finally, the unit transaction fee f on Uniswap V2 is 0.3%."},{"assumption":"Base gas fee is calculated as base fee per gas multiplied by gas used (median 5.6 USD).","type":"data","stated":"explicit","quote":"We use the median of the base fee per gas in blocks with arbitrage and the mode of the gas used in our empirical study"}]},"datasets":{"status":"available","items":[{"name":"WETH/USDT swap transactions on Uniswap V2","provider":"Uniswap V2 (on‑chain)","kind":"crypto-onchain","access":"public","period":"2024-10-27 to 2024-12-01","universe":"WETH/USDT pair","quote":"We collect 100,000 swap transactions for the WETH/USDT pair on Uniswap V2 from October 27, 2024 to December 1, 2024."},{"name":"Blockchain block data (reserves, base fee per gas)","provider":"Ethereum blockchain via Infura","kind":"crypto-onchain","access":"public","period":"2024-10-27 to 2024-12-01","universe":"Blocks containing the above swaps","quote":"includes the block number, block generation time, WETH/USDT reserve amounts on Uniswap V2 at block generation, and the base fee per gas."},{"name":"Second‑by‑second ETH price data","provider":"Binance","kind":"market-prices","access":"public","period":"2024-10-27 to 2024-12-01","universe":"ETH/USD","quote":"We collect second‑by‑second ETH price data from Binance, the largest CEX for cryptocurrency trading.","known_id":"binance"}],"known":[{"id":"binance","name":"Binance","provider":"Binance","kind":"market-prices","access":"public","quote":"28 compute the ex ante proﬁt and loss (P&L) using equation (3), where the gas fee g and trading amount di are extracted from the swap data, and the prices pi are taken from Binance, i ∈{A, B}.","mentions":6,"used_in":6},{"id":"the-graph","name":"The Graph (subgraphs)","provider":"The Graph","kind":"crypto-onchain","access":"public","quote":"The WETH/USDT pair address on Uniswap V2 is obtained via the subgraph https://thegraph.com/explorer/subgraph/ianlapham/uniswapv2.","mentions":2,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Unit transaction fee f = 0.3% on Uniswap V2 and gas fees (base fee per gas × gas used, median 5.6 USD).","quote":"Trading on a DEX incurs a proportional fee. Let f be the unit trading fee","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Unit transaction fee f = 0.3% on Uniswap V2 and gas fees (base fee per gas × gas used, median 5.6 USD).","card_quote":"Trading on a DEX incurs a proportional fee. Let f be the unit trading fee","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"Pure symmetric equilibria do not exist in any of the three reversion settings; unique mixed symmetric equilibria can be characterized.","kind":"theoretical","quote":"We show that pure symmetric equilibria do not exist, but unique mixed equilibria can be characterized."},{"finding":"Under low inventory risk, the no‑revert setting yields higher arbitrageur profits, while auto‑revert and selectable‑revert improve market efficiency.","kind":"theoretical","quote":"Comparative analysis reveals that under low inventory risk, the no-revert setting favors arbitrageurs in terms of profit, while auto-revert and selectable-revert settings enhance market efficiency."},{"finding":"Under high inventory risk, the no‑revert and selectable‑revert settings dominate the auto‑revert setting in both profitability and efficiency.","kind":"theoretical","quote":"Under high inventory risk, the no-revert and selectable-revert settings dominate the auto-revert setting in both profitability and efficiency."},{"finding":"Empirically, gas fees increase with price discrepancies and liquidity, and trading amounts rise with both price discrepancies and gas fees.","kind":"other","quote":"We empirically confirm that arbitrageurs face positive inventory risk and validate our model's implications: gas fees increase with price discrepancies and liquidity, while trading amounts rise with both price discrepancies and gas fees."},{"finding":"Regression results show that relative swap amount (conditional on trading) increases with price discrepancy and gas fee, and gas fee increases with price discrepancy and liquidity.","kind":"other","quote":"We find that the relative trading amount, conditional on trading, increases with price discrepancy and gas fee, and that the gas fee, conditional on trading, increases with price discrepancy and liquidity."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":41,"citations":{"cited_by_in_corpus":6,"by_independent_authors":6,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":5,"unclassified":1},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:33:12Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}