Adolescent depression and adult labor market outcomes

The prevalence of adolescent depression has more than doubled in the past ten years. Youth with depression do worse in the labor market as adults, but it is unclear whether adolescent depression plays a causal role in determining labor market outcomes. This dissertation uses a mediation analysis framework to estimate the direct and indirect effects of adolescent depression on adult earnings and wages. Adolescent depression leads to many adverse outcomes, including poor performance in school and a higher risk of mental health disorders. I treat adult depression and educational attainment as mediators of the effect of adolescent depression on labor market outcomes. First, I conduct a literature review on the relationship between adolescent depression and labor market outcomes. I establish a conceptual framework for thinking about the long-term impacts of adolescent depression and discuss the challenges in identifying causal effects. Next, I produce an in-depth descriptive analysis of how adolescent depression relates to adult earnings and wages. I use mediation analysis and the difference method to estimate the direct and indirect effects of adolescent depression. This approach reconciles many seemingly conflicting results in the previous literature. To address several issues of endogeneity, I then use instrumental variables and a system of equations to identify causal effects of adolescent depression on earnings and wages. I estimate the indirect effects through years of education and adult depression, as well as the leftover ‘direct’ effect. I find that adolescent depression lowers educational attainment and increases the likelihood of adult depression, leading to lower average earnings of about 5% and lower average wages of about 3.7%. These results are robust to identification strategy and alternative measurements of several variables. In contrast, using instruments for identification drops the direct effect of adolescent depression closer to zero, suggesting that findings of a direct pathway are driven by omitted variables bias. The findings of this dissertation imply that there are large economic benefits to better preventing and treating adolescent depression. The long-term labor market consequences of adolescent depression can also be avoided by addressing the effects of adolescent depression on mediating outcomes like educational attainment and adult depression