Statistical physics and anomalous macroeconomic fluctuations.

Summary How can disturbances at the level of the individual become large fluctuations at the level of the economy? This question, despite being studied for decades, is still open. In this book, I investigate this puzzle using methods from statistical physics. Starting from an in-depth analysis of power-law distributions, I show that a clear understanding of their origin and properties allows us to better understand their socio-economic consequences. I then propose a model of a network economy, where firms depend on each other to produce, so that the very nature of their interactions can make them susceptible to amplify fluctuations. I then provide an empirical study of the statistical properties of growth rates and a framework for studying their dynamics. In the final part I focus on models that exhibit non-trivial collective phenomena because they consider imitation or memory effects in the decisions taken by individuals, showing the need to take into account the possible complexity of the different constituent parts of economic models.