Search

The julia package integrates the Julia programming language into Stata. Users can transfer data between Stata and Julia, issue Julia commands to analyze and plot, and pass results back to Stata. Julia's econometric ecosystem is not as mature as Stata's or R's or Python's. But Julia is an excellent environment for developing high-performance numerical applications, which can then be called from many platforms. For example, the boottest program for wild bootstrap-based inference (Roodman et al. 2019) and fwildclusterboot for R (Fischer and Roodman 2021) can both call the same Julia back end. And the program reghdfejl mimics reghdfe (Correia 2016) in fitting linear models with high-dimensional fixed effects but calls a Julia package for tenfold acceleration on hard problems. reghdfejl also supports nonlinear fixed-effect models that cannot otherwise be fit in Stata--though preliminarily, as the Julia package for that purpose is immature.

Summan, Nandi, and Bloom (2023), hereafter SNB, find that exposure during infancy to India's Universal Immunization Programme (UIP) increased wages and per-capita household expenditure in early adulthood. SNB regress these outcomes on a treatment indicator that depends upon year and district of birth while controlling for age at follow-up. Because year of birth and age are nearly collinear, SNB's identifying variation does not come from the staggered introduction of the UIP, but rather from the progression of time during the follow-up period. Within the 12-month follow-up period, those interviewed later were more likely to have been treated and, on average, reported higher wages and household expenditure. Wages and household expenditure, however, rose by at least as much in a control group composed of people too old to have been exposed as infants to the UIP as in the treated group. SNB's results are best explained by inflation, economic growth, and non-random survey sequencing during the follow-up survey period.

Hjort and Poulsen (2019) frames the staggered arrival of submarine Internet cables on the shores of Africa circa 2010 as a difference-in-differences natural experiment. The paper finds positive impacts of broadband on individual- and firm-level employment and nighttime light emissions. These results largely are not robust to alternative geocoding of survey locations, to correcting for a satellite changeover at end-2009, and to revisiting a definition of the treated zone that has no clear technological basis, is narrower than the spatial resolution of nearly all the data sources, and is empirically suboptimal as a representation of the geography of broadband.

This paper reanalyzes Khanna (2023), which studies labor market effects of schooling in India through regression discontinuity designs. Absent from the data are four districts close to the discontinuity; restoring them cuts the reduced-form impacts on schooling and log wages by 57% and 63%. Using regression-specific optimal band-widths and a robust variance estimator clustered at the geographic unit of treatment makes impacts statistically indistinguishable from 0. That finding is robust to varying the identifying threshold and the bandwidth. The estimates of general equilibrium effects and elasticities of substitution are not unbiased and have effectively infinite first and second moments., Comment: 18 pages + references + appendices; 2 figures + 3 tables in main text

Duflo (2001) exploits a 1970s schooling expansion to estimate the returns to schooling in Indonesia. Under the study's difference-in-differences (DID) design, two patterns in the data--shallower pay scales for younger workers and negative selection in treatment--can violate the parallel trends assumption and upward-bias results. In response, I follow up later, test for trend breaks timed to the intervention, and perform changes-in-changes (CIC). I also correct data errors, cluster variance estimates, incorporate survey weights to correct for endogenous sampling, and test for (and detect) instrument weakness. Weak identification-robust inference yields positive but imprecise estimates. CIC estimates also tilt positive.

Many interventions in global health save lives. One criticism sometimes lobbed at these interventions invokes the spirit of Malthus. The good done, the charge goes, is offset by the harm of spreading the earth's limited resources more thinly: more people, and more misery per person. To the extent this holds, the net benefit of savings lives is lower than it appears at first. On the other hand, if lower mortality, especially in childhood, leads families to have fewer children, life-saving interventions could reduce population. This document critically reviews the evidence. It finds that the impact of life-saving interventions on fertility and population growth varies by context, and is rarely greater than 1:1. In places where lifetime births/woman has been converging to 2 or lower, saving one child's life should lead parents to avert a birth they would otherwise have. The impact of mortality drops on fertility will be nearly 1:1, so population growth will hardly change. In the increasingly exceptional locales where couples appear not to limit fertility much, such as Niger and Mali, the impact of saving a life on total births will be smaller, and may come about mainly through the biological channel of lactational amenorrhea. Here, mortality-drop-fertility-drop ratios of 1:0.5 and 1:0.33 appear more plausible. But in the long-term, it would be surprising if these few countries do not join the rest of the world in the transition to lower and more intentionally controlled fertility.

This paper reviews the research on the impacts of alcohol taxation outcomes such as heavy drinking and mortality. Where data availability permits, reviewed studies are replicated and reanalyzed. Despite weaknesses in the majority of studies, and despite seeming disagreements among the more credible one--ones based on natural experiments--we can be reasonably confident that taxing alcohol reduces drinking in general and problem drinking in particular. The larger and cleaner the underlying natural experiment, the more apt a study is to detect impacts on drinking. Estimates from the highest-powered study settings, such as in Alaska in 2002 and Finland in 2004, suggest an elasticity of mortality with respect to price of -1 to -3. A 10% price increase in the US would, according to this estimate, save 2,000-6,000 lives and 48,000-130,000 years of life each year.

This paper critically reviews the research on the impact of immigration on employment and wages of natives in wealthy countries--where "natives" includes previous immigrants and their descendants. While written for a non-technical audience, the paper engages with technical issues and probes one particularly intense scholarly debate in an appendix. While the available evidence is not definitive, it paints a consistent picture. Industrial economies can generally absorb migrants quickly, in part because capital is mobile and flexible, in part because immigrants are consumers as well as producers. Thus, long-term average impacts are probably zero at worst. And the long-term may come quickly, especially if migration inflows are predictable to investors. Possibly, skilled immigration boosts productivity and wages for many others. Around the averages, there are distributional effects. Among low-income "native" workers, the ones who stand to lose the most are those who most closely resemble new arrivals, in being immigrants themselves, being low-skill, being less assimilated, and, potentially, undocumented. But native workers and earlier immigrants tend to benefit from the arrival of workers different from them, who complement more than compete with them in production. Thus skilled immigration can offset the effects of low-skill immigration on natives and earlier immigrants.

This paper reviews the research on the impacts of incarceration on crime. Where data availability permits, reviewed studies are replicated and reanalyzed. Among three dozen studies I reviewed, I obtained or reconstructed the data and code for eight. Replication and reanalysis revealed significant methodological concerns in seven and led to major reinterpretations of four. I estimate that, at typical policy margins in the United States today, decarceration has zero net impact on crime outside of prison. That estimate is uncertain, but at least as much evidence suggests that decarceration reduces crime as increases it. The crux of the matter is that tougher sentences hardly deter crime, and that while imprisoning people temporarily stops them from committing crime outside prison walls, it also tends to increase their criminality after release. As a result, "tough-on-crime" initiatives can reduce crime in the short run but cause offsetting harm in the long run. A cost-benefit analysis finds that even under a devil's advocate reading of this evidence, in which incarceration does reduce crime in U.S., it is unlikely to increase aggregate welfare.