Hallmarks of Deception in Asset-Exchange Models

TL;DR

This paper analyzes how probabilistic cheating affects wealth distribution and dynamics in asset exchange models, revealing critical thresholds and signatures of deception.

Contribution

It provides an analytical framework for detecting hidden cheaters and identifies key thresholds where wealth and behavior change abruptly.

Findings

Steady-state wealth distribution derived analytically.

Detection of cheaters from wealth variance.

Identification of critical cheating probabilities causing phase transitions.

Abstract

We investigate the transient and steady-state dynamics of the Bennati-Dragulescu-Yakovenko money game in the presence of probabilistic cheaters, who can misrepresent their financial status by claiming to have no money. We derive the steady-state wealth distribution per player analytically, and show how the presence of hidden cheaters can be inferred from the relative variance of wealth per player. In scenarios with a finite number of cheaters amidst an infinite pool of honest players, we identify a critical probability of cheating at which the total wealth owned by the cheaters experiences a second-order discontinuity. Below this point, the transition probability to lose money is larger than the probability to gain; conversely, above this point, the direction is reversed. We further establish a threshold cheating probability at which cheaters collectively possess half of the total wealth in the game. Lastly, we provide bounds on the rate at which both cheaters and honest players can gain or lose wealth, contributing to a deeper understanding of deception in asset exchange models.