Nonlinear analysis of employment in waste management

This paper contributes to the ongoing debate on the employment effects of waste management policy. We examine the dynamic linkage between waste tonnage and employment in the region of Paris (France), at the point when the waste management policy by delegation of service is adopted. To account for the presence of possible policy shifts, we propose the implementation of nonlinear causality tests based on the smooth transition autoregressive regression (STAR) framework. Using weekly data for four waste streams over the period January 2015–June 2017, the linearity tests reveal the presence of nonlinearity in most of the data series. When applying nonlinear Granger tests, our results provide strong support for nonlinear dependencies between waste tonnage and employment across different waste streams. For instance, we find strong statistical evidence that the causal relationship is consistently bidirectional in the miscellaneous and outdoor garbage waste streams. From a policy point of view, our findings suggest that waste management practices should factor the presence of these strong linkages and how they would affect environmental jobs creation.