On Trade in Bilateral Oligopolies with Altruistic and Spiteful Agents

Michele Lombardi, Michele Lombardi 0002, Simone ToninarXiv 2018 · arXiv

Introduction

Traditional economic theory often operates under the assumption of homo economicus—the perfectly selfish agent who seeks only to maximize their own consumption. However, real-world behavior frequently deviates from this model. Humans exhibit "interdependent preferences," where an individual's utility depends not only on their own holdings but also on the welfare of others. This research investigates how such social preferences—specifically altruism and spite—influence trade within a bilateral oligopoly.

The Strategic Framework: Bilateral Oligopoly and Social Utility

The study focuses on a bilateral oligopoly, a market structure characterized by a small number of strategic players on two sides (e.g., two groups of sellers trading distinct commodities). To model this, the authors utilize the Shapley-Shubik mechanism. In this system, prices are not set by an external auctioneer but are determined endogenously by the ratio of total bids to total offers.

To incorporate social preferences, the authors employ Edgeworth Utility Functions. In this framework, an agent's total "social" utility is the sum of their "internal" utility (derived from their own consumption) and the weighted internal utilities of all other agents in the market. These weights (represented by $\gamma$) define the agent's disposition:

  • Altruism ($\gamma > 0$): The agent gains utility when others are better off.
  • Spite ($\gamma < 0$): The agent gains utility when others are worse off.
  • Selfishness ($\gamma = 0$): The agent is indifferent to others' welfare.

The Dual Nature of Altruism

One of the paper's most significant contributions is the distinction between the effects of altruism based on its direction.

Cross-Side Altruism: A Catalyst for Trade

When agents care about the welfare of those on the opposite side of the market (cross-side altruism), trade is incentivized. By offering goods or bidding more, an agent directly improves the utility of their trading partners. This creates a "pro-trade" environment where the existence of a trade equilibrium—a state where goods are actively exchanged—is mathematically guaranteed. The authors prove that this holds true even in strategic settings where agents have the power to move prices.

Same-Side Altruism: The Market Killer

Conversely, caring for one's "peers" (same-side altruism) can lead to market collapse. In a Shapley-Shubik model, increasing one's offer typically lowers the market price. While this benefits buyers, it hurts other sellers on the same side. An altruistic seller, realizing that their participation might "cannibalize" the profits of their peers, may choose to withhold trade entirely. This paradoxically results in a "non-trade" equilibrium where no exchange occurs, leaving everyone with only their initial endowments.

Spite and Mathematical Instability

Spitefulness introduces even more disruption. Spite directed at the opposite side leads agents to restrict trade to prevent the "enemy" from gaining any advantage. More technically, spite toward the same side (competitive spite) can make payoff functions non-concave. In game theory, non-concavity often means that a stable Nash equilibrium—where no player has an incentive to deviate—simply does not exist for any positive volume of trade.

Key Results and the "Non-Trade" Trap

The authors provide a rigorous existence theorem, proving that trade equilibria always exist under cross-side altruism, generalizing previous results for purely selfish markets. However, they also highlight the "non-trade trap." Through numerical examples, they demonstrate that under same-side altruism or spite, the only stable equilibrium is often the trivial one: zero trade.

The collapse of trade in these models happens through two primary failure modes:

  1. Negative Optimality: The first-order Kuhn-Tucker conditions for utility maximization become negative, meaning it is never rational to start trading.
  2. Structural Instability: The mathematical properties required for equilibrium (like concavity) vanish, preventing the formation of a stable trade strategy.

Discussion and Future Directions

The current findings are based on "corner endowments," where agents start with only one type of commodity. It remains an open question whether these negative results persist if agents begin with a mix of goods (interior endowments). Additionally, the study notes that while selfish oligopolies converge toward perfect competition as the number of agents increases, this convergence is not guaranteed when social preferences are present. Finally, the model assumes fixed social weights, leaving the door open for future research into reciprocal altruism, where social ties evolve based on the behavior of others.

Summary updated

BibTeX
@misc{lombardi2018trade,
  author = {Michele Lombardi and Michele Lombardi and Simone Tonin},
  title = {{On Trade in Bilateral Oligopolies with Altruistic and Spiteful Agents}},
  year = {2018},
  eprint = {1803.03994},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1803.03994},
}
Figure