Paper: arXiv 2512.19838

Authors: Fayçal Drissi, Xuchen Wu, Sebastian Jaimungal

Abstract

We develop an economic model of decentralised exchanges (DEXs) in which risk-averse liquidity providers (LPs) manage risk in a centralised exchange (CEX) based on preferences, information, and trading costs. Rational, risk-averse LPs anticipate the frictions associated with replication and manage risk primarily by reducing the reserves supplied to the DEX. Greater aversion reduces the equilibrium viability of liquidity provision, resulting in thinner markets and lower trading volumes. Greater uninformed demand supports deeper liquidity, whereas higher fundamental price volatility erodes it. Finally, while moderate anticipated price changes can improve LP performance, larger changes require more intensive trading in the CEX, generate higher replication costs, and induce LPs to reduce liquidity supply.

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 employs advanced stochastic control, forward-backward stochastic differential equations (FBSDEs), differential Riccati equations, and variational methods, indicating very high mathematical complexity. However, it is purely theoretical with no backtests, datasets, or implementation details, resulting in low empirical rigor.

Research Flowchart

  flowchart TD
  R["Research Goal<br>Liquidity in Decentralized Exchanges"] --> M["Methodology<br>Rational Risk-Averse LP Model"]
  M --> I["Key Inputs<br>Risk Aversion, Volatility, Demand"]
  I --> C["Computational Process<br>Cost-Benefit Analysis & Optimization"]
  C --> F["Key Findings<br>Thinner Markets, Risk Management, Volume Impact"]
  F --> O["Outcomes<br>Conditions for DEX Liquidity Viability"]

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 structural model of liquidity provision in decentralized exchanges, endogenising market outcomes and characterising how liquidity depth, rewards and risks arise from risk preferences, costly risk management, price volatility, private information and strategic liquidity demand.

Source sentence
We develop a structural model in which DEX market outcomes are endogenised. Our analysis characterises how the size, rewards, and risks of DEX liquidity provision arise from the interplay between risk preferences, costly risk management, fundamental price volatility, access to private information, and strategic liquidity demand.

Assumes 6

  • The trading function f is three times continuously differentiable on the positive quadrant with strictly positive partial derivatives. model
    Source sentence
    (i) f ∈C3((0, ∞)2) and has strictly positive partial derivatives.
  • For each y>0, f(·,y) is surjective, yielding a unique level function φ(y,κ) for each liquidity depth κ. model
    Source sentence
    (ii) For each y > 0, f(·, y) : (0, ∞) →(0, ∞) is surjective. Thus, for each κ > 0, the level set f(x, y) = κ2 admits a unique solution x = φ(y, κ) .
  • The ratio R = ∂2f/∂1f satisfies R∂1R−∂2R>0 everywhere and is decreasing in κ. model
    Source sentence
    (iii) R := ∂2f ∂1f satisfies R ∂1R −∂2R > 0 everywhere, and is decreasing in κ.
  • The derivative of the level function φ satisfies lim_{y→0}∂1φ(y,κ)=−∞ and lim_{y→∞}∂1φ(y,κ)=0. model
    Source sentence
    (iv) ∂1φ satisfies the limits limy↓0 ∂1φ(y, κ) = −∞ and limy↑∞∂1φ(y, κ) = 0 .
Show 2 more
  • Liquidity takers pay a proportional fee π∈(0,1) on each trade in the DEX. execution
    Source sentence
    8 convexity of the level function φ, liquidity takers pay a proportional fee π ∈(0, 1) when transacting in the DEX.
  • A penalty parameter ϕ>0 scales the aversion to net inventory exposure, with higher ϕ implying greater risk aversion. model
    Source sentence
    This penalty is scaled by a parameter governing the degree of risk aversion of the LP. The parameter ϕ > 0 scales this penalty: larger values correspond to greater aversion to holding non-zero net exposure Qν t + Yt.

Data

No dataset named in the text.

Costs

Modeled Costs are part of the reported headline results.

Liquidity takers incur a proportional fee π∈(0,1) on each transaction, which determines fee revenue and trading costs.

Source sentence
8 convexity of the level function φ, liquidity takers pay a proportional fee π ∈(0, 1) when transacting in the DEX.

Finds 5

  • An equilibrium exists under general trading functions. theoretical
    Source sentence
    We establish existence of equilibrium under general trading functions.
  • DEX liquidity depth κ balances amplified inventory exposure and fee revenue, with the equilibrium level reflecting this trade‑off. theoretical
    Source sentence
    Thus, in stage one, the LP anticipates that liquidity depth simultaneously (i) amplifies inventory exposure, risk-offsetting costs, and adverse selection costs, and (ii) increases fee revenue. The equilibrium level of κ at this stage reflects the balance between these opposing forces.
  • The ratio of risk aversion to replication costs governs the aggressiveness of replication: greater relative risk aversion reduces risk but also lowers equilibrium liquidity and its mean profitability. theoretical
    Source sentence
    The ratio governs the aggressiveness of replication: greater relative risk aversion reduces risk but also lowers equilibrium liquidity and its mean profitability.
  • Private information has a non‑monotonic effect: moderate price movements increase liquidity via speculative benefits, while large movements reduce liquidity due to anticipated adverse selection and replication costs. theoretical
    Source sentence
    Private information has a non-monotonic effect. For moderate price movements, speculative benefits increase liquidity.
  • Deeper liquidity κ raises exposure to price fluctuations and expected replication costs, creating incentives to reduce liquidity provision, which are offset by fee revenue from price‑sensitive liquidity demand. theoretical
    Source sentence
    Deeper liquidity κ also increases the magnitude of reserves Yt and therefore amplifies exposure to price fluctuations dFt. When inventory risk management through dynamic trading is costly, greater exposure implies higher expected future replication costs.

Checked by others

Supported by

  • Equilibrium Investment with Random Risk Aversion: (Non-)uniqueness, Optimality, and Comparative Statics (2025)
    On the question: How does higher risk aversion affect investment risk and market liquidity?
    Compare the findings

    This paper: The ratio of risk aversion to replication costs governs the aggressiveness of replication: greater relative risk aversion reduces risk but also lowers equilibrium liquidity and its mean profitability.

    That paper: A larger risk aversion under the reverse hazard rate order always leads to less risky investment.

    Both agree higher risk aversion reduces risk (A: less risky investment; B: lower risk). B further finds it also lowers equilibrium liquidity and mean profitability.

Cited by 3 papers in the archive (3 by independent authors)

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 (53)

  1. Incorporating signals into optimal trading (2019)
  2. Optimal fees for liquidity provision in automated market makers — → 2508.08152
  3. Convex analysis and variational problems (1999)
  4. Who wins ethereum block building auctions and why? (2024) — link
  5. Flash boys 2.0: Frontrunning in decentralized exchanges, miner extractable value, and consensus instability (2020)
  6. The Price of Liquidity: Implied Volatility of Automated Market Maker Fees — → 2509.23222
  7. Optimal Dynamic Fees in Automated Market Makers — → 2506.02869
  8. Optimal exit time for liquidity providers in automated market makers — → 2509.06510
and 45 more
  1. Models of market liquidity: Applications to traditional markets and automated market makers (2023)
  2. Decentralized finance and automated market making: Predictable loss and optimal liquidity provision (2024)
  3. Predictable losses of liquidity provision in constant function markets and concentrated liquidity markets (2023)
  4. Broker-Trader Partial Information Nash-Equilibria — → 2412.17712
  5. The need for fees at a dex: How increases in fees can increase dex trading volume (2022)
  6. The conceptual flaws of decentralized automated market making (2023)
  7. Informed Liquidity Provision on Decentralized Exchanges (2023)
  8. Decentralized exchange: The uniswap automated market maker (2021)
  9. Decentralized finance: Protocols, risks, and governance — → 2312.01018
  10. Cryptofinance (2016)
  11. DeFi and the Future of Finance (2021)
  12. Blockchain disruption and smart contracts (2019)
  13. Smart contracts and decentralized finance (2023)
  14. Replicating Market Makers (2021) — link
  15. The Evolution of Decentralized Exchange: Risks, Benefits, and Oversight (2024)
  16. Optimal trading in automatic market makers with deep learning — → 2304.02180
  17. Do Longer Block Times Impair Market Efficiency in Decentralized Markets? (2025)
  18. Just-in-time liquidity on the uniswap protocol (2022)
  19. The paradox of just-in-time liquidity in decentralized exchanges: More providers can sometimes mean less liquidity — → 2311.18164
  20. Automated market making and arbitrage profits in the presence of fees — → 2305.14604
  21. Hedging with temporary price impact (2017)
  22. Axioms for automated market makers: A mathematical framework in fintech and decentralized finance (2025)
  23. Coexisting exchange platforms: Limit order books and automated market makers (2025)
  24. Optimal exit from Uniswap v3 and best expected return for a liquidity provider (2025)
  25. The economics of constant function market makers (2025)
  26. Clustering Uniswap v3 traders from their activity on multiple liquidity pools, via novel graph embeddings (2024)
  27. Quantitative modelling and analysis of the Automated Market Maker Uniswap (2024)
  28. Finding the right curve: Optimal design of constant function market makers (2023)
  29. Learning from DEFI: Would automated market makers improve equity trading? (2024)
  30. Automated market making and loss-versus-rebalancing (2022) — link
  31. Arbitrage on decentralized exchanges — → 2507.08302
  32. Optimal design of automated market makers on decentralized exchanges — → 2404.13291
  33. DeFi Arbitrage in Hedged Liquidity Tokens — → 2409.11339
  34. Decentralised finance and automated market making: Execution and speculation — → 2307.03499
  35. Strategic bonding curves in automated market makers (2024)
  36. A closed-form execution strategy to target volume weighted average price (2016)
  37. The Financial Mathematics of Market Liquidity: From optimal execution to market making (2016)
  38. Algorithmic and high-frequency trading (2015)
  39. Mean-field games with differing beliefs for algorithmic trading (2020)
  40. Mean field games with partial information for algorithmic trading (2018) — link
  41. Latency and liquidity risk (2021)
  42. Trading algorithms with learning in latent alpha models (2019)
  43. Optimal Execution: A Review (2022)
  44. Incorporating order-flow into optimal execution (2016)
  45. Non-blow-up conditions for Riccati-type matrix differential and difference equations (2000)

13 of 53 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.