Timo Ehrig

For phone, office and adress information, please click here .

New Economic Theory: Reflexive Market Dynamics

How do economic agents form expectations regarding future asset prices and the development of macroeconomic quantities, when there is a fundamental uncertainty about the market process, like after financial crises? How does the formation of expectations fold back to the realized economic process, and in particular, to the selection of one of multiple possible equilibria in the economic process?
In a joint project with Juergen Jost, our starting point is that the formation of expectations can be usefully studied as a game-like situation of mutual anticipation of economic agents, if the key uncertainties that drive the development are endogenous, that is, if there is uncertainty about the actions of other economic agents. For instance, after the financial crisis of 2008, there was uncertainty about how governments, banks, and firms would behave, and the anticipation of this behavior was key to the formation of expectations regarding of individual agents.
In our view, a key shortcoming of the currently established modeling paradigms to study macroeconomic development, that is, Dynamic Stochastic General Equilibrium and Rational Expectations models, is that here price movements are assumed to be caused by events exogenous to the economy, but not by endogenously caused by changes in the higher order beliefs of the interacting agents.
In our model, individual agents entertain higher order beliefs regarding the expectations of other economic agents, which are the basis for their own expectation formation. These beliefs are updated in a learning process. The learning process is more complex as in rational expectations equilibria (REE) models: Prices also contain information about the higher order beliefs of the agents. Agents cannot fully distinguish if price movements are caused by a change in information about fundamentals, or a change in expectations of the other agents.
Scholars have already articulated the idea that macroeconomic expectations are the result of mutual anticipation of the economic agents (compare Morris, S., Shin, H. 2000. Rethinking Multiple Equilibria in Macroeconomic Modeling. NBER Macroeconomics Annual,Vol. 15, pp. 139-161.). However, we argue that the current macroeconomic models that are based on this idea are not developed far enough. In particular, Morris and Shin do not discuss how prices mediate the respective expectation formation and mutual anticipation of interacting agents, and how price movements that unfold over time mediate expectation formation.
Our contribution is to offer such a in which these dynamics are captured. In our model, expectation formation is based on mutual anticipation of the agents and their actions, and prices take a central mediating role. In our model, rationality of the agents is common knowledge. Yet this does not imply that prices reveal information. This result has first been obtained by Ben-Porath and Heifetz (2010. Common Knowledge of Rationality and Market Clearing in Economies with Asymmetric Information. Northwestern University: Discussion Paper). We extend this discussion by showing how agents nevertheless learn about fundamentals and higher order beliefs from price series.
The project has been presented or will be presented at the World Economic Congress in Beijing, 2011, and the American Economic Association Meeting in Chicago, 2012.
We also collaborate with Gerd Gigerenzer and Konstantinos Katsikopoulos to study if higher order beliefs matter in bank decisions in the real world. For this purpose, we are empirically studying banks.