NYU node

Leader: Jess Benhabib

Several faculty at NYU are interested in the broad subject of macroeconomic fluctuations and their relation to learning, expectation coordination, multiple equilibria, coordination failure, sunspots and rare events with large deviations. The interested faculty so far are Jess Benhabib, Alberto Bisin, Xavier Gabaix and Thomas Sadzik.

Jess Benhabib has worked on macroeconomic indeterminacies, multiple equilibria and sunspot equilibria in the context of business cycles, economic growth and monetary policy. He is currently interested in learning models  along the lines of Evans and Honkapohja and  implication for  macroeconomic fluctuations in the context of rare events, large deviations and heterogeneous agents, and for  problems of macroeconomic policy coordination.

Alberto Bisin’s interests in macroeconomics concern mostly the study of the role of financial market frictions as an amplification mechanism of business cycles and as a filter to monetary policy in the context of heterogeneous agents.  They extend as well to the study of the role of financial market frictions in explaining wealth inequality and in affecting redistributive financial policies and on  behavioral economics and on the transmission of culture has important implication in macroeconomics, e.g., on savings behavior and entrepreneurship.

Xavier Gabaix is a behavioral economist interested in macroeconomics and finance. He’s currently working on a tractable model of bounded rationality. “In A Sparsity-Based Model of Bounded Rationality”, he proposes a model with boundedly rational features in which the decision-maker (DM) builds a simplified representation of the world. Crucially, this representation is “sparse,” i.e., uses few parameters that are non-zero, or differ from the usual state of affairs. The DM may imperfectly maximize, based again on a penalty related to sparsity.  Sparsity is formulated so as to lead to well-behaved, convex maximization problems. The model is a tractable algorithm that can be used with paper and pencil in many situations of interest. For instance, it offers a way to model boundedly rational dynamic programming. I apply it to a variety of prototypical economic situations: choosing an action while considering only subset aspects of the problem; selecting a consumption bundle with imperfect understanding of prices; optimal pricing with boundedly rational consumers — which generates a novel mechanism for price rigidity; life-cycle consumption and investment problems; portfolio choice with imperfect understanding of the structure of returns. Ongoing work extends the model to many agents, who form boundedly rational expectations about other agent’s actions.

Thomas Sadzig would like to investigate the relationship between the asset price bubbles and the coordination of expectations. In particular, recent literature (following up on Abreu and Brunnermaier 2003) focuses on the bubbles as arising out of miscoordination of beliefs about the fundamentals, while still maintain standard equilibrium assumption (full coordination of expectations about the actions). Drawing on the ideas from his recent paper on coordination in learning, he would like to relate bubbles more directly to miscoordinated expectations about agents’ actions, learning rules. For example, would small miscoordination of beliefs about agents’ actions lead to lasting bubbles? Can bubbles exist while agents coordinate their beliefs on small action sets, or sets of learning rules? Answering those, and other questions in the spirit of Eductive Stability (see Guesnerie 2005 for a review) would hopefully shed more light on the asset price bubbles.

The common interests of this group on expectation formation and macroeconomic fluctuations should fit in well with the project on expectational coordination (or lack thereof) in financial markets.