Mobility in the urban labor market: A panel data analysis for Mexico

In this article, we develop a dynamic discrete choice model that can explain the dynamics of sector choice in urban Mexico. We study the mobility patterns among three states, formal sector employment, informal sector work, and not working, and analyze how these patterns vary across groups with different characteristics and family resources. We use data from Mexico's quarterly Urban Employment Survey. We focus on eight panel waves drawn in 1999-2000 but also compare these with estimates for panels covering 1992-93 and 1994-95. The period 1992-93 was a time of steady economic growth; the 1994-95 period is characterized by the economic downturn after the peso crisis. After some years of recovery, the 1999-2000 period is again characterized by economic growth. It seems interesting to compare the results for these three time periods, particularly since the nature of the difference between the formal and informal sector may have changed because of institutional reforms and the growth of foreign-owned in-bond manufacturing industries. The model estimates are used to predict the probabilities of changing from one state to another for benchmark individuals with given observed and unobserved characteristics. These predicted probabilities are used to test the null hypothesis that the transition patterns between sectors are in line with the symmetric view of formal and informal sector jobs. To be precise, we show that the null hypothesis that formal and informal sector jobs have the same entry and access rates implies restrictions on the state probabilities and the transition rates for both sectors. On the basis of our model estimates, we test restrictions for the benchmark workers.