Israeli node

Leader: Elchanan Ben-Porath

The Israeli group consists of game theorists who are interested in applying solution concepts that have been developed in the game-theoretical literature to problems that involve the functioning of .nancial markets and competitive economies. Three more specific issues that members in the group have worked on are: (1) The information conveyed by prices when there is common knowledge of rationality but where agents may have heterogenous beliefs on the relationship between the prices and the states of nature. (2) Adaptive heuristics in dynamic economies. (3) Applications of the theory of global games to problems in .financial markets and Macroeconomics. The members that are currently affiliated with the group are: Elchanan Ben-Porath, Aviad Heifetz, and Ady Pauzner. Sergiu Hart plans to participate in the activities of the group but since he is now fully financed by an ERC grant prefers not to be listed as a member of the group. Following is a brief description of the research of members in the group that is relevant to the theme of the project.

Ben-Porath and Heifetz study exchange economies with incomplete information and characterize the set of outcomes that is consistent with common knowledge of rationality when different agents may have different beliefs about the relationship between the prices and the states of nature. In a second paper this result is applied to an economy where agents trade in contingent assets. The main result characterizes the set of prices that are consistent with common knowledge rationality in terms of the information partitions and initial endowments of the agents. Both papers highlight the dramatic difference between the predictions of Rational Expectations Equilibrium which typically has a unique solution and the multiplicity of the outcomes that are consistent with common knowledge of rationality.

In a series of papers Sergiu Hart has studied the implications of adaptive heuristics, notably regeret-matching, in repeated games (most of this work is joint with Andreu Mas-Colell). A heuristic is a rule of behavior that is simple – a .rule of thumb”. A heuristic is adaptive if it induces a behaviour that reacts to what happens in the environment in directions that loosely speaking seem better. Sergiu is interested in extending this approach to the analysis of dynamic economies.

Ady Pauzner.s research has focused on extending the theory of global games (which offers a theoretical tool that selects among multiple equilibria) and applying it to Financial markets and Macroeconomic models. In particular, Ady (in a joint work with Itay Goldstein) developed a model which selects among the two equilibria in the classical bank runs model of Diamond and Dybvig. The model relates the probability of a bank run to the parameters of the demand deposit contract that is offered by the bank to the depositors. Another paper (also, with Goldstein) uses the same theoretical tools to investigate the possibility that a financial crisis will cross country borders even when there are no economic dependencies between the countries – apart from the fact that the global investors diversify their investment portfolios between the countries.