Abstract
This special issue of the Journal of Human Resources showcases an emerging literature on the economics of child mental health. Until recently, the severe stigma associated with mental health issues meant that few people felt comfortable talking about them or realized how very prevalent they are (Bharadwaj, Pai, and Suziedelyte 2017). Arguably, this changed with the COVID-19 pandemic, which precipitated widespread discussions of a child and youth mental health crisis (United States Public Health Service, Office of the Surgeon General 2021).
One reason to focus on children’s mental health is that it so often predicts worse educational outcomes. For example, Currie and Stabile (2006, 2009) find, using U.S. and Canadian data, that ADHD and other common conditions, such as mood disorders, predict a higher probability of delinquency, grade repetition, and special education, as well as reduced test scores. Using sibling fixed effects models, Fletcher (2010) finds that a one standard deviation increase in teen depressive symptoms is associated with a 25–30 percent increase in the incidence of dropping out of high school. Rossin-Slater et al. (2020) show that school shootings cause subsequent increases in depression treatment among children in affected schools, while Cabral et al. (2020) show that school shootings worsen schooling attainment, decreasing the probability that children graduate from high school, attend college, or finish college.
Several studies also show a direct relationship between childhood mental health and adult labor market outcomes. Currie et al. (2010) use registry data from the Canadian province of Manitoba and find that children with ADHD and conduct disorders are much more likely (30–100 percent depending on the child’s age) to be on welfare after age 18. These effects are much larger than those found for common physical health conditions, such as asthma and major injuries. Fletcher (2014) reports negative effects of childhood ADHD on adult labor market outcomes. Smith and Smith (2010) use retrospective survey data from the U.S. Panel Study of Income Dynamics and find that in addition to lower educational attainment, adults who suffered from mental health problems before the age of 16 have family incomes 20 percent lower than their own siblings, with a lifetime difference of $300,000. Some of this negative impact on household income may come from a lower probability of being married; another important mechanism is a reduction in weeks worked.
Similarly, Goodman, Joyce, and Smith (2011) use data from the National Child Development Study, which followed 17,634 children born in Great Britain during a single week in March 1958, and find that children with psychological problems in childhood had 28 percent lower family incomes by age 50. As in the United States, reductions in labor supply and in marriage are important mechanisms. Biasi, Dahl, and Moser (2021) use Danish registry data to document that people with specific diagnoses of depression, bipolar disorder, or schizophrenia have earnings 31, 36, and 73 percent less than their siblings. Much of the earnings penalty comes from being less likely to have any earnings at all. People with these diagnoses are also more likely to receive government disability payments. One takeaway from these studies is that mental health problems are generally more predictive of lower educational and labor market outcomes than poor physical health in childhood.
These results linking labor market performance and mental health should not come as a surprise. Currie and Madrian (1999) show that the importance of mental health as a determinant of labor supply was already well known in the 1980s and 1990s, which makes it even more puzzling that there has been so little general acknowledgment of this fact among either researchers or policymakers. Mental health problems are a leading cause of working days lost, as well as of nonparticipation in the labor market.1 Poor mental health also affects productivity on the job—many U.S. workers indicate that they have a significant number of bad mental health days.2 An astonishing 40 percent consistently report that they had some bad mental health days in the past month, while more than 10 percent report that they had ten or more such days.
Until recently, few economists explicitly studied childhood mental health. However, concepts that are closely related to mental health have been included under the broader umbrella category of “noncognitive skills.” Heckman and Rubenstein (2001) emphasized that people with similar levels of cognitive skill (as measured by standardized test scores) can have vastly different outcomes and point out that other types of skills must be important. This argument was further developed by Heckman, Stixrud, and Urzua (2006), who showed that measures of noncognitive skills predict education, earnings, and risky behaviors. A large literature about the importance of noncognitive skills has since followed. As a practical matter, many different measures have been included under this rubric, from Big Five personality traits (extroversion, agreeableness, openness, conscientiousness, and neuroticism; see Borghans et al. 2008), to “grit,” to social skills and self-control (Zhou 2016).
Some research on noncognitive skills focuses on constructs that are much more closely linked to mental health, such as “internalizing” behaviors (such as those linked to anxiety and depression) and “externalizing” behaviors (linked to ADHD and conduct disorders). Attanasio et al. (2020) compare cohorts of British children born in 1970 and 2000 and find that measures of these skills at age five are important predictors of adult outcomes.
More generally, mental health can be thought of as something that affects noncognitive skills. For example, Goodman et al. (2011) find that childhood mental illness is negatively associated with the Big Five traits of conscientiousness and agreeableness. Moreover, specific mental health conditions are associated with specific deficits in noncognitive skills. For example, Edin et al. (2021) use Swedish registry data to measure changes in the returns to the noncognitive skills of military recruits over time. They find that deficits in social maturity, focus, internal motivation, or stress tolerance are associated with worse outcomes and that the strength of this relationship has increased over time. In turn, people with specific mental illnesses are more likely to have deficits in these areas—children with autism spectrum disorder struggle with social skills, children with ADHD lack focus, those with depression have difficulties with internal motivation, and those with anxiety lack stress tolerance. These observations suggest that the literature on the growing importance of noncognitive skills (see Deming 2017) can be interpreted as further evidence supporting the long-term impacts of childhood mental health conditions.
A second strand of the existing literature looks at the effectiveness of individual medical treatment for child mental health problems. Some of the best available evidence comes from studies of ADHD treatment. For example, Currie, Stabile, and Jones (2014) consider a policy in the Canadian province of Quebec that extended prescription drug coverage to previously uncovered children. The policy change increased the uptake of ADHD medications but did not improve educational outcomes. Dalsgaard, Nielsen, and Simonsen (2014); Chorniy and Kitashima (2016); and Chorniy (2018) instrument an individual child’s medication status with area-level measures of provider’s propensities to prescribe. The first two papers find positive effects on health (for example, a reduction in injuries), but the third confirms that there is no effect on educational outcomes.
There is less information about the efficacy of treatment for other disorders. Observational epidemiological studies tend to show that people with mental illnesses have poorer outcomes, even when treated, but it might still be the case that treatment improves outcomes relative to nontreatment. One shortcoming of the medical literature evaluating mental health treatments is that randomized controlled trials typically focus only on short-term medical outcomes. For example, a clinical study of a new antidepressant or of therapy for depression would typically follow individuals for a few months and would focus mainly on whether there was a reduction in depression symptoms as measured using an instrument such as the Hamilton Depression Rating Scale. These studies are largely silent on the impacts of mental health treatment on outcomes such as educational attainment and labor force participation.
Using large-scale registry data, Biasi, Dahl, and Moser (2021) find that the labor market penalty associated with bipolar disorder declined after lithium became available as a treatment for the disease, which they attribute to the beneficial effects of treatment. In a cross-country setting, Ludwig, Marcotte, and Norbert (2009) find that the rise in the use of selective serotonin reuptake inhibitors (SSRIs) during the 1990s was associated with a decline in suicide, especially for youth aged 15–24. Conversely, Busch, Golberstein, and Meara (2014), using data on more than 100,000 adolescents, finds that FDA warnings regarding the safety of adolescent use of antidepressants in 2004 resulted in a reduction in treatment, a reduction in GPA, and an increase in delinquency, tobacco, and illicit drug use. Cuddy and Currie (2020, 2023) use data from a large U.S. private insurer to examine the treatment of adolescent depression and anxiety. They find that much of the treatment offered does not meet professional standards and that children treated with drugs that raise a “red flag” are more likely to suffer from self-harm and have higher total medical costs. Cowan and Hao (2021) report that although the Affordable Care Act increased the use of mental health care among college students from poor backgrounds, there was no corresponding improvement in student educational outcomes, substance use, or self-assessed mental health. In sum, the existing literature evaluating the treatment of child mental health problems comes to mixed conclusions. There is some evidence of positive effects, but one cannot assume that all treatment is beneficial.
The literature to date makes a strong case that child mental health is one of the most important determinants of children’s future economic outcomes, but that treatments currently available are not always effective and are sometimes even harmful if applied improperly. The papers in this volume take the next step by looking at a variety of additional questions raised by the existing literature. These questions can be summarized as follows:
Does the increase in reported child mental health problems represent an increase in the actual incidence of these conditions, or is it driven by increasing recognition of an existing problem?
The existing literature shows that child mental health conditions have future economic consequences. But to what extent do poor economic conditions in childhood cause mental health problems?
A logical follow-up question is whether there are intergenerational correlations in mental health problems and, if so, whether there are interventions that can break these links. A related question is how much of the well-documented intergenerational link in socioeconomic status might be attributable to intergenerational linkages in mental illness?
After the family, the school is the most important place in the lives of most children. This observation leads to questions about what role schools can play in either exacerbating or addressing the consequences of mental illness? Relatedly, many children with mental health issues eventually end up in jail, and it is important to evaluate the mental health services that are provided there.
Finally, are there successful interventions that can be scaled up to address child mental health issues at a population level?
In what follows we offer a brief overview of each contribution and how it addresses one or more of these questions.
The Role of Measurement
In “To What Extent Are Trends in Teen Mental Health Driven by Changes in Reporting? The Example of Suicide-Related Hospital Visits,” Adriana Corredor-Waldron and Janet Currie look at one of the most alarming trends in recent data about child mental health: the rise in reported suicidal behaviors among teens since 2010 (Corredor-Waldron and Currie 2024). They point out that the increases in the reporting of such behaviors could be affected by increases in the extent of screening for mental health problems, more careful recording of diagnoses, and by declining stigma surrounding mental health problems.
Taking New Jersey as a case study and using all hospital visits for children 10–18 years old from 2008–2019, they show that there were two inflection points in suicidal behaviors: an uptick in 2012 followed by a sharp rise after 2016. They show first that both increases are accounted for by increases in diagnoses of suicidal ideation. The rise after 2012 was concentrated among girls and coincided with a U.S. Preventive Services Task Force recommendation to screen adolescent girls for depression. This recommendation was given additional force by the Affordable Care Act’s requirement that insurers pay for U.S. Preventive Services Task Force recommended services. The rise after 2016 coincided with changes in the recommended coding of suicidal ideation as a secondary diagnosis. It was in hospitals’ interests to adopt the new coding standards because they often resulted in higher reimbursements. Rates of self-harming behaviors, attempted suicides, and completed suicides were essentially flat in New Jersey teens over this period.
These results show how difficult it can be to measure trends in mental health and suggest that underlying suicide-related behaviors among children, while alarmingly high, may not have risen as sharply as reported rates suggest. This work complements previous research showing that the diagnosis of child mental illness can be extremely sensitive to the incentives facing providers (Chorniy, Currie, and Sonchak 2018), to information about the diagnostic status of other family members (Persson and Rossin-Slater 2018; Persson, Qiu, and Rossin-Slater 2021), to race (Coker et al 2016), or to insurance status (Cowan and Hao 2021).
Economic Factors and Child Mental Health
Two papers explore the relationship between economic factors and children’s mental health. Previous work has explored the relationship between economic hardship at the cohort level and future child well-being. For example, Adhvaryu, Fenske, and Nyshadham (2019) find that increases in cocoa prices in cocoa-producing regions of Ghana during a person’s early childhood reduce the probability of severe adult mental distress. But there has been little work investigating this issue at the individual level.
In “The Effect of Household Earnings on Child School Mental Health Designations: Evidence from Administrative Data,” Lauren Jones, Mark Stabile, Kourtney Koebel, and Jill Furzer use administrative data from the Canadian province of British Columbia from 2000 to 2017 (Jones et al. 2024). They were able to link parental tax records with public school records for children born between 1992 and 2000 and for the tax years 2000–2015. They focus on families that suffered income losses, especially in the Great Recession of 2008–2009, instrumenting household income losses with census-tract-level measures. The main outcome is whether the child is flagged as having a special need due to a mental health problem in the school records.
Using propensity score matching methods, they find that relative to children whose families did not experience earnings losses, children who experienced such losses had a 20 percent higher rate of acquiring a new mental health designation. This effect is very persistent and grows over time, especially if children experienced the family earnings loss when they were aged 10 or younger.
In “Child Mental Health, Family Circumstance, and Long-Term Success: The Effect of Household Income,” Randall Akee, William Copeland, and Emilia Simeonova look at the opposite case: What happens to child mental health when the family gains a new source of income? (Akee, Copeland, and Simeonova 2024). The authors use data from a long-running natural experiment in which American Indian households received income transfers from a casino’s revenues while other non-American Indian households living in the same area did not. The children were initially aged 9, 11, or 13 when the payments started, so they were exposed to transfers over differing windows during their childhoods. The payments were quite large relative to family income (bigger, for example, than the Earned Income Tax Credit).
Using a difference-in-differences analysis, they find that “treated” children had fewer depression and anxiety symptoms at age 30. Adults with more years of treatment exposure also had greater economic well-being when measured around age 30. Untreated children showed strong persistence in measures of mental health from adolescence through age 30, while in treated children persistence was greatly attenuated. The estimated effects are slightly stronger for treated children who experienced fewer mental health symptoms before the transfers began at ages 16 and 21, but by age 30, all affected children showed reduced symptoms of anxiety and depression and improved economic outcomes.
Together these papers suggest that income matters for child mental health—losses can cause or exacerbate problems, while income gains reduce them. Moreover, both effects are persistent. Hence, these two papers suggest that mental health may be one of the mechanisms underlying the intergenerational transmission of economic status. Poverty causes mental health problems in children, which in turn lowers their incomes as adults, so that their own children are also more likely to have mental health problems due to poverty. The next two papers in the special issue explicitly tackle these intergenerational linkages.
Intergenerational Effects of Mental Health Problems
In “(Breaking) Intergenerational Transmission of Mental Health,” Aline Bütikofer, Rita Ginja, Krzysztof Karbownik, and Fanny Landaud use rich administrative Norwegian data that allow them to track the incidence of mental health diagnoses in parents and children, as well as in both blood relatives and other relations (Bütikofer et al. 2024). In their main analysis, the children were 13–18 years old at the time of first diagnosis, while the mothers and fathers were 25–30 at the time when they were first diagnosed, but the results are similar when they use other age ranges. The paper is one of the first to document intergenerational linkages in mental health (but see also Shields 2013).
They find strong evidence of intergenerational correlations, especially in low-income families, and offer a nuanced discussion of what this might mean. For example, while it is true that child mental health is highly correlated with parent’s mental health, parental mental health problems also predict other sorts of health problems in children, such as injuries. Moreover, children’s mental health problems are also correlated with those of family members who are not blood relatives. These findings suggest that the results cannot simply reflect genetic factors, but must also reflect social factors, such as knowledge of mental health conditions, attitudes about mental illness and treatments, differences in provider’s screening propensities, or those that work through income effects as discussed above.
Bütikofer et al. (2024) also evaluate a Norwegian policy that started in 2007, which was aimed at helping young children of parents with mental health problems or substance abuse. The program mainly involved ensuring that these children received priority for a wide range of existing social services, such as home visiting and parent training programs. They find that the initiative reduced the size of the correlation in mental health conditions across generations by approximately 40 percent, but this effect was concentrated among families in which at least one parent was college educated. This paper suggests that it is possible to at least reduce the intergenerational transmission of poor mental health conditions by targeting the children of parents with mental health conditions, though further research is needed in terms of the mechanisms.
In “Domestic Violence Reports and the Mental Health and Well-Being of Victims and Their Children,” Manudeep Bhuller, Gordon Dahl, Katrine Løken, and Magne Mogstad focus on social responses to domestic violence as one very specific mechanism that could cause an intergenerational correlation in mental health issues (Bhuller et al. 2024). They also use rich Norwegian administrative data that allow them to link offenders to victims and their children over time. They set up a difference-in-differences framework in which women whose cases will be reported to the police in the future serve as controls for those whose cases are reported to the police in the current period.
They find that calling the police causes large changes in the home environment, including a jump in partner dissolutions and declines in household income. A domestic violence report increases mental health visits by 35 percent for women and by 19 percent for their children in the year of the event. This effect decays over time for the women but remains large for the children. A complementary regression discontinuity analysis shows that the children also experience declines in test scores and are less likely to complete the first year of high school. The children are also more likely to receive child protective services and to commit a crime.
These results indicate that the experience of domestic violence is associated with a higher incidence of mental health problems in both mothers and children. However, since domestic violence tends to be persistent, it is likely that in many cases children suffered from ongoing violence in their households, which eventually culminated in a call to the police. It is possible that this index incident was particularly heinous and that the subsequent negative effects mainly reflect the severity of that incident. But it is also possible that the police intervention and subsequent social service response and/or the family breakup was traumatizing. For example, Doyle (2007, 2008) shows that for the marginal child, removal to foster care has a significant negative effect on future outcomes. Given the prevalence of domestic violence, it is important to understand these mechanisms better so social service agencies can be sure that they are helping and not harming children.
The Role of Schools and Jails
In “Societal Disruptions and Childhood ADHD Diagnosis during the COVID-19 Pandemic,” Seth Freedman, Kelli Marquardt, Dario Salcedo, Kosali Simon, and Coady Wing highlight the central role that schools play in children’s mental health by suggesting diagnosis and treatment (Freedman et al. 2024). Using both nationwide data on private health insurance claims and comprehensive electronic medical records from Indiana, they focus on new diagnoses of ADHD among children 5–12 years old and ask how the rate changed during the pandemic.
They find that the pandemic initially reduced new ADHD diagnoses by 8.6 percent among boys and 11.0 percent among girls nationwide through February 2021. They further show that the extent of the decline in new diagnoses was related to the extent of school closures in fall 2020 as measured using cell phone location data from SafeGraph. These declines in diagnosis occurred despite a rapid recovery in the number of doctor visits after an initial fall in visits at the start of the pandemic. They find differing effects on Black, white, and Hispanic children, with white and Hispanic children seeing the greatest declines in the number of new diagnoses.
The authors discuss whether these changes are likely to have increased or decreased welfare. They show that the new diagnosis rate for boys is closer to the “true” ADHD incidence rate estimated by Marquardt (2022), suggesting that on the margin, the reduction in diagnoses may have resulted in fewer type 1 errors. It is more likely that a reduction in diagnosis for girls is harmful, given that girls are thought to be underdiagnosed to begin with. Turning to the results by race and ethnicity, it may be the case that the reduction in diagnosis is also welfare-improving for white children, given that white children are much more likely than Black children to be diagnosed with ADHD. Hence, the results leave open the question of whether these changes in diagnoses were positive or negative, and for whom.
The paper “In-Person Schooling and Youth Suicide: Evidence from School Calendars and Pandemic School Closures,” by Benjamin Hansen, Joseph Sabia, and Jessamyn Schaller, points to a darker role for schools (Hansen, Sabia, and Schaller 2024). While school plays a central role in most children’s lives, there are some children who have a miserable time there. The authors point out that youth suicides are consistently seasonal, with a sharp fall between May and June during summer vacations and a spike in October. Suicides fall in December during the holidays and rise in January, February, and March. This seasonal pattern is not seen in young adults but only in school-age children. Bacher-Hicks et al. (2022) show that Google Trends searches on “bullying” show a similar pattern. Although it is difficult to make a direct connection between the data on bullying and the data on suicides, it seems reasonable to suppose that the two are related. Children who report being bullied are twice as likely to report suicidal ideation or to have suicide attempts (Tang et al. 2020) Also, younger children are bullied more and generally show larger seasonal swings in suicide rates.
As further evidence of a connection, the authors show that there was a sharp decline in searches for “bullying” or “cyber-bullying” after the pandemic shut down many schools in March 2020, especially among children 12–15, which coincided with a 25 percent decline in youth suicides. Using data on foot traffic to the schools from SafeGraph, the authors further show that the pandemic-induced decline in suicides was smaller when children were spending more time in school. Finally, leveraging county-level variation in the timing of school reopenings, they find that returning to in-person instruction increased youth suicides by 12–18 percent from the March 2020 low.
If being in school is causing some children to consider, attempt, or complete suicide, then one policy response in addition to anti-bullying programs might be to improve mental health services in schools. In “Effects of School-Based Mental Health Services on Youth Outcomes,” Ezra Golberstein, Irina Zainullina, Aaron Sojourner, and Mark Sanders examine the impacts of physically locating mental health services in schools (Golberstein et al. 2024). They analyze 19 years of administrative data from Hennepin county (Minneapolis), Minnesota and combine it with both student surveys and information about changes in policy. The data cover about 255 schools and 120,000 students each year.
Using the staggered rollout of in-school services across the county to identify the effects, they find that locating licensed clinical mental health workers in schools increases the use of mental health services, improves disciplinary outcomes, and reduces the incidence of self-reported suicide attempts. There is weaker evidence that school based mental health services reduced suspensions and juvenile justice involvement. There were no significant effects on average attendance, standardized test scores, or self-reported substance use, however. Hence, this innovation improved mental health outcomes, and there is little evidence of any crowd-out of instructional resources when more resources were devoted to mental health.
Unfortunately, jails and prisons are other institutional settings that are often called on to provide youth mental health services. Inmates have much higher rates of mental health issues than the general population, leaving prisons to function as mental hospitals of last resort. In “Adverse Impacts of Mental Health Needs Assessment on Jail Outcomes: Evidence from Transition Age Youth and Adults,” Scott Cunningham, Jonathan Seward, Karen Clay, and Vivian Vigliotti evaluate one attempt to improve services by establishing special mental health courts that can refer juveniles to services (Cunningham et al. 2024). The setting is a large urban jail with about 2,150 inmate bookings per day. Youth 17 and older make up about one-third of the bookings. Inmates charged with a misdemeanor receive a mental health assessment from a randomly assigned clinician within 36 hours of booking. Those found to have moderate or serious mental health issues are diverted to a separate mental health docket.
Using a “judges” design that exploits random assignment to clinicians with different scoring propensities, the authors find that the marginal person who scored higher on the assessment has a significantly longer length of stay in jail (even though those with misdemeanors ought to be released on bond) and are significantly more likely to attempt suicide. It is possible that youth diverted to the mental health docket find themselves with an indefinite wait for treatment, which, instead of helping them, contributes to a deterioration in their mental condition. Ironically, the results suggest that it might be better to reduce the number of people referred to the mental health docket until services can be improved.
Interventions to Improve Child Mental Health in Developing Countries
Up to this point, the papers in the volume have focused on child mental health in rich countries, including the United States, Canada, and Norway. While there are important differences in the way that mental health services are delivered in these countries, these differences may pale into insignificance compared to the gulf between the mental health services available in most rich countries and those available in many poorer countries.
In “Improving Mental Health of Adolescent Girls in Low- and Middle-Income Countries: Causal Evidence from Life Skills Programming,” Manisha Shah, Sarah Baird, Jennifer Seager, Benjamin Avuwadah, Joan Hamory, Shwetlena Sabarwal, and Amita Vyas report on a series of ambitious programs aimed at improving a variety of noncognitive skills (Shah et al. 2024). They note that it is not uncommon to provide female adolescents in developing countries with life skills training, safe spaces, and education on navigating relationships with boyfriends and parents and that many of these programs have been evaluated using randomized controlled trials. Most previous evaluations of these types of interventions have focused on outcomes such as increases in educational attainment and delays in age at first birth. The question the authors address is whether these interventions also improve mental health. The paper focuses on 6,000 girls aged 10–19 in Bangladesh, Ethiopia, Jordan, and Tanzania. The authors note that girls have higher rates of depression than boys, and they hypothesize that this might be due in part to restrictive gender norms that emerge in adolescence.
The interventions evaluated are all manualized and include: Growth Mindset, Act With Her, Empowerment and Livelihood for Adolescents (ELA), and Goal Setting. Although some of these programs have been shown to have positive results on some outcomes, such as reducing teen childbearing, the results suggest that more research into how they affect mental health is necessary. For example, in Tanzania, the goal setting intervention alone increased depression rates, possibly because setting goals without the agency to achieve them can be deeply frustrating. The goal setting intervention in conjunction with ELA was successful in reducing depression. In Ethiopia, the Act With Her intervention was successful in improving outcomes when implemented alone but appeared to reduce well-being when coupled with transfers to the girls. This is a puzzling result because in other contexts, transfers alone have been shown to improve the mental health of adolescent girls (Baird, de Hoop, and Özler 2013). The authors hypothesize that giving transfers only to female participants may have engendered hostile responses from male peers. This result raises the interesting question of whether it is possible to improve the situation of women without also educating men.
In “Trajectories of Early Childhood Skill Development and Maternal Mental Health,” Dilek Sevim, Victoria Baranov, Sonia Bhalotra, Joanna Maselko, and Pietro Biroli build on an experimental intervention to treat maternal depression in rural Pakistan (Sevim et al. 2024). The intervention treated perinatally depressed women and was delivered by peer volunteers through home visits and group sessions. The intervention had previously been shown to be effective in reducing maternal depression. This study focuses on the effects on the children at 6, 12, and 36 months of age.
They show that, in the untreated control group of depressed women, mental health is strongly related to child outcomes. However, they do not find that the intervention improved child outcomes in the treatment group on average. They report a larger improvement in children’s skills when moving from severe to moderate depression than they do when moving from moderate to mild depression. The treatment leads to improvements in skills, particularly for children whose mothers did not recover from depression. Overall, the intervention brings outcomes for treated mothers closer to outcomes for mothers who were not depressed at baseline.
Conclusion
This volume provides a sampling of a rapidly growing literature about the economics of child mental health. The questions addressed in these studies touch on issues central to economics, including the way that incentives facing agents can distort measurement, the relationship between income and health, the key factors underlying well-known intergenerational correlations in economic status, the role of schooling in the development of human capital, and the evaluation of programs aimed at enhancing human capital development. Given our increasing knowledge about the prevalence and profound impact of childhood mental health problems, and the growing importance of noncognitive attributes including mental health, it is likely to remain a fertile field of study for many years to come.
Acknowledgments
Sponsored by the NOMIS Foundation and the Center for Health and Wellbeing at Princeton University.
Footnotes
↵1. For example, the U.K.’s Health and Safety Executive reports that in 2017–2018, 57 percent of worker days lost in the United Kingdom were due to stress, depression, and anxiety, compared to 25 percent lost because of musculoskeletal issues, such as back pain (Health and Safety Executive, U.K. Department for Work and Pensions 2020).
↵2. For example, since 2002, the U.S. General Social Survey has asked employed individuals about the number of days in the past month that their mental health was “not good.” The specific question is: “Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?”
- Received July 2023.
- Accepted July 2023.
This open access article is distributed under the terms of the CC-BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) and is freely available online at: https://jhr.uwpress.org.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.