Pretrial detention accounts for a growing share of jail incarceration, yet the large literature on bail reform focuses almost entirely on monetary bail and low-level offenders, leaving reforms that target violent offenders largely unexamined. This paper studies the repeal of the Presumptive Denial of Bail statute in Virginia, which made defendants charged with specified violent offenses presumed ineligible for bail. Using administrative court records from the Virginia Court System and a difference-indifferences design comparing defendants charged with treated and control offenses, the author estimates the effects of repeal on case outcomes, pretrial misconduct, and recidivism. The author finds that the repeal increased pretrial release among treated defendants by 6.4%, and lowered conviction and incarceration rates by 11.4% and 11.7%, respectively. The author argues this improvement in case outcomes is driven by a reduction in the coercive pressure detention places on defendants to accept unfavorable plea deals. Although bail laws are designed to prevent pretrial misconduct and recidivism, increasing access to bail does not increase failure to appear for court hearings or recidivism. These findings suggest that restoring judicial discretion over pretrial detention, even for violent charges, improves case outcomes without compromising court appearance or public safety.

The tort reform movement in the United States has had significant effects on the civil justice system and public perceptions of the law. Although tort law has been reformed multiple times, what is commonly known as the tort reform movement generally refers to a political movement launched in the mid-1980s by large corporations and insurance companies to make it more difficult for people to sue. This article explores the movement's history, its direct impacts on legal doctrine and judicial practices, and its indirect consequences for legal consciousness and perceptions of the courts. Tort reformers strategically propagated misleading narratives that emphasized a litigation crisis and decried the competence of juries, influencing public opinion and shaping legislative and judicial reforms that restricted access to justice. Despite significant empirical evidence debunking many of the tort reformers’ claims, these narratives continue to resonate. This article advocates for renewed scholarly attention to these effects and to other, potentially more far-reaching, impacts of tort reform, such as the movement's impacts on the rights of politically and legally vulnerable groups, public confidence in the civil justice system, and views of legal institutions and actors more broadly.

This article provides global context that will help American judges better understand how efforts to undermine judicial independence play out. Several established democracies have recent experience of institutions working to limit judicial independence when they are dissatisfied with court rulings. Part 1 defines “judicial independence.” Judicial independence is “the ability of courts to exercise their power ‘free of influence or control by other actors.’” The author argues that judicial independence does not necessarily mean that courts are the final word on matters of constitutional interpretation or that public opinion plays no role in the trajectory courts take over time. Part 2 explains why societies might perceive that protecting judicial independence is valuable. They might do so because broad sections of the public respect courts and view their decisions as legitimate, such that they expect officials in other branches of government to obey even decisions they dislike. Parts 4 and 5 describe how governments have taken steps to limit the independence of courts they perceived as thwarting their objectives and how one has largely succeeded in eroding judicial independence.

Preparing more students for well-paying, high-opportunity careers increasingly requires helping them continue into postsecondary education and training after high school. The partnerships included in this report demonstrate that community colleges and K-12 schools can make greater use of educational structures that already exist on a large scale—high school Career and Technical Education (CTE) pathways, career academies, and community college CTE programs—by coherently integrating dual enrollment into those programs. Although the strategies for doing this differ based on structure, they share a common purpose: Helping more students move from career exploration in high school to postsecondary education and training that leads to high-opportunity careers. Whether dual enrollment is embedded in traditional CTE pathways, career academies, or college CTE programs for high school students, the goal is the same—to transform existing educational pathways into stronger on-ramps to college and career opportunity. Demand continues to grow for both dual enrollment and high-quality career programming in high school. The examples in this report point toward a shared vision: Dual enrollment should offer more than just the opportunity to earn college credit and career programming should do more than just prepare students for a first job after high school.

After high school, many young people hit stretches of time when they’re not engaged in school, training, or work. These periods—called disconnection—are costly for young people and for society. But little is known about how long disconnection typically lasts or how it shapes young people’s broader well-being. This study uses survey data from the American Youth Panel to trace patterns of disconnection among 18- to 22-year-olds, identify who is most likely to experience the longest periods of disconnection, and examine how disconnection relates to social interaction and mental health. Key findings include that long periods of disconnection are common among American youth, with nearly one in four reporting disconnection lasting at least six months and 11% reporting disconnection of a year or more. In addition, nearly one in four disconnected young men reported that their mental health was “not good” every day of the prior month, and around two-thirds of young men experiencing disconnection reported no in-person time with friends in the prior week, a stark contrast to their peers who had never been disconnected. The authors conclude that for disconnected young men, especially, education and employment programs alone may miss critical needs around social isolation and well-being.

One in five children in the U.S. speaks a language other than English at home, and about half of these students are classified as Multilingual/English Learners (ML-ELs). ML-ELs require specialized language support services, which are intended to be temporary, and each state sets its own criteria for determining when a student is English proficient and can be reclassified as no longer ML-EL. As of 2023, 18 states include non-English language proficiency assessments in the reclassification process, including Florida. This study explores what happens to secondary school ML-ELs when a state brings academic content knowledge in addition to English proficiency into the ML-EL reclassification process. Findings include that reclassification based on academic English language arts criteria increased the likelihood of successfully completing an advanced course (e.g., Advanced Placement) in high school by 7 percentage points. At the same time, students reclassified in eighth grade struggled in their core high school classes and were 8-10 percentage points more likely than peers who remained ML-EL to fail core courses and 10 percentage points less likely to graduate on time. Per the authors, these mixed findings suggest needs for greater course access for ML-ELs and increased support in their transition out of ML-EL status.

In 2024, the state of California raised the state minimum wage to $20 per hour for employees at limited-service fast-food chains with at least 60 establishments nationwide, including franchises. This paper analyzes the effect of the minimum wage increase on employment, finding negative general effects with considerable variation among chains of different sizes. For example, effects on employment were small and statistically insignificant for the 20 largest chains affected by the policy but large and significant for smaller chains. In addition, the total number of limited-service fast food restaurants fell after the minimum wage increase, with effects driven by exits of smaller chains. The authors conclude that the overall effects of the minimum wage increase have been negative on employment. For the larger chains, however, the shorter-run effects are insignificant, suggesting that adverse employment effects may appear slowly for firms of different sizes. For the smaller chains, the negative effects on employment, establishment count, and entry (i.e., new firms) are already apparent. Moreover, the decline in entry among smaller chains may have helped deter employment declines among larger chains, a pattern which may account for insignificant observed effects on larger firms that potentially change in the future.

Americans have grown more negative toward data centers since the beginning of 2026 according to a nationally representative survey of U.S. adults (n=10,548). Findings include that rural Americans have become more likely to say data centers are mostly bad for the environment (from 32% in January to 50% in August), home energy costs (32% to 46%) and the quality of people living nearby (28% to 45%). This matters because most data centers are currently being planned or developed in rural areas. Democrats and Democratic-leaning independents are still more likely than Republicans and Republican leaners to view the impacts of data centers negatively. However, Republicans’ views have turned more negative, with 42% now saying say that data centers are mostly bad for the quality of life of people nearby, up from 24% in January. In addition, more Democrats now say the effects of data centers on local jobs and tax revenue are mostly bad than mostly good. Although more Republicans say the effects of data centers on local jobs and tax revenue are mostly good, they are 7 percentage points more likely than in January to say the effects in both areas are mostly bad.

The menopausal transition is a critical window for the health of women and may serve as an inflection point for the onset of chronic disease. During this transition, a woman’s ovaries become less efficient at producing estrogen and progesterone, and women may experience symptoms such as hot flashes, night sweats, insomnia or sleep disturbances, and abnormal uterine bleeding. Key findings include that in 2025, nearly one in two women ages 40–60 who were not currently pregnant, postpartum, or in surgical menopause reported menopausal symptoms in the past 12 months (49.4%). White non-Hispanic women (54.3%) were more likely to report menopausal symptoms compared with Black non-Hispanic (45.3%), Hispanic (44.3%), and Asian non-Hispanic (36.1%) women. Women with a high school diploma or GED, some college or associate degree, or a bachelor’s degree or more were more likely to report menopausal symptoms compared with women with less than a high school diploma. The percentage of women who report menopausal symptoms varied by urbanization level of their place of residence. Among women with menopausal symptoms in the past 12 months, approximately 15.6% indicated that they were currently taking a prescription medication for those symptoms.

Although generative artificial intelligence (GenAI) has been used as a source of general health information, its specific pertinence to mental health remains unclear. This study examines the relationship between the use of GenAI for health issues and anxiety and depression using data from a nationally representative sample of nearly 100,000 U.S. college students. Key findings include that the odds of screening positive for clinically significant anxiety and depression were 52% and 46% greater for people who reported using GenAI tools like ChatGPT or Claude for health-related questions, and these associations remained consistent when controlling for prior anxiety and depression diagnoses. According to the authors, the results raise questions about what happens when people repeatedly turn to GenAI when worried about their health. For example, people who are uncertain about symptoms may ask for reassurance from an AI system, receive an answer, and then develop additional questions or uncertainties. According to the authors, repeated cycles of questions could reinforce health anxiety or other distress and explain the study’s findings.