Paper: arXiv 2512.04603
Authors: Alexander Barzykin, Robert Boyce, Eyal Neuman
Abstract
As the FX markets continue to evolve, many institutions have started offering passive access to their internal liquidity pools. Market makers act as principal and have the opportunity to fill those orders as part of their risk management, or they may choose to adjust pricing to their external OTC franchise to facilitate the matching flow. It is, a priori, unclear how the strategies managing internal liquidity should depend on market condions, the market maker’s risk appetite, and the placement algorithms deployed by participating clients. The market maker’s actions in the presence of passive orders are relevant not only for their own objectives, but also for those liquidity providers who have certain expectations of the execution speed. In this work, we investigate the optimal multi-objective strategy of a market maker with an option to take liquidity on an internal exchange, and draw important qualitative insights for real-world trading.
Complexity vs Empirical Score
- Math Complexity: 9.0/10
- Empirical Rigor: 4.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper presents a complex stochastic control problem with a Hamilton–Jacobi–Bellman quasi-variational inequality (HJBQVI) requiring numerical solution, indicating high mathematical density. However, the work is purely theoretical with no backtests, code, or real-world data implementation details, placing it firmly in the theoretical ‘Lab Rats’ category.
Research Flowchart
flowchart TD A["Research Goal:<br>Optimal MM Strategy for Internal Liquidity"] B["Key Inputs:<br>Market Conditions, Risk Appetite,<br>Client Algo Parameters"] C["Methodology:<br>Multi-Objective Optimization Model"] D["Process:<br>Simulation & Numerical Analysis"] E["Outcome 1:<br>Dynamic Pricing Thresholds"] F["Outcome 2:<br>Optimal Inventory Management"] A --> B B --> C C --> D D --> E D --> F
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 proposes a quantitative model of a market maker who streams prices to external clients while optimally timing trades with internal liquidity providers, and derives the optimal multi‑objective strategy.
Source sentence
In this work, we present the first quantitative investigation of internal exchange management from the market maker’s perspective. We develop a model in which the market maker continuously streams a price ladder of multiple sizes to external, liquidity‑taking clients while optimally timing trades with liquidity‑providi
Assumes 5
- The mid‑price follows a Brownian motion: St = S0 + σWt. model
Source sentence
Let (St)t≥0 be the price process of a risky asset such that St = S0 + σWt, where W is a standard Brownian motion and S0, σ are positive constants.
- OTC client order arrivals are independent Poisson processes with intensity λ exp(−κzδ). execution
Source sentence
Market buy and sell orders of OTC clients are modelled by independent counting processes
- Internal exchange limit order arrivals follow a Poisson process with intensity µ and unit jump size, occurring only when no order is present. execution
Source sentence
... we restrict arrivals to occur only when no order is currently present.
- Replenishment after the dealer consumes the last unit occurs with probability p (Bernoulli). model
Source sentence
(Rt)t≥0 represents the replenishment process of an order immediately after the dealer consumes the last unit of liquidity.
Show 1 more
- The internal order price offset is ρ = ˜ρ − ξ, where ξ > 0 is a constant fee per unit. model
Source sentence
From the modeling perspective this can be incorporated by letting ρ = ˜ρ−ξ, where ˜ρ is the price offset relative to the mid chosen by the client, and ξ > 0 is a constant representing the fee per unit.
Data
- Simulated trajectories synthetic unknown
Source sentence
the number of simulated trajectories is 5,000.
Costs
Modeled
A constant fee per unit ξ is included in the price offset ρ = ˜ρ−ξ.
Source sentence
From the modeling perspective this can be incorporated by letting ρ = ˜ρ−ξ, where ˜ρ is the price offset
Finds 4
- The optimal strategy consistently outperforms the naïve benchmark strategy in P&L. performance
Source sentence
We observe that the optimal strategy consistently outperforms the naïve benchmark, and that the market maker’s profits are slightly more sensitive to the client’s pricing offset ˜ρ ...
- Under the optimal strategy, internal order fill times are longer than under the naïve benchmark. performance
Source sentence
Although execution times under the optimal strategy are slower than those of the naïve strategy, the latter substantially reduces the market maker’s P&L.
- Fees improve both P&L and internal‑exchange turnover for the optimal strategy, while margins mainly improve P&L but reduce volume. performance
Source sentence
... In contrast, the optimal strategy benefits from fees, improving both P&L and turnover.
- In iceberg scenarios, ask‑side quotes are lower when the market maker holds a short inventory, reflecting a willingness to accumulate inventory via internal liquidity. theoretical
Source sentence
... that, in the iceberg scenario, the ask‑side quotes are lower near the execution region (i.e., when the market maker has a short position greater than one unit, q < −1)
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 58 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers 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.