Many people engage in self-employment, yet there exists a dearth of data on these arrangements. This paper addresses this gap by creating a novel dataset of self-employment roles to examine heterogeneity in self-employment arrangements. The approach uses non-public 2016 Health and Retirement Study (HRS) data on employer names and locations and narrative descriptions of industry and occupation for older workers reporting self-employment to classify self-employment of older adults in the United States into entrepreneurial roles (own/run; manage; independent). Using this classification system along with the breadth of information collected in the HRS, this work finds substantial differences in demographic characteristics, work characteristics, income, and benefits, as well as substantial variation in quality of life and retirement expectations. The paper also links the classification to administrative records on self-employment and wage employment to identify discrepancies by role across data sources.
The paper found that:
- Of the self-employed in the HRS, business management and ownership are associated with being male and having higher education; greater labor income and wealth; and expectations of working longer.
- Independent roles are associated with being more likely to identify as retired, while business management roles are associated with better mental health.
- Self-employment in general is associated with being less likely to say they would like to leave work but stay employed for money or health insurance and with having a more favorable view of work, though disparities by self-employment exist across role as well.
- Further work linking to administrative records suggests substantial discrepancies between survey responses and administrative records.
The policy implications of the findings are:
- Meaningful differences in the nature of work and characteristics of the self-employed exist.
- One-size-fits-all policies will not address the needs of this diverse group.
- Combining survey and administrative records is critical for identifying the prevalence and characteristics of heterogeneity among the self-employed and formulating appropriate policy.