Women in the workforce

Material Information

Women in the workforce education, employment and entrepreneurship
McCullough, Theresa E
Place of Publication:
Denver, Colo.
University of Colorado Denver
Publication Date:
Physical Description:
vi, 111 leaves : ; 29 cm

Thesis/Dissertation Information

Master's ( Master of Arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Economics, CU Denver
Degree Disciplines:
Committee Chair:
Helburn, Suzanne
Committee Members:
Morris, John
Beckman, Steven


Subjects / Keywords:
Women -- Employment -- United States ( lcsh )
Businesswomen -- United States ( lcsh )
Businesswomen ( fast )
Women -- Employment ( fast )
United States ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Economics.
General Note:
Department of Economics
Statement of Responsibility:
by Theresa E. McCullough.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
28246883 ( OCLC )
LD1190.L53 1992m .M32 ( lcc )

Full Text
Theresa E. McCullough
B. S. Columbia College, 1982
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Theresa E. McCullough
has been approved for the
Department of
Date J

McCullough, Theresa E. (M.A., Economics)
Women in the Workforce: Education, Employment and
Thesis directed by Professor Suzanne Helbum
Over the past few decades, women's participation in the
labor force, higher education and business has increased substan-
tially. This thesis analyzes same of the social and economic
trends, such as income, marriage, divorce, and birth rates, which
have had a significant inpact on the number of women in the labor
force and their educational achievements. It also examines
growth patterns in the number of wamen-owned businesses, their
contribution to economic development, income and job creation.
There also is a discussion of the policy issues which affect
business women. These issues include access to capital, procure-
ment, data collection and technical training. Until recently,
most public policy has been basically ineffective. Currently,
however, a number of public and private initiatives are being
implemented at both the national and state levels.

The thesis includes a discussion of the rate of women-owned
businesses by state. Given its population size, Colorado had the
highest number of women-owned businesses of any state in 1987.
Two econometric models are developed in this thesis to determine
the economic, demographic and political characteristics of states
with a relatively higher rate of female entrepreneurs. This
analysis suggests that per capita income growth, unemployment
levels, political affiliation, and the percentage of women in
state legislatures are among the variables which have a signifi-
cant relationship to the number of women-owned firms.
In addition to testing the state determinants of women-owned
firms, a third regression model is conducted to test the strength
in the relationship between the number of women in the labor
force and social factors. Variables such as income, education,
birth rates and marriage rates do indeed have a significant
impact on women's labor force patterns.
This abstract accurately represents the contents of the
candidate's thesis. I recammer

1. INTRODUCTION.......................................... 1
Female labor Force Characteristics............... 6
Marriages and Dissolutions....................... 8
Birth Rates......................................11
Employment Trends................................24
Income. .........................................28
Women's Income by Educational Attainment. ... 29
3. WQMEN-OWNED BUSINESSES................................35
Growth in Wamen-Owned Firms......................36
Industry Characteristics and Receipts............39
Employees of Wamen-Owned Firms...................45
Geographic location of Wamen-Owned Businesses . 46
and Wamen-Owned Businesses in Colorado
Policy Issues....................................54
Access to capital...........................56
Data collection.............................57
Technical training..........................59

Theoretical Discussion...........................61
Description of the Model.........................64
List of Variables...........................65
Methodology of Models............................68
Mbdel 1.....................................69
Model II.................................69
Model III...................................70
Mbdel 1.....................................70
Mbdel II....................................77
Mbdel III...................................82
5. CONCLUSIONS...........................................86
OTHER READINGS......................................... . 94

This thesis summarizes findings of a study of two trends
affecting the lives of American women. The first is the rapid
increase in the number of working women. The second is the
persistent rise in the number of female entrepreneurs during the
This thesis is divided into five chapters. Following this
first introductory chapter, the second offers a discussion of the
labor market participation of women and a summary of the changes
in social and economic patternsmarriage, childbearing, income,
and educationwhich contribute to women's increased
participation in the workforce. Over the past twenty years,
greater numbers of women have elected to postpone or to decline
marriage and childbearing. One effect of this trend is that it
offers women more time to pursue education and career goals.
Current studies show that in addition to working more, women are
attaining higher levels of education. More importantly, many
women are entering nan-traditional fields of study such as
management, law and medicine. This trend is important because
women's broader educational background results in greater

occupational diversification, greater potential for career
advancement and, hopefully, higher wages.
Because of these demographic and educational trends, women
now have broader occupational choices than before. One career
option open to an increasing number of women is entrepreneurship.
Over the past two decades, the number of women-owned businesses
has skyrocketed. The purpose of Chapter 3 is to identify the
number of businesses owned by women, the types of businesses they
own, and the economic implications of this trend at the national
and state level. This chapter also includes a brief discussion
of the policy implications resulting from the increase in women-
owned enterprises.
The four primary policy issues related to the development in
opportunities for women-owned businesses:
availability of capital
technical assistance
data collection
The rate of women's business ownership varies greatly by
state. Data analyzed in Chapter 3 shows that in 1987, Colorado
had the highest number of women-owned businesses per 100,000
people of any state. The fourth chapter, therefore, provides a

cross-sectional analysis of the determinants of women-owned
businesses by state and an analysis of the determinants of the
female labor force participation rate. Two models are designed
to identify determinants of increasing entrepreneurship between
1982 and 1987 losing economic, demographic, political and region
variables. A third model tests the relationships between income,
marriage rates, divorce rates, and childbirth rates and the
female labor force participation rate.
The data suggests that states with relatively higher incomes
per capita have more women-owned businesses. In addition, these
states are predominantly Republican, have a high female labor
force participation rate and low unemployment. Further, the data
shows that the majority of women-owned businesses are located in
the West. Other important determinants of the rate of female
entrepreneurship include state business failure rates, age, and
the percentage of women in the state's legislature.
Results of the third test imply a significant relationship
between the female labor force participation rate and all social
and economic variables except the marriage rate.
The last chapter is reserved for conclusions. The most
fundamental social changes of the past decade include the rapid
flow of women into the labor market, the decline in marriage and

fertility and the broadening of women's educational and occupa-
tional opportunities.
Although social attitudes are changing, most women feel it
necessary to balance the dual roles of family and career. The
balancing act often requires a great deal of flexibility in
scheduling daily activities. Entrepreneurship has proven to be
one career choice offering this kind of flexibility. Because of
women's scheduling demands, and broadening educations and occupa-
tions, there is a large number of women opening businesses in
both traditional and non-traditional industries. In addition to
social and occupational changes, the regression analysis shows
that economic and political factors also contribute to women's
entrepreneurial success.

This chapter presents an overview of changes in social and
economic patterns which have contributed to the increase in the
female labor force participation rate.
The factors include marriage rates, divorce rates, child-
birth rates, income and education. These factors are important
because there are strong links between social changes that have
occurred over the past few decades and the types of occupations
and businesses women have entered. Lower rates of marriage and
childbirths witnessed throughout the 1970s and 1980s and the high
divorce rate, for example, all suggest a decline in marriage and
families as a life priority. This reordering of life priorities
has allowed women more freedom and time to pursue education and
career goals. As a result, the proportion of women's lives spent
in marriage and child rearing has declined and the number of
years spent working has increased.
Women's educational achievements over the past few decades
have also contributed to the increase in the number of working
women. Women are not only earning more than half the bachelor's
and master's degrees, but their educations are becoming more

diverse, providing greater occupational choice and relatively
higher incomes.
Women's incomes tend to be sixty percent to seventy percent
of men's income. However, there is evidence that the wage gap
may be less for women entering non-traditional occupations.
According to research conducted by Victor Fuchs, the group of
women achieving the greatest gains relative to men are unmarried,
white, young, and well educated (Fuchs 1988).
Fgmalft Tahnr Form Characteristics
Prior to World War II, women held jobs most often before
they were married or after their children were raised. In 1960,
the female labor force participation rate (FLFER) was highest
among women aged 35-64 whose children were older or grown. This
pattern has changed drastically over the past three decades.
Marriage and childbirth no longer restrict women from working as
it once did. By 1985, 71% of the women in the labor force were
25-45; half the mothers with children under age four are in the
labor force (Fuchs 1988).
The FLFER increased from 36% in 1960 to 57% in 1988. It is
expected to reach 63% by the year 2000. Meanwhile, the aging of
the population and early retirements have resulted in a slight

decline in the male labor force participation rate from 80% in
1970 to 76% in 1988 (Statistical Abstract 1990). As a result,
women are becoming a relatively more important component of the
labor force. Women now compose 44% of the national labor force
and have taken two-thirds of the new jobs in this country in the
past two decades (NWBC 1990).
The number of working women has increased not only for
single women but also for married women and for women with chil-
dren. There is also an increase in the number of women working
full time.
In 1988, 77% of the women in the labor force worked full
time, 63% were mothers and over half (55%) were married. The
percentage of working mothers with pre-school children increased
from 12% in 1950, to 45% by 1980. Meanwhile, the number of
working women with children aged seven to seventeen more than
doubled from 28% to 62% (Mclaughlin et al. 1988). In the 1980's,
one-half of the new mothers returned to work before their child
reached their first birthdayfour times as many as in 1960
(Fuchs 1988).
The FLFER in Colorado has followed the national trend. In
1970, 37% of the Colorado workers were women; by 1988 that figure

topped 44%. The female labor force participation rate in
Colorado was 62% in 1988 and is expected to exceed 66% by the
year 2000. fColorado Workforce 2000 Oct. 1988).
Although it is impossible to identify the order of cause and
effect, there are several social factors that have contributed to
the increase in the number of working women. These factors
include declining marriage and birth rates, increased divorce
rates, educational advancement, and income.
Marriage and Dissolutions
Although marriage is still the norm, women have been
postponing marriage and are more likely than women in past
generations to remain single.
The Colorado marriage rate has been consistently higher than
the national rate since 1960 until 1987 and 1988 when it began to
fall slightly below the national rate. The marriage rate in
Colorado averaged about eleven marriages per 1,000 people during
the 1970s. The marriage rate increased to 12.5 per 1,000 people
fcy 1981 but it decreased to 9.5 by 1988 r Colorado vital
Statistics 1991).

The median age for marriages has increased since the 1970s,
in both Colorado and the nation. For example, the national median
age for women marrying for the first time in 1970 was 21, but it
had increased to 23 by 1986. The trend was also true for men.
In 1970, the median age for men marrying for the first time was
twenty-three; by 1986 it increased to twenty-five. In Colorado,
the mean age for grooms in 1988 was 31 years compared with
29 years for women; yet, about half of all marriages were among
people in their twenties (Colorado vital Statistics 1990).
Marriage dissolutions have been high since the 1960's.
Couples are not as likely to remain in unsatisfying marriages as
they once did. It is also more socially acceptable and much
easier to dissolve a marriage now than it was in the past since
the majority of the states instituted the no-fault divorce stat-
utes. During the 1950s and 1960s, the national dissolution rate
was low and stable. But between 1965 and 1975, dissolutions
nearly doubled from 2.5 per 1,000 people to 4.8. The national
dissolution rate peaked in 1979 at 5.3 before declining to 4.8 in
1988 (Colorado vital statistics 1991).
Dissolutions in Colorado have exceeded the national rate
since the 1960s. The highest rate was recorded in 1977 at 7.6

per 1,000 people. It has been declining since then. By 1988, It
had declined to 5.6 (Colorado vital statistics 1990).
The recent trends in marriage and divorce and childbearing
have had a profound effect on the lives of women. The lower
rates of marriage, the rise in the age of first marriages, and
the increased divorce rates all point to a decline in marriage as
a life priority. As a result, the proportion of women's lives
spent in marriage has declined while the number of years spent
working has increased. In 1970, a sixteen-year-old woman could
expect to spend 23 years of her life working. By 1980, a
sixteen-year-old woman could expect to spend more than 29 years
working (McLaughlin et al. 1988). The combined effects of lower
marriage and birth rates allow women more freedom and time to
pursue education and career goals. For same women, however, high
divorce rates have a negative effect, forcing them into the labor
market unprepared. Many women, lacking training or education,
are forced to take low-paying, low-skilled jobs.

Birth Rates
Birth rates declined in both Colorado and in the United
States from 1960 until the mid-1970s. Bates have remained low
compared to rates in the 1950s and early 1960s, but have fluc-
tuated since then, averaging slightly higher in Colorado than for
the nation as a whole. In 1960, the Colorado birth rate was 24.5
per 1,000 women. It dipped to 15.4 in 1974 before rising again
to 16.2 in 1988 (Colorado Vital Statistics 1991).
Traditionally, women were not encouraged to invest in their
education or to plan careers because they were expected to devote
most of their lives to their marriage and families. If they
worked, it was usually for short periods of time, either before
marriage or after their children were raised. Low expectations,
no encouragement and narrow educational opportunities kept women
in low paying, low skilled occupations. Even educated women such
as teachers, nurses and secretaries, earned relatively low
This pattern is beginning to change. According to a report
issued by the Colorado Women's Foundation, an adequate education

was viewed as a key element to improving the economic self-
sufficiency of women. Education was listed as the highest
priority need of the women interviewed for the Foundation study.
Women have realized the overwhelming importance of higher
education to their employment potential and to their income.
During the 1970s and 1980s, many young women placed more
importance on investing in education and planning for life-long
careers and less importance in marriage and childbearing.
Marriage and childbearing have became less of a constraint on
women's educational opportunities. Women, especially older than
thirty years, have made education a priority (McLaughlin et al.
Although about half the high school graduates are women,
until recently they did not participate equally in higher educa-
tion. Currently, however, women compose more than fifty percent
of the enrollment in many colleges and universities, and they are
tiie majority in two-year community colleges. Women now comprise
over half of the enrollees and degree recipients at all levels of
higher education except at the doctorate and first professional
(MD, DDS, JD) level (U.S. National Center for Education Statis-
tics, annual).

College enrollment levels have increased for women of all
ages. From 1965 to 1983, enrollment of women aged 18 to 19
increased from 38% to 50%; enrollment for women 25 to 29 in-
creased from 3% to 9%; for women 30 to 34, enrollment rose from
2% to 7%. By 1988, one-fourth of the women enrolled in higher
education were over thirty (Mclaughlin et al. 1988).
Women have traditionally earned degrees in education, for-
eign language, health science, home economics and library sci-
ence. More recently, they have been entering fields of study
traditionally dominated by men. The most drastic changes have
been in law, dentistry and medicine, but many women are earning
degrees in business management, agriculture and architecture.
There is even signs of improvement, albeit slow, in the number of
women earning degrees in math and sciences.
As presented in Table 2-1, in 1970 only 8% of the medical
degrees in 1970 were awarded to women; by 1987, 32% of the medi-
cal degrees were awarded to women. Nearly one-fourth on the
dentistry degrees were earned by women in 1987, up from less than
one percent in 1970. law degrees received by women increased
from five percent in 1970 to over forty percent in 1987.

Table 2-1
Degrees Conferred in Medicine, Law and Dentistry
1960 to 1987
% Women
% Women
% Women
Source: U.S. Bureau of the Census, Statistical Abstract of
the United States: 1990 (110 edition), Washington D.C., 1990
In addition, Table 2-2 shows Earned Degrees conferred by
field of study and level of degree for 1971 and 1987. Women's
relative gain is remarkable in several areas. For example, at
the undergraduate level, women earned 47% of the business degrees
in 1987, up from 9% in 1971. They earned one-third of the de-
grees in agriculture and 37% of the degrees in architecture; 46%
of the math degrees. Women earned 28% of the degrees in physical
science in 1987, up from 14% in 1971. In addition, women

studying life sciences earned 49% of the degrees in 1987, up
from 29% in 1971. Declines are apparent in the number of women
earning degrees in foreign language, home economics and library
Table 2-2
Earned degrees conferred by Field of Study
and Level of Degree 1971 1987
Field of Studv Total Degrees Awarded 1971
Bachelors, total 839,730
Agricultural/natural 12,672
Architecture; en- 5,570
vironmental design
Ethnic studies 2,582
Business/management 114,865
Communication 10,802
Computer; informa- 2,388
tion sciences
Education 176,614
Engineering 50,046
Foreign language 19,945
Health sciences 25,190
Home economics 11,167
law 545
Letters 64,933
Liberal/general 5,461
Library science 1,013
Life science 35,743
Mathematics 24,801
Military 357
Multidisciplinary 8,306
Percent Female 1971 Total Degrees Awarded 1987 Percent Female 1987
43.4 991,339 51.5
4.2 14,991 31.2
11.9 8,922 37.3
52.4 3,340 61.7
9.1 241,156 46.5
35.3 45,408 60.0
13.6 39,664 34.6
74.5 87,115 76.2
0.8 93,074 13.7
74.6 10,184 72.6
77.1 63,206 85.5
97.3 14,942 92.5
5.0 1,178 68.6
65.5 37,133 65.8
29.0 21,365 56.4
92.0 139 85.6
29.1 38,114 48.5
38.0 16,489 46.4
0.3 383 6.8
28.4 16,402 53.7

Field of Study______
Physical science
Protective services
Public affairs
Social science
Visual and per-
forming arts
Masters total
Architecture; en-
vironmental design
Ethnic studies
Computer; informa-
tion sciences
Foreign language
Health sciences
Home economics
Library science
Life science
Physical science
Total Degrees Awarded 1971 Percent Female 1971
1,621 34.7
11,890 25.5
21,412 13.8
2,045 9.2
37,880 44.5
6,252 60.2
155,236 36.8
30,394 59.7
230,509 40.1
2,457 5.9
1,705 13.8
1,032 38.3
26,481 3.9
1,856 34.6
1,588 10.3
88,952 56.2
16,443 1.1
4,755 65.5
5,445 55.9
1,452 93.9
955 4.8
11,148 60.2
549 44.3
7,001 81.3
5,728 33.6
5,191 29.2
2 83.0
1,157 10.9
218 29.8
4,036 27.1
6,367 13.3
Total Degrees Awarded 1987 Percent Female 1987
4,107 60.2
11,686 30.1
19,974 28.4
12,930 38.3
42,868 68.9
14,161 68.0
96,185 44.0
36,223 61.9
289,557 51.2
3,523 30.1
3,142 34.0
851 46.9
67,496 33.0
3,937 59.2
8,491 29.4
75,501 74.0
22,693 12.6
1,746 70.4
18,426 78.9
2,070 87.6
1,943 26.8
6,123 65.0
1,126 59.2
3,815 79.1
4,954 48.7
3,321 39.1
0 2.4
3,041 42.0
476 55.3
5,989 34.5
5,652 24.9

Field of Studv Total Degrees Awarded 1971 Percent Female 1971 Total Degrees Awarded 1987 Percent Female 1987
Protective services 194 10.3 1,019 29.4
Psychology 4,431 37.2 8,204 65.2
Public affairs 8,215 49.2 17,032 63.7
Social science 16,476 28.5 10,397 39.5
Visual and per- 6,675 47.4 8,506 55.8
forming arts
Doctorates total 32,107 14.3 34,102 35.2
Agricultural/natural 1,086 2.9 1,049 17.0
Architecture; en- 36 8.3 92 28.3
vironmental design
Ethnic studies 144 16.7 132 44.7
Business/management 807 2.9 1,098 23.6
Communication 145 13.1 275 42.5
Computer; informa- 128 16.4 374 13.9
tion sciences
Education 6,403 21.2 6,909 54.9
Engineering 3,638 0.6 3,820 6.9
Foreign language 781 38.0 441 58.3
Health sciences 459 16.3 1,213 53.5
Home economics 123 61.0 297 78.1
law 20 0.0 120 34.2
letters 1,857 28.0 1,181 56.4
Life science 3,645 16.3 3,423 35.0
Mathematics 1,199 7.8 725 17.4
Multidisciplinary 80 13.8 276 37.0
Philosophy 866 5.8 1,658 13.3
Physical science 4,390 5.6 3,672 17.3
Psychology 1,782 24.0 3,123 53.3
Public affairs 185 23.8 398 45.7
Social science 3,659 13.9 2,916 30.5
Visual and per- 621 2.24 792 43.6
forming arts
Source: U.S. National Center for Education Statistics, of Education Statistics. Annual. Diaest

Women's educational trends in Colorado mirror the national
image. For example, Colorado's women earn over 55% of the asso-
ciate degrees, and half of the bachelor's degrees and master's
degrees. Although they earn about 48% of the first professional
degrees, they were awarded only 30% of the doctorate degrees.
See Table 2-3.
Table 2-3 Degrees Conferred by Gender for Colorado, 1989-1990
Degree Level Total Male Female % Female
Certificate 2,089 913 1,170 56
Associate 3,870 1,744 2,122 55
Bachelor's 14,126 6,760 7,363 52
Master's 3,887 1,814 2,046 53
1st Prof. 419 219 200 48
Doctor's 578 404 173 30
Source: CCHE. Scorecard. Jan. 1911.
Degrees conferred in Colorado for 1989-1990 are shown in
Table 2-4. Although there are more women earning degrees in
math, engineering and computer science, their percentages are
low. In 1990, women earned 42% of the bachelor's degrees in
math, fifteen percent in engineering, and 23% of the computer
science degrees. Women need more encouragement to study these
fields because demand for these skills in the labor market is

Table 2-4
Degrees by Gender for 1989/1990
Degree Level Total Male Female % Female
Certificate 122 61 61 50
Associates 54 36 18 33
Bachelors 63 49 14 22
Masters 12 8 4 33
Doctors 4 4 0 0
Certificate 2 1 1 50
Associates 31 4 27 87
Bachelors 118 56 62 53
Masters 29 18 11 38
Doctors 20 18 2 10
NATURAL RESOURCES-AGRI. Associates 10 9 18 10
Bachelors 122 88 34 28
Masters 30 18 12 40
Doctors 9 8 1 11
Associates 1 0 1 100
Bachelors 111 75 36 32
Masters 99 58 41 41
Bachelors 49 17 32 65
Certificate 22 9 13 59
Associates 265 109 156 59
Bachelors 2 ,927 1,469 1,457 50
Masters 636 396 236 37
Doctors 15 8 7 47
Certificate 338 43 295 87
Associates 405 73 332 82

Degree Level Total
Certificate 63
Associates 4
Bachelors 843
Masters 65
Certificate 4
Associates 11
Bachelors 305
Masters 83
Doctors 8
Certificate 32
Bachelors 810
Masters 1,265
Doctors 75
Bachelors 1,092
Masters 482
Doctors 105
Certificate 168
Associates 402
Bachelors 374
Bachelors 139
Masters 29
Doctors 8
Certificate 530
Associates 186
Male Female % Fema!
10 53 84
2 2 50
286 557 66
21 41 63
3 1 25
7 4 36
236 69 23
57 26 31
7 1 13
2 30 94
192 618 76
318 932 74
37 38 51
933 159 15
414 67 14
95 10 10
151 17 10
318 82 20
345 29 8
38 101 73
6 23 79
3 5 63
190 338 64
38 148 80

Degree Level Total Male Female % Female
Bachelors 153 22 153 100
Masters 39 10 29 74
Doctors 5 4 1 20
health sciences
Certificate 158 4 154 97
Associates 389 48 341 88
Bachelors 551 87 464 84
Masters 203 23 180 89
1st prof. 283 139 144 51
Doctors 11 4 7 64
Bachelors 286 16 270 94
Masters 35 3 32 91
Doctors 3 2 1 33
HOME ECONOMICS VOC Certificate 25 0 25 100
Associates 80 7 73 91
Certificate 102 10 92 90
Associates 53 4 49 92
1st prof. 136 80 56 41
Bachelors 599 222 376 63
Masters 94 30 63 67
Doctors 7 5 2 29
Associates 1,210 500 710 59
Masters 10 0 10 100
Bachelors 674 306 368 55
Masters 74 40 33 45
Doctors 46 36 10 22

Degree Level Total Male Female % Female
Bachelors 320 187 133 55
Masters 69 52 15 22
Doctors 14 13 1 7
INTERDISCIPLINARY STUDIES Bachelors 547 202 345 63
Masters 52 26 26 50
Doctors 10 7 1 10
Bachelors 82 38 44 54
Masters 10 3 7 70
PHHOSOEHY AND RELIGION Bachelors 58 38 20 34
Masters 8 8 0 0
Doctors 3 1 2 67
Associates 232 145 87 38
Bachelors 257 194 63 25
Masters 120 97 23 19
Doctors 121 96 25 21
Certificate 31 25 6 19
Bachelors 853 231 622 73
Masters 135 43 92 68
Doctors 43 14 29 67
Certificate 202 160 41 20
Associates 146 111 35 24
Bachelors 125 72 53 42
Masters 5 4 1 20
Bachelors 147 30 117 80
Masters 80 41 39 49
Doctors 11 10 1 9

Degree Level
Bachelors 1,800
Masters 145
Doctors 30
Certificate 37
Associates 21
Certificate 122
Associates 152
Certificate 162
Associates 112
Certificate 5
Associates 106
Bachelors 80
Bachelors 561
Masters 78
Doctors 30
Source: CCHE. SURDS File.
Male Female % Fema
1,034 846 45
84 61 42
20 10 33
29 8 22
19 2 10
117 4 3
149 3 2
93 31 25
67 45 40
5 0 0
98 6 6
60 20 25
237 323 58
36 42 54
15 15 50
It was projected by the National Center for Education Sta-
tistics that women's participation in higher education will level
off in the 1990s at the bachelor's and master's levels as women
earn half of all these degrees. However, growth is expected to
continue in the proportion of female doctorates. For many women,

education is the cornerstone for occupational progress, relative-
ly higher wages, and economic independence.
Employment Trends
As a result of the opportunities now available to women in
higher education, we have seen a number of women entering occupa-
tions historically dominated by men.
Although the majority of women are still employed in tradi-
tional professions, such as administrative support and household
services, many women are rapidly entering nan-traditional occupa-
tions such as management, architecture, and construction trades.
Table 2-5 reflects the occupational changes of men and women
between 1983 and 1988. As shown, there is a large percentage of
women entering fields such as management, sales, protective
services, construction and transportation. Although women still
dominate in administrative support and household services, Table
2-6 shows that almost half of the professionals are women and
that there is same growth in the percentage of female mechanics,
craftsmen, and equipment handlers.

Table 2-5
Percent of Women in Selected Occupation
Total % Total %
Occupation 1983 Women 1988 Women
Managerial 17,451 40.91 21,770 45.03
Exec./Admin./Manag. 8,116 34.15 10,725 41.84
Prof. Specialty 9,334 46.79 1,045 48.12
Total % Total %
Occupation 1983 Women 1988 Women
Tech. Sales & Admin. 21,642 62.46 24,931 62.83
Technicians 2,574 44.52 2,960 44.36
Sales 6,313 38.97 7,741 41.62
Admin. Support 12,755 77.70 14,230 78.21
Service Occupations 7,321 49.15 8,669 50.20
Private Household 278 96.04 328 97.56
Protective 1,453 9.57 1,747 11.68
Other Services 5,591 57.11 6,594 58.05
Precision Prod., Craft 9,964 7.87 11,175 8.29
Mechanics and Repair 3,538 3.39 3,850 3.56
Construction Trades 3,011 1.49 3,691 1.87
Other 3,415 18.13 3,635 19.83
Operators, Fabricators 13,319 26.17 14,763 25.31
Machine Op. Assembly 6,991 40.81 7,407 39.87
Trans. & Material 3,358 4.73 3,852 5.61
Handlers, Equip., 2,970 15.96 3,505 16.23
Agriculture, Forestry 1,280 11.17 1,382 11.65
Source: U.S. Bureau of the Census, Statistical Abstract of the
United States: 1990 (110 Edition) Washington D.C., 1990.

Table 2-6
Occupations of Full-Time Wage Workers, by Gender, 1983-1988
Exec., Admin., Manag.
Prof. Specialty
Tech. Sales & Admin. Supp.
Admin. Support
Service Occupations
Private Household
Other Services
Precision Prod. Craft
Mechanics, Repair
Construction Trades
Operators, Fabricators
Machine Operators, Assemb.
Trans. & Material Moving
Handlers, Equip., Cleaners
Agriculture, Forestry
Exec., Admin., Manag.
Prof. Specialty
Number of Number of
Workers Workers %
1983 1988 Change
7,139 9,802 37
2,772 4,487 62
4,367 5,315 22
13,517 15,664 16
1,146 1,313 15
2,460 3,222 31
9,911 11,129 12
3,598 4,352 21
267 320 20
139 204 47
3,193 3,828 20
784 926 18
120 137 14
45 69 53
619 721 16
3,486 3,737 7
2,853 2,953 4
159 216 36
474 569 20
143 161 13
10,312 11,968 16
5,344 6,238 16
4,967 5,730 15

Occupation Number of Workers 1983 Number of Workers 1988 % Change
Tech. Sales & Admin. Supp. 8,125 9,267 14
Technicians 1,428 1,647 15
Sales 3,853 4,519 17
Admin. Support 2,844 3,1019 9
Service Occupations 3,723 4,317 16
Private Household 11 8 -3
Protective 1,314 1,543 17
Other Services 2,398 2,766 15
Precision Prod. Craft 9,180 10,249 12
Repair Mechanics, Repair 3,418 3,713 9
Construction Trades 2,966 3,622 22
Other 2,796 2,914 4
Operators, Fabricators 9,833 11,026 12
Machine Operators, Assemb. 4,138 4,454 7
Trans. & Material Moving 3,199 3,636 14
Handlers, Equip., Cleaners 2,496 2,936 18
Agriculture, Forestry 1,137 1,221 7
Source: Statistical Abstract. 1990.

In addition to lifestyle changes, economic factors such as
personal income also contribute to the number of working women.
Rising wages, for example, make the opportunity cost of staying
at home higher. Barbara Bergmann, however, argues that real
wages have not increased significantly for either men or women
(Bergmann 1986). Many couples find that a two-income family is
necessary to maintain a desired standard of living. Whether
married or not, most women earn 60% to 70% of men's wages.
According to research conducted by Victor Fuchs, the group of
women achieving the greatest relative gain are unmarried, white,
young and well educated.
Women's Income by Educational Attainment
Some of the most disappointing women's income data is that
which shows women's median income as a percentage of men's by
educational attainment. Table 2-7 shows that in 1970, women with
at least four years of college earned 44% of men's wages. By
1988, women with four years of college were earning only 57% of
men's wages. Although the table shows 13% growth in income over
eighteen years, it is still disappointing that highly educated
women still earn less than 60% of men's wages.

Table 2-7
Women's Median Income as a Percent of Men's
By Educational Attainment
High School College
1-3 4 1-3 4 54-
Year Years Years Years Years Years
1988 45 46 53 57 59
1987 44 45 51 54 59
1986 44 42 42 51 60
1985 44 43 49 51 59
1984 44 42 49 48 60
1983 43 43 47 48 59
1982 41 41 43 48 58
1981 39 38 42 44 56
1980 37 36 42 44 55
1979 36 34 39 41 52
1978 36 36 40 43 53
1977 37 40 44 46 55
1976 36 40 41 46 57
1975 37 38 41 48 58
1974 35 37 40 45 57
1973 33 37 39 45 56
1972 34 38 38 44 57
1971 34 40 36 46 60
1970 33 39 38 44 59
Source: Trends in Income. 1990.

One issue that should be considered when analyzing the data
in this table is that the data is highly aggregated; that is, it
reflects all working women, all ages and all occupations. The
aggregated data also includes women working part tine. It is
possible that the volume of women earning very low wages is so
much greater than the number of women beginning to earn higher
wages that average increases are not reflected in this type of
aggregated income analysis. The next few tables show median
weekly earnings by gender, age, and occupation.
Table 2-8 shows that women's weekly earnings as a percentage
of men's was 70% in 1988, up from 67% in 1983. Young women
between sixteen and twenty-four were among those having the most
narrow wage gap, but women twenty-five years and older earned
69% of men's wages in 1988, up from 66% in 1983.
The table also shows that the percentage of women's weekly
earnings to all workers increased only one percent between 1983
and 1988 when it was reported at 82%. Compared to all workers,
women sixteen to twenty-four years of age earned 61% of the
median weekly earnings of all workers. Women twenty-four years
and older were earning 87% of median earnings of all workers in
1988, up from 84% in 1983.
Table 2-9 shows that women working in non-traditional fields
such as mechanics, equipment handlers and agriculture earn a

relatively higher wage. Similarly, the data shows that women
working in fields such as construction, protective services, and
professional specialities were among the highest paid.
Overall, however, data such as that shown in Table 2-10 show
that women's aggregated income compared to national averages
failed to exceed 70% between 1970 and 1988.
Table 2-8
Median Weekly Earning by Gender, 1983-1988
1983 1985 1987 1988
All Workers 313 343 373 385
Women 252 277 303 315
16 to 24 years old 197 210 226 235
25 years and older 267 296 321 335
Mien 378 406 433 449
16 to 24 years old 223 240 257 261
25 years and older 406 442 477 487
% of women's weekly
earnings to men's 67 68 70 70
16 to 24 years old 88 88 88 90
25 years and older 66 67 67 69
1983 1985 1987 1988
% of women's weekly
earnings to all workers 81 81 81 82
16 to 24 years old 63 61 61 61
25 years old 85 86 86 87
Source: Statistical Abstract 1990

Table 2-9
Women's Weekly Full-Time Wages as a Percent of Men s
Occupation 1983 1988
Managerial 69 70
Exec., Admin., Manag. 64 63
Prof. Specialty 73 75
Tech. Sales & Admin. Support 64 65
Technicians 71 75
Sales 52 54
Admin. Support 69 73
Service Occupations 68 70
Private Household
Protective 70 82
Other Services 81 82
Precision Prod., Craft Repair 66 68
Mechanics and Repair 89 89
Construction Trades 79
Other 60 60
Operators, Fabricators 66 68
Machine Operators Assembly 63 64
Trans. & Material Moving 76 73
Handlers, Equip., Cleaners 84 83
Agriculture, Forestry 85 86
Source: Statistical Abstract 1990

Table 2-10
Women's Income Compared to the National in 1988 Dollars
Year Per Capita Income Women's Median Income Percent
1988 13,123 8,884 68
1987 12,904 8,638 67
1986 12,596 8,214 65
1985 12,108 7,935 66
1984 11,759 7,820 67
1983 11,341 7,608 67
1982 11,112 7,285 66
1981 11,129 7,166 64
1980 11,193 7,072 63
1979 11,459 6,957 61
1978 11,313 7,130 63
1977 10,829 7,377 68
1976 10,498 7,122 68
1975 10,142 7,125 70
1974 10,132 7,025 69
1973 10,379 7,008 68
1972 10,042 6,925 69
1971 9,379 6,609 70
1970 9,100 6,408 70
Source: Trends in Income 1990
Women have historically earned less money than men. There
are several reasons for this fact. Among them are occupational
segregation, discrimination, impact of marriage, and the desire
for families. As Victor Fuchs states, "most women face a serious
dilemma. If they devote themselves to housework and child care,
they are likely to have to assume a subordinate role in a

hierarchial marriage. Moreover, full-time homemakers run the
risk that divorce will leave them with inadequate skills for
earning a living at a paid job... .If women seek to compete with
men in the labor market, the experience to date suggests that
most will have to make sacrifices with respect to marriage and
children" (Fuchs 1988).
Although many women have changed their family and work
priorities, marriage is still the norm and most women must bal-
ance their roles as wife, mother, student and employee, often
simultaneously. Therefore, women are increasingly seeking flexi-
bility in the structure of their lives, to allow them more oppor-
tunities to maximum their potential, both inside and outside of
family life. One career option offering the kind of flexibility
many women need and that more and more women are choosing is
Entrepreneurship has also became a logical choice for many
educated, skilled women with the desire to work for themselves.
The next chapter provides an analysis of the trends in female

Across the nation the number of woman-owned businesses has
increased six-fold over the past few decades, and gross receipts
of these firms has more than doubled. Most women own small
service and retail businesses which are typically
undercapitalized and hone based.
This chapter includes an analysis of the number of women-
owned businesses, the types of businesses they own and their
income. It also includes a discussion of why so many women have
chosen to start a business and same of the public policy implica-
tions of this growing trend. The format includes a comparison of
national trends with those in Colorado.
The analysis in this chapter shows women-owned businesses
are becoming more diverse. The types of businesses women own is
broadening more rapidly in non-traditional industries such as
construction, manufacturing and professional services. These are
also the industries where women earn relatively higher wages and
where greater employment opportunities exist.
Entrepreneurship offers the kind of flexibility many women
need to balance families and careers. Other women open a busi-

ness because of their education, experience and the perceived
limitations of working for others.
Growth in Women-Owned Firms
Since the 1970's, the number of women-owned businesses have
increased at a rapid clip. According to the 1987 Economic Census
of Women-Owned Businesses released in 1991, the number of women-
owned firms between 1982 and 1987 increased four times faster
than all United States firms. Women-owned firms increased 58%,
from 2.6 million to 4.1 million. See Table 3-1. Meanwhile, the
number of all United States firms increased by 14% to 14 million.
Receipts adjusted for inflation jimped 141%, from $98.3 billion
to $237 billion for women-owned firms; adjusted receipts for all
United States firms rose 76%.
Table 3-1
Growth of Women-owned Businesses
1970 1982 1987
Percent of U.S. <5% 24% 30%
Number (millions) 2.6 4.1
Receipts (billions adjusted for inflation) $98.3 $237.0
Percentage of Total U.S. Receipts 10% 14%
Source: Economic Census of Women-Owned Businesses. 1982, 1987

As shown in Table 3-2, 90% of all women-owned businesses are
sole proprietorships, four percent are partnerships, and nearly
six percent are S corporations. Sole proprietorships are unin-
corporated businesses owned by an individual. Partnerships are
unincorporated businesses owned by two or more persons, each of
wham shares in its ownership. A subchapter S corporation is
defined by the Internal Revenue Service to be a legally incorpo-
rated business having 35 or fewer shareholders. These sharehold-
ers electbecause of tax advantagesto be taxed as individual
shareholders rather than as corporation. In prior years, the
size limit was ten or fewer shareholders, so the non-subchapter S
corporations' share has probably increased. A firm is considered
women-owned if the sole owner or one-half or more of the partners
were women. A corporation fell in this class if half or more of
its shares were owned by women.
Although the census data used in this report is the most
comprehensive federal source of information on women-owned
businesses, it undercounts the number of women-owned firms be-
cause it does not include the generally larger "C" corporations.
In 1990, however, the Federal Reserve Board of Governors, with
SBA Cooperation, conducted a survey, the National Survey of small
Business Finances (NSSBF), to determine the financial needs of
small businesses. According to the NSSBF, there were 184,100 C

corporations owned by women in 1987, with total receipts of
approximately $198.8 billion. C corporations are generally
larger than S corporations and carry different legal, shareholder
and and tax liabilities. The combined census and SEA data reveal
that in 1987 there were approximately 4.3 million women-owned
firms with total receipts of $477 billion.
Table 3-2 Percent of Wamen-Owned Firms and All U.S. Firms by Legal Form of Organization, 1982-1987
Firms By Legal Form % Wamen- Owned Firms 1982 1987 % All U.S. Firms 1982 1987
Individual Proprietorship 92 90 92 89
Partnership 5 4 5 5
Subchapter S Corporation 3 6 3 1
Firms By t prpi tv-ithi % Wamen- Owned Firms 1982 1987 % All U.S. Firms 1982 1987
Receipts By Legal Farm
Individual Proprietorship 50 29 45 30
Partnership 20 11 30 16
Subchapter S Corporation 30 60 25 54
Source: Economic Census of Wamen-Owned Businesses 1982. 1987.
Note: Does not include C corporations.

Industry Charaotgristics and Receipts
Women own 30% of the nation's businesses and generate 14% of
the national receipts. The mismatch of ownership and receipts
has been the focus of much controversy. The low percentage of
receipts most likely reflects the volume of young start-ups as
well as the high concentration of service business. Women own
businesses in all industries, but they are most concentrated in
services, retail trade and finance, insurance and real estate.
Women owned 38% of the nation's service businesses, over one-
third of the nation's retail trade businesses, and more than a
third of the finance, insurance and real estate firms.
Table 3-3 shows the industrial distribution of women-owned
firms compared to all United States firms. Women own a signifi-
cantly higher percentage of businesses in services and retail

Table 3-3
Distribution of Wamen-Owned Firms by Industry, 1987, Compared
to All U.S. Firms
Industry % Wamen-Owned Firms % All U.S.
Agriculture 1.17 2.6
Mining .64 .9
Construction 2.29 12.0
Manufacturing 2.28 3.1
TEU 1.94 4.3
Wholesale Trade 2.08 3.2
FIRE 10.63 9.0
Retail trade 19.41 16.4
Services 55.14 43.0
Note: TEU: Transportation and Public Utilities
FIRE: Finance, Insurance and Real Estate
Source: Economic Census of Wamen-Owned Businesses. 1982
, 1987.
The types of services most commonly provided by women are
those with which women have been traditionally associated and
which have little or no start up costs. Traditional businesses
include personal services such as cleaning and tailoring,
nursing, education and social services, which is primarily child
care. Among the types of services shown in Table 3-4, there was
tremendous growth in the number of firms offering day care. In
1982, there were an estimated 2,330 day care businesses national-
ly owned by women; by 1987, there are 269,000. The enormity of

this figure reflects not only a tremendous increase in demand for
day care, but also the increase in regulation and changes in
reporting of child care facilities.
Table 3-4 also shows a dramatic increase in the number of
women offering business and legal services. In 1982, there were
57,860 firms offering business services, which include advertis-
ing, computer data processing, property management, and others;
by 1987, there were 690,500. legal services nearly doubled,
increasing from 23,000 to 42,000. Other types of services show-
ing substantial growth are motion picture production studios,
recreation and medical services.
Table 3-4
Changes in National Service Firms Owned by Women, 1982-1987
Service 1982 1987 % Change
Ml services L,401,776 2,269,028 62
Educational 82,613 104,187 26
Hotels/Lodging 17,487 22,211 27
Personal Services 394,566 561,695 42
Legal 23,333 41,925 80
Misc. Repair 13,273 24,027 81
Health 123,111 235,318 91
Auto Repair Recreation 10,991 23,481 114
44,951 99,504 121
Museums 5 13 160
Motion Pictures 2,927 7,953 172
Business Services 57,860 690,494 1,093
Social Services (child care) 2,300 269,197 11,454
Source: Economic Census of Woman-Owned Businesses. 1982, 1987

Table 3-5 lists the largest types of women-owned businesses
in terns of receipts. Surprisingly, auto service firms owned by
women outpaced all other types of businesses in terns of re-
ceipts. After adjustment for inflation, auto service receipts
increased 148% between 1982 and 1987, averaging $822,585 per
fim. (Receipts in current 1987 dollars were $965,000.) Growth
of average receipts per fim was also high for wholesalers of
nondurable goods and for food services, which increased 117% and
58%, respectively. Next to auto services, receipts increased
most in wholesale firms offering nondurable goods.
Table 3-5
Ten Largest Services in Receipts of Woman-Owned Firms in U.S.,
1982, 1987
Adjusted to 1982 Dollars
# of Receipts Receipts/
Industry 1982 Firms ($ min inn) Firm
Eating and Drinking Est. 66,811 6,684 100,043
Food Stores 37,635 6,047 160,675
Personal Services 419,113 5,500 13,123
Wholesale Nondurable 22,231 5,297 238,271
Auto Services 14,353 4,754 331,220
Real Estate 225,551 4,733 20,984
Health Services 128,389 3,989 31,070
Wholesale Durable 12,021 3,893 323,850
Special Trade 47,219 2,497 52,881
Apparel 29,130 2,446 83,968

Industry 1982
Wholesale Nondurable
Misc. Retail
Auto Services
Business Services
Wholesale Durable
Food Stores
Eating and Drinking Est.
Real Estate
Personal Services
Health Services
Changes in Selected In-
dustries, 82-87
Eating and Drinking Est.
Food Stores
Personal Services
Wholesale Nondurable
Auto Services
Wholesale Durable
Real Estate
Health Services
Source: Economic Census of
Real Real
# Of Receipts Receipts
Finns ($ million) Firm
39,514 20,450 517,532
546,353 18,049 33,035
20,942 17,227 822,585
690,494 16,129 23,359
42,999 16,011 372,359
48,469 12,290 253,556
90,848 12,067 132,829
335,429 10,767 32,101
561,695 8,764 15,603
235,318 8,193 34,815
% Change Real % Change
Finns Avg. Receipts
36 33
29 58
34 19
78 117
46 148
258 15
49 53
83 12
Businesses 1982, 1987
Although most women own service businesses, the types of
services women are providing are broadening. In addition, Table

3-6 shows that growth in wamen-owned firms is greatest in non-
traditional industrial sectors, such as agriculture, transporta-
tion, manufacturing and wholesale trade. Although growth in
these industries is apparent, the percent increase in the number
of firms is magnified because the figures were very low to begin
The types of businesses women own in agriculture include
veterinary and animal services, landscaping and commercial fish-
ing. Most woman-owned transportation firms are in trucking,
warehousing and freight forwarding. Women are manufacturing a
broad range of products, from apparel to metal and glass prod-
ucts. The largest percentage of firms, though, were in printing
and publishing. Over two-thirds of the wamen-owned wholesale
firms distribute nondurable goods.
Table 3-6 also shows that between 1982 and 1987, income
growth was greatest in wholesale trade, manufacturing, construc-
tion and transportation.

Table 3-6
Percentage Change in Number of Wamen-Owned Firms and Receipts,
% Change Firms % Change Receipts
All Industries 57 141
Retail Trade 27 103
Mining 33 -26
Construction 60 279
FIRE 77 138
Services 77 98
THJ 105 188
Manufacturing 109 397
Agriculture 146 140
Wholesale Trade 167 297
NOTE: FIRE: Finance, Insurance and Real Estate
THJ: Transportation, Public Utilities
Source: Eoonamic Census of Wamen-Owned Businesses 1982, 1987
Employees of Wamen-Owned Firms
Mary of the businesses owned by women are very small firms,
with no employees. In 1987, only 15% of the wamen-owned firms
had employees. By comparison, 25% of all united States firms
have employees. It is the firms with employees that are
generating the highest volume of receipts. Indeed, women-cwned
firms with employees accounted for 86% of the total receipts of
wamen-owned firms.
A survey conducted in 1982 revealed that most women owned
young businesses with no employees, and that the business were

operated from the home by married women who worked part-time and
did not rely on their business as a sole source of income fChar-
acteristics of Wamen-Owned Businesses 1980). These characteris-
tics, however, may not be true for today's businesswoman. Many
are opening offices, expanding their employment base and working
full time, especially in manufacturing, wholesale trade and
Wcmen-owned manufacturing firms, for example, averaged
fourteen employees in 1987, up from nine in 1982; wholesalers
averaged eight, up from six in 1982; and construction firms
employed five in 1987, compared to four in 1982. In 1987, ser-
vice businesses averaged four employees and retailers five.
By contrast, between 1982 and 1987, the average number of employ-
ees in services and retail trade was unchanged.
Geographic Location of Wamen-Owned Businesses and
Wcmen-Owned Businesses in Colorado
Women are starting businesses by the thousands in all indus-
tries across the nation. This section looks at the growth of
wamen-owned businesses by state, with particular focus on Colora-
Table 3-7 shows the number of firms and percentage of wamen-
owned businesses for each state. Hawaii ranked first showing

that 36% of the states are businesses owned by women; In Colorado
and Maryland (ranked second and third), women owned 34% of the
states' businesses.
It appears from the data that western states host many of
the nation's wamen-cwned businesses.
Table 3-7
WanaKXmsd Firms Gtnparad to All U.S. Firms ky State 1987
Wamai Firms Tbtal Firms Wamai lecoat: Vfcnm Baoeipts $ Minima Tbtal U.S. Ifeoedpts Vfcnm Efeiuat
State NLnker NLnber of Tbtal $ Minima of Tbtal
thited States 4,114,787 13,695,480 30.04 278,138 1,994,808 13.94
Alakaia 48,018 178,119 26.96 3,624 26,519 13.67
Alasla 13,976 48,784 28.65 829 4,983 16.64
Arizona 60,567 191,908 31.56 2,911 22,698 12.82
Arkansas 35,469 134,766 26.32 2,008 13,739 14.62
Q=Tlifnmia 559,821 1,809,252 30.94 31,027 246,406 12.59
ChlfiTrio 89,411 262,597 34.05 4,261 27,008 15.77
Qxnscdcut 60,924 196,537 31.00 5,320 35,024 15.19
Eela-are 9,727 30,976 31.04 753 4,662 16.15
M. D.C. 10,907 29,244 37.57 774 6,509 11.89
Florida 221,361 735,810 30.08 16,828 99,289 16.95
Oacmgia 88,050 305,382 28.83 5,874 48,818 12.03
ISwaii 21,696 60,928 35.61 857 6,522 13.14
IMd IB,973 68,006 27.90 813 6,524 12.46
miraia 177,057 573,973 30.85 13,884 99,147 14.00
Thfliara 89,949 294,570 30.54 8,913 50,672 17.59
Ibwa 53,592 174,121 30.78 2,905 18,199 15.96
Knsas 53,505 169,593 31.55 2,661 17,086 15.57
Icuisiana 53,454 193,806 27.58 3,265 22,732 14.36
55,852 204,723 27.28 2,962 21,473 13.79
Lfeine 23,922 88,208 27.12 1,635 10,407 15.71

Warn Dotal Womai Warn Dotal U.S. Warn
liras linrs Kuwait Raceipts RaceiptB RaxHt.
Stabs N liter NLrter cf Dotal $ MiUdcrs $ Minims cf Dotal
*&rylard 81,891 244,071 33.55 5,509 41,392 13.31
mpj ha 111,376 356,780 31.22 11,140 66,364 16.79
Michigan 133,958 426,656 31.40 7,889 63,900 12.35
Muresota 88,137 280,249 31.45 4,991 34,852 14.32
Mississippi 28,976 112,245 25.81 2,062 13,095 15.75
Missouri. 87,658 293,131 29.90 5,349 37,898 14.11
Mntara 17,747 63,623 27.89 930 5,621 16.55
N&aasla 32,285 102,811 31.40 1,649 12,027 13.71
Nkrada 18,831 59,784 31.50 1,414 9,933 14.24
Narlfenp^ire 22,713 79,771 28.47 1,858 13,083 14.20
Nbw Jers^ 117,373 406,792 38.85 13,554 94,111 14.40
Newlfedoo 25,397 82,253 30.88 1,166 7,282 16.01
New Yadc 284,912 930,669 30.61 29,970 208,303 14.39
N. GraLdra 93,532 329,373 28.40 6,813 45,357 15.02
N. EEtota 12,689 42,717 29.70 572 4,128 13.86
Chio 154,084 521,123 29.57 8,872 72,188 12.29
CHabaia 63,690 223,676 28.47 2,948 19,592 15.05
Qtsgn 58,941 185,151 31.83 4,279 22,113 19.35
^T^lvania 167,362 595,653 28.10 13,339 116,167 11.48
Hide Island 14,517 52,780 27.50 1,340 10,048 13.34
S. carddra 42,604 149,190 28.56 2,950 18,552 15.90
S. EBfaota 13,374 47,829 27.96 726 5,135 14.14
Daressea 67,448 251,255 26.84 4,226 33,913 12.46
Tfeas 298,138 1,025,617 29.07 13,385 115,560 11.58
Utah 29,810 100,186 29.75 1,392 9,894 14.07
\fenrmt 13,802 45,243 30.51 766 5,191 14.76
Virginia 94,916 297,541 31.73 5,952 41,261 14.43
Wtiiingtm 90,285 286,224 31.54 4,689 36,872 12.72
W. Virginia 22,549 78,026 28.90 1,114 7,048 15.81
Wisocnsin 69,185 239,185 28.93 4,667 32,068 14.55
pairing 10,796 34,573 31.23 524 3,435 15.25
Source: Wrmnin ftww 1987

Table 3-8 shows the number of women owned businesses as a
percentage of population, by state. Colorado has more women-
owned businesses per 100,000 people than any other state. De-
spite the Colorado recession, the number of woman-owned
businesses increased thirty-eight percent from 1982 to 1987, from
64,675 to 89,411. Their receipts increased one hundred thirty-
three percent, from $1.8 billion to $4.2 billion. In 1987, women
owned over thirty-four percent of the state's businesses and
accounted for sixteen percent of the state's gross business
Table 3-8
Number of Wamen-Owned Firms as Percent of Total Population by
# of Population Firms/ 100,000
State Firms (1,000) Population
Colorado 89,411 3,293 2,715
Alaska 13,976 524 2,667
Vermont 13,802 547 2,523
Wyoming 10,796 490 2,203
Montana 17,747 809 2,194
Oregon 58,941 2,723 2,165
Kansas 53,505 2,475 2,162
New Hampshire 22,713 1,056 2,151
Minnesota 88,137 4,244 2,077
Nebraska 32,285 1,594 2,025
California 559,821 27,653 2,024
Maine 23,933 1,186 2,018
Hawaii 21,696 1,082 2,005
Washington 90,285 4,542 1,988
Oklahoma 63,690 3,259 1,954

# of Population Firms/ 100,000
State Firms (1,000) Population
Connecticut 60,924 3,212 1,897
North Dakota 12,689 671 1,891
South Dakota 13,374 709 1,886
Nevada 18,831 1,006 1,872
Florida 221,361 12,022 1,841
Maryland 81,891 4,536 1,805
Arizona 60,567 3,400 1,781
Texas 298,138 16,781 1,777
Utah 29,810 1,680 1,774
Washington DC 10,987 621 1,769
Missouri 87,658 5,107 1,716
New Mexico 25,397 1,494 1,700
Indiana 89,949 5,530 1,627
New York 284,912 17,835 1,597
Virginia 94,416 5,914 1,596
New Jersey 117,373 7,674 1,529
Illinois 177,057 11,584 1,528
Delaware 9,727 648 1,501
Arkansas 35,469 2,388 1,485
Rhode Island 14,517 986 1,472
Arkansas 35,469 2,388 1,485
Rhode Island 14,517 986 1,472
North Carolina 93,532 6,409 1,459
Michigan 133,958 9,205 1,455
Wisconsin 69,185 4,807 1,439
Kentucky 53,454 3,723 1,436
Ohio 154,084 10,816 1,425
Georgia 88,050 6,227 1,414
Pennsylvania 167,362 11,942 1,401
Tennessee 67,448 4,855 1,389
Louisiana 55,852 4,448 1,256
South Carolina 42,604 3,425 1,244
West Virginia 22,549 1,898 1,118
Alabama 48,018 4,084 1,176
Mississippi 28,976 2,624 1,104
Source: Economic Census of WOmen-Owned Businesses. Current
Population Report, Series P-25

Table 3-9 shows the number of films and receipts for wcmen-
owned firms in Colorado. In 1987, 60% of the women-owned firms
in Colorado were services, up from 50% in 1982. Wholesale
trade, transportation and services firms showed the greatest
growth. The percentage of retail trade businesses owned by women
declined from 26% to 17%. Much like United States averages,
Colorado's average receipts per firm, adjusted for inflation,
were highest for manufacturing, finance and construction firms.
Table 3-9
Wamen-Owned Businesses by Industry for Colorado
Constant Receipts, 1982 Dollars
Percent Average
1987 # of Firms of Total Receipts
All Industries 89,411 100.00 40,589
Agriculture 1,980 2.21 44,731
Construction 2,273 2.54 141,811
Manufacturing 2,313 2.59 87,197
TFU 1,539 1.72 112,514
Wholesale Trade 1,818 2.03 243,639
Retail Trade 15,075 16.86 71,630
FIRE 8,022 8.97 35,785
Selected Services 53,216 59.52 18,257
Not Classified 3,175 3.55 19,507
Percent Average
1982 # of Firms of Total Receipts
All Industries 64,675 100.00 28,287
Agriculture 1,324 2.05 74,099
Construction 1,907 2.95 77,629
Manufacturing 1,806 2.79 36,425
TFU 726 1.12 80,722
Wholesale Trade 529 0.82 222,195

# of Firms
of Total
Retail Trade 16,697 25.82 40,403
FIRE 6,161 9.53 19,574
Selected Services 32,049 49.55 15,385
Not Classified 3,476 5.37 15,272
% Change % Change
Change, 1982-1987 in Firms in Receipts
All Industries 38 53
Agriculture 50 -40
Construction 19 83
Manufacturing 28 139
TEU 112 39
Wholesale Trade 244 10
Retail Trade -10 77
FIRE 30 83
Services 66 19
Not Classified -9 28
TEU: Transportation, Public Utilities
FIRE: Finance, Insurance, Real Estate
Source: Economic Census of Women-Owned Businesses 1982, 1987
In Colorado, there was a decline in the average number of
employees in women-owned agriculture and construction firms;
these were among the industries hardest hit by the recession.
There was also a decline in the average number of employees in
service firms. Other industries showed either stable employment
or slight increases. One may question whether or not recession

and its subsequent layoffs from larger corporations cause more
people to start their own businesses. David Birch, in his book
Job Creation in America, argues that 98% of all new jobs are
created by firms with less than twenty employees, and that more
new small businesses are created as a result of layoffs from
large corporations (Birch 1988).
Based on its population size, Colorado ranks at the top of
the nation for the number of woman-owned firms. The large per-
centage of women-owned businesses may be, in part, attributable
to same of the state's assets. First, Colorado has one of the
most highly educated populations in the nation. Over half of its
degree recipients are women. Second, Colorado has a relatively
high standard of living, which means that people have more income
to spend on goods and services. Third, the state's economy has
stabilized since the 1980's, and major statewide economic devel-
opment efforts have been initiated. Hie increase could also be
due to the economic downswing in Colorado in the 1980's. Hie
massive layoffs from large firms not only produced a net
outmigration from the state, but also may have prompted many
skilled people to start their own business. Fourth, the female
labor force participation rate in Colorado is higher than the
national average. Finally, several public and private organiza-

tions have recently developed education and training programs
designed to assist women in business. The efforts focused on
strategies to assist women business owners have prompted a number
of policy issues which will be discussed in the next section.
Policy Issues
Growth in the number of female entrepreneurs gained a
national interest in the 1970s. In 1977, President Carter
appointed an Interagency Task Force on Women Business Owners and
created a National Women's Business Enterprise Policy. The
policy called for affirmative action by federal agencies to
foster women-owned businesses and create programs to assist
Although the Policy is still in effect, there is little
evidence that it responds to the needs of women's businesses.
Past progress of women in business is attributable to their own
efforts rather than any public policy action. (Clark and James
1992, p. 35). The Women's Business Council, however, is
currently developing a long-run federal policy to aid women
business owners. There are four primary issues being discussed
with respect to women enterprise development.

access to capital
data collection
technical training
The Carter order also included a provision to increase
government prime contracts to women. Through assistance from the
Small Business Administration, prime contracts awarded to women
increased from $181 million in 1979 to $1.9 billion in fiscal
year 1988 (New Economic Realities 1988).
Although there has been an increase in the volume of prime
contracts awarded to women, they still receive about 1% of all
prime contracts awarded by the government. There are several
reasons why women receive so few government contracts.
First is the types of businesses women own. Since most
women own very small businesses with no employees and work in
traditional services and retail trade, they may be neither quali-
fied nor equipped to submit a proposal.
Second, business owners who do own larger businesses and
have products or services for which the government contracts for

find the certification and bidding process riddled with time-
consuming and inefficient complications. A third reason is that
there has been no data base available to government purchasing
departments which list certified woman-owned businesses.
Since 1980/ federal agencies have been required to set
procurement goals and policies for contracting to women. Each
agency, however, is allowed to set its own goals and monitor
them. National policy does state that contractors use their
"best efforts" to increase opportunities to women to participate
in subcontracting.
The certification procedure is currently being revisited by
policy makers, in Colorado, for example, legislation has been
introduced which would, among other things, simplify the certifi-
cation procedures, eliminate on-site inspections, and provide a
complete data base listing to state purchasing departments.
Access to Capital
The second policy issue, access to capital, is one which
plagues most small businesses. Capital access for small business
growth and development is a national concern. There are a number
of reasons for this capital shortfall. The first is the

perception of small businesses by traditional lending
institutions such as banks. Small businesses are perceived as
high risk (60% of all small businesses fail within the first two
years) and as having an unacceptable rate of return. In
addition, small business owners often lack technical and
managerial experience. This lack of experience contributes to
the risk factor. Another reason why banks do not like to grant
loans to small businesses is that these small businesses often
lack tangible collateral; banks are reluctant to grant loans
based on receivables. loan policies of banks and other lending
institutions revolve around big businesses having sufficient
collateral which will produce higher returns to the institution.
With respect to federal action, there is no federal fi-
nancing or guarantee programs which give women preferential
treatment for loans. Despite the lack of funding available to
women entrepreneurs, the number of their businesses have contin-
ued to expand.
Data Collection
The third policy issue important to business women's prog-
ress is data collection. Existing data sources are criticized

for being incomplete and untimely. The Bureau of Census, Econom-
ic Census of Wamen-Owned Business, is tabulated every five years.
Data estimating the number of wamen-awned businesses and their
income is understated because, as discussed earlier, it does not
count C corporations. This is an important amission. Generally,
C corporations are larger firms which generate higher incomes and
employ more people. A 1990 Federal Reserve Board of Governors
report, which identified over 184,000 wcmen-awned C corporations
whose income approached $2 billion indicates how important this
classification of businesses are to the overall count. By
comparison, income generated from combined proprietorship,
partnership and S corporations neared $237 billion in 1987.
Reasons for this amission vary from inadequate funding for
Census Bureau personnel to methodological complications in iden-
tifying women who own a business. The Census Bureau has been the
target of federal budget cuts while simultaneously receiving
greater demand for information. One of the methodological diffi-
culties has been in identifying women-owned businesses from
personal income tax forms. Businesses are generally counted as
male owned because the man's name appears first on tax forms.
Another methodological problem is one of definition. The Census

Bureau defines a women-owned business one in which the female
principle must own one-half of the business and manage the daily
operations of that business. Other federal agencies define a
women-owned business as one in which at least 51% of the business
is owned by a woman.
In addition to being incomplete, the Census data has a
three-year lag between the time the data is collected and when
the reports are available. The 1987 data, for example, became
available in late 1990. The inefficiences in data collection
inhibit the analysis of women's business progress and the ability
to direct policy.
Technical Training
The last policy issue is another problem faced by all small
businesses in generalthe lack of adequate managerial or techni-
cal skills. Realizing the increasing importance of all small
business to the economy, many public and private initiatives have
been implemented to assist small business owners. The assistance
includes improving managerial skills, such as marketing, ac-
counting, bockkeeping, and hiring personnel. Other programs
include advice on government contracting and how to negotiate
with financial institutions.

One of the most important federal actions occurred in 1988
when Congress passed the Women's Business Ownership Act. It is
the only such law devoted to women. The Act authorizes $10
million over a three-year period to finance demonstration proj-
ects which offer training and counseling in finance, management
and marketing. The Act also established the National Women's
Business Council which has the task of conducting hearings across
the nation and developing a long-run federal strategy for aiding
women-owned businesses. The Act also established a small loan
program which is administered by the Small Business Administra-
tion. Although the Small Business Administration does not give
preferences to women, it's loan policy is attractive to small
businesses in services where women are highly represented.
(NWBC 1990).

In 1987, Colorado had the highest rate of business ownership
by women of any other state. One question arising from this
observation is, why? What are the economic and political
characteristics that explain this fact? The purpose of this
chapter is to identify these state-level characteristics. It
presents findings of a cross-sectional regression analysis which
was designed to explain the determinants of female
entrepreneurship and recent growth in the number of women-owned
firms. The dependent variable is the number of women-owned
businesses per 100,000 people. The theoretical discussion below
identifies potential explanatory variables.
Theoretical Discussion
There are a number of facorseconomic, political and
demographicthat contribute to business growth and development.
Each group of variables is used in different combinations within
three models used in this analysis.
A state's economic health is often measured by characteris-
tics such as income and unemployment rates. The economic vari-

ables used in this analysis include personal income per capita
and business failure rates. Higher income should be conducive to
business formation of any type. Whether higher incomes would
spur female entrepreneurship relative to male entrepreneurship is
an open question, but our prior belief was that the relatively
recent phenomenon of female entrepreneurship would be more evi-
dent in regions with a healthy business climate. Higher incomes
might release women from traditional roles as well as encourage
lenders to take more risk. Higher business failure rates are
generally indicative of high risk and might discourage business
formation and reduce lending to non-traditional borrowers. As we
will see, this prior belief is not supported by the data.
In addition to income, this analysis examines unemployment
rates as well as the increase in female entrepreneurship.
Unemployment might spur women to take risks and might induce
unemployed women to start their own small business, but our prior
belief was that business formation and lending would be more
likely in states with low unemployment.
Other demographic variables used in the analysis include
female labor force participation rate (FLEER), percentage of
civilian non-institutional labor force, ratio of non-white
females to the population and the percentage of women over sixty-
five to the papulation.

Our hypothesis is that working age white women are more
likely to became entrepreneurs. The reason age may be important
is obvious but race may not be. Nan-whites are more likely to be
in traditional roles, have lower incomes and are more likely to
be discriminated against by lenders.
lastly, demographic variables such as marriage rates, di-
vorce rates, childbirth rates, and female higher education en-
rollment levels, are included in the discussion. Data in Chapter
2 indicates that these variables have a significant effect on the
FLFPR. The models test whether these variables have any addi-
tional effect on women's entrepreneurship. Since they affect the
rate at which women enter the labor force, they may also affect
business growth.
Models I and II also incorporate two political and two dummy
variables. The first political variable is the ratio of female
legislators in each state. The second is the majority party
affiliation in each state. Our prior belief about the effect of
party is ambiguous: Republican states may be more supportive of
entrepreneurship in general, but Democrats may be more supportive
of women in particular. Female representatives should act as
role models for female entrepreneurs, help encourage acceptance
of female leaders, and help increase awareness of women's

progress (or lack of it) in the workplace and in business. These
representatives nay encourage more women to participate in public
policy decision making and nay be effective in changing the
economic environment. On the other hand, causality nay in
reality go the other way. States with more female entrepreneurs
might elect more women into political office. There nay be more
women-owned businesses in Republican states which tend to have a
relatively higher per capita income and investment potential.
lastly, the model includes two dummy variables. One is a
regional variable and the other is the year. According to the
data reviewed in Chapter 3, there appears to be more women-owned
businesses in the Western states than Eastern. The variable East
will test the significance of this regional observation in a
multi-variant context. The second dumny variable, year, shows
the significance of change in the level of women-owned businesses
for each time period, 1982 and 1987.
Description of the Model
The models used in this analysis follow the general linear
form of a multiple regression:
Yi = Bg + + ^2^2i. + + Ej_ where:
Yi = the dependent variable

Bq = the intercept
Bi/ B2 ... Bfc = the coefficient of the independent
Ej_ = error term
List of variables
Dependent Variable:
Yj_ = Women-Owned Businesses per 100.000 people. (U.S.
Bureau of the Census, Economic Census. Census of Wamen-Owned
Businesses: 1982, 1987).
Independent Variables:
Xi = Personal income per Capita. Dolaars. (Statis-
tical Abstract of the United States; 1984, 1989). Average 1982 =
$12,000; 1987 = $16,000 (adjusted to 1982 dollars).
X2 = Business Failure Rate. Per 100,000 Firms. (Dunn
& Bradstreet; Business Failure Reports 1982, 1987).
X3 = Total state Unemployment Rate. Percent. (Depart-
ment of labor, U.S. Bureau of Labor Statistics, Geographic Pro-
file of Employment and Unemployment, 1988).
X4 = State Female* Tal-or Force Participation Rate.
Percentage of civilian non-institution population 16 years old
and over. (Statistical Abstract, 1990).

Xs = Civilian Non-Institnt-.irmai Female Population Rate.
Percent of population over 16 years old and over. (Current
Population Report, Series P-25).
Xg = Female Non-White Population Rate. Thousands.
(Current Population Report, Series P-25).
X7 = Percentage of Women Over 65 Years Old. Thousands.
(Current Population Report, Series P-25).
Xg = Marriage Rate per 1,000 People. (U.S. National
Center for Health Statistics, vital statistics of the United
States, annual).
Xg = Divorce Rate per i -non Marriages. (U.S. National
Center for Health Statistics, vital statistics of the United
States, annual).
X10 = Birth Rate per 1.000 Women age 15-45. (U.S.
National Center for Health Statistics, Vital Statistics of the
United States, annual).
Xu = Percent of Students in Higher Education that are
Women. Per 1,000 students. (Digest of Education Statistics,
Xj2 = Percent of State Legislator that are Women.
Number of women in legislature over total legislative body.
(Statistical Abstract, 1989).

X13 = Percent of Democrats in State Legislature.
1,000. (Statistical Abstract, 1989).
X14 = East. A dummy variable which takes the value of
one for states east of the Mississippi River and zero for states
west of the Mississippi River.
X15 = Year. A dummy variable which takes the value of
one for the year 1987 and zero for 1982.
Data for the dependent variable, wcmen-owned businesses, was
taken from the Economic Census of Wamen-Owned Business. The data
is reported by the Census Bureau and is reported every five
years, for years ending in two and seven. The first economic
census of women-owned businesses was taken in 1972. However,
data taken in 1972 and 1977 are not comparable to the 1982 and
1987 data because of improvements and methodology used by the
Census Bureau in collecting and processing the data. The methods
also changed between 1982 and 1987, making same of the data
comparatively unreliable. The Bureau of Census, however, ad-
justed portions of the 1982 figures to be more accurately compa-
rable to 1987. The data used for the dependent variable was
adjusted by the Census Bureau to be comparable to 1987 figures.

Data for the independent variables comes from a variety of
sources as shown in the previous section. All data for the
independent variables is recorded for the years 1982 and 1987.
Methodology of the Models
The goal of the econometric study is to test the theories
about characteristics of individual states which have relatively
higher percentages of female entrepreneurs.
Three models are presented in this analysis of state charac-
teristics. In the first two models, the dependent variable is
the number of wcmen-owned businesses per 100,000 people. The
independent variable for Models I and H fall into four catego-
ries: economic, demographic, political and dummies.
Prior to establishing the variables to be used for Models I
and II, initial trial tests (not shown) of all the independent
variables indicated that same variables, specifically percentage
of women in higher education, marriage rates, female population
rate, divorce rate and birth rate, did not add significant ex-
planatory value to Models I and II. This is an interesting
result. Since these variables are assumed to significantly
affect the FLFER, it was expected that they would also affect the
number of women in business. Since these variables were insig-

nificant in the initial set of regressions, they were excluded
from the two models of female entrepreneurship. They are
included, however, in a third model which estimates the
determinants of FLFPR. The results will be explained later.
Model I
There are four tests in the first model. Test 1 presents
the relationships of all the independent variables to the rate of
woman-owned businesses for 1982 and 1987. The second test
includes only the political variables. Test 3 includes all but
the the political variables. By examining the groups of
variables separately, we are able to determine each group's
individual strength.
Test 4 in Model I includes all variables except percentage
of Democrats in legislature. The variable was deleted because
the t-statistics was instable.
Model II
In Model II, the dependent variable is the rate of warnen-
owned businesses in 1987. The independent variables include the
rate of woman-owned businesses in 1982 to account for past histo-
ry. The number of observations drops to 49.

There are three tests included in Model II. As in Model 1,
each test shows different combinations of the independent vari-
ables. Both regression Models I and II are tested and corrected
for heteroscadasticity and multicollinearity.
Model III
Model III identifies variables which contribute to the
state-level FLFER. The independent variables in this regression,
while insignificant in the first two models, are important in
predicting state-level FLFER.
Model I
The first model, shown in Table 4-1, shows same very
interesting results. It shows that state per capita income is
not a significant determinant of women-owned businesses for any
of the specifications tested. This is an unexpected result, but
Model II will demonstrate that the change in female entrepreneur-
ship between 1982 and 1987 is positively associated with income,
confirming our prior belief. The significance in Model II also
may reflect women's need for expansion capital for existing,
growing businesses.

Table 4-1, Model I
Determinants of Wamen-Owned Businesses; 1982, 1987
Dependent Variable: Number of Wamen-Owned Businesses per
100,000 people
Variable Test 1 Test 2 Test 3 Test 4
Intercept 44.09 13.73 43.81 39.24
t-stat 2.62 14.47 2.48 2.39
Income per capita .0002 .0003 .0001
t-stat 1.1480 1.6290 1.1130
Business failures .0134 .0171 .0138
t-stat 3.3100* 4.1020* 3.4200*
Unemployment -.4600 -.5411 -.4937
t-stat -3.4540* -3.9740* -3.7740*
FLFPR .0496 .1228 .0592
t-stat .7907 1.7510 .9532
Female population -.5449 -.5992 -.4576
t-stat -1.6760 -1.7680 -1.4490
Female non-white -.0907 -.1611 -.1169
t-stat -1.7600 -2.3040* -2.4060*
Age >65 -64.2400 -59.6700 -65.1500
t-stat -2.7990* -2.3600* -2.7490*
Women in legis. .1411 .2510 .1396
t-stat 4.0020* 6.8890* 3.9410*
Dems in legis. -.0068 -.0404
t-stat -.5742 -3.7140*
Year 2.3960 3.5030 2.2900 2.2800
t-stat 5.1120* 7.0600* 4.5700* 4.9570*
East -1.2780 -2.4250 -1.1360 -1.4480
t-stat -2.5020* -4.9120* -2.1350* -2.9140*
R2 .8002 .6587 .7685 .7997
Wald test F-stat 3.5190* 11.3000
DP=87, Prob. of
R2=0 .0185 .0000
* Indicates significance at 5% level.
Note 1: 98 observations (heteroscadasticity consistent co-
variance matrix)

Although the income per capita variable proved to be insig-
nificant, the analysis shows that there is a strong association
between the rate of wamen-owned businesses and business failure
rates. The positive association was surprising but several
hypotheses suggest themselves after the fact. The association
may be commensurate with the large volume of new starts experi-
enced during the 1980's. It also may be associated with female
entrepreneurship because wamen-owned businesses are concentrated
in services and retail trade, two industries which report the
highest incidence of failures nationwide. It should not be
assumed from this analysis, however, that there are comparatively
more business failures for wamen-owned firms than for all United
States firms. The business failure rate computed by Dun &
Bradstreet is not gender specific. Further, high failure rates
may be associated with high entry rates and easy entry would
promote female entrepreneurship.
As expected, Model I reflects an inverse relationship be-
tween wamen-owned businesses and unemployment rates across speci-
fications, indicating that the rate of wamen-owned businesses is
higher in states with low unemployment rates.
Another interesting point about the first model, although it
is reasonable to expect that the FLFER would contribute to the
level of entrepreneurship, the t-values for the female labor

force rate and the female population rate were insignificant
across specifications. Other demographic variables such as the
percentage of non-white females in the population and the age
constraint were more consistent with expectations. The t-statis-
tic for percentage of non-white women, although insignificant in
test one, has a negative sign and is significant when the
political variables are dropped. The age variable indicates
clearly that women of retirement age are not likely to own a
The unexpected results of the FLEER and the female popula-
tion rate and the non-white population rate raises same questions
about the possible correlation among the independent variables.
As a result, these three variables were tested for
multicollinearity. Multicollinearity exists when two or more
independent variables are linearly related. If the variables are
linearly related, the results can be unreliable or can show a
reverse sign.
An examination of the correlation matrix of all the vari-
ables shows that the FLEER, female population rate and female
non-white population ratio are associated with each other. This
indicates the possibility that while the independent contribu-
tions of each are insignificant, the three together are jointly

Therefore, the Wald test is applied to determine the
strength of their relationship as a group. Application of the
Wald test indicates that they do indeed became significant.
In addition to testing for multicollinearity, the model is
tested for heteroscadasticity, which often appears in cross-sec-
tional analysis. Heteroscadasticity exists when the variances in
the error terms are correlated with same other variable.
An initial regression was run with number of wamen-owned
businesses as the dependent variable. It was found that large
statistics produced large absolute errors. Dividing the depen-
dent variable by population removed this flaw. No other sources
of heteroscadasticity were found. As a precaution, the White
procedure to correct for heteroscadastic disturbances was used.
Use of the procedure had no material effect on the results.
The dummy variables in Model I provide a look at the rela-
tionships of the independent variables holding year and region
constant. The high t-statistic for the year indicates a signifi-
cant difference in the level of wamen-owned businesses from 1982
to 1987. The regional variable confirms that a higher percentage
of those firms are located in the west. The regional variable
raises same interesting questions. Further research could be
conducted to identify the nature of the association. Perhaps

western states are experiencing same economic change making
business ownership more attractive.
Same of the most surprising relationships were discovered
while testing the political variables. For example, the t-sta-
tistic for the percentage of women in the legislature is very
high. Likewise, the t-statistic for the percentage of Democrats
in the legislature is high and negative in Test 2 but insignifi-
cant when the demographic variables are included. This fragility
suggests spurious correlation: rich, white states with low unem-
ployment are more conducive to female entrepreneurship but less
likely to be democratic.
The link between business and politics is certainly plausi-
ble, although the precise nature of the link may be disputed. It
is possible that one party's policies are more conducive to the
concerns of women-owned businesses. For example, it is possible
that one party's policies are more conducive to business forma-
tion. Republicans are generally perceived as pro-business and
for providing greater support for economic growth and business
development. They are also perceived as having higher incomes
which is required for business investment and expansion. It is
also possible that successful women in visible political posi-
tions serve to increase awareness of the business and personal

progress same women have made. Politically involved women also
may serve as role models encouraging more women to run for office
or became politically active.
The persistence of the political factors is tested further
by applying them with the other variables in different combina-
tions. Test two in Table 4-1 shows the strength of the political
variables while excluding the demographic and economic factors.
Test three reinstates the demographic and economic variables
without the political influence. The t-values of the variables
in test two and three do not change dramatically except for
female non-white population. The sign is still negative but in
Test 3 the t-value is now significant. Possibly this result
reflects the low percentage of women-owned businesses run by
Test four presents the results when only one political
variable, percentage of women in legislature, Is used. This
estimation is presented due to the possibility of spurious
correlation between Democrats and the demographic variables.
In summary, states are more likely to have high numbers of
women-owned businesses if the unemployment rate is low, if they
are located in the west, and have a high rate of business fail-
ures. These states are most likely Republican and have a rela-
tively higher percentage of women in the legislature.

Model II
There are same very important differences between the re-
sults of the first and second models. Model II, shown in Table
4-2, produces the significance of the change in the level in the
of women-owned businesses from 1982 to 1987.
There are three tests in this model. As in Model I, the
first test utilizes all of the independent variables. Test 2
analyzes the strength of the political factors alone while test
three examines the association of the economic and demographic
variable barring the political influence.

Table 4-2, Model II
Determinants of Women-Owned Business, 1987
Dependent Variable: Number of Women-Owned Businesses per
100,000 people
NUniber of observations = 49
Variable Test 1 Test 2 Test 3
Intercept .8204 2.7510 -4.2680
t-stat .0831 2.3360 -3.8170
Income per capita .0002 .0002
t-stat 2.1490* 2.8700*
Business failures .0008 -.0019
t-stat .2058 -.4954
Unemployment -.2139 -.1218
t-stat -2.2530* -1.1250
FLFER .0130 .0440
t-stat .2976 ' 1.0740
Female population .0015 .0639
t-stat .0085 .3177
Female non-white -.1231 -.1020
t-stat -5.5150* -- -3.8420*
Age >65 1.7990 7.6870
t-stat .2661 .7758
Women in legislature .0542 .1120
t-stat 2.5070* 5.2370*
Democrats in legislature .0201 -.0063
t-stat 2.6410* -.6627
East .7256 1.2130 .9640
t-stat 2.0070* 3.4020* 2.5180*
WOBR (1982) 1.0280 .9869 1.0740
t-stat 14.0400* 14.8400* 16.8100*
R2 .9743 .9419 .9679
Wald Test F-stat 11.3000* 6.6030
DF=39 PTOb. of R2 0 .0000 .0010 indicates significance at 5% level Note 1: 49 observations (heteroscadasticity consistent co-
variance matrix)

One of the major differences between models one and two lie
in the economic variables. For example, income per capita was
weak in model one but model two suggests that its importance
changed significantly in the five-year time period. The results
suggest that the change in the number of women-owned businesses
is positively related to state per capita income. It is possible
that income becomes significant as the types of businesses owned
by women broadens or as they mature. Higher income levels also
is important to ease of entry into business. The change in
the rate of women-owned businesses is now insignificantly related
to business failure rates.
The unemployment rate in Test 1 of Model II is consistent
with the results obtained in Model I. But the t-value for unem-
ployment loses its strength when the political variables are
eliminated in Test 3.
As noted in the discussion of model one, the female labor
force and female population rate are insignificant. The female
non-white population rate, however, remains negative and highly
significant. The higher negative value reflected in the
coefficient of the non-^white female population may suggest that
between these two time periods, 1982 and 1987, the percentage of
minority women-owned firms declined.

As in Model I, the Wald test was applied to the three vari-
ables, female population rate, FLEER and female non-white
population. Here, the results of the Wald test indicate that the
combination of these variables is significantly associated with
increases in women-owned businesses.
The age variable in Model 1 is insignificant, apparently
indicating that while there were fewer female headed businesses
in states with an older population, age is a less important
variable in more recent data.
One of the roost important results of the analysis is that
the percent of female legislators significantly contribute to
female entrepreneurship, and that this relationship persists over
a variety of functional forms. Model I showed a surprising
positive effect on the level of women-owned businesses and Model
II demonstrates the association applies to change as well. It is
difficult to avoid the conclusion that the percent of women in
the legislature promote women-owned businesses, while it remains
possible that women advance in both political and economic arenas
simultaneously, our results clearly show that to the extent
female labor force participation captures this rising, tide that
women in the legislature provide an additional impetus to women-
owned businesses.

The effect of Democrats In the state legislature has proven
to be highly sensitive to specification. The most conservative
approach would be to simply discount all results associated with
this variable. However, the sensitivity to structure is inter-
pretable. The level of women-owned businesses was negatively
associated with Democratic legislatures only if the socio-
economic variables were excluded. This suggests Democrats served
as a proxy for these socio-economic variables and were not
causally associated with entrepreneurship. However, Model II
shows that if socio-economic variables are included, then
democrats contribute to the growth of women-owned businesses.
All these facts are explained by the hypothesis that Democrats
are strongest in poorer states with more minorities and higher
unemployment, all of which reduce female entrepreneurship. But,
holding these factors constant, Democrats are more inclined to
promote women-owned businesses.
Another example of the differences between Model I and Model
II is observed in the regional variable. Table 4-1 shows that
there are more women-owned firms in the western states than

eastern. While this may be true, the results in Table 4-2 sug-
gest that growth in the number of women-owned firms was greatest
in the east.
In general, Table 4-2 suggests that the dramatic change in
the rate of women-owned businesses is fostered by income growth,
sustained low unemployment, and continued growth in the number of
women in political positions. It also appears that growth in the
number of women-owned firms is in democratic and eastern states.
Model III
The most unexpected results from the first two models is the
lack of association between the rate of women-owned businesses
and the rate of women in the labor force. However, the Wald
tests indicate that this failure may be due to the close
association of labor force participation and other socio-economic
variables. For this reason, we believe an analysis of the
determinants of participation rates retains interest as a
possible contributor for female entrepreneurship and as a topic
that is important in its own right. The last model presents an
analysis of the relationship between the FLFER and selected
social and income variables.

Table 4-3, Model III
Determinants of the Female Labor Force Participation Pate
Dependent Variable: Female labor Force Participation Pate
Variable Value
intercept t-stat 140.4 9.393*
Income per capita t-stat .0013 8.201*
Women higher educ. t-stat 2.893 4.292*
Marriage rate t-stat .0835 4.472*
Divorce rate t-stat -.1517 -.7268
Female population t-stat -228.6 -7.217*
Birth rate t-stat -1.263 -5.538*
R2 .6072
Walt Test F-stat 1.75
* indicates significance at 5% level
As shown in Model III, there are same very strong associa-
tions between the FLFER and the social variables. For example,
the results show a strong relationship between the FLFER and the
percentage of women in higher education, indicating that as more
women advance their educational goals,they are more likely to
work and to stay in the labor force.

Secondly, strong positive association between the FLEER and
the marriage rate confirms the fact that marriage alone no longer
eliminates women from the labor market. The value of the associ-
ation between the FLEER and the divorce rate is not what was
expected. The high divorce rate is assumed to be a major factor
driving women into the work forcemany unexpectedly and unpre-
pared. However, the co-efficient shown in this model is
insignificant. The surprisingly low association raises questions
about the method of the testing procedure. Perhaps if the
variables were tested in a time series model, the results would
be stronger. Either way, further analysis would be useful.
Another variable showing a strong relationship to the FLEER
is the birth rate. As would be expected, the birth rate is
inversely related to the FLEER. This strong negative association
probably reflects the time women take off after having their
Fourth, the model shows that the percentage of women in the
population shows a negative relationship with the FLEER. This
may reflect lower returns to women when women are relatively
The last variable included in this model is per capita
income, which has the strongest relationship with the FLEER. The
possibility exists that the FLEER, per capita income and the

female population rate are closely correlated. Application of
the Wald test, however, shows that there is no significant
correlation among these variables.
Except for the divorce rate, the results of this model
confirm the assumptions made in Chapter 2 about the determinants
of the female labor force.

The most fundamental changes of the past decades are the
rapid flow of women into the labor force and the steep decline in
fertility. Other changes in marriage, divorce and education are
derived from greater choices about work and fertility. Of the
many factors offered as explanation to the changing work force,
no one stands out as the major cause. It is the combination of
the factors occurring simultaneously that set forth change among
the inter-related behaviors.
Changes in the proportion of working women are fundamental
to the changing relationship between family and the economy. No
longer limited by marriage and family responsibilities, women are
more likely to remain single, postpone marriage, limit family
size, and change the period in time in which they have a family.
Women are now free to increase their time commitment to education
and work.
Although women have changed their priorities regarding
family and work, most women must still balance their roles as
wife, mother, student and employee, often simultaneously. Since

most women who work have children, there will became an increased
demand to commercialize child care and household responsibili-
Education and social trends are associated with changes in
female labor force participation and how women choose to struc-
ture their lives. Bather than accommodating other families,
women are choosing a life structure which allows them to maximize
their potential both inside and outside of family life. Women
must be educated to accommodate the labor market demands of the
future. Continuing support for women and their educational
achievements will promote women's employment, occupational diver-
sification and, hopefully, a higher standard of living.
A common response by women seeking ways to balance families
and careers is entrepreneurship. Entrepreneurship has also
became a logical choice for many educated, skilled women with the
desire to work for themselves. As women's educational and occu-
pational experience broadens, so will the types of industries
women open their businesses in. Women's educational, occupation-
al and entrepreneurial accomplishments are likely to continue.
As women's businesses that survive grow, they will continue to
play an increasingly important economic role across the nation.

Women's businesses, especially those in non-traditianal indus-
tries vail continue to generate more income and employ more
people. The United States now has a large number of women busi-
ness owners with business experience in all major sectors of the
United States economy. Their continued success can be acceler-
ated and confirmed through proper economic and political action.
Although the econometric analysis in this thesis were exper-
imental, the results confirm the complex nature of the interrela-
tionship among the variables. As shown in Model III, variables
which directly contribute to the increase in the FLFER may not
directly affect the level of female entrepreneurship. For exam-
ple, the percent of women in higher education, marriage and birth
rates all contribute to the rising proportion of working women
but did not contribute to their participation in business. The
FLFER, however, is among the variables which are associated with
the level of women business owners. But the strength of the
association is difficult to determine because it was found to be
significant when used in combination with other demographic
Models I and II confirm that a state's economic health is
important to the development of women's businesses. Per capita

income, for example, proved to be an important determinant of the
growth in women-owned businesses between 1982 and 1987. The
state's unenplcyment levels also were important to business
growth. The association between women-owned businesses and the
social variables appeared weak when the variables are examined
individually, but their impact in concert was significant. By
using the Wald test, we determined that the FLFER, the non-^white
population rate and the female papulation rate were important to
women's business formation, but it is not possible to determine
the weight of the variables individually in this model.
The relationship between female entrepreneurship and the
political variables is the most interesting result. The percent
of women in the legislature showed a strong positive association
to the rate of women-owned businesses in both Models I and II.
This result raises several questions. Does the percent of women
in business increase as a result of women's political activities?
Or do business women became more politically involved?
The results of this analysis suggest that public policy
efforts could be directed at encouraging women to participate
more actively in politics. Questions about the interrelationship
between the variables could be studied further in a multiple

equation model. For example, we could conduct further research
utilizing the percent of state legislators that are women, and
the FLEER as dependent variables in a multiple equation regres-
sion model. These regressions could be conducted using same of
the other variables presented in Chapter 4. By testing the
variables in a multiple equation model, we could identify more
accurately the individual strength of the economic, demographic
and political variables on the actions women are pursuing in the
labor force, business and political participation.

Bergmann, Barbara R. 1986. The Economic Emergence of Women.
New York: Basic Books, Inc.
Birch, David L. 1987. Job Creation in America: How are Small
Companies Putting the Most People to Work? New York: Free
MacMillian, Inc.
OCHE, Scorecard. January 1991. See Colorado Commission on
Higher Education.
OCHE, SURDS File. February 1991. See Colorado Commission on
Higher Education.
Characteristics of Woman-Owned Businesses. 1980. See U.S.
Bureau of the Census 1980.
Clark, Thomas A., and James, Franklin J. February 1992. Women-
Owned Businesses: Dimensions and Policy Issues. Economic
Development nna-rteriy. Sage Publications, Vol. 6, No. 1.
Colorado Commission on Higher Education. January 1991. Score-
card on Public Higher Education: How the Public Higher
Education System and Its Students Perform on Selected
Colorado Commission on Higher Education. February 12, 1991.
SURDS Degree Granted File, FY 1989-1990. Degrees Granted
bv Sex and Ethnicity.
Colorado Department of Health, Health Statistics and Vital
Records Division. September 1990. Colorado Vital
Statistics 1989.
Colorado Department of labor and Employment. October 1988.
Workforce 2000 and Beyond: Jobs and Workers for Colorado's
Future. Denver, Colorado.
Colorado Vital Statistics. 1989. See Colorado Department of

Current Papulation Reports, Series P-25. See U.S. Department of
Digest for Education Statistics, annual. See U.S. National
Center for Education Statistics.
Dun and Bradstreet Corporation. 1982-1990. Business Failure
Economic Census of Wamen-Owned Businesses. 1982, 1987. See
U.S. Department of Commerce.
Fuchs, Victor E., 1988. Women's Quest for Equality. Cambridge
Mass.: Harvard Press.
McLaughlin, Steven D., Barbara D. Melber, John O. G. Billy,
Denise M. Zimmerle, Linda D. Wings, and Terry R. Johnson.
1988. The Changing Lives of American Women. University
of N. Carolina: The Hearst Corporation.
National Women's Business Council. Dec. 1990. second Annual
Report to the President and Congress.
New Economic Realities. 1988. See U.S. House of Representa-
NWBC, 1990. See National Women's Business Council.
Statistical Abstract. 1990. See U.S. Bureau of the Census.
Trends in Income. 1990. See U.S. Department of Commerce.
U.S. Bureau of the Census. 1980. 1977 Economic Census: Selected
Characteristics of Women-Owned Businesses. WB77-2
(Washington, DC: U.S. Government Printing Office).
U.S. Bureau of the Census. 1984, 1989, 1990. Statistical
Abstract of the United States: 1990 (Eds. 104, 109, 100).
Washington, DC.
U.S. Bureau of Tabor Statistics. 1988. Geographic Profile of
Employment and Unemployment. Washington, D.C.: Government
Printing Office.

U.S. Department of Commerce, Bureau of the Census. 1982, 1987.
Economic Census of Wamen-Owned Businesses. (Washington
D.C.: U.S. Government Printing Office).
U.S. Department of Commerce, Bureau of the Census. 1982-1988.
Current Population Reports. Population Estimates and
Projections, State Population and Household Estimates by
Age, Sex and Components of Change. 1982-1988. Series
P-25, No. 1044.
U.S. Department of Commerce, Bureau of the Census. 1990. Cur-
rent Population Reports, Series P-60, No. 167, Trends in
Income, bv Selected Characteristics: 1947 to 1988. U.S.
Government Printing Office, Washington, D.C.
U.S. Department of Education. Office of Educational Improve-
ment. Digest of Educatjnnai statistics. NCES 91-660.
U.S. House of Representatives, 1988. New Economic Realities:
The Rise of Women Entrepreneurs: A Report of the Committee
nn small. Business. (Washington, D.C.: U.S. Government
Printing Office)
U.S. National Center for Education Statistics, annual. Digest
of Education Statistics.
U.S. Office of Labor and Employment. 1987. Employment and
Women's Foundation of Colorado, 1987. An Assessment of the
Needs of Colorado's Women and Girls. Denver, Colo.
Workforce 2000. 1988. See Colorado Department of labor and

Adelman Clifford. The NSL/72 Women a-h Thirty Something;
Paradoxes of Attainment. Office of Research U.S. Department of
Education 12-90 Working Paper OR 90-521.
Blackwood, Christopher. June 8, 1987. Colorado Economic
Analysis. Boettcher & Company.
Colorado Commission on Higher Education. FY 1988 89 May
1990. Status of Minorities in Colorado Public Higher Education.
Colorado Commission on Higher Education. 1984, 1985, 1986,
1987. Status of Minorities in Colorado Public Higher Education.
Colorado Department of Highways. 1988. Disadvantaged
Business Enterprise: A Program of the Colorado Department of
Colorado Department of labor and Employment. March 1990.
Boulder-Denver Region. Occupational Employment Outlook 1990-
Colorado Department of Labor and Employment. 1990. labor
Market Information. Colorado tahnr Market Directory.
Colorado Division of labor and Employment. 1990. labor
Market Information. Colorado Tai->nr Force Review.
Colorado Office of Regulatory Reform. 1990. Greater Denver
Chamber of Commerce. The Colorado Business Start-Up Kit.
Colorado Office of State Planning and Budgeting. 1990. The
1995 Report; The Draft Five Year Economic Development Strategic
The Corporation for Enterprise Development. October 1988.
The Entrepreneurial Economy Review, Vol. 7, No. 3.