Abstract

Motivated by the growth of various networked systems as potential market places, we study market models wherein, owing to the size of the markets, transactions take place between largely unknown agents. In such scenarios, intermediaries or brokers play a significant role in a transaction. We analyze market behavior in large networks wherein all sellers are not known to the buyers and vice-versa and depend on intermediaries to conduct any transactions. In such markets, we study a specific case where buyers wish to purchase goods from trusted sources at minimal prices. Sellers wish to maximize selling price. Brokers attempt to maximize profit by aiding in trade by acting as intermediaries; brokers have an advertising budget. We show the existence of competitive equilibria in such layered broker markets. We also describe efficient algorithms to compute these equilibria. We give polynomial-time distributed mechanisms to reach the equilibrium for two extreme cases of the brokers’ advertising budget constraints.