How Fixed-Amount Transactions and Liquidity Constraints Amplify Wealth Inequality: A Kinetic Model Deviating from the Maximum Entropy Benchmark

Jihyuan Liuh · 2025

Read How Fixed-Amount Transactions and Liquidity Constraints Amplify Wealth Inequality: A Kinetic Model Deviating from the Maximum Entropy Benchmark by Jihyuan Liuh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-08202.txt.

Introduction

Jiyuan Lyu gives an exact mean field solution for the Mesa Boltzmann wealth model, a standard agent based example of money exchange. The paper replaces simulation with exact analysis for a model that is usually studied only on a computer. Instead of local trades, the mean field version uses global random matching with synchronous updates.

A master equation for the wealth distribution, solved through probability generating functions, yields the steady state in closed form. Variance, the Gini inequality coefficient, and tail behavior follow directly from that solution. Simulations of the matching agent model agree closely with the formulas, giving an analytical benchmark for checking related simulations.

Catalog record

CREATORJihyuan Liuh
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE28 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-08202

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

How Fixed-Amount Transactions and Liquidity Constraints Amplify Wealth Inequality: A Kinetic Model Deviating from the Maximum Entropy Benchmark Jihyuan Liuh Abstract This paper investigates the emergence of wealth inequality through a minimalist kinetic exchange model that incorporates two fundamental economic features: fixed-amount transactions and hard budget constraints. In contrast to the maximum entropy principle, which predicts an exponential Boltzmann-Gibbs distribution with moderate inequality (Gini ≈ \approx 0.333) for unconstrained wealth exchange, we demonstrate that these realistic trading rules drive the system toward a highly unequal steady state. We develop a self-consistent mean-field theory, deriving a master equation where agent income follows a Poisson process coupled to the poverty rate. Numerical solution reveals a stationary distribution characterized by a substantial pauper class ( p 0 ≈ 55.1 % p_{0}\approx 55.1\% ), high Gini coefficient ( G ≈ 0.649 G\approx 0.649 ), and exponential tail—significantly deviating from the maximum entropy benchmark. Agent-based simulations confirm these findings ( p 0 ≈ 45.2 % p_{0}\approx 45.2\% , G ≈ 0.618 G\approx 0.618 ). We identify the poverty trap as the key mechanism: the liquidity constraint creates asymmetric economic agency, where zero-wealth agents become passive recipients, unable to participate in wealth circulation. This work establishes that substantial inequality can emerge spontaneously from equal-opportunity exchanges under basic economic constraints, without requiring agent heterogeneity or multiplicative advantage, providing a mechanistic foundation for understanding poverty as an emergent property of exchange rules. Keywords: Wealth Inequality, Poverty Trap, Kinetic Exchange Models, Maximum Entropy Principle, Mean-Field Theory, Econophysics, Agent-Based Modeling, Stationary Distribution, Gini Coefficient JEL Codes: O33. 1 Introduction 1.1 The Pervasiveness of Wealth Inequality The unequal distribution of wealth is a ubiquitous and persistent feature of nearly all modern economies [ 9 ] . Understanding the fundamental mechanisms that give rise to such macroscopic inequality from microscopic interactions remains a central challenge across disciplines, including economics, statistical physics, and complexity science. 1.2 Theoretical Background: Kinetic Models and the Maximum Entropy Benchmark A powerful framework for tackling this problem originates from statistical mechanics, employing agent-based stochastic models inspired by kinetic theory. In these models, wealth is redistributed among a population through simple, local exchange rules. The most prominent classes are the kinetic wealth exchange models [ 2 , 5 ] and the Yard-Sale model [ 3 ] . In their simplest forms, with unconstrained random exchanges, these models often lead to a stationary distribution that can be derived from the principle of maximum entropy [ 7 ] . This principle states that the probability distribution which best represents the current state of knowledge, constrained only by the conservation of total wealth, is the one with the largest Shannon entropy. This yields the exponential Boltzmann-Gibbs distribution [ 5 ] . For a system with a mean wealth of 1 unit, this distribution takes the form p_{k}=(1/2)^{k+1} and is characterized by a low level of fundamental inequality , with a Gini coefficient of approximately G\approx 0.333 . This can be considered the null model or benchmark for inequality arising from pure randomness under wealth conservation, absent any other structural constraints. 1.3 Research Gap and Problem Formulation However, many canonical models employ exchanges where the transaction amount is a fraction of an agent’s wealth or a randomly varying amount [ 2 , 8 ] . While mathematically tractable, this arguably differs from a common real-world scenario: many everyday transactions have a fixed, minimum unit or a typical scale (e.g., the price of a staple good). Furthermore, a crucial, often implicit constraint in economic interactions is liquidity : an agent cannot spend what they do not have. This simple, hard budget constraint can have profound implications for wealth dynamics. This leads to our core research questions: When a closed economic system is governed by 1) fixed-amount transactions and 2) a hard budget constraint , does its macroscopic steady state still align with the maximum entropy benchmark? If not, how do these specific rules alter the distribution of wealth, and what is the underlying mechanism? 1.4 Our Work and Central Thesis In this paper, we introduce and analyze a minimalist wealth exchange model that incorporates these two features. Our central thesis is that the interplay between fixed transactions and the liquidity constraint spontaneously generates a poverty trap , leading to a stationary state that significantly deviates from the maximum entropy benchmark . This emergent state exhibits a stable impoverished class and a Gini coefficient substantially higher than the theoretical baseline of 0.333, demonstrating that substantial inequality can arise from stochastic, equal-opportunity exchanges under fundamental liquidity constraints, without any assumed heterogeneity. We establish this result through a multi-faceted approach: we develop a self-consistent mean-field theory leading to a master equation, provide a precise numerical solution for the stationary distribution, and validate our findings with agent-based simulations . The convergence of all three methods confirms the robustness of our findings and provides a clear, mechanistic explanation for the emergence of poverty and heightened inequality in this parsimonious framework. 2 Model and the Maximum Entropy Benchmark 2.1 Model Definition We consider a closed economic system consisting of N agents, indexed by i=1,2,\dots,N . The total wealth is conserved and normalized such that the initial wealth per agent is 1, yielding a total wealth of N and a mean wealth \langle w\rangle=1 . The dynamics proceed in discrete time steps (rounds). In each round, the following process occurs simultaneously for all agents: 1. Expenditure Decision: Each agent i checks their current wealth, w_{i}(t) . If w_{i}(t)\geq 1 , the agent is active and must send a fixed payment of exactly 1 unit. If w_{i}(t)=0 , the agent is inactive and makes no payment. 2. Recipient Selection: Each active agent selects a target recipient j uniformly at random from the set of all N agents (including themselves). The payment of 1 unit is transferred from agent i to agent j . This parallel update rule is crucial for mathematical tractability and avoids artifacts from sequential updates. The system state is described by the wealth vector \mathbf{w}(t)=(w_{1}(t),w_{2}(t),\dots,w_{N}(t)) , which evolves stochastically under two key constraints: • Wealth Conservation: \sum_{i=1}^{N}w_{i}(t)=N for all t . • Non-Negative Wealth: w_{i}(t)\geq 0 due to the hard budget constraint. The interplay between the fixed transaction amount and the liquidity constraint is the central feature of this model. 2.2 Maximum Entropy Principle and the Ideal Benchmark Distribution To establish a theoretical baseline, we apply the principle of maximum entropy [ 7 ] . This principle identifies the least biased probability distribution consistent with known constraints, which corresponds to the most random or disordered macrostate. For a large population of agents with non-negative integer wealth k , we seek the distribution p_{k} that maximizes the Shannon entropy H=-\sum_{k=0}^{\infty}p_{k}\ln p_{k}, subject to the following two constraints: 1. Normalization: \sum_{k=0}^{\infty}p_{k}=1 2. Mean Wealth Constraint: \sum_{k=0}^{\infty}kp_{k}=1 Using the method of Lagrange multipliers, we define the Lagrangian: \mathcal{L}=-\sum_{k}p_{k}\ln p_{k}+\lambda_{0}\left(1-\sum_{k}p_{k}\right)+\lambda_{1}\left(1-\sum_{k}kp_{k}\right). Taking the variational derivative with respect to p_{k} and setting it to zero yields: \frac{\partial\mathcal{L}}{\partial p_{k}}=-\ln p_{k}-1-\lambda_{0}-\lambda_{1}k=0. Solving for p_{k} : p_{k}=e^{-1-\lambda_{0}}e^{-\lambda_{1}k}=\frac{1}{Z}e^{-\lambda_{1}k}, where Z=e^{1+\lambda_{0}} is the partition function. This is the form of the Boltzmann-Gibbs distribution [ 5 ] . The constants are determined by the constraints. From the normalization condition: \sum_{k=0}^{\infty}p_{k}=\frac{1}{Z}\sum_{k=0}^{\infty}(e^{-\lambda_{1}})^{k}=\frac{1}{Z}\cdot\frac{1}{1-e^{-\lambda_{1}}}=1\quad\Rightarrow\quad Z=\frac{1}{1-e^{-\lambda_{1}}}. From the mean wealth constraint: \sum_{k=0}^{\infty}kp_{k}=\frac{1}{Z}\sum_{k=0}^{\infty}k(e^{-\lambda_{1}})^{k}=\frac{1}{Z}\cdot\frac{e^{-\lambda_{1}}}{(1-e^{-\lambda_{1}})^{2}}=1. Substituting Z : (1-e^{-\lambda_{1}})\cdot\frac{e^{-\lambda_{1}}}{(1-e^{-\lambda_{1}})^{2}}=\frac{e^{-\lambda_{1}}}{1-e^{-\lambda_{1}}}=1.

Read the complete text