Symbolic Dynamics in a Matching Labor Market Model

In this paper we apply the techniques of symbolic dynamics to the analysis of a labor market which shows large volatility in employment flows. In a recent paper, Bhattacharya and Bunzel [1] have found that the discrete time version of the Pissarides-Mortensen matching model can easily lead to chaotic dynamics under standard sets of parameter values. To conclude about the existence of chaotic dynamics in the numerical examples presented in the paper, the Li-Yorke theorem or the Mitra sufficient condition were applied which seems questionable because they may lead to misleading conclusions. Moreover, in a more recent version of the paper, Bhattacharya and Bunzel [2] present new results in which chaos is completely removed from the dynamics of the model. Our paper explores the matching model so interestingly developed by the authors with the following objectives in mind: (i) to show that chaotic dynamics may still be present in the model for standard parameter values; (ii) to clarify some open questions raised by the authors in [1], by providing a rigorous proof of the existence of chaotic dynamics in the model through the computation of topological entropy in a symbolic dynamics setting.