Making Sense of Sensitivity: Extending Omitted Variable Bias

We extend the omitted variable bias framework with a suite of tools for sensitivity analysis in regression models that: (i) does not require assumptions about the treatment assignment nor the nature of confounders; (ii) naturally handles multiple confounders, possibly acting non-linearly; (iii) exploits expert knowledge to bound sensitivity parameters; and, (iv) can be easily computed using only standard regression results. In particular, we introduce two novel sensitivity measures suited for routine reporting. The robustness value describes the minimum strength of association unobserved confounding would need to have, both with the treatment and the outcome, to change the research conclusions. The partial R2 of the treatment with the outcome shows how strongly confounders explaining all the residual outcome variation would have to be associated with the treatment to eliminate the estimated effect. Next, we offer graphical tools for elaborating on problematic confounders, examining the sensitivity of point estimates, t-values, as well as “extreme scenarios”. Finally, we describe problems with a common “benchmarking” practice and introduce a novel procedure to instead formally bound the strength of confounders based on comparison to observed covariates. We apply these methods to a running example that estimates the effect of exposure to violence on attitudes toward peace.

Live Births and Fertility amidst the Zika Virus Epidemic in Brazil

Abstract: In late 2015, the Brazilian Ministry of Health classified the increase in congenital malformations associated with the Zika Virus (ZIKV) a public health emergency. The risk of ZIKV-related congenital syndrome posed an exogenous threat to reproductive outcomes that could result in declining numbers of live births and potentially fertility. Using 2014-2016 monthly microdata on live births from the Brazilian Information System on Live Births, in this talk I examine live births and fertility trends amidst the ZIKV epidemic in Brazil. Findings suggest a decline in live births that is stratified across socioeconomic status and geographic lines, especially nine months after the call for pregnancy postponement. While declines in total fertility rates were small, fertility trends estimated by age and socioeconomic status suggest important differences in how Zika might have impacted Brazil’s fertility structure. Further findings using monthly data by municipality suggest that the epidemic resulted in a significant decline in fertility even when controlling for characteristics of the municipality. The findings highlight the importance of understanding how exposure to the risk of a health threat directed at fetuses has led to declines in fertility.

The Other Side of Assimilation: How Immigrants are Changing American Life

The immigration patterns of the last three decades have profoundly changed nearly every aspect of life in the United States. What do those changes mean for the most established Americans—those whose families have been in the country for multiple generations? The Other Side of Assimilation shows that assimilation is not a one-way street. Jiménez explains how established Americans undergo their own assimilation in response to profound immigration-driven ethnic, racial, political, economic, and cultural shifts.

Co-sponsored with the Center for the Study of International Migration and the Race and Ethnicity Sociology Working Group

Essays in Education and Crime in Colombia

This dissertation contains three essays in applied microeconomics. In the first chapter paper, I test whether the return to college education is the result of human capital accumulation or instead reflects the fact that attending college signals higher ability to employers. The second chapter provides evidence that parental incarceration increases children’s educational attainment. Finally, in the third chapter I derive a new expression that extends the Local Average Treatment Effect concept, to a setting with two sources of unobserved treatment heterogeneity.

"Renal Relationships: Understanding Living Kidney Donor Relationship Patterns"

Abstract: Who do we turn to in times of need? Traditionally, social support research has shown a strong preference to rely on strong ties in these scenarios - often, even when weak ties might be better positioned to help. However, this conclusion has recently been challenged by Small (2017), who argues that people often rely on weak ties for emotional support in stressful times, preferring to avoid more complicated strong ties. This suggests that the types of ties we activate in times of need varies by the situation. In this study, we apply this framework to the study of living donor kidney transplantation (LDKT), effectively asking: How does this behavior differ when the stakes are potentially life and death? Using a variety of primary and secondary datasets, we compare the distribution of LDKT ties to the distribution of ties who would be likely able to help, then seek to explain these relative utilization patterns as a function of medical fundamentals, social/spatial relationships, and qualitative reasoning invoked by survey respondents. Our preliminary findings show that LDKT patterns are primarily driven by social relationship quality, and far less by medical fundamentals such as the potential donors' health or genetic relationship to the patient.

"Can Labor Market Discrimination Explain Racial Disparities in Schooling? Evidence from WWII"

Abstract: Can the racial gap in labor market earnings explain black-white disparities in the schooling of the next generation? To answer this, we exploit the large increase in labor demand in markets that received WWII defense industry contracts. This increase in labor demand combined with a policy that prohibited discrimination by race and ethnicity in the defense industries resulted in significant increases in African American earnings and declines in the racial gap in earnings between 1940 and 1950. This was achieved largely via occupational upgrading among African Americans into semi-skilled professions. In contrast with women, whose progress in the labor market was largely reversed in short order, this occupational upgrading persisted for African Americans. We argue that this persistence is consistent with declines in statistical discrimination. Moreover, we find that in these same labor markets, the next generation of African Americans invested relatively more in their human capital, as measured by greater years of schooling and a decline in the black-white schooling gap. We explore three reasons why reductions in the black white earnings gap might lead to reductions in the black white schooling gap of the next generation. First, this would relax the financial constraint faced by many African American families, allowing their children to remain in school longer. Second, occupational upgrading might have increased the returns to human capital among African Americans. Finally, there may be political responses that result in changes in public funding and provision of schooling and other public goods that affect the human capital accumulation of the next generation of African Americans. We find evidence consistent with the first explanation only. We conclude that efforts to further reduce the racial gap in schooling might consider labor market interventions.