Un enfoque de microsimulación para medir la segregación del ingreso

Residential segregation by income can be defined as the extent to which, within a given neighborhood, individuals belonging to different social groups live in neighborhoods characterized by different social compositions. In the United States and United Kingdom over the last 40 years, numerous measures of segregation have been proposed and tested. These measures require income data from the decennial census at a fine spatial resolution to work. This is problematic for countries such as Mexico, where income data are not collected nor reported in the census. To address this, different methodologies have been developed to analyze income distribution with national census data and income surveys. Our research uses microsimulation techniques to construct a synthetic population of any Mexican city. The technique combines an income survey with the census data from 2020. The method of Iterative Proportional Fitting (IPF) clones individuals from the survey to allocate them in the areal units as defined from the census. The result is a synthetic population that includes individual level sociodemographic information anchored at the minimum spatial geographies that the Mexican census considers. The method is tested and validated for five Mexican cities.