Abstract:In coordination games and speculative over-the-counter financial markets, solutions depend on higher-order average expectations: agents' expectations about what counterparties, on average, expect their counterparties to think, etc. We offer a unified analysis of these objects and their limits, for general information structures, priors, and networks of counterparty relationships. Our key device is an interaction structure combining the network and agents' beliefs, which we analyze using Markov methods. This device allows us to nest classical beauty contests and network games within one model and unify their results. Two applications illustrate the techniques: The first characterizes when slight optimism about counterparties' average expectations leads to contagion of optimism and extreme asset prices. The second describes the tyranny of the least-informed: agents coordinating on the prior expectations of the one with the worst private information, despite all having nearly common certainty, based on precise private signals, of the ex post optimal action.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is that in coordination games and speculative over - the - counter financial markets, the results depend on higher - order average expectations. Specifically, agents' expectations are based not only on their direct knowledge of the state of the world, but also on their understanding of the expectations of other counterparties, and on these counterparties' views of the expectations of other counterparties, etc. The goal of the paper is to conduct a unified analysis of these objects and their limits, applicable to general information structures, prior beliefs, and counterparty relationship networks. The author achieves this goal by introducing an interaction structure that combines the network and agents' beliefs and analyzing it using the Markov method. This method allows classic beauty contests and network games to be embedded in a model and unifies their results.
The paper mainly focuses on the following points:
1. **Unifying and generalizing network and asymmetric information results**: Study how, when agents have the same information but may have different prior beliefs, the consensus expectation becomes a weighted average of the agents' different prior expectations, with the weights determined by the eigenvector centrality in the network. In addition, when there is asymmetric information but agents have a common prior belief, the consensus expectation is equal to the prior expectation of the external state, independent of other features of the network structure and information structure.
2. **Contagion of optimism**: Study how, when each agent, given a signal, believes that its average counterparty is more optimistic about the value of an external random variable than itself, a small second - order optimism can lead to a very high consensus expectation, that is, close to the highest possible expected value \(y\), and is achieved through the contagion mechanism of higher - order expectations.
3. **The tyranny of the least - informed**: In the context where agents have heterogeneous prior beliefs about the external state but a common interpretation of the signal, study how the consensus expectation approximates the prior expectation of the agent with the least precise private information, even if all agents have very precise private signals about the state.
Through these studies, the paper aims to reveal how information and beliefs interact and spread through the structure of higher - order expectations in complex social and economic networks.