Essays on economic inequality

The last years have seen a surging interest in inequality in our society and in the world, with particular emphasis on economic inequality. Both long-run trends and the recent economic crisis have contributed to an increase in the gap between the rich and the poor, raising new questions on why this happened and on whether society should (or shouldn't) counteract these forces. The three main chapters of this dissertation aim at understanding the reasons behind inequality in the probability of being unemployed, inequality in test scores in fluenced by parental investment, and persistence of employment status across generations. Chapter 1 studies the concentration of lifetime unemployment and its determinants. Using panel data from the US, I document three new stylized facts on unemployment. First, 10% of workers account for two-thirds of unemployment in prime-age. Second, young unemployment predicts prime-age unemployment. Third, differences in job-finding rates between the most unemployed and the rest increase over the life cycle, while differences in separation rates shrink. I show that a model of heterogeneity across workers and information frictions, in which agents learn workers' types from their labor market history, is quantitatively consistent with all these facts. I find information frictions to be responsible for the whole decrease in job-finding rates of the most unemployed workers over the life cycle. The concentration and persistence of prime-age unemployment are mainly explained by heterogeneity across workers, while information frictions have a negligible role. Chapter 2 focuses on the relation between returns to parental investment and the way parents decide to spend their time. Time allocation choices of households and test scores of children exhibit regular patterns in the data. First, households spend more than twice as much time in child care when their child is under 6 years old than in later years, and reduce labor supply by almost 10 % in the same years. Second, college-educated parents spend more time with their children than noncollege ones. Third, cognitive test scores are intergenerationally correlated. In order to analyze the mechanisms behind these facts, I build a model of parental choices and both cognitive and noncognitive skills formation, which embeds the technology of skills formation estimated by Cunha, Heckman and Schennach (2010). I find that the model, jointly with the properties of the technology, can account qualitatively and quantitatively for the bulk of such patterns. I also use the model to simulate the effect of applying the German scheme of child allowances to the US economy. Consistently with the empirical literature, such a policy is found to have little effect on the intergenerational persistence of cognitive skills. Chapter 3, jointly developed with Salvatore Lo Bello, investigates the relation between the employment status of parents and the employment prospects of off spring. We study how parental links affect employment prospects, using monthly job histories from the BHPS. We motivate our empirical strategy by means of a stylized model of intergenerational transmission of networks. We find that having the father employed rather than unemployed increases the employment rate by about 8 p.p. and the monthly job finding probability by at least 50% (5-6 p.p). The effect is even larger when the father and the off spring work in the same occupational group. The empirical evidence suggests that such results are due to informational advantages rather than human capital transmission, direct hiring or common shocks.Programa Oficial de Doctorado en EconomíaPresidente: Espen Moen; Secretario: Matthias Kredler; Vocal: Ludo Visscher