An agent-based approach to evaluate the impact of economic dismissals facilitation on the French labor market

National audienceThe El Khomri law (also called " Work law ") has triggered a lot of conflicting judgements among French economists in 2016. However no model has been used to evaluate its effects ex ante. We have developed over the past 10 years a calibrated agent-based model of the French labor market, in order understand better such a complex system, and do policy analysis and design. The model integrates the hetero-geneity of agents, and their decisions (firms and workers) based on a search theoretical framework to generate gross flows. It gives rational microeconomic foundations to behavior , albeit with decision rules using bounded rationality rather than optimal rules in an equilibrium with rational expectations which would not make sense in such a complex system. Instead agents form anticipations based both on their own history and a detailed algorithm is developed to model the firms'computation of expected in-tertemporal profits for different scenarios of future own demand, with the possibility of loss aversion, to decide on hiring or not, and the type of labor contracts. We introduce several main institutions of the labor market and specially the two main labor contracts, Open-Ended contracts and Fixed Term Contracts. The WorkSim model simulates the gross flows between inactivity, unemployment and these two types of employment, with a consistent accounting system. It is calibrated by a powerful algorithm to set 63 parameters in order to fit 64 aggregate real variables. The calibration reveals an important loss aversion on part of firms, which is determinant of the hiring decisions. If we experiment with a low loss aversion, the unemployment diminishes by several points. We then analyze the facilitation of the economic dismissals allowed by the El Khomri law. We find that it has little effect on global unemployment but benefits the young who crowd out the seniors. This result is based on the substitution by the employers of Open Ended Contracts which become more precarious to Fixed Duration Contracts which become less useful, and the fact that young workers are more often in Fixed Term Contracts than the other age categories. The labor market is very deeply transformed in terms of flows also. When aggregate demand is endoge-nously changed, the experiment shows that the employment and the unemployment react more strongly than before the law, yielding a higher aggregate flexibility