View article

In this chapter we develop a heterogeneous interacting agents (HIAs) model suitable to replicate well-known empirical regularities of industrial dynamics, such as the power law distribution of firm size (Okuyama et al., 1999; Ramsden and Kiss-Haypal 2000; Axtell, 2001; Gaffeo et al., 2003) and the Laplace distribution of firm growth rates (Stanley et al., 1996; Bottazzi and Secchi, 2003).

In our model the analysis of the business cycle is related with the properties and evolution of the power law distribution of firm size. The power law distribution was originally discovered by Vilfredo Pareto who maintained that the distribution of personal incomes above a certain threshold v0 follows a heavy-tailed distribution (Pareto, 1897). In particular, he found that the probability of observing an income Y greater than or equal to y is proportional to a power of y, that is Pr (Y≥ y)∞ ya, with α close to 1.5. Our aim is to propose a suitable agent-based model to analyse the evolution of power law distributions and discover how interaction among heterogeneous agents works and influences the relation between the distribution of agents' variables and business cycle fluctuations. For instance, a major consequence of a heavy-tailed firm size distribution is that small idiosyncratic shocks can generate large aggregate fluctuations even in the absence of aggregate shocks; therefore, understanding how firm size distribution changes and interacts with the business cycle is an important goal for economic research with considerable consequences for policy intervention. 1