¿Desaparece la clase media en México?: Una aplicación de la polarización por subgrupos entre 1984 y 2000.

This paper empirically applies the statistical approaches to the phenomenon of polarization generated by Esteban, et al. (1999) and Gradín (2000) in order to quantify the evolution of the middle class in Mexico and the role of various household attributes in the formation of groups during 1984-2000. It is assumed that the formation of extreme groups and the clustering process in every society is determined not only by equivalent income, but also by socioeconomic characteristics of the household. Micro-data of the National Survey of Household Income and Expenditure (ENIGH) is used, so that household disposable equivalent income is related to the attributes of the household head. Once the social groups are quantified, an ordered probit model is settled out so influences of characteristics are attached to them and respective probabilities and marginal effects are obtained. Findings reveal that both a huge gap between poor and rich incomes and the effect that education induces to separate the sub-populations groups, lead to increases in polarization engendering a weaker middle class in the distribution. It is also found that a higher effort is required in order to improve household conditions within the Mexican society between 1984 and 2000.