We estimate the causal impact of military conscription on long-term beliefs and personality traits. To address potential endogeneity concerns, we exploit the conscription lottery in Argentina. We combine administrative data from the conscription lottery with data from a survey we designed on beliefs and personality traits. We find that men who were conscripted are more likely to adopt a military mind-set and that the effect is long lasting. Given the many people who go through military conscription, our results are useful for understanding how personality traits and beliefs are formed for a very salient part of the world's population.

Can investing in failing schools help them improve? This paper studies this question using a natural experiment based on a 2017 lawsuit settlement that allocated substantial resources to the lowest-performing schools in the Los Angeles Unified School District (LAUSD). Using a difference-in-differences design, I compare 50 secondary schools that received an increase of 13.5% on average in their annual budgets for three years, to nearby public, noncharter schools that received no settlement funding. The intervention mandated hiring of additional staff members and allocating of funds for professional development, but allowed discretionary spending on initiatives for high-need students. I find that, in line with the intent of the settlement, schools hired more personnel, including instructional staff such as teachers and counselors, and support personnel such as paraprofessionals and school service staff, all effects statistically significant. Settlement schools achieved a 0.75 percentage points reduction in suspension rate relative to unfunded schools, reflecting a marked improvement in performance outcomes. These reductions were particularly notable given that the settlement triggered demographic sorting, with the treated schools losing students overall and shifting toward higher concentrations of economically disadvantaged students and lower Black enrollment. A simple bounding exercise that accounts for demographic sorting indicates that the settlement had meaningful effects on suspension rates, suggesting real improvements on noncognitive dimensions of schooling. Survey evidence suggests that two key mechanisms for lower suspensions were improvements to school climate as reported by staff and students, and enhancements to educators’ capacity and disciplinary approaches.

High temperatures have been found to worsen mental health, but evidence from lowand middle-income countries remains scarce, where adaptive capacity is limited. Using nationwide weekly administrative data covering the entire population and a survey representative of urban Mexico, we estimate heat effects on self-reported well-being, emergency department visits for mental illness, and suicide using a unified empirical design. Comparing hotter and cooler weeks within the same locations and calendar week, we find that one extra day above 30°C (86°F) increases suicides by 3.2% and emergency department visits for mental illness by 2.4%, and lowers reported mental well-being. The effect sizes exceed comparable U.S. estimates but are attenuated in municipalities served by psychiatrists, suggesting protection through access to care. Gender patterns differ from U.S. studies: women experience larger emergency department visit increases, while suicide effects are concentrated among men. Our results indicate that heat imposes mental health burdens from reduced well-being to clinical demand and mortality, highlighting mental health as a key component of climate damages and a priority for adaptation investments in health systems.

We study the effects of a major immigration reform on alleged and substantiated maltreatment of Hispanic children using administrative data from child protective services agencies. Secure Communities ties federal immigration enforcement to local law enforcement, effectively increasing the likelihood of deportation for undocumented immigrants who are arrested for a crime. We exploit the staggered rollout of Secure Communities across counties to estimate a dynamic treatment effect model. We find that Secure Communities implementation increased the number of Hispanic children per 1000 found to be victims of child maltreatment as well as the likelihood that maltreatment allegations for Hispanic children are substantiated, consistent with increased average severity of investigated cases.

The rise of AI-powered tools for mental health support has generated enthusiasm about their potential to expand access to mental health support at low cost. Yet important questions remain about effectiveness, user engagement, and whether such tools might exacerbate feelings of isolation or crowd out traditional forms of care. We evaluate the effects of an AI-powered digital mental health app in a randomized controlled trial among women experiencing at least some mild psychological distress in Mexico. Access to the app significantly improved mental health outcomes during the 8-week study period: it reduced symptoms of anxiety, depression, and stress; improved subjective well-being, feelings of isolation, and sleep quality; and increased the prevalence of healthful behaviors. In addition, treated respondents were more likely to seek traditional psychotherapy. Engagement with the app was initially high, though use declined over time. However, effects persisted, potentially consistent with "as-needed" use. Our findings suggest that scalable digital tools can improve mental health outcomes, benefiting both individuals with limited sources of support and those who already have access to other forms of care.

Public school choice has evolved rapidly in the past two decades, as districts roll out new magnet, dual-language, and themed programs to broaden educational opportunity. We use newly collected national data to document that opt-in (voluntary) systems: (i) are the modal design; (ii) are harder to navigate; and (iii) have participation that is concentrated among more advantaged students. These facts suggest a striking inconsistency: districts have largely adopted centralized assignment algorithms to broaden access, but most rely on optional participation that fragments public education. We study the implications of this design choice in the Los Angeles Unified School District, the largest opt-in system in the country, combining two decades of administrative data, randomized lotteries, and quasi-experimental expansions in access. Participation is highly selective, consistent with national evidence, and lottery estimates suggest that the students with the lowest demand for choice schools are the ones who gain the most from attending. Opt-in participation therefore embeds a selection mechanism that screens out high-return students and leaves many effective programs with unused capacity. To evaluate system-level implications, we estimate a structural model linking applications, enrollment, and achievement. Choice schools are vertically differentiated and generate meaningful gains, but the opt-in participation rule—through high application costs and negative selection on gains—prevents these benefits from reaching the students who need them most. Counterfactual simulations make the design stakes clear: information and travel-cost reductions have limited effects, whereas reforms that change the participation architecture eliminate core inefficiencies and deliver the largest district-wide achievement gains. These results underscore that system design—not school effectiveness alone—shapes who benefits from public school choice and to what extent.