Determinants of job quality: Persistence and dynamic segmentation in the labor market of a developing country

Abstract

We adapt the multidimensional poverty methodology to the study of job quality dynamics using a unique household survey panel, for a developing country like Chile. This data allows us to control for the history of the worker in addition to the usual socio-demographic and economic characteristics. We estimate a dynamic panel model with random effects and find existence of better job quality among larger and unionized firms but neither gender, public/private status nor economic sectors are significant determinants of job quality. Moreover, the data reveals the significance of labor history in predicting job quality and therefore the existence of persistence in job quality. A worker can move between different size of firms and economic sectors, but those jobs will be of similar quality (for better and for worse). Even though our sample of workers corresponds to those with slightly better working conditions (dependant workers that have contributed to pensions at least once in their labor history) job quality mobility seems to be weak suggesting the existence of job quality segmentation. Having been in a low quality job induces an increase in the probability of continuing to be in a low quality job in 20-40 percent. This effect is important while the second biggest marginal effect comes from firm size (being in a big firm reduces the probability of being in a low quality job), and it’s 10 percent at the most. We show evidence that this result of persistence could be indicative of job quality dynamic segmentation