Paper: arXiv 2403.18177

Authors: Cheuk Yin Lee, Shen-Ning Tung, Tai-Ho Wang

Abstract

We study how trading fees and continuous-time arbitrage affect the profitability of liquidity providers (LPs) in Geometric Mean Market Makers (G3Ms). We use stochastic reflected diffusion processes to analyze the dynamics of a G3M model under the arbitrage-driven market. Our research focuses on calculating LP wealth and extends the findings of Tassy and White related to the constant product market maker (Uniswap v2) to a wider range of G3Ms, including Balancer. This allows us to calculate the long-term expected logarithmic growth of LP wealth, offering new insights into the complex dynamics of AMMs and their implications for LPs in decentralized finance.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 2.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper is heavily mathematical, utilizing stochastic calculus, reflected diffusion processes, and deriving explicit formulas via parabolic PDEs with Neumann boundary conditions, placing it at the high end of complexity. However, it lacks any empirical validation, backtests, or implementation details, focusing purely on theoretical analysis and model derivation.

Research Flowchart

  flowchart TD
  A["Research Goal: Analyze LP Wealth Growth<br>in Geometric Mean Market Makers"] --> B["Key Methodology: Stochastic<br>Reflected Diffusion Processes"]
  B --> C["Data: Arbitrage-Driven Market<br>Trading Fee Structures"]
  C --> D{"Computational Process"}
  D --> E["Model G3M Dynamics<br>Under Arbitrage"]
  D --> F["Extend Tassy & White Findings<br>to Balancer & G3Ms"]
  E --> G["Calculate Long-Term<br>Expected Logarithmic Growth"]
  F --> G
  G --> H["Key Outcomes: New Insights on<br>LP Profitability & AMM Dynamics"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper develops a tractable analytical framework to study the long‑term logarithmic growth rate of liquidity‑provider wealth in Geometric Mean Market Makers, explicitly accounting for continuous‑time arbitrage and transaction fees.

Source sentence
The goal of this paper is to address this question by developing a tractable analytical framework for studying the long-term growth of liquidity provider (LP) wealth in G3Ms, explicitly accounting for the interplay between continuous-time arbitrage and transaction fees. Our approach is motivated by the observation that, in a G3M, every trade moves the pool price toward the external market price, generating fee

Assumes 3

  • All trades are executed by arbitrageurs; no noise traders are present. model
    Source sentence
    Assumption A: No Noise Traders. We assume all trades are executed by arbitrageurs, with no noise traders present.
  • An external reference market with infinite liquidity and zero transaction costs exists, providing a frictionless price oracle. market
    Source sentence
    Assumption B: Frictionless External Reference Market. We assume the existence of an exter-nal reference price with infinite liquidity and no transaction costs, representing an aggregated oracle across all available liquidity sources — DEXs, CEXs, and routing solvers — rather than the price on any single venue.
  • Arbitrageurs monitor and act on arbitrage opportunities continuously and instantaneously. execution
    Source sentence
    Assumption C: Continuous Arbitrage. We assume arbitrageurs continuously monitor and act on arbitrage opportunities instantaneously.

Data

No dataset named in the text.

Costs

Not applicable The paper reports no trading or execution result; the cost the extractor found is a model parameter.

The paper-text extraction found a cost parameter inside a model (Fees are incorporated as a no‑arbitrage band that generates income for LPs while exposing them to adverse selection risk.); the classifier found no reported trading result, so costs are not applicable.

Finds 3

  • Even under the worst‑case assumption of only adversarial order flow, the LP wealth growth rate can outpace that of a constant‑rebalanced portfolio when the fee tier is chosen appropriately. theoretical
    Source sentence
    In particular, our results show that, from a portfolio construction perspective, short-term arbitrage losses can be transformed into long-run volatility harvesting through the fee mechanism, with the LP wealth growth rate outperforming the excess growth rate of a constant-rebalanced portfolio in Stochastic Portfolio Theory (SPT) [Fer02, KF09] — which itself assumes frictionless trading — when the fee tier is chosen
  • Liquidity in the pool consistently increases due to the transaction fees collected, as shown by the derived liquidity dynamics equation. theoretical
    Source sentence
    The liquidity ℓof the pool changes as follows: dℓ ℓ= wdx x + (1 −w)dy y = ( (1 −γ)(1 −w)dy y if dy > 0, (1 −1 γ)(1 −w)dy y if dy < 0. (10) This equation reveals that the liquidity consistently increases due to the transaction costs collected by the pool. Remark 2.3.
  • The long‑term growth rate of LP wealth can be expressed in closed form using the ergodic properties of a reflected diffusion, linking fee structures, market volatility, and portfolio weights. theoretical
    Source sentence
    The long-term growth rate of LP wealth can then be expressed in terms of the ergodic properties of this reflected diffusion, yielding explicit formulas that connect fee structures, market volatility, and portfolio weights. As a direct consequence, we obtain the following result for the long-term growth rate of LP wealth in the G3M model.

Checked by others

Supported by

  • Generalizing Impermanent Loss on Decentralized Exchanges with Constant Function Market Makers (2023)
    On the question: How do transaction fees affect pool depth and liquidity in constant function market makers?
    Compare the findings

    This paper: Liquidity in the pool consistently increases due to the transaction fees collected, as shown by the derived liquidity dynamics equation.

    That paper: Trade fees increase the pool depth (KX > K) because fees are added to incoming assets.

    Both findings state that fees raise pool depth (KX > K) and cause liquidity to grow; B’s dynamics equation confirms A’s fee‑induced depth increase.

  • Optimal Decisions for Liquid Staking: Allocation and Exit Timing (2025)
    On the question: How does fee accumulation affect liquidity provider wealth and optimal staking decisions?
    Compare the findings

    This paper: Liquidity in the pool consistently increases due to the transaction fees collected, as shown by the derived liquidity dynamics equation.

    That paper: Numerical analyses show that fee accumulation can make remaining in the pool beneficial despite price declines.

    Both findings (1) show fee accumulation drives wealth growth in G3Ms and (2) shows it can keep pool participation optimal even with price declines.

Cited by 1 paper in the archive

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (46)

  1. The Adoption of Blockchain-based Decentralized Exchanges (2021) — link
  2. Decentralized Finance & Blockchain Technology (2023)
  3. Automated market makers and decentralized exchanges: a DeFi primer (2022) — link
  4. Uniswap v2 Core (2020)
  5. Balancer: A non-custodial portfolio manager, liquidity provider, and price sensor (2019)
  6. Quantifying Loss in Automated Market Makers (2022) — link
  7. Automated Market Making and Loss-Versus-Rebalancing (2022) — link
  8. Automated Market Making and Arbitrage Profits in the Presence of Fees — → 2305.14604
and 38 more
  1. Brownian Models of Performance and Control (2013) — link
  2. Growth rate of a liquidity provider’s wealth in $xy=c$ automated market makers (2020)
  3. An arbitrage driven price dynamics of Automated Market Makers in the presence of fees — → 2401.01526
  4. Predictable Losses of Liquidity Provision in Constant Function Markets and Concentrated Liquidity Markets (2023) — link
  5. Decentralized Finance and Automated Market Making: Predictable Loss and Optimal Liquidity Provision (2024) — link
  6. An analysis of Uniswap markets (2019) — link
  7. Improved Price Oracles: Constant Function Market Makers (2020) — link
  8. Liquidity Provider Returns in Geometric Mean Markets (2021) — link
  9. Automated market makers and their implications for liquidity providers (2024) — link
  10. Stochastic Portfolio Theory (2002) — link
  11. An explicit formula for the Skorokhod map on [0,a] (2007) — link
  12. A stochastic model for order book dynamics (2010) — link
  13. Price dynamics in a Markovian limit order market (2013) — link
  14. A stochastic partial differential equation model for limit order book dynamics (2021) — link
  15. Lifetime Portfolio Selection under Uncertainty: The Continuous-Time Case (1969) — link
  16. Optimum consumption and portfolio rules in a continuous-time model (1971) — link
  17. Stochastic Portfolio Theory: an Overview (2009) — link
  18. On the asymptotic behavior of solutions of parabolic equations (1983) — link
  19. Central Limit Theorems and Large Deviations for Additive Functionals of Reflecting Diffusion Processes (2013) — link
  20. Asymptotic replication with modified volatility under small transaction costs (2016) — link
  21. Weighted variance swaps hedge against impermanent loss (2023) — link
  22. Model-Free Hedging of Impermanent Loss in Geometric Mean Market Makers with Proportional Transaction Fees (2024) — link
  23. A mathematical framework for modelling CLMM dynamics in continuous time — → 2412.18580
  24. Continuous martingales and Brownian motion (1999) — link
  25. Once more about the connection between elliptic operators and Itô’s stochastic equations (1985)
  26. Mimicking the one-dimensional marginal distributions of processes having an I t ^o differential (1986) — link
  27. One-dimensional stochastic differential equations with singular and degenerate coefficients (2005)
  28. Time inhomogeneous stochastic differential equations involving the local time of the unknown process, and associated parabolic operators (2018) — link
  29. Diffusion processes with boundary conditions (1971) — link
  30. Stochastic differential equations with reflecting boundary conditions (1984) — link
  31. Brownian motion and stochastic calculus (1991) — link
  32. Diffusions and elliptic operators (1998)
  33. Equilibrium Liquidity and Risk Offsetting in Decentralised Markets — → 2512.19838
  34. Coexisting Exchange Platforms: Limit Order Books and Automated Market Makers (2025) — link
  35. The Viability of Blockchain Markets under Discrete Clearing and Paid Priority (2025) — link
  36. The AMM Challenge (2024)
  37. Pricing and Hedging for Liquidity Provision in Constant Function Market Making — → 2603.01344
  38. Optimal Fees for Liquidity Provision in Automated Market Makers — → 2508.08152

6 of 46 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.