Date of Degree


Document Type


Degree Name





Miles Corak

Committee Members

Nuria Rodríguez-Planas

Jonathan Conning

Subject Categories

Growth and Development | Income Distribution | Labor Economics


Intergenerational mobility, Social mobility, Inequality of Opportunity, Mexico, COVID-19


This dissertation consists of three chapters on the relationship between immutable characteristics of a person (also known as circumstances) and the intergenerational social mobility experienced by them, as well as their human capital accumulation. It pays particular attention to the role played by the economic resources of the household of origin in producing differences in the economic outcomes across persons.

In the first chapter, I show that the decomposition of intergenerational persistence indicators into their structural and positional components offers a clearer understanding of the determinants of heterogeneity in subnational mobility rates. The crucial element for the separate analysis of positional and structural mobility is the use of regionally defined instead of nationally defined quantiles. This constitutes a departure from the current consensus in the estimations of mobility rates at the subnational level in economics. Using the Mexican case as an example, I show that there are no significant differences across the country’s regions in terms of positional mobility. This contrast with the existing results and their interpretations, particularly regarding intergenerational mobility in the south region of the country. This highlights the importance of incorporating positional measures into the battery of tools used for intranational analysis.

In the second chapter, I study the difference in intergenerational mobility patterns between the indigenous and non-indigenous Mexican populations, analyzing the role played by the variability in skin tones inside both groups. I employ a composite sample based on two recent surveys with the same sampling frame and questionnaire design. I obtained three main results based on rank-rank regressions, transition matrices, and quantile rank regressions. First, the indigenous population faces a higher degree of intergenerational persistence and remains concentrated at the bottom of the distribution of economic resources. Secondly, the indigenous population faces less upward mobility from the bottom of the distribution but similar rates of persistence to the non-indigenous population at the top. Furthermore, there is a skin tone gradient for the indigenous and non-indigenous populations, where lighter skin tones tend to remain at a higher rank in the distribution of economic resources than darker skin tones.

In the third chapter, I propose a framework for analyzing the short and long-run effects of temporary educational disruptions on children’s learning progression. The framework includes continuous parental investments, filling a gap in the literature on the Potential Pedagogical Function. Similarly, it considers the role of economic resources as part of households’ attenuation capacity. I take this framework to the data to estimate the potential effects of the COVID-19 pandemic in Mexico. The potential persistent loss in learning lies between 20%-90% of learning acquired during a typical school year, depending on the effectiveness of the policies implemented in 2020 and 2021.