Coordinators: Prof. Jean-Louis Arcand (CFD) and Prof. Max-Olivier Hongler (Laboratory of Microengineering for Manufacturing, EPFL)

The purpose of this research project is to re-inject the stochastic element into our fundamental understanding of the sources of global economic heterogeneity, with a particular focus on poverty traps. While the literature on poverty traps has been very ably summarised by Azariadis and Stachurski (2005), with an emphasis on the importance of non-ergodic behaviour and the role played by coordination failures (Matsuyama 1991, Krugman 1991), several potentially promising lines of research, based on recent results in statistical mechanics, remain unexplored by economists. The aim of the proposal is to identify key stylised features enabling a swarm of interacting economic agents to dynamically escape poverty traps. By studying stochastic agent dynamics, we shall explicitly take into account the ubiquitous presence of stochastic fluctuations, embodied in a random environment, the availability of noisy information sources, and bounded rationality concerns. In a nutshell, our proposal is based upon the founding principles of off-equilibrium thermodynamics.