Gender Differences in Full-Time Self- Employment
Suzanne Heller Clain
This analysis reveals interesting gender differences in full-time self-employment. Women who choose full-time self-employment have personal characteristics that are less highly
valued in the marketplace than women who work full-time in wage-and-salary employ- ment. The reverse is true for men. It is unclear whether the gender gap in self-employment
income is the result of different supply decisions made by women, or greater constraints andor discriminatory elements faced by women. There is some suggestion that women
may place a higher value on nonwage aspects of self-employment than men do.
© 2000 Elsevier Science Inc.
Keywords: Self-employment; Gender differences JEL classification: J23, J16, J30
I. Introduction
In this paper, gender differences in the propensity for self-employment vs. wage-and-salary employment and in the levels of earnings in each type of employment are investigated, using
econometric techniques. It is found that women who choose self-employment have personal characteristics that are less highly valued by the market than women who choose
wage-and-salary employment; the reverse is true for men. Certain personal characteristics appear to affect self-employment earnings differently for women than for men. It is
unclear whether the resulting gender gap in self-employment earnings is the result of different supply decisions made by women or the result of greater constraints andor
discriminatory elements faced by women. Finally, the observed gender differences in the gap between self-employment earnings and potential wage-and-salary income suggest that
self-employed women may place a higher value on the nonwage aspects of self- employment than self-employed men do.
Associate Professor, Villanova University, Villanova, PA. Address correspondence to: Suzanne Heller Clain, Associate Professor, Department of Economics, Villanova
University, 800 Lancaster Avenue, Villanova, PA, 19085.
Journal of Economics and Business 2000; 52:499 –513 0148-6195 00 –see front matter
© 2000 Elsevier Science Inc., New York, New York PII S0148-61950000032-1
A review of the relevant literature is presented in Section II. A discussion of the methodology used in this paper is contained in Section III. Section IV provides a
description of the data used, whereas the empirical findings are outlined in Section V. The main conclusions and implications of this research appear in Section VI.
II. Literature Review
The issue of self-employment has attracted the attention of economic researchers in the past decade for various reasons. Some economists have been drawn to the topic by the
expansion of the self-employed sector that began in the U.S. in the mid-1970s Blau, 1987; Evans and Leighton, 1989; Devine, 1994a, 1994b. Others have been interested in the
self-employment of marginalized groups Moore, 1983a; Bates, 1997; Borjas and Bronars, 1989; Fairlie and Meyer, 1996. Common threads running throughout this body of
research concern the role of self-employment in economic advancement and growth, and the impact of public policies on self-employment Moore, 1983b; Blau, 1985 and 1987;
Yuengert, 1994.
In many of these studies, the empirical portion of the analysis was applied exclusively to men Rees and Shah, 1986; Evans and Leighton, 1989; Evans and Jovanovic, 1989;
Yuengert, 1994. Rees and Shah explained their decision to omit all females from their study by characterizing full-time self-employment as “predominantly a male preserve.”
1
Moore 1983a, Devine 1994a, and Fairlie and Meyer 1996 stand apart from these others by applying their analyses to women as well as men. The figures reported in these
studies confirm that men have been more prone to full-time, full-year self-employment than women.
2
However, the growth in the self-employment rates has made the study of women more viable and more crucial.
3
In fact, Devine 1994b focused exclusively on women, in an effort to explore gender-specific reasons for the substantial increase in the
nonagricultural female self-employment rate between 1975 and 1987. The information on self-employed men and women found in these latter studies does
not provide a complete picture of gender differences in self-employment, however. In the study by Fairlie and Meyer, malefemale comparisons were secondary to the primary
interest in ethnic and racial self-employment differences. Even so, although these authors did explore ethnic and racial differences in the self-employment rates of both men and
women, their analysis of ethnic and racial differences in earnings was restricted to men, because sample sizes were too small for women. In testing for employer discrimination
against women, Moore focused on femalemale earnings ratios in self-employment and in wage-and-salary employment. He did not explore gender differences in self-employment
rates, because he felt such differences would reflect imperfections in the capital markets, and not provide a true test of employer discrimination.
1
Some of these authors used household data and elected to focus exclusively on men Rees and Shah, 1986; Yuengert, 1994. Others used the National Longitudinal Survey of Young Men, which did not include data on
women Evans and Jovanovic, 1989; Evans and Leighton, 1989.
2
In Moore’s 1978 sample, 6.7 of men and 2.5 of women were self-employed. Fairlie and Meyer found self-employment rates to be 10.8 for men and 5.8 for women in 1989. In 59 of the 60 ethnicracial groups
studied by Fairlie and Meyer, the self-employment rates of men exceeded the self-employment rates of women. The exception was the Vietnamese group.
3
Devine reported that the full-time, full-year self-employment rate for women men had increased from 3.2 to 5.8 10.9 to 12.8 between 1975 and 1990
500 S. H. Clain
An extensive effort by Devine 1994a provides greater detail on gender differences in self-employment. She measured gender differences in self-employment rates by race, age,
marital status, years of schooling, and fullpart-time and fullpart-year status. She recorded gender differences in occupation and industry distributions and earnings levels, for
self-employed workers as well as wage-and-salary workers. In particular, she reported that among full-time, full-year workers in 1990, self-employed women earned 73 of the
annual income of female wage-and-salary workers, whereas self-employed men earned 107 of the annual income of male wage-and-salary workers.
The results of Devine’s descriptive statistics and the writings of previous researchers are suggestive of the causes for these gender differences. According to Devine, there are
noticeable differences in the personal characteristics of self-employed men and women. For example, self-employed men are more likely to be in high-paying occupations e.g.,
executive, administrative, and managerial; precision production, craft, and repair and industries e.g., construction than self-employed women. Self-employed women are more
likely to be in service occupations and industries. Self-employed men are, on average, more educated than self-employed women. Moreover, self-employed men are more likely
than self-employed women to have incorporated their self-employment business.
After controlling for some differences in personal characteristics, Moore 1983a nevertheless found that femalemale earnings ratios in self-employment were much lower
than femalemale earnings ratios in wage-and-salary employment. Moore suggested that the finding could be evidence of substantial consumer discrimination against women.
The idea that discrimination is a source of differences between the wage-and-salary and self-employment sectors has also received the attention of other researchers in other
applications. Borjas and Bronars 1989 developed the theory of consumer discrimination in the context of self-employment more fully. They proceeded to test it on data consisting
of White, Black, Asian, and Hispanic men. The results of estimations of earnings functions showed an income gap between self-employed Blacks Hispanics and Whites
that remained, even after controlling for differences in demographic characteristics. These results were consistent with the implications of their theoretical model.
4,5
In this paper, the econometric techniques used in these other settings are applied to microdata, in an effort to contribute to the understanding of gender differences in
self-employment. The methodology of the analysis is described in greater detail in the following section.
III. Methodology
