Female self-employment in the United States: an update to 2012

After seeing considerable increases in the 1970s and the 1980s, the share of women in self-employment and the female self-employment rate leveled off in the 1990s and remained relatively unchanged through 2012. While this recent 20-year trend may not be remarkable, the characteristics of female business owners have changed considerably over time. The majority of empirical research on self-employed workers either analyzes self-employed men and women together, as a similar “type” of worker, or simply omits women from the study of self-employment.¹ By contrast, the present article uses the Annual Social and Economic Supplement of the Current Population Survey (CPS) to examine changes and trends in the demographic and socioeconomic characteristics of self-employed women from 1993 to 2012. It finds that these female workers continue to be a unique labor market group deserving separate treatment in analysis.

New trends in female self-employment suggest a positive shift in opportunities for women, especially for those who differ from the “typical” self-employed women of the past. For instance, the percentage of female minorities in self-employment doubled from 1993 to 2012, and more divorced women and women without young children have become self-employed. Earnings trends have also been favorable. Although female business owners still have lower mean earnings compared with other worker groups, between 1993 and 2012 the gender gap in hourly earnings among the self-employed shrank by nearly 20 percentage points for full-time workers and by 17 percentage points for part-time workers.

In addition, several characteristics distinguish self-employed women from their male self-employed counterparts as well as from their female wage and salary² counterparts. For example, relative to other worker groups, the female self-employed have higher mean levels of education but are in poorer health and more likely to be disabled. Their work schedules are highly variable and, in 2012, 38 percent of them worked part time for noneconomic reasons, compared with 17 percent of self-employed men and a quarter of wage and salary women.

The section below discusses the data and the sample used in the analysis. The following sections examine the trends in female self-employment over the 1993–2012 period and the characteristics that make self-employed women similar or dissimilar to other worker groups. Finally, a brief section on the top occupations and industries of self-employed women is offered and conclusions are drawn.

Data

The analysis uses 1993–2012 data from the Annual Social and Economic Supplement of the CPS, a monthly survey of about 60,000 households. This annual supplement collects data on demographics, work experience, income, and benefits, including government assistance and health insurance coverage. In 1994, a CPS redesign changed the measurement of employment characteristics such as class of worker, part-time versus full-time work, and industry and occupation distributions.³ Thus, this analysis is limited to data collected since the survey redesign in order to make the best use of comparable data over time.

The study sample is comprised of civilian nonfarm workers age 16 or older who were employed in the job they held the longest during the year. Workers are defined as employed if they worked 1 or more weeks during the reference year, a work period that may include paid time off. Workers are defined as self-employed if they reported being self-employed in the job they held the longest during the reference year. This definition does not include workers reporting self-employment income from a second job but includes workers earning wage and salary income from a second job. Self-employed workers report themselves as unincorporated or incorporated. The sample excludes unpaid workers in nonagricultural family businesses.

Trends in female self-employment

Female representation in self-employment and the self-employment rate. In 2012, about one in three self-employed workers was female. After a consistent rising trend in female self-employment from the 1970s to the early 1990s,⁴ the female-to-male ratio in self-employment remained relatively flat, at approximately 35 percent. (See table 1.) In 2012, 1 in 14 employed women was self-employed, down from about 1 in 12 women in 1993.

View Chart Data

Nonagricultural self-employment rates, by gender, annual averages, 1993–2012 (percent)

Year	Women's self-employment rate	Men's self-employment rate
1993	8.1	12.7
1994	7.9	12.3
1995	7.4	12.1
1996	7.7	12.5
1997	7.3	12.1
1998	7.0	11.6
1999	7.1	11.6
2000	7.0	11.2
2001	6.8	11.0
2002	7.3	12.2
2003	7.4	12.2
2004	7.6	12.7
2005	7.4	12.6
2006	7.5	12.7
2007	7.2	12.5
2008	7.3	12.2
2009	7.2	12.0
2010	7.1	11.7
2011	6.9	11.4
2012	7.1	11.5
Source: Current Population Survey, Annual Social and Economic Supplement.

The self-employment rate, defined as the ratio of nonagricultural self-employed workers to nonagricultural workers of both employment sectors (i.e., self-employment and wage and salary), fell from 10.5 percent in 1993 to 9.4 percent in 2012. (See table 2.) In 2012, the female self-employment rate was 7.1 percent and the male rate was 11.5 percent. Figure 1 depicts the nearly 20-year trend in self-employment rates by gender, with shaded areas representing recessions. While both male and female rates trend slightly downward, female self-employment was less affected by economic downturns than was male self-employment, a differential effect particularly evident in the most recent recession. In the 3 years before the Great Recession of 2007–2009, male self-employment averaged 12.7 percent; between 2006 and 2012, it dropped by 1.2 percentage points. The female self-employment rate fell only 0.4 percentage points in this 6-year period and made up twice as much ground in 2012 relative to the men’s rate.

View Chart Data

Nonagricultural incorporated and nonincorporated self-employment rates, by gender, annual averages, 1993–2012

Year	Women's incorporated self-employment rate	Men's incorporated self-employment rate	Women's unincorporated self-employment rate	Men's unincorporated self-employment rate
1993	2.0	4.2	6.0	8.5
1994	1.9	4.3	5.9	8.1
1995	1.4	4.0	6.0	8.1
1996	1.7	4.1	5.9	8.4
1997	1.6	4.3	5.6	7.8
1998	1.6	4.0	5.4	7.6
1999	1.8	4.2	5.4	7.4
2000	1.6	4.0	5.4	7.2
2001	1.7	4.3	5.0	6.7
2002	1.9	4.7	5.4	7.5
2003	2.0	4.8	5.4	7.4
2004	2.1	4.9	5.5	7.8
2005	2.1	4.9	5.3	7.7
2006	2.1	5.0	5.3	7.7
2007	2.2	5.1	5.1	7.4
2008	2.2	4.9	5.1	7.3
2009	2.2	4.7	5.0	7.3
2010	2.1	4.6	5.0	7.2
2011	1.9	4.6	5.0	6.8
2012	2.1	4.7	5.0	6.8
Source: Current Population Survey, Annual Social and Economic Supplement.

Incorporated vs. unincorporated self-employment. Self-employed workers report themselves as either incorporated or unincorporated. Incorporation, which allows business owners to reduce their personal liability and income taxes, has risen among the self-employed since the 1960s.⁵ However, in the past two decades, the progression toward incorporation was seemingly driven by self-employed men. Between 1993 and 2012, the incorporated self-employment rate rose about half a percentage point for men but remained mostly unchanged for women. (See figure 2.) In the same time, the unincorporated self-employment rate fell 1.7 percentage points for men but only 1.0 percentage point for women. In 2012, 2.1 percent of all female workers were incorporated self-employed, compared with 4.7 percent of all male workers.

Characteristics of self-employed women

Age. About 34 percent of self-employed women in 2012 were age 55 or older, up nearly 13 percentage points since 1993. (See table 3.) A similar trend is found for self-employed men and is likely correlated with the trend of gradual retirement through “bridge jobs” (i.e., jobs that bridge the gap between a career job and full retirement). Self-employment is one type of a bridge job that is common among older individuals. This interpretation is supported by table 4, which shows that self-employment rates increase with age, especially among men.⁶

Race, Hispanic origin, and citizenship. The middle panels of table 3 present data on race and citizenship. Compared with the wage and salary sector, the self-employment sector has a higher percentage of Whites; however, this proportion fell considerably between 1993 and 2012, especially for self-employed women. Notably, the concentration of female minorities in self-employment doubled during this period.

Self-employment rates are above average for Whites and Asians, regardless of gender. (See table 4.) Between 1993 and 2012, self-employment rates increased for all female minority groups; among men, rates increased only for Blacks and those of Hispanic origin. Foreign-born U.S. citizens and noncitizens continue to have above-average self-employment rates, and foreign-born citizens are increasingly entering both the wage and salary and the self-employment sectors, but especially the latter.

Marital status and children. While the share of married female workers in 2012 was about 50 percent in the wage and salary sector, it was more than 66 percent in the self-employment sector. (See table 3.) A similar distribution is found for men. This pattern may be due to the fact that self-employment often gives married people greater work schedule flexibility and the opportunity for joint production, while still allowing them to enjoy the earnings stability and benefits coverage provided by their spouse. However, in recent decades, self-employment has become more common among unmarried women, in particular divorced women, who make up one-fifth of self-employed women and whose self-employment rates are slightly above average.

In 2012, nearly 20 percent of self-employed women under age 55 had young children (6 years and younger) at home, the highest proportion across all worker groups. The share of workers with young children decreased over the study period for all workers, and especially for self-employed women. Table 4 indicates that the self-employment rate for women with young children was above average in 1993 but below average in 2012.

Education. Although mean levels of educational attainment rose in the 1993–2012 period for all worker groups, the increase was greater for women. (See table 5.) At the end of the period, self-employed women had the highest mean level of education—14.3 years, compared with 14.1 years for both self-employed men and wage and salary women. More than half of self-employed women in 2012 had at least an associate’s degree, but less than half of the workers in other groups had at least this level of education. About 16 percent of self-employed workers of either gender held an advanced degree, compared with 12.2 percent of wage and salary women and 11.1 percent of wage and salary men. Self-employment rates fell for nearly all educational levels, consistent with the decline in the overall self-employment rate; however, the drop was considerably large for self-employed men holding professional degrees. (See table 6.) Because of this shift, in 2012, the self-employment rate for professional degree holders was about double the average rate for both genders.

Health and disability status. In 2012, self-employed women reported the poorest health of all worker groups. Among these female workers, 8.9 percent reported fair or poor health, compared with 8.3 percent for self-employed men, 6.8 percent for wage and salary women, and 6.1 percent for wage and salary men.⁷ (See table 5.) In addition, self-employed women had the highest proportion of disabled workers—5.1 percent, compared with 3.0 percent for self-employed men and about 2 percent for wage and salary workers of either gender.⁸ In 2012, 0.8 percent of women in both employment sectors received disability compensation through the U.S. Department of Veteran Affairs, the Supplemental Security Income program, or another source. In terms of their disability characteristics, self-employed women are more similar to wage and salary women and less similar to men in either employment sector. Table 6 shows an increasing self-employment rate among female disabled workers, a rate that was more than twice the average female rate in 2012. By contrast, this trend is not evident for self-employed men.

Job characteristics and earnings

Work schedules. In 2012, nearly half of self-employed women worked less than 35 hours a week, and the majority of them worked part time for noneconomic reasons (e.g., to take care of children and other family members, attend school or training, or contribute a secondary income to their households). As shown in table 7, self-employed women worked the lowest number of mean weekly hours and weeks per year. Interestingly, their work schedules exhibited the highest variation—a variation reflecting a heterogeneous group of workers putting forth different levels of effort depending on their nonwork responsibilities and limitations, as well as personal ambitions. In terms of work schedules, self-employed women are more similar to women working in the wage and salary sector and less similar to self-employed men, who generally work the most mean weekly hours and weeks per year among all worker groups.

Health insurance coverage. Not surprisingly, business owners have the lowest shares of employer-provided health insurance coverage and the highest shares of private- and government-provided health insurance coverage. In 2012, less than 16 percent of self-employed women and a quarter of self-employed men had employer-provided health insurance. (See table 7.) About 40 percent of self-employed workers had private- and government-provided insurance. The high share of government-provided insurance likely reflects the fact that self-employed people are relatively older. Further, self-employed workers, especially female ones, are more likely to have health insurance coverage through a plan of a spouse or another family member. While a third of self-employed women in 2012 were insured through a family member, this share had come down from nearly 42 percent in 1993. This change is consistent with the upward trend in self-employment among unmarried women.

Earnings. Comparing earnings across employment sectors can be problematic. Although self-employed workers exhibit greater income variation than do wage and salary workers, they benefit from nonpecuniary benefits that are not reflected in their reported earnings. In addition, self-employed workers, in particular those who are incorporated, often report low or zero earnings.⁹ Thus, this article compares the earnings of wage and salary workers with those of unincorporated self-employed workers only.

Table 8 shows the annual, weekly, and hourly earnings (measured in 2013 dollars) of workers in 1993 and 2012, by employment sector and gender.¹⁰ Among the worker groups presented in the table, self-employed women in 2012 earned the lowest mean earnings. With respect to mean hourly earnings, however, part-time self-employed women are a notable exception. While these women had mean weekly earnings similar to those of wage and salary women, they also worked fewer hours. In 2012, their mean hourly earnings were $19.52, compared with $16.87 for wage and salary women. Overall, the earnings differential between the employment sectors was the largest for full-time workers. In 2012, the mean annual earnings of self-employed women were more than $12,000 less than the mean annual earnings of women working in the wage and salary sector.

All worker groups experienced an increase in earnings over the 1993–2012 period, albeit to a different degree. On average, self-employed women in both full- and part-time status experienced the most dramatic increase in mean earnings. Between 1993 and 2012, self-employed women working full time increased their mean annual earnings by 32 percent, their mean weekly earnings by 39 percent, and their mean hourly earnings by 53 percent. The percent increases were even steeper for part-time self-employed women; these female workers increased their mean annual earnings by 51 percent, their mean weekly earnings by 64 percent, and their mean hourly earnings by 52 percent. Wage and salary women also experienced considerable gains in earnings, although their percent increases were about half those of self-employed women.

The female-to-male earnings ratio—that is, women’s mean earnings as a percentage of men’s—continues to be smaller in the self-employment sector than in the wage and salary sector. (See table 9.) In 2012, the ratio was 66.2 (59.1) percent for full-time (part-time) self-employed workers and 75.3 (71.0) percent for full-time (part-time) wage and salary workers. Nevertheless, self-employed women have made considerable strides in closing the gap; between 1993 and 2012, the gender gap in hourly earnings dropped by about 20 percentage points for full-time workers and by about 17 percentage points for part-time workers.

Occupation and industry

The top panel of table 10 lists the most common occupations of incorporated and unincorporated self-employed women. The data suggest that occupations are more disparate among incorporated business owners, with managerial and professional occupations being the most common within this group. By contrast, unincorporated business owners tend to work in a few highly concentrated occupations. Of all self-employed women in 2012, nearly 7 percent were unincorporated childcare workers, 5 percent were unincorporated hairdressers/cosmetologists, and about 4 percent each were unincorporated housekeepers, general managers, and retail sales managers.

The bottom panel of table 10 lists the most common major industries of incorporated and unincorporated self-employed women. The pattern of industry concentration is generally the same for both types of female workers. In 2012, about a quarter of self-employed women worked in professional industries and another fifth worked in educational and health services.

ALTHOUGH THE SHARE OF WOMEN in self-employment and the female self-employment rate have remained relatively steady in recent decades, several notable within-group trends have emerged. For instance, after the Great Recession, the female self-employment rate fell only about half as much as the male self-employment rate did. Female business owners saw considerable gains in earnings, yet remained the lowest-earning workers in 2012. In addition, their sociodemographic characteristics changed over time. In 2012, self-employed women were more diverse in terms of race, marital status, presence of young children, health, and mean educational level.

In some ways, self-employed women are similar to their self-employed male counterparts. Like male business owners, and despite the observed trend toward greater diversity in female self-employment over the 1993–2012 period, they are still more likely to be older, white, married, better educated, and with less health insurance coverage. In other ways, they are similar to their female counterparts working in the wage and salary sector. Both self-employed and wage and salary women work fewer hours and earn less than their respective male counterparts do.

Despite these similarities, some characteristics distinguish self-employed women from all other worker groups. Female self-employed workers are more likely to have a disability, diverse work schedules, and lower mean annual earnings. Those working part time have a somewhat better earnings position, which gives them both a higher value on their time and the ability to benefit from joint production.