A fresh look at self-employment, stress and health: accounting for self-selection, time and gender

Purpose: Past research on self-employment and health yielded conflicting findings. Integrating predictions from the Stressor-Strain Outcome model, research on challenge stressors and allostatic load, we predict that physical and mental health are affected by self-employment in distinct ways which play out over different time horizons. We also test whether the health impacts of self-employment are due to enhanced stress (work-related strain) and differ for man and women. Design/methodology/approach: We apply non-parametric propensity score matching in combination with a difference-in-difference approach and longitudinal cohort data to examine self-selection and the causal relationship between self-employment and health. We focus on those that transit into self-employment from paid employment (opportunity self-employment) and analyze strain and health over four years relative to individuals in paid employment. Findings: Those with poorer mental health are more likely to self-select into self-employment. After entering self-employment, individuals experience a short-term uplift in mental health due to lower work-related strain, especially for self-employed men. In the longer-term (four years) the mental health of the self-employed drops back to pre-self-employment levels. We find no effect of self-employment on physical health. Practical implications: Our research helps to understand the nonpecuniary benefits of self-employment and suggests that we should not advocate self-employment as a “healthy” career. Originality/value: This article advances research on self-employment and health. Grounded in stress theories it offers new insights relating to self-selection, the temporality of effects, the mediating role of work-related strain, and gender that collectively help to explain why past research yielded conflicting findings.