Space-time income distribution dynamics in Mexico

The uneven income distribution in Mexico has frequently being subject to both scrutiny and concern. This phenomenon seems to have a spatial counterpart much less explored and increasingly complex. This paper introduces a novel approach to analyse space-time integration of income distribution dynamics in Mexico. We draw on recent extensions in Exploratory Space-Time Data Analysis (ESTDA) and directional statistics, to investigate spatial dynamics and income movement patterns between Mexican states over 70 years. We use spatial randomization to test directional co-movements in spatially contiguous units as evidence of spatial dependence in income distribution dynamics. Strong evidence of spatial integration of income distribution dynamics is found along the study period. Our results suggest that, in many instances, states do not act independently and we identify a range of integrated movements. The lack of independence itself has important implications for income distributional outcomes as higher mobility will not necessarily imply mixing in the distribution, which is more likely to occur among neighbouring states within a similar income strata. This can be illustrated by the findings of two spatial, yet, contrasting effects: spatial integration in (low) top-down mobility, on the one hand, and negative spatial dependence on the other. While the latter is mostly explained by losses in states and distributional gains of the neighbours with income mixing (rank shifts among neighbouring states) within a group of states, the former phenomena obey a different spatial process where a state and neighbours move in the same direction in the distributional space. Altogether, our findings suggest the presence of slow spatial change in Mexico, mainly due to the presence of heterogeneous, spatially integrated dynamics influencing net change in income distributional outcomes.