Content area
Abstract
Disentangling the relationship between migration, development and inequality has been a challenge for social scientists for more than forty years. In this study I explore three aspects of migration and inequality, to further refine theoretical and methodological approaches to the problem. Critical for untying the empirical knots are studies that take a longitudinal approach and are sensitive to selectivity bias. All three of my studies examine data with temporal depth and incorporate attention to selectivity bias.
In the first study, I compare statistical models predicting the distribution of migration across states in Mexico. These models are hierarchical logit models, conditional logit models, and nested logit models. This paper shows that the results of interstate migration analyses are sensitive to the model used. Different models have different accuracy and can lead to different conclusions with respect to the importance of the factors affecting the level and directionality of migration.
The second chapter answers to the question do past migration, historic, administrative and infrastructure ties, and trade between states contribute to the explanation of the directionality of interstate migration in Mexico? These variables have been identified in the economic sociology and historic-structural literature of migration as key for understanding why migrants from some states go to certain destinations, but have not been considered in past quantitative studies of migration. The results in this chapter show that past migration and historic, administrative and infrastructure ties between states help to understand the directionality of migration in 1975–1980, while increased trade between states helps to understand the directionality of migration in 1995–2000.
The last chapter answers to the questions: do migrants differ from nonmigrants in their household transfer behavior in the presence of government welfare policies? And how do regional differentials in the probability of migration affect cash transfer policies? The results of this chapter show that transfer behavior differs between migrants and nonmigrants, that public transfer programs are more likely to crowd-out nonmigrant transfers than migrant remittances, and that it is possible that public transfer programs are less effective in areas of low migration than in areas of high migration.
