Skip to main content


Office-Based Mental Healthcare and Juvenile Arrests
Monica Deza, City University of New York and NBER
Thanh Lu, Cornell University
Johanna Catherine Maclean, Temple University and NBER

In this study, Deza, Lu, and Maclean investigate the effect of local access to office-based mental healthcare providers (e.g., psychologists and psychiatrists in private practice) on juvenile arrest outcomes. Specifically, the researchers estimate the effect of changes in the number of office-based mental healthcare providers per county on county-level juvenile arrest outcomes in the United States over the period 1999 to 2016. Office-based treatment accounts for 59% of mental healthcare received by juveniles and is therefore an important modality to study. Deza, Lu, and Maclean find that better access to office-based mental healthcare reduces juvenile arrest outcomes. In supplementary analyses, the researchers document that these reductions are attributable to improved mental health among juveniles. Their findings provide support for increasing local access to mental health treatment as a means to reduce juvenile arrest outcomes. Crime imposes substantial costs on society and individuals, and interventions during early life can have more pronounced effects than those received at later stages, therefore their results imply increased juvenile access to mental healthcare may have an unintended benefit for the current and future generations.


This paper was distributed as Working Paper 29465, where an updated version may be available.

A Gift of Health: The Duke Endowment’s Impact on Hospital Care and Mortality
Alex Hollingsworth, Indiana University and NBER
Krzysztof Karbownik, Emory University and NBER
Melissa A. Thomasson, Miami University and NBER
Anthony Wray, University of Southern Denmark

Does access to hospital care affect infant and later-life mortality? Hollingsworth, Karbownik, Thomasson, and Wray study this question in the context of a large-scale expansion in hospital care in North Carolina that was facilitated by funding from the Duke Endowment starting in the mid-1920s. Difference-in-differences estimates imply that counties exposed to Duke funding experienced reductions in infant mortality and mortality at older ages. These gains are qualitatively larger for African Americans compared to Whites. Plausible mechanisms include improved access to nursing care and higher-skilled physicians. On the other hand, Hollingsworth, Karbownik, Thomasson, and Wray do not find evidence of additional gains from new pharmaceutical treatments available to hospital patients.

Bad Lighting: Effects of Youth Indoor Tanning Prohibitions
Christopher S. Carpenter, Vanderbilt University and NBER
Brandyn F. Churchill, Vanderbilt University
Michelle M. Marcus, Vanderbilt University and NBER

Indoor tanning beds (ITBs) emit UV light at high intensity and have been classified as carcinogenic to humans by the World Health Organization since 2009. Carpenter, Churchill, and Marcus are the first to study the role of state laws prohibiting youths from indoor tanning using a difference-in-differences research design. The researchers find that youth ITB prohibitions reduced population search intensity for tanning-related information. Among white teen girls, ITB prohibitions reduced self-reported indoor tanning and increased sun protective behaviors. They also find that youth ITB prohibitions significantly reduced the size of the indoor tanning market by increasing tanning salon closures and reducing tanning salon sales.


This paper was distributed as Working Paper 29443, where an updated version may be available.

"I Quit": The Role of Schedule Volatility in Employee Turnover
Alon Bergman, University of Pennsylvania
Guy David, University of Pennsylvania and NBER
Hummy Song, University of Pennsylvania

Bergman, David, and Song examine how employer-driven volatility in workers' schedules impacts their decision to voluntarily leave their job. Using time-stamped work log data of home health nurses, the researchers construct and study a novel measure of schedule volatility. This measure may be endogenous to the worker's decision to quit. Hence, they instrument for schedule volatility using time off taken by coworkers. Bergman, David, and Song find that higher levels of schedule volatility substantially increase workers' likelihood of quitting. Through policy simulations, the researchers illustrate how schedule volatility, and employee turnover, could be mitigated. Bergman, David, and Song conclude by discussing the generalizability of their results to other industries.

Towering Intellects? Sizing Up the Relationship Between Height and Academic Success
Stephanie G. Coffey, Syracuse University
Amy Ellen Schwartz, Syracuse University

Do tall students do better in school? A robust literature documents the links between physical characteristics and economic outcomes. Tall adults earn more, while obese ones earn less, and obese children perform worse at school. However, Coffey and Schwartz know little about the potential academic origins of the height earnings gradient. Height may affect outcomes through two broad channels. First, height may reflect underlying individual characteristics, such as childhood (dis)advantage or epi-genetic factors which shape academic performance directly (through cognitive ability, say) or indirectly (through health). Second, height may affect social interactions or opportunities. For example, shorter students may be disproportionately bullied or, conversely, taller students included in sports. These suggest height relative to classmates may be particularly salient and a key driver of the well documented positive effects of relative age. There is, unfortunately, a dearth of research in this area, due, in part, to the scarcity of data linking student academic outcomes and height. In this paper, Coffey and Schwartz use unique student-level longitudinal data from New York City (NYC) to estimate the height premium in academic performance in grades 3 through 8, exploring the sensitivity of results to alternative height measures and specifications including school and/or student fixed effects and examining the role of race, gender and relative age.
In a parsimonious model, Coffey and Schwartz find a statistically and substantively significant height gradient in math and ELA – up to .048 sd. increase in achievement for each sd. increase in height. The researchers estimate a positive height gradient for boys and girls and among students of all races. Differences at the extremes important – the tallest students perform as much as .2 sd. better than the shortest. Height is not meaningfully related to absenteeism, suggesting results are not due to contemporaneous health problems, while controlling for obesity (BMI) increases estimated effects, suggesting physical attractiveness may be relevant. Student fixed effects models yield substantially smaller estimates – particularly among girls, providing support for the hypothesis that both height and cognitive ability are shaped by gene-environment interactions in the early years (Case and Paxson, 2008; Schick and Steckel, 2015). Still, after controlling for underlying, time-invariant differences and observed, time-varying differences the tallest boys (girls) score up to .11 (.073) sd. better than the shortest. Coffey and Schwartz also find evidence that ordinal height rank relative to peers may have a small effect on achievement conditional on cardinal height. Moving from being the shortest boy (girl) within grade is associated with an increase in ELA achievement of .054 (.085) sd conditional on cardinal height. This paper contributes to a long-standing literature on the effect of age-within-grade on achievement. Their estimates suggest that failing to account for relative height may upwardly bias the relationship between relative age and achievement by up to 25%, and a one sd. increase in height is associated with an increase in achievement roughly equal to the one associated with a six month increase in age.

Uber and Alcohol-Related Traffic Fatalities
Michael L. Anderson, University of California, Berkeley and NBER
Lucas W. Davis, University of California, Berkeley and NBER

Previous studies of the effect of ridesharing on traffic fatalities have yielded inconsistent, often contradictory conclusions. In this paper Anderson and Davis revisit this question using proprietary data from Uber measuring monthly rideshare activity at the Census tract level. Most previous studies are based on publicly-available information about Uber entry dates into US cities, but the researchers show that an indicator variable for whether Uber is available is a poor measure of rideshare activity — for example, it explains less than 3% of the tract-level variation in ridesharing, reflecting the enormous amount of variation both within and across cities. Using entry Anderson and Davis find inconsistent and statistically insignificant estimates. However, when the researchers use the more detailed proprietary data, Anderson and Davis find a robust negative impact of ridesharing on traffic fatalities. Impacts concentrate during nights and weekends and are robust across a range of alternative specifications. Overall, their results imply that ridesharing has decreased US alcohol-related traffic fatalities by 6.1% and reduced total US traffic fatalities by 4.0%. Based on conventional estimates of the value of statistical life the annual life-saving benefits range from $2.3 to $5.4 billion. Back-of-the-envelope calculations suggest that these benefits may be of similar magnitude to producer surplus captured by Uber shareholders or consumer surplus captured by Uber riders.


This paper was distributed as Working Paper 29071, where an updated version may be available.

The Impact of Legal Abortion on Maternal Health: Looking to the Past to Inform the Present
Sherajum Monira Farin, Georgia State University
Lauren Hoehn-Velasco, Georgia State University
Michael F. Pesko, Georgia State University

From 1959 to 1980, abortion-related mortality declined by 97%, and maternal mortality fell by 86%. In this study, Farin, Hoehn-Velasco, and Pesko question whether the legalization of abortion over 1969-1973 explains a portion of this maternal mortality decline. Their results suggest that legal abortion reduced non-white maternal mortality by 30-40%, with little impact on overall or white maternal mortality. Farin, Hoehn-Velasco, and Pesko also find that early state-level legalizations were crucial, and explain more of the observed mortality decline than the Roe v. Wade decision itself. Overall, their findings suggest that legal abortion substantially improved maternal health for disadvantaged groups.

Cigarette Taxes, Smoking, and Health in the Long Run
Andrew I. Friedson, University of Colorado Denver
Moyan Li, Indiana University
Katherine Meckel, University of California, San Diego and NBER
Daniel I. Rees, Universidad Carlos III de Madrid
Daniel W. Sacks, Indiana University

Medical experts have argued forcefully that using cigarettes harms health, prompting the adoption of myriad anti-smoking policies. The association between smoking and mortality may, however, be driven by unobserved factors, making it difficult to discern the underlying long-term causal relationship. In this study, Friedson, Li, Meckel, Rees, and Sacks explore the effects of cigarette taxes experienced as a teenager, which are arguably exogenous, on adult smoking participation and mortality. A one-dollar increase in teenage cigarette taxes is associated with an 8 percent reduction in adult smoking participation and a 6 percent reduction in mortality. Mortality effects are most pronounced for heart disease and lung cancer.


This paper was distributed as Working Paper 29145, where an updated version may be available.

Ambulance Taxis: The Impact of Regulation and Litigation on Health Care Fraud
Paul J. Eliason, Brigham Young University
Riley J. League, Duke University
Jetson Leder-Luis, Boston University
Ryan C. McDevitt, Duke University
James W. Roberts, Duke University and NBER

Eliason, League, Leder-Luis, McDevitt, and Roberts study the relative effectiveness of administrative regulations, criminal enforcement, and civil lawsuits for combatting health care fraud. Between 2003 and 2017, Medicare spent $7.7 billion on 37.5 million regularly scheduled, non-emergency ambulance rides for patients traveling to and from dialysis facilities, with dozens of lawsuits alleging that Medicare reimbursed rides for patients who did not meet the requirements for receiving one. Using a novel data set and an identification strategy based on the staggered timing of regulations and lawsuits across the United States, Eliason, League, Leder-Luis, McDevitt, and Roberts find that a regulation requiring prior authorization for ambulance reimbursements reduced spending much more than criminal and civil lawsuits did. Despite the sharp drop in both ambulance transports and the companies that provide them following prior authorization, patients’ health outcomes did not change, indicating that most rides were not medically necessary. Their results suggest that administrative actions have a much larger impact than targeted criminal enforcement, providing novel evidence that regulations may be more cost-effective than ex post ligation for preventing health care fraud.


This paper was distributed as Working Paper 29491, where an updated version may be available.

The Effect of Hospital Postpartum Care Regulations on Breastfeeding and Maternal Time Allocation
Emily C. Lawler, University of Georgia
Katherine G. Yewell, University of Louisville

Lawler and Yewell study the effects of state hospital regulations intended to increase breastfeeding by requiring certain standards of care during the immediate postpartum hospital stay. The researchers find that these regulations significantly increased breastfeeding initiation by 3.8 percentage points (5.1 percent) and the probability of breastfeeding at 3 and 6 months postpartum by approximately 7 percent. Lawler and Yewell also provide evidence that these breastfeeding-promoting policies significantly increased maternal time spent on child care, crowding out time spent on formal work. Observed reductions in employment are concentrated among mothers with infants between 0 and 3 months of age.



Jorge Aguero, University of Connecticut
Ege Aksu, City University of New York
Carolina Arteaga, University of Toronto
Alicia Atwood, Vassar College
Reagan Anne Baughman, Univeristy of New Hampshire
Panka Bencsik, University of Chicago
Ashley Bradford, Indiana University
Anne M. Burton, The University of Texas at Dallas
Stephanie G. Coffey, Syracuse University
Maoyong Fan, Ball State University
Sherajum Monira Farin, Georgia State University
Katherine A. Harris-Lagoudakis, Iowa State University
Thomas A. Hegland, Agency for Healthcare Research and Quality
Laura E. Henkhaus, Vanderbilt University
Lauren Hoehn-Velasco, Georgia State University
Kandice A. Kapinos, RAND, UT Southwestern Medical
Jessica H. Kiser, Vanderbilt University
Matthew Knowles, Vanderbilt University
Conor Lennon, University of Louisville
Chuxin Liu, City University of New York
Shaoying Ma, The Ohio State University
Laura Nettuno, Vanderbilt University
Alfredo R. Paloyo, University of Wollongong
Joaquin A. Rubalcaba, University of North Carolina at Chapel Hill
Kaylyn R. Sanbower, Emory University
Keith F. Teltser, Georgia State University
Lucy Xiaolu Wang, University of Massachusetts Amherst
Katherine Wen, Cornell University
Gal Wettstein, Boston College
Lindsey Woodworth, University of South Carolina

More from NBER

In addition to working papers, the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter, the NBER Digest, the Bulletin on Retirement and Disability, and the Bulletin on Health — as well as online conference reports, video lectures, and interviews.



  • Lecture
Alan J. Auerbach, the Robert D. Birch Professor of Economics and Law at the University of California, Berkeley, and...
  • Article
Long-time NBER research associates Joshua Angrist, David Card, and Guido Imbens have been awarded the 2021 Nobel Prize in Economic Sciences in recognition of...
  • Lecture
The credible estimation of causal effects is a central task of applied econometrics. Two tools for this purpose that...