This study examines whether a supply-side intervention on prescription drugs, the introduction of an abuse-deterrent formulation of OxyContin, increased the homicide rate in the United States. If so, could this effect be mitigated? First, leveraging cross-state variation in pre-reformulation OxyContin exposure, difference-in-difference estimates show that OxyContin reformulation led to a transitory increase in the homicide rate. This effect is strongest among victims between 15 and 24, which is plausibly due to an increase in illicit drug distribution associated with the rise in post-reformulation demand for illicit opioids. The study then explores the role of medical marijuana laws in moderating post-reformulation homicide rate and finds that medical marijuana legalization is associated with a reduction in post-reformulation homicide rate. This effect is strongest among states where the pre-reformulation OxyContin misuse rates are high.

We explore the firm internationalization’s impact on firm credit ratings in emerging economies. Adopting Chinese data from 2009 to 2018, we document that firm internationalization varies negatively with its credit ratings, indicating that emerging debt market participants are risk averse and prioritize the risks involved in firm internationalization endeavors. This association is amplified for firms operating in host countries with lower institutional quality, decreased cultural distance from home countries, and when firms do not hold tax haven subsidiaries. We observe that the main association is consistent when alternative dataset (India, Russia, and Brazil) or proxy (cost of debt) is applied.

Between 1999 and 2021, the opioid crisis has caused nearly 650000 opioid-related overdose deaths, with little sign of slowing down (CDC). However, the overdose death toll unlikely captures all the hidden costs of the opioid crisis. In this paper, I study whether a supply-side intervention on prescription drugs, the introduction of an abuse-deterrent formulation of OxyContin, increased suicide victims. Using the National Survey of Drug Use and Health (NSDUH) and the National Vital Statistics System (NVSS) multiple Causes of Death data and leveraging cross-state variation in pre-reformulation OxyContin exposure, difference-in-difference estimates show that this intervention led to an increase in post-reformulation suicide victims. Moreover, this effect persists for victims whose underlying cause of death is not directly drug-related, suggesting psychological distress might play an essential role in this increase.

In recent years, the United States has been gradually adopting a harm reduction approach to its drug control policies, reducing penalties associated with the possession of drugs. The Safe Neighborhoods and Schools Act, known as California’s Proposition 47, is an example. Passed by California voters in November 2014 and became effective immediately since, it reclassified simple possession (possession with no intent to distribute) of “hard drugs,” such as methamphetamine, from a wobbler offense (a crime that can be punished as either a misdemeanor or a felony) to a misdemeanor. In this paper, I study whether this legislative change affects the prevalence of common sexually transmitted diseases (STDs). The primary data source for measures of STDs is the National Center For HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (NCHHSTP) Atlas-Plus, from which I obtained county-level prevalence data for chlamydia, gonorrhea, and primary and secondary syphilis. Exploring the timing of the implementation of this policy and using counties of neighboring states (Arizona, Nevada, Oregon) as the control group, difference-in-difference regression estimates show that there is little evidence that suggests Proposition 47 increased the prevalence of these conditions. On the other hand, event study estimates show that there is a temporary increase in the prevalence of chlamydia after 2014.

Extracurricular activities are an integrated part of many students’ secondary school experience and benefit their accumulation of human capital, especially non-cognitive skills. Peer characteristics likely influence a student’s decision to participate in extracurricular activities, yet such empirical evidence is limited. This paper studies how a high school student’s participation in non-academic and non-athletic activities affects her peer height. To answer this question, I use a student in-home survey and parent study from wave I of the National Longitudinal Study of Adolescent Health (ADD HEALTH) and use a cross-cohort, within-school research design to explore the plausibly exogenous variation in peer height to identify its effect on participation in the school newspaper or student council. Fixed effects estimates show that having taller peers is associated with a lower probability of participating in social activities. This effect vanishes after controlling for the socioeconomic status of peers’ parents, suggesting that it could be mainly attributed to skill variations among students’ families. Heterogeneous analysis reveals that male students are only influenced by the height of their male peers, negatively affecting their participation in social activities. Conversely, female students are unaffected by the height of their female peers but are more likely to participate in social activities if they have taller male peers.