CULTURE, EMPLOYMENT, AND VOLATILITY: THREE ESSAYS ON HISPANIC LABOR

This dissertation is a collection of three essays that investigate several topics on Hispanic men’s and women’s labor. Chapter 1, entitled “The Effects of Cultural Beliefs on Fertility and Work Decisions of Hispanic Women,” investigates the impact of cultural beliefs on the fertility and labor outcomes of US-born Hispanic women. : I use data on over 60,000 U.S.-born Hispanic women from the 2006-10 waves of the American Community Survey, combined with aggregate variables from their country of ancestry to proxy effects of culture on their own fertility and labor decisions. Results show the effects of culture on fertility and work decisions for Hispanic women to be qualitatively and quantitatively different from prior research based on descendants from non-Hispanic samples. The effects of culture on U.S.-born Hispanic women imply that they realize the tradeoffs between family size and work and adjust their family size accordingly and increase the number of weeks they work but not the number of hours per week. Additionally similar proxies for the husband’s cultural beliefs may either reinforce or offset his wife’s beliefs, depending on whether the couple shares the same country of origin. Hence, the cultural differences between Hispanics subethnicities and other groups result in different fertility and work outcomes, and this should give pause to agglomerate ethnic groups into monolithic groups such as “Hispanic”, “Asian”, or “Other” if the data permit. Additionally, it should urge researchers to further explore the link between cultural variation and other economic outcomes. Chapter 2, entitled “The Employment Effects of the 2007-09 Recession on Hispanic Groups” study examines effects of the recession on levels of employment among men and women of different Hispanic subethnicities using data from the Current Population Survey. As employment was rising rapidly for some Hispanic subgroups prior to the recession, the decomposition method of Engemann and Wall (2010) is used to separately identify the recession’s effects from ongoing changes in the employment-population ratios. As may be expected, Hispanic groups that were concentrated disproportionally in jobs and in parts of the country that were hardest hit by the housing-market bust saw their employment-to-population ratios decline sharply relative to the pre-recession trend. Yet the data show a considerable amount of heterogeneity across Hispanic subethnicities that cannot be traced to housing-market factors, suggesting that different Hispanic groups are moving along different employment paths unique to the group’s immigration history and experience settling in the U.S. The results underline that the Hispanic population is not monolithic and that analysts should disaggregate Hispanics into subethnicities whenever sufficient data permit. Chapter 3 entitled “Trends in Hispanic Earnings Volatility” investigates differences in income volatility between Hispanic men and women, versus white and black non-Hispanic men and women, and among Hispanics of different national origins, by examining which groups face the largest risks of experiencing a large drop in economic resources, how relative risks have changed over time, and what the patterns tell us about the sources of differential trends. Using panel data from the National Longitudinal Survey of Youth (1979), this essay compares two measures of earnings volatility: a volatility decomposition developed by Gottschalk and Moffitt (1994) and the standard deviation of arc percent change to document changes over time in the cross-sectional distribution of income changes. The NLSY shows that, for all groups, earnings volatility is lower today than in 1979. This finding is counter to previous research using other data sets (the PSID, CPS, SIPP, and LEHD) and likely reflects a life-cycle effect, whereby young people settle into stable jobs over the first 10-20 years of their careers and their earnings volatility falls. However, the data show some differences in levels of earnings volatility across subethnicities that are invariant to the volatility measure used, and hold up when individual characteristics are controlled for, namely that earnings volatility is relatively high for Puerto Ricans and non-Hispanic blacks and relatively low for Cubans; in the middle of the range, earnings volatility is similar for Mexicans, other Hispanic groups, and non-Hispanic whites. Further research would be valuable for explaining the sources of these differences across groups, and the extent to which policies could help attenuate earnings volatility among groups for which it is relatively high.