This paper studies the impact of adult prosecution on recidivism and employment trajectories for adolescent, first-time felony defendants. The authors use extensive linked Criminal Justice Administrative Record System and socio-economic data from Wayne County, Michigan (Detroit). Using the discrete age of majority rule and a regression discontinuity design, the authors find that adult prosecution reduces future criminal charges over 5 years by 0.48 felony cases (a 20% decrease) while also worsening labor market outcomes: 0.76 fewer employers (a 19% decrease) and $674 fewer earnings (a 21% decrease) per year. The authors develop a novel econometric framework that combines standard regression discontinuity methods with predictive machine learning models to identify mechanism-specific treatment effects that underpin the overall impact of adult prosecution. Using these estimates, the authors consider four policy counterfactuals: (1) raising the age of majority, (2) increasing adult dismissals to match the juvenile disposition rates, (3) eliminating adult incarceration, and (4) expanding juvenile record sealing opportunities to teenage adult defendants. All four scenarios generate positive returns for government budgets. When accounting for impacts to defendants as well as victim costs borne by society stemming from increases in recidivism, the authors find positive social returns for juvenile record sealing expansions and dismissing marginal adult charges; raising the age of majority breaks even. Eliminating prison for first-time adult felony defendants, however, increases net social costs. Policymakers may still find this attractive if they are willing to value beneficiaries (taxpayers and defendants) slightly higher (124%) than potential victims.

States across the country have increasingly adopted statistical risk assessment tools in multiple stages of their criminal legal systems with the hope of reducing incarceration without increasing crime. These tools use various characteristics to estimate an individual’s future risk of recidivism, and judges consider the results of these assessments when determining levels of custody or community supervision for convicted individuals. Despite much debate amongst academics and activists on the utility and fairness of these tools, one critique seems beyond debate: the tools are built for men, not women. These tools are based on criteria, statistics, and theory drawn from the experiences of men and thereby result in inaccurate and inequitable sentencing when applied to women. When women are sentenced according to the higher rates of violence and recidivism that are associated with men, they are often incarcerated or under supervision longer than justified by their gender-specific risk to society. The unfairness of these assessments is specifically concerning when one considers that, as of 2019, 1.2 million women in the United States were under the supervision of the criminal legal system, with approximately 58% of them leaving at least one minor child at home. Separate risk assessment tools for men and women can combat the inaccurate sentencing of women. While many commentators have argued for separate tools for men and women, they have not sufficiently addressed how such an approach would survive legal, theoretical, and policy hurdles. This report argues (1) that gender-specific assessments could survive an equal protection challenge; (2) that such assessments for women should be implemented despite the need for further research and work on the conflation of sex and gender and the utilization of a gender binary in the United States criminal legal system; and (3) that they could be adapted for women defendants without opening the floodgates to a demand for assessments designed for every conceivable category of criminal defendant.

Earlier this year, a Fortune Magazine headline proclaimed that “The next era of work will be about skills – not pedigree,” suggesting that employers are moving away from standard indicators of potential employee success, like college degrees and years of experience, and prioritizing tangible skills. At the same time, the rise in artificial intelligence (AI) bots that can write like Shakespeare and win art competitions has led to concerns that technology will make many current jobs obsolete. These changes represent a seismic shift in the future of the workplace. Rather than job-specific knowledge and skills, students will need to do what AI bots can’t do—be curious and creative. There is no indication that our current education system is preparing students for this new reality. Fully 28 years ago, U.S. surveys revealed that kindergarten teachers knew that curiosity and creativity were even more predictive of later academic success than was teaching the alphabet or teaching how to count, and a more recent study provided evidence for this intuition. Yet, children report that curiosity is inconsistent with what they are expected to do in school, and observations of preschool and elementary classrooms support this report. Given this reality, some are asking how to fill the curiosity and creativity gap between what schools are doing and what the workplace needs for the future. A recent book “Making Schools Work,” lays out a road map for a new way forward for schools. The book presents scientific data and suggests that one way to transform the system is to teach in the way that human brains learn, to create active learning environments that are meaningful and engaging. Developing student curiosity and creativity will be central to that charge. Cultivating curiosity and creativity will lead to happier students who are more prepared to learn and meet the challenges of the future.

This study integrates an intersectional framework with data on 15,000 U.S. ninth graders from the High School Longitudinal Study of 2009 to investigate differences in ninth-grade math course placement at the intersection of adolescents’ learning disability status, race, and socioeconomic status. The study examines the increased liability perspective in which the negative relationship of a learning disability would be exacerbated for youth who are higher status in terms of their race and/or socioeconomic status. Descriptive results support an increased liability perspective, with the negative relationship between a learning disability and math course placement larger for adolescents more privileged in terms of their race and/or socioeconomic status. Adjusted results suggest that the lower math course placements of youth with learning disabilities are due to cumulative disadvantage rather than disability-related inequities in the transition to high school for youth of diverse racial and socioeconomic backgrounds. In addition to demonstrating the importance of intersectional perspectives, this study provides a roadmap for future studies by introducing the new perspective of increased liability to be used in conjunction with the widely employed perspective of multiple marginalization, which describes the negative relationship of a learning disability that would be exacerbated for youth who are higher status in terms of their race and/or socioeconomic status.

This report examines current teacher shortages in the United States and examines seven key areas of focus for supporting recruitment, preparation, support, and retention in teaching. According to the U.S. Department of Education, all 50 states reported shortages in more than one area for the 2022–23 school year. Among them were especially widespread shortages of special education teachers, science teachers, and math teachers. To handle the shortages, schools have increased class sizes, canceled course offerings, added duties to the responsibilities of existing teachers, and hired underqualified individuals to fill the positions—all of which undermine students’ learning. To address this, the report recommends: (1) increasing educators’ net compensation through tax credits, housing subsidies, and salary incentives; (2) strengthening recruitment by making teacher preparation debt-free through expanding service scholarships and loan forgiveness programs to fully cover the cost of comprehensive preparation at the undergraduate and graduate levels; (3) supporting improved preparation by expanding high-retention pathways into teaching such as Grow Your Own programs which recruit and train school staff and community members to become certificated teachers; (4) providing high-quality mentoring for all beginning teachers; (5) increasing investments that enable educators to expand and share expertise; (6) incentivizing the redesign of schools to support teaching and learning; and (7) rethinking school accountability to center measures of school progress around authentic measures of school quality and equity that inform improvement, rather than around punitive metrics that make it more difficult to recruit and retain teachers willing to work in the highest-need schools.

The Census Bureau released the 2022 Census of Governments — Organization, a compilation of the total count and types of all local governments, including special districts, in the 50 states and the District of Columbia. The Census Bureau classifies local governments (primarily funded by taxpayer dollars) as either general purpose (counties, municipalities, and townships) or special purpose (independent school districts and special district governments). Special district governments operate independently from a local county or municipality. They have the legal power to collect their own tax revenues necessary to provide services that benefit our communities, such as irrigation, water treatment, and soil and water conservation. The publication shows that 16,020 or 41% of the nation’s 39,555 special district governments in 2022 supported parts of the economy related to natural resources. In comparison, there were fewer (38,542) special district governments in 2017 but slightly more (16,145) devoted to natural resources. In general, the number of natural resources special districts remained relatively consistent over the five-year period. Illinois had the highest number of local government entities (6,930) in 2022, while the District of Columbia had the lowest number (2). Florida had 1,947 local government entities (including 1,374 special districts) in 2022.

In this report, the authors create benchmarks for comparison with the U.S. Department of the Air Force's accession cohorts by estimating the fraction of the eligible (and propensed) population, using ten mutually exclusive categories of gender and race and ethnicity. The benchmarks provide a measure of progress on diversity and inclusion in the force and a comparison to clearly identify whether a demographic's overrepresentation or underrepresentation can be attributed to specific eligibility standards or propensity to serve, or both. The report found that eligibility requirements limit racial and ethnic minority representation, but propensity to serve offsets barriers to eligibility for these minorities. Body mass index, height, and education and aptitude requirements are the most important barriers to both enlisted and officer eligibility, but these requirements affect the eligibility of gender and racial and ethnic groups differently. Considering gender and race and ethnicity jointly, no minority group meets the demographic benchmarks of the U.S. population that is both eligible and has a propensity to serve across the three accession sources (enlisted, the United States Air Force Academy, and Reserve Officers' Training Corps or Officer Training School). The report recommends that benchmark department accessions and consider gender and race and ethnicity jointly. Examining the intersection of gender and race and ethnicity allows for a more accurate view of the effects of eligibility requirements on demographic distributions by sifting out patterns that might otherwise be obscured by the large representation of White men and women.

In this brief, the authors provide a literature review of three simulation studies that model the potential impacts of baby bonds, with a focus on outcomes relating to racial wealth equity. Baby bonds are publicly funded child trust accounts that target children from low-wealth or low-income families. When the children reach adulthood, they can use the funds for wealth-building activities, such as purchasing a home or starting a small business. The brief also reviews the literature of related early life wealth-building programs (e.g., child development accounts) to assess outcomes that may be achievable with baby bonds policies. The three simulations analyzed all find that baby bonds would reduce Black-White racial wealth inequities, though they differ in scale. Zewde (2020) predicts the biggest improvement in racial wealth inequities—predicting that at the median, the White/Black gap would be reduced to a factor 1.4 to 1 ($79,000 to $58,000), compared to of 15.8 to 1 ($46,000 to $2,900). Weller et al. (2021) estimate that the White/Black wealth gap will reduce to a factor of approximately 2.7 to1 by 2060, still leaving a gap of $1.37 million. Mitchell and Szapiro (2020) predict baby bonds would narrow the White/Black wealth gap to a factor of about 3.4 to 1 at the median for adults ages 18–25, with a remaining gap of over $90,000.

In 2021, 140 million emergency department (ED) visits occurred in the United States. During that year, about 4% of children had two or more ED visits in the past 12 months, and 18% of adults had visited the ED in the past 12 months. This report presents characteristics of ED visits by age group, sex, race and ethnicity, insurance, and mentions of COVID-19, using data from the 2021 National Hospital Ambulatory Medical Care Survey. Key findings include that the overall ED visit rate was 43 visits per 100 people in 2021. ED visit rates were highest for infants under age 1 year (103 visits per 100 infants) and adults aged 75 and over (66 per 100 people). The ED visit rate for Black or African-American non-Hispanic people (81) was the highest among the selected racial and ethnic groups. The ED visit rate for patients with private insurance was lowest compared with all other primary expected sources of payment, and the rate for patients with Medicaid was highest. In 2021, a COVID-19 diagnosis was confirmed for 3.8% of all ED patient visits.

While Artificial Intelligence (AI) algorithms have achieved performance levels comparable to human experts on various predictive tasks, human experts can still access valuable contextual information not yet incorporated into AI predictions. Humans assisted by AI predictions could outperform both human-alone or AI-alone. In this study, the authors conducted an experiment with professional radiologists that varied the availability of AI assistance and contextual information to study the effectiveness of human-AI collaboration and to investigate how to optimize it. Findings revealed that (i) providing AI predictions does not uniformly increase diagnostic quality, and (ii) providing contextual information does increase quality. Radiologists do not fully capitalize on the potential gains from AI assistance, as they may partially underweight the AI’s information relative to their own and not account for the correlation between their own information and AI predictions. In light of these biases, the authors designed a collaborative system between radiologists and AI. Results demonstrate that, unless the documented mistakes can be corrected, the optimal solution involves assigning cases either to humans or to AI, but rarely to a human assisted by AI.

Safety-net hospitals (SNHs) are ideal sites to deliver addiction treatment to patients with substance use disorders, but the availability of these services within SNHs nationwide remains unknown. This study examines differences in the delivery of different substance use disorders programs in SNHs vs. non-SNHs across the U.S. and to determine whether these differences are increased in certain types of SNHs depending on ownership. This cross-sectional analysis used data from the 2021 American Hospital Association Annual Survey of Hospitals to examine the associations of safety-net status and ownership with the availability of substance use disorders services at acute care hospitals in the U.S. A total of 2,846 hospitals were included: 409 were SNHs and 2,437 were non-SNHs. The lowest proportion of hospitals reported offering inpatient treatment services (791 hospitals [27%]), followed by medications for opioid use disorder (1,055 hospitals [37%]), and outpatient treatment services (1,087 hospitals [38%]). The majority of hospitals reported offering consultation (1,704 hospitals [60%]) and screening (2240 hospitals [79%]). The SNHs were significantly less likely to offer substance use disorders services across all five categories of services (screening, consultation, inpatient services, outpatient services, and medications for opioid use disorder). With the exception of medications for opioid use disorder, public or for-profit SNHs did not differ significantly from their non-SNH counterparts. However, nonprofit SNHs were significantly less likely to offer all five substance use disorders services compared with their non-SNH counterparts. These findings add to a growing body of research suggesting that SNHs may face additional barriers to offering substance use disorders programs. Further research is needed to understand these barriers and to identify strategies that support the adoption of evidence-based substance use disorders programs in SNH settings.