{"paper_id":"2404.03976","title":"A theoretical framework for fees in AMMs","authors":["Abe Alexander","Lars Fritz"],"paper_date":"2024-04-05","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2404.03976v4","page_url":"https://thequant.space/flowcharts/a-theoretical-framework-for-fees-in-amms/","scores":{"empirical_rigor":3,"math_complexity":7,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper models the dynamics of arbitrage activity for a concrete AMM pool and studies its sensitivity to the fee choice to maximize value retention, mapping the dynamics to a random walk with a specific reward scheme.","quote":"To that end, we model the dynamics of arbitrage activity for a concrete implementation of a pool and study its sensitivity to the choice of fee aiming to maximize the value retention. We manage to map the ensuing dynamics to that of a random walk with a specific reward scheme that provides a convenient starting point for further studies.","summary":"In the ever evolving landscape of decentralized finance automated market makers (AMMs) play a key role: they provide a market place for trading assets in a decentralized manner. For so-called bluechip pairs, arbitrage activity provides a major part of the revenue generation of AMMs but also a major…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Trading is exclusively via arbitrage.","type":"model","stated":"explicit","quote":"For the purpose of this paper, we thus model the trading activity of an AMM based on the assumption that trading is exclusively via arbritrage.","support":"full"},{"assumption":"An infinite‑liquidity source (CEX) provides a quote at every point in time.","type":"model","stated":"explicit","quote":"To that end we simulate a source of infinite liquidity which provides a quote at every point in time.","support":"full"},{"assumption":"The external quote follows a Brownian motion with drift and no further assumptions.","type":"model","stated":"explicit","quote":"This quote follows Brownian motion with a drift, while we make not further assumptions.","support":"full"},{"assumption":"The AMM is a Uniswap v2‑style constant‑product pool (xA·xB = L²).","type":"model","stated":"explicit","quote":"For simplicity, we assume that the AMM is described as a Uniswap v2 type constant function pool xA ∗xB = L2 (1) where xA is the number of tokens of type A and xB the number of tokens of type B. While we choose for a v2-style pool for simplicity, the analysis applies equally to a v3 style pool.","support":"full"},{"assumption":"Transaction costs are negligible (TXN = 0).","type":"execution","stated":"explicit","quote":"We are assuming for now that transaction costs are negligible (TXN = 0), meaning we have a condition α 1 −f 1 + piA∆FL(1 −f) −(1 + ffl) > 0 .","support":"full"},{"assumption":"The fee on the liquidity side (ffl) is set to zero.","type":"model","stated":"explicit","quote":"(11) In the remainder of this paper we will discuss ffl = 0 which is realistic on many platforms.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Synthetic price series (Brownian motion) for CEX","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"For the purpose of our simulation, the CEX price pCEX will follow Brownian motion. To this end, we have studied the following case: we simulate an equidistant price series of 1000 points with a starting price of ps = pCEX = 1 and a volatility per time step of σ = 0.001.","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"n/a","evidence_card":"no","evidence_card_quoted":true},"conflict_kind":"applicability","card_detail":null,"card_quote":"We are assuming for now that transaction costs are negligible (TXN = 0), meaning we have a condition α 1 −f 1 + piA∆FL(1 −f) −(1 + ffl) > 0 .","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"For a given price mismatch there is an optimal fee that lets the AMM retain two‑thirds of the arbitrage gains while the arbitrageur receives one‑third.","kind":"theoretical","quote":"It turns out, that for a given price mismatch, there is an optimal fee choice which allows the AMM to retain 2/3 of the gains, while the arbitrageur has to content with 1/3.","support":"full"},{"finding":"The arbitrage threshold gain is proportional to the square of the fee and inversely proportional to price impact.","kind":"theoretical","quote":"Furthermore, we find that the threshold gain in arbitrage processes (threshold defined as the minimal price difference required for successful arbitrage) is proportional to the square of the fees and inversely proportional to the price impact (non-linear price).","support":"full"},{"finding":"The threshold condition for arbitrage is α > 1 + ffl/(1‑f), which simplifies to α > 1 + f/(1‑f) when ffl = 0.","kind":"theoretical","quote":"(10) We find that there is a threshold condition on α for a set AMM fee f which needs to be overcome defined by α > 1 + ffl 1 −f .","support":"full"},{"finding":"Simulation shows the average return is highly sensitive to fees when fee < volatility, and levels off for fees higher than volatility.","kind":"performance","quote":"We find that the average return for a given volatility σ as a function of the fee f is quite sensitive to fees for f < σ while it levels off for fees higher than the volatility, see Fig. 4.","support":"full"},{"finding":"A reasonable fee choice is slightly higher than the volatility; the exact fee value is not critical as long as it is not too high.","kind":"performance","quote":"We find that it is a reasonable choice to choose the fee according to the volatility, ideally slightly higher.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[{"paper":{"paper_id":"2404.13291","title":"Optimal Design of Automated Market Makers on Decentralized Exchanges","paper_date":"2024-04-20","authors":["Xue Dong He","Chen Yang","Yutian Zhou"],"empirical_rigor":2.5,"page_url":"https://thequant.space/flowcharts/optimal-design-of-automated-market-makers-on-decentralized-exchanges/"},"question":"What fee level should be set in AMMs relative to exchange‑rate volatility?","this_paper_finding":"A reasonable fee choice is slightly higher than the volatility; the exact fee value is not critical as long as it is not too high.","other_paper_finding":"The optimal unit trading fee is relatively insensitive to the mean and individual asset volatilities when the exchange‑rate volatility is held constant.","other_paper_quote":"Using numerical studies, we find that the optimal unit trading fee is increasing in the volatility of the fundamental exchange rate of the two assets but is not sensitive to the mean and fundamental volatility of each individual asset in the pool, holding the fundamental exchange rate volatility constant. Because the mean and volatility of each individual asset, when holding the fundamental exchange rate volatility","note":"Both findings state the optimal fee is only slightly above volatility and is largely insensitive to mean or individual asset volatilities, as long as it isn’t too high.","confidence":0.445}],"contradicts":[],"qualifies":[],"compared_with":21,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"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":2,"extracted_at":"2026-11-14T19:58:23Z"},"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/"}