FAMILY CARE AND HOUSEHOLD PRODUCTION: IMPLICATIONS FOR ADULT CHILDREN BY GENDER, RACE AND ETHNICITY

My dissertation consists of three chapters on unpaid care work and its implications for adult children by gender, race and ethnicity. The first chapter studies the effect of family care on adult children's employment and earnings, and its implications for the gender gap in labor market outcomes among adult children. I first provide empirical evidence that caring for an elderly parent results from decision-making within a family rather than at the individual level. Regarding the gender gap in family care, I document that daughters with less attachment to the labor market and lower earnings provide the brunt of family care. The gender gap in family care is most salient in sibling groups with mixed gender composition. Motivated by the empirical evidence, I structurally estimate a model of strategic interactions between a daughter and a son in mixed-gender sibling groups. Adult children face different opportunity costs in terms of wages and have heterogeneous preferences or perceived care responsibility for family care to a parent with long-term care (LTC) needs. I find that the heterogeneity in preferences explains the gender gap in family care among adult children more substantially than the heterogeneity in opportunity costs. Using life-cycle profiles of parents' LTC needs and adult children's wages, I simulate the long-run trajectories of employment and family care of adult children. In a counterfactual scenario, I quantify that daughters face a 4.6 percent drop in lifetime earnings due to family caregiving compared to a 1.5 percent drop for sons. The second chapter empirically investigates the effect of family care on adult children's employment by race and ethnicity. Due to lower access to quality formal care and differences in norms and traditions, minority populations rely more heavily on family care than non-minority populations do. Despite the growing diversity in the demography of the older population and their family caregivers, we know little about the racial and ethnic differences in family care patterns over time and their impact on the economic outcomes of caregivers. My study intends to fill this gap. Using the pooled 1998-2019 Health and Retirement Study, I first provide a descriptive analysis documenting the disability and family care trajectories of elderly individuals aged 50 and over across racial and ethnic groups. I then employ a recursive bivariate probit model to examine how family care affects adult children's employment by race and ethnicity. I find that non-Hispanic Black and Hispanic elderly individuals have higher levels of LTC needs and rely more on family care provision over their lifespans, compared to non-Hispanic Whites. Family care lowers adult children's employment by 5 to 9 percentage points depending on care type. This effect is 3.3 to 8.4 percentage points for non-Hispanic Whites and 11 to 13 percentage points for non-Hispanic Blacks. These effects are stronger for adult children with non-married parents and those aged less than 40. The third chapter examines whether the gendered division of household labor persists intergenerationally from parents to children. Suppose an adult child grows up in a household with the mother as the sole provider of housework or the father who shares housework equally with the mother. Do children show a similar division of household labor in their marriages in adulthood? To investigate this question, we draw on social psychology literature on childhood socialization and the development of gender role perceptions by focusing on the 'modeling effect' of parents during childhood. Using the questionnaire asking two generations of married couples their time spent on housework in the 1968-2019 Panel Study of Income Dynamics, we link the adult children split off from the original cohort of families and examine the intergenerational persistence in the gendered division of household labor. We first use an event study approach to explore the gender gap in housework hours between adult children and their spouses in response to the arrival of a first child. We then employ a multinomial logit model to show the relationship between childhood socialization and the division of household labor. Having a mother as the sole provider of housework across any childhood period strongly predicts a less egalitarian division of household labor for adult children and their spouses. Being exposed to working mothers or fathers who were more involved in housework does not significantly predict adult daughters' division of household labor. However, both indicators imply a more egalitarian division of household labor for adult sons and their spouses.