ESSAYS ON THE ECONOMICS OF GENDER AND PARENTING

Abstract

In my first chapter, I present the first causal evidence on the effect of the entry of women into occupations on the wages of those occupations. In particular, I examine the effect of changes in the gender composition of an occupation on wages for men and women within the occupation. To determine the causal effect of a change in gender composition, I construct a shift-share instrument by using the dramatic increase in the relative educational attainment and workforce participation of women from 1960-2010 to instrument for changes in the gender composition of occupations with different levels of "exposure" to increased female work and education. I find evidence that a 10 percentage-point increase in the fraction of females within an occupation leads to an 8 percent decrease in average male wage and a 6 percent decrease in average female wage in the concurrent census year. Over the 10 years following the change in the gender composition, I find that the effect of such an increase in the fraction of females persists for male workers and grows for female workers, leading to an 8 percent decrease in male wages and an 11 percent decrease in female wages. I present suggestive evidence attributing this finding to effects of gender composition on the prestige and amenity value of occupations. In my second chapter, (co-authored with Eleonora Patacchini and Marco Battaglini) we study the effect of hearing cases alongside female judicial colleagues on the probability that a Federal judge hires a female law clerk. Federal judges are assigned to cases and to judicial panels at random and have few limitations on their choices of law clerks: these two features make the Federal court system a unique environment in which to study the effect of professional interactions and beliefs in organizations. For our analysis, we constructed a unique dataset by aggregating federal case records from 2007-2017 to collect information on federal judicial panels, and by merging this data with judicial hiring information from the Judicial Yellow Books, a directory of federal judges and clerks. We find that a one standard deviation increase in the fraction of co-panelists who are female increases a judge’s likelihood of hiring a female clerk by 4 percentage points. In my third chapter, I develop and test a model that explains differences in parenting style by socioeconomic status. Spanking, severe discipline, and other forms of "Authoritarian Parenting" are more common among low income and black parents than among high income and white and Hispanic parents. Although they are associated with increased short-term obedience, these "Authoritarian" parenting strategies are also associated with lower levels of cognitive development, self-esteem, school performance and "moral internalization." I construct a multi-stage parenting model in which children choose levels of school effort and delinquent behavior while heavily discounting future consequences, and parents altruistically regulate their children both by disciplining them and investing in their self-control. The model predicts that parents employ more discipline when the negative effects of child delinquent behavior are large. I test this model by measuring the effect of school safety (which influences the cost of child delinquent behavior) on parenting practices, measured in the Los Angeles Family and Neighborhood Survey (LA.FANS). Because school safety affects parents only through its effect on children, the relationship between school safety and parenting style should not reflect parents’ stress, social isolation, or economic circumstances. Controlling for family, neighborhood and school characteristics, I find that a 1 standard deviation increase in school disorder is associated with a 0.11 standard deviation increase in harsh parental discipline, with larger effects for poor and Black households. I argue that this effect is driven by parents’ concern about the cost of their child’s misbehavior.