Profit equitably: An investigation of market maker's impact on equitable outcomes

TL;DR

This paper investigates how market makers influence fairness in trading environments, demonstrating that they support equitability but with a trade-off in profits, and proposes reinforcement learning to optimize both fairness and profitability.

Contribution

It introduces a simulation-based approach to quantify market equitability and applies multi-objective reinforcement learning to balance fairness and profit for market makers.

Findings

Market makers support trading equitability.

A negative correlation exists between market maker profits and equitability.

Reinforcement learning can optimize policies for fairness and profitability.

Abstract

We look at discovering the impact of market microstructure on equitability for market participants at public exchanges such as the New York Stock Exchange or NASDAQ. Are these environments equitable venues for low-frequency participants (such as retail investors)? In particular, can market makers contribute to equitability for these agents? We use a simulator to assess the effect a market marker can have on equality of outcomes for consumer or retail traders by adjusting its parameters. Upon numerically quantifying market equitability by the entropy of the price returns distribution of consumer agents, we demonstrate that market makers indeed support equitability and that a negative correlation is observed between the profits of the market maker and equitability. We then use multi objective reinforcement learning to concurrently optimize for the two objectives of consumer agent equitability and market maker profitability, which leads us to learn policies that facilitate lower market volatility and tighter spreads for comparable profit levels.