Many papers use fixed effects (FE) to identify causal impacts of an intervention. In this paper we show that when the treatment status only varies within some groups, this design can induce nonrandom selection of groups into the identifying sample, which we term selection into identification (SI). We begin by illustrating SI in the context of several family fixed effects (FFE) applications with a binary treatment variable. We document that the FFE identifying sample differs from the overall sample along many dimensions, including having larger families. Further, when treatment effects are heterogeneous, the FFE estimate is biased relative to the average treatment effect (ATE). For the general FE model, we then develop a reweighting-on-observables estimator to recover the unbiased ATE from the FE estimate for policy-relevant populations. We apply these insights to examine the long-term effects of Head Start in the PSID and the CNLSY. Using our reweighting methods, we estimate that Head Start leads to a 2.6 percentage point (p.p.) increase (s.e. = 6.2 p.p.) in the likelihood of attending some college for white Head Start participants in the PSID. This ATE is 78% smaller than the traditional FFE estimate (12 p.p). Reweighting the CNLSY FE estimates to obtain the ATE produces similar attenuation in the estimated impacts of Head Start.

This paper studies human capital responses to the availability of the Deferred Action for Childhood Arrivals (DACA) program, which provides temporary work authorization and deferral from deportation for undocumented, high-school-educated youth. We use a sample of young adults that migrated to the United States as children to implement a difference-in-difference design that compares noncitizen immigrants (“eligible”) to citizen immigrants (“ineligible”) over time. We find that DACA significantly increased high school attendance and high school graduation rates, reducing the citizen-noncitizen gap in graduation by 40 percent. We also find positive, though imprecise, impacts on college attendance. (JEL H52, I21, I26, J13, J15, J24)

This paper examines the effect of the female-to-male wage ratio, “relative wage,” on women's spouse quality, marriage, and labor supply over three decades. Exploiting task-based demand shifts as a shock to relative pay, I find that a higher relative wage (i) increases the quality of women's mates, as measured by higher spousal education, (ii) reduces marriage without substitution to cohabitation, and (iii) raises women's hours of work. These effects are consistent with a model in which a higher relative wage increases the minimum non-pecuniary benefits (“quality”) women require from a spouse and therefore reduce marriage among low-quality husbands.

While a growing literature shows that women, relative to men, prefer greater investment in children, it is unclear whether empowering women produces better economic outcomes. Exploiting plausibly exogenous variation in US suffrage laws, we show that exposure to suffrage during childhood led to large increases in educational attainment for children from disadvantaged backgrounds, especially Blacks and Southern Whites. We also find that suffrage led to higher earnings alongside education gains, although not for Southern Blacks. Using newly digitized data, we show that education increases are primarily explained by suffrage-induced growth in education spending, although early-life health improvements may have also contributed. (JEL H75, I21, I22, J13, J15, J16, N32)