Relationship between subjective well-being and depressive disorders: Novel findings of cohort variations and demographic heterogeneities

This paper uses a large-scale nationally representative dataset, the Chinese General Social Survey, to examine the relationship between subjective well-being and depressive disorders. Statistical results indicate that higher levels of subjective well-being help decrease perceived depression. Robustness checks are carried out using different types of explanatory and dependent variables, various regression models, penalized machine learning methods, instrumental variable approaches, and placebo tests, all of which lend further credence to the above findings. Based on it, heterogeneities in the relationship between subjective well-being and self-rated mental disorders are explored. In respect of variations in age cohorts, it is found that the absolute values of happiness's estimated coefficients are smaller in the 20-30 and 30-40 age groups, while that in the 40-50 age group increase substantially. In older cohorts, the estimates remain at higher levels while fluctuating to some degree. Furthermore, the significantly negative interaction between happiness and age proves that age amplifies subjective well-being's effect on perceived depressive disorders. With age increasing, the impact of happiness on reducing perceived depression tends to be stronger. Therefore, for older people, subjective well-being plays a more important role in suppressing self-rated depression. Heterogeneities of the relationship between happiness and perceived depressive disorders in subgroups with different demographic characteristics are also investigated. It is found that the negative correlation between subjective well-being and self-rated depression is stronger among those with higher educational levels, living in urban areas, being members of the Communist Party of China, having pensions, and owning more housing assets. However, gender, ethnic identity, religious belief, and marital status exert no significant moderating effects.