This paper is devoted to investigating the matching process for Tunisia using desaggregated data by assuming that the rising of the unemployment rate result from regional disparities which yield variation of matching efficiencies across regions. Since most econometric aspects of spatial heterogeneity can be handled by means of the standard panel data methods, we focus our discussion on the new technique: Panel Smooth Transition Regression models (PSTR). The distinction is that we can compute regional specific sensibilities for 23 regions over the period 1984-2004. Given this objective, we consider three structural factors that allow explaining the regional imbalances. Contrary to the previous econometric techniques of the matching theory, estimates of the coefficients depend, of three transition variables. The results show that women insertion, the qualification share and population density significantly contribute to explain the asymmetry of the matching process across regions. Our main conclusion is that the hiring in Tunisia is driven essentially by the stock of vacancies about is the region. The willingness of job seekers, obviously, remains low, although it is different between regions and seems relatively important in urban zones.
Modelling nonlinearities in commodity prices using smooth transition regression models with exogenous transition variables
