Previous research on educational mismatch concentrates on estimating its labor market consequences but with a focus on wage and salary workers. This paper examines the far less studied influence of mismatch on the self-employed. Using a sample of workers in science and engineering fields, results show larger earnings penalties for mismatch among the self-employed but no diminution in job satisfaction. Moreover, the reasons for mismatch among the self-employed differ dramatically by gender.
The public and private costs of education are huge, causing economists, policy makers, and the public to be concerned about whether or not workers utilize the skills acquired during education in the labor market. Responding to this concern, researchers examine the causes and effects of mismatches in the skills required for the job and the skills acquired during education. While mismatch can be in the type of skills or simply the quantity (over- or under-education), the research finds that mismatch generates lower earnings, lower job satisfaction, and higher turnover, ceteris paribus. These findings appear robust to differences in country, time period, or whether the data analysis is cross-sectional or panel in nature.
Thus, while the effects of mismatch are fairly well established, the research so far has focused only on wage and salary workers, meaning that there has been no research to date on the relationship between mismatch and self-employment. Since self-employment is often seen as a driver of economic growth and particularly in employment growth (see for example, Birch, 1987 and Neumark et al., 2008, for the US; Burges, 1991, for Australia; Audretsch & Fritsch, 2003, for Germany although Haltiwanger, Jarmin, & Miranda, 2010, offers a contrasting view), the study of how mismatch interacts with self-employment enriches our understanding of both educational mismatch and this critical area of policy interest.
Research on educational mismatch has focused on the causes of mismatch and the consequences of mismatch on labor market outcomes for wage and salary workers. This paper is the first to explicitly consider differences in educational mismatch and whether workers are in the wage and salary sector or are self-employed, focusing on differences between men and women in those sectors. Controlling for a list of demographic characteristics we find that the self-employed are more likely to report being mismatched, particularly if the self-employed worker is a woman. Furthermore, mismatch among the self-employed is associated with larger earnings decreases for the moderately mismatched (compared to the decrease for moderately mismatched wage and salary workers) and smaller relative earnings discounts for the severely mismatched (again compared to the discount for severely mismatched wage and salary workers). However, these declines in earnings are not associated with even further decreases in subjective well-being as measured by job satisfaction. The negative correlation between job satisfaction and severe mismatch for the self-employed is much smaller than for wage and salary workers who are severely mismatched.
This paper, however, is just a first step in this research. An important next step would be to examine these interactions in a panel context where educational mismatch might be an indicator of changing sectors. While the self-employed are shown to have greater rates of educational mismatch, it is unclear whether this was a state that occurred when the worker became self-employed or whether it occurred before or after joining this sector. Panel data would help to identify some of these patterns as workers change from one sector to another. Furthermore, it would be interesting to see, as in previous literature, whether mismatch itself generates some incentive to change jobs or sectors. Finally, data on positive job attributes of self-employment, such as hours flexibility, would be interesting to examine to see if these explain some of the mismatch-job satisfaction differences between male and female self-employed workers.
