By Minji Bang
The rise of the ‘gig economy’ means that understanding patterns of self-employment is more important than ever for designing tax benefits and subsidies that affect business activities. In fact, self-employment represents as many as 1 in 5 jobs.
We find that gender differences in self-employment patterns are mostly driven by the differences across marital status. Married women are more likely to be self-employed than single women. On average, self-employed married women work fewer hours than men and single women, regardless of employment type, and married women who are employer-employed. These differences carry over to earnings.
In this blog, we document the key gender differences in self-employment,1 including hours worked and hourly earnings conditional on marital status using the National Longitudinal Survey of Youth 1979 Sample (NLSY79). This data follows 12,686 individuals who were aged 14 to 21 in 1979 from 1979 to 2014.
Figure 1 shows that older people are more likely to be self-employed than younger people. Also, married people are more likely to be self-employed than single people regardless of gender and age. Finally, women are less likely to be self-employed than men at all ages.
Gender
Marital Status
Source: National Longitudinal Survey of Youth 1979 Sample
Digging a little deeper, Figure 1 also shows that married men have different self-employment patterns than married women. In fact, married men have the highest probability of being self-employed. A similar portion of young married and young single men are self-employed. However, middle-aged married men are more likely to be self-employed than middle-aged single men. In contrast, at almost any age, married women are more likely to be self-employed than single women.
Figure 2 shows that gender also matters for hours worked. On average, men at every age work more hours than women in both types of employment. However, both men and women spend less hours at work when they are self-employed. Notice that for both men and women, self-employed individuals have more volatile working hours over the life cycle compared to employer-employed individuals.
Gender
Marital Status
Source: National Longitudinal Survey of Youth 1979 Sample
Married men work about 14 percent2 more hours in self-employment jobs than single men while married women work 11 percent less hours than single women in self-employment jobs. This difference may reflect that married women often perform as a primary caregiver at home. In fact, self-employed married women work even fewer hours (22 percent less than self-employed single women) when they have children under the age of five.3 Those who are single and are self-employed exhibit no gender difference in working hours. In contrast, for singles who are employer-employed there is a persistent working hour difference with men working 11 percent more than women at all ages.
It is also important to consider hourly earnings (Figure 3). On average, men have higher hourly earnings regardless of employment type. However, the gender earnings gap is more than twice as large for those who are self-employed than for those who are employer-employed.
Gender:
Marital Status:
Source: National Longitudinal Survey of Youth 1979 Sample
Age and marital status also have an impact on hourly earnings. For singles who are self-employed, young men out earn young women by about 40 percent with the advantage fading to less than 20 percent for older self-employed singles. In contrast, for married individuals who are self-employed, there is a persistent and substantial gender earnings gap (50 percent or more) at almost all ages. The gender earnings gap is smaller and present at all ages for those who are employer-employed (12 percent single, 30 percent married).
-
We define a self-employed individual as a person who is self-reported as being self-employed without having any employer-employment jobs within a year. It is important to note we exclude individuals with government jobs from our analysis. ↩
-
Differences are computed based on weighted average within each group where the weight is the number of observations. ↩
-
In the future, we will explore how self-employed people allocate their time in housework in more detail. ↩
-
Hourly earnings are normalized using Average Wage Index with base year = 1994. ↩