Mental health disparities by socioeconomic status over the life cycle

Mental health problems are widespread among the general population. At any point in time approximately 15-20% of the working age population is affected by a common mental health problem, such as anxiety or depression, and one in two people experiences mental ill-health during life (OECD, 2014). Poor mental health leads to huge direct and indirect economic costs both for individuals and at the macro level through its negative effects on well-being, productivity, wages and employment. The OECD (2014) estimates that, at the country level, the annual costs of mental ill-health may exceed 4% of GDP and finds that in all member countries mental health care should improve.

The need to alleviate the (economic) burden of mental ill-health requires a proper understanding of how mental health problems are distributed across the population. To this end, different fields of study have indicated the importance of relations between mental health and variables of socioeconomic status (SES), such as wealth, income, education, occupation and labor market status. However, at least two major shortcomings in the existing literature impede a more thorough understanding of mental health disparities by socioeconomic status. First, empirical studies have paid little attention to the dynamic nature of mental health developments in relation to variables of SES. Hence, there exists hardly any empirical work on how life cycle developments of mental health may vary among individuals of different SES, and vice versa. Second, there exists no comprehensive framework to explain how differences in SES can influence decisions of economic agents that affect mental health, thereby leading to disparities in mental health outcomes. This paper develops new insights in the study of mental health by addressing both of these issues.

The first part of the paper presents stylized facts on the empirical life cycle trajectories of mental health using longitudinal data from the Netherlands. Here the well-established MHI-5 score is used as a measure of mental health. On average mental health scores slightly improve over working life and mental health scores are higher for those with higher income. The data do not indicate significant differences in mental health developments across work types (e.g. no significant differences in mental health developments between manual and non-manual jobs exist). A persistent U-shaped development exists for mental health scores over 'working life' for individuals who are out of the labor market, regardless of income level. This U-shape contrasts sharply with the relatively stable mental health scores of individuals with jobs.

The second part aims to incorporate the empirical findings in a comprehensive theoretical framework. We present a conceptualization of mental health as a form of capital to develop a theory of mental health based on the Grossman (1972) framework. Here, 'mental health behavior' consists of an individual’s decisions to (i) invest in mental health and (ii) accept job-related psychosocial stress, which yields a wage premium but negatively affects mental health. This framework is deployed to reflect on the empirical findings and raises new questions such as whether job-related psychosocial stress is compensated differently across skill levels.