Essays on microeconomic causal inference in welfare, education, and health

My dissertation consists of three applied microeconomic papers unified by the techniques of causal inference and the themes of welfare, education, and health. Family features prominently. The first two chapters study homeless families, while the third shifts focus to breastfeeding, a topic of general familial interest. Chapter 1 investigates the educational effects of New York City's policy of placing homeless families in shelters near their children's schools. I find that proximity augments homeless students' educational outcomes. Homeless K--8 graders whose families are placed in shelters in their school boroughs have 8 percent (2.4 days) better attendance, are a third (18 percentage points) less likely to change schools, and exhibit higher rates of proficiency and retention. Homeless high schoolers have 5 percent (2.5 days) better attendance, 29 percent (10 pp) lower mobility, and 8 percent (1.6 pp) greater retention when placed locally. These results proceed from novel administrative data on homeless families observed in the context of a scarcity-induced natural experiment. A complementary instrumental variable strategy exploiting homeless eligibility policy reveals a subset of proximity-elastic students benefit considerably more. Panel evidence demonstrates homelessness does not cause educational impairment as much as reflect large preexisting deficits. Chapter 2 situates neighborhood-based homeless shelter placements in the context of whole-family outcomes. Again using an original administrative dataset in the context of a scarcity-based natural experiment in New York City, I find that families placed in shelters in their neighborhoods of origin remain there considerably longer than those assigned to distant shelters. Locally-placed families also access more public benefits and are more apt to work. A fixed effects model assessing multi-spell families confirms these main results. Complementary instrumental variable and regression discontinuity designs exploiting policy shocks and rules, respectively, suggest difficult-to-place families---such as those that are large, disconnected from services, or from neighborhoods where homelessness is common---are especially sensitive to proximate placements. Better targeting through improved screening at intake can enhance program efficiency. The practice of assigning shelter based on chance vacancies ought to be replaced with a system of evidence-based placements tailored to families' resources and constraints. In Chapter 3, I retain emphases on families and education and study the long-term effects of breastfeeding. Despite consensus among medical authorities about the desirability of breastfeeding, causal evidence about its effects is surprisingly scant. Using a thorough collection of empirical approaches and detailed longitudinal data spanning five decades, I investigate a comprehensive set of outcomes with greater breadth and continuity than previous work. On average (per OLS), breastfeeding is associated with modest and persistent cognitive advantages from childhood through young adulthood---even after controlling for an extensive set of confounding forces. Accounting for breastfeeding duration strengthens these relationships and uncovers favorable labor market and fertility linkages as well. But there is no evidence for enduring health benefits. At the same time, a novel extended family fixed effects analysis comparing differentially breastfed siblings and cousins finds little association between breastfeeding and any outcome. I argue these findings are not mutually exclusive by providing evidence that, contrary to conventional wisdom, the divergent estimates are the consequence of considerable negative selection into the subset of families contributing to fixed effects identification.Ph.D.Includes bibliographical reference