The emergence of social inequality: a co-evolutionary analysis.

Oliveira, Fernando S.

Abstract

Social inequality is an important issue at the core of economic thought. Based on a stylized pie-sharing game, this article proposes a co-evolutionary computational model to study the interaction between social classes, aiming to answer the following questions: Why do social classes exist, and how do they affect social efficiency and individual effectiveness? Why are wealth inequality and social exclusion persistent? From a methodological perspective, this article extends the pie-sharing game to include a network of interactions and classes, social mobility, evolving wealth, and learning agents. The results show that social inequality is self-emergent. Surprisingly, in the simulations, the existence of social classes increases individual effectiveness, mainly benefiting the poor. Nonetheless, flatter societies (where social classes exist) have a higher average individual effectiveness. As expected, it was observed that wealth inequality is persistent in hierarchical societies, and the upper classes keep a higher proportion of wealth. Furthermore, the analysis extends the bargaining game to include social mobility, showing that, surprisingly, it increases the robustness of the class system and wealth inequality. Finally, the simulation of dual societies shows that these are an evolutionary equilibrium: social exclusion is persistent and accepted by wealthy and poor individuals, and resources are efficiently used.

Citation

OLIVEIRA, F.S. 2023. The emergence of social inequality: a co-evolutionary analysis. Journal of economic behavior and organization [online], 215, pages 192-206. Available from: https://doi.org/10.1016/j.jebo.2023.06.003

• Journal Article Type: Article
• Acceptance Date: Jun 7, 2023
• Online Publication Date: Sep 23, 2023
• Publication Date: Nov 30, 2023
• Deposit Date: Sep 29, 2023
• Publicly Available Date: Sep 29, 2023
• Journal: Journal of economic behavior and organization
• Print ISSN: 0167-2681
• Electronic ISSN: 1879-1751
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 215
• Pages: 192-206
• DOI: https://doi.org/10.1016/j.jebo.2023.06.003
• Keywords: Agent-based model; Evolutionary games; Inequality; Reinforcement learning
• Public URL: https://rgu-repository.worktribe.com/output/2093042