Are estimates typically closer to the true parameter value when those estimates are published in highly-ranked economics journals? Within literatures, we find that the distribution of estimates does not appreciably differ by journal rank. Therefore, regardless of what the true parameter value is that literatures are attempting to estimate, it cannot be that the estimates in high-ranked journals are substantially closer to it. We discuss a number of possible explanations and implications for a variety of audiences.

I study the effects of the Deferred Action for Childhood Arrivals program (DACA) on labor market outcomes among potentially eligible immigrants. DACA allowed undocumented immigrants to participate in the labor market without fear of deportation, which might be expected to increase the probability of working and allowing workers to move to higher-skilled occupations. However, using a regression discontinuity design, I find very little to no effects on the probability of working and the likelihood of working in high-skilled jobs among DACA-eligible immigrants. The confidence intervals permit modest effects on these variables, but rule out large ones. Overall, my results suggest that temporary legal status had limited effects for DACA-eligible immigrants.

Fear of immigration enforcement may deter undocumented parents from seeking government benefits for their US citizen children. This paper examines the effect of providing legal status to parents through the Deferred Action for Childhood Arrivals (DACA) program on health insurance coverage among US-born children. Using a regression discontinuity design, I find that DACA eligibility among likely undocumented mothers increases Medicaid enrollment among their US-born children by 4 to 7 percentage points. I do not find evidence to support a similar effect among US-born children with likely undocumented fathers.