Content uploaded by Michelle Budig
Author content
All content in this area was uploaded by Michelle Budig on Jan 15, 2014
Content may be subject to copyright.
137
8. Exploring the heterogeneity of
women’s entrepreneurship:
the impact of family structure and
family policies in Europe and the
US
Vartuhí Tonoyan, Michelle Budig and
Robert Strohmeyer
INTRODUCTION
Recent work suggests that both self- employed women and men represent
a heterogeneous group with regard to their occupational pro le. For
instance, in the US and Southern Europe, the majority of self- employed
women are unskilled workers in the service sector (Arum, 1997; Arum
and Müller, 2004; Budig, 2006; McManus, 2001). In contrast, the share of
professional (female and male) and highly compensated self- employed (for
example, lawyers, doctors, architects and engineers) is the highest in con-
servative and highly regulated corporatist welfare states in Western Europe
(such as Germany, Austria and the Netherlands) (Arum and Müller, 2004;
Lohmann, 2004). However, prior work did not systematically consider the
heterogeneity of women’s self- employment when examining the e ects
of individual determinants on women’s entry into entrepreneurship.
Rather, empirical studies usually pool all types of female self- employment
together, without taking into account whether self- employed work is being
performed in professional or non- professional/low- skilled occupations.
Against this background, our study makes three new contributions.
First, we explore whether determinants predicting women’s entrepreneur-
ship vary by the type of occupation entered. We disaggregate entrepre-
neurship into professional (high- skilled) and non- professional (low- skilled
and unskilled) occupations. Our primary focus is to understand whether
family structure (motherhood and partner’s occupational status; see Budig,
2006) has a di erential impact on the quali cation pro le of self- employed
M2401 - BRUSH TEXT.indd 137M2401 - BRUSH TEXT.indd 137 2/8/10 16:12:082/8/10 16:12:08
138 Women entrepreneurs and the global environment for growth
women. We analyse the pathways into di erent forms of entrepreneurship
across 23 countries to understand how country- speci c institutional envi-
ronments may in uence women’s entry into high- skill entrepreneurship
versus low- skill entrepreneurship. Speci cally, we seek to examine how
family policies (such as publicly funded childcare, maternity leave length)
might a ect self- employed women’s occupational pro le.
Next, we elaborate on the theoretical arguments and develop hypoth-
eses. We then provide a description of the sources of data for Europe and
the US, as well as variables, methods, and ndings. Finally, we conclude
with a discussion of our results and limitations and o er implications for
future research and public policy.
THEORETICAL BACKGROUND AND HYPOTHESES
In the US, non- professional wage work (compared to professional occu-
pations) is more likely to lack employer- sponsored childcare, to pay
wages too low to fund private daycare, and to entail closely supervised
and in exible schedules. Recent research demonstrates that workers with
lower wages and occupational status are less likely to have access to family-
friendly bene ts (Budig, 2006). In addition, women in non- professional
jobs are less likely to be in control of the pace and timing of their work,
factors which are positively associated with reduced work and family con-
ict (Thomas and Ganster, 1995). Considering this, self- employment may
o er female non- professional wage workers more autonomy and greater
exibility in setting the time and place of work. Past research partially sub-
stantiates this argument (see, for example, Presser, 1995). Given this, we
expect that motherhood and thus women’s responsibilities for childcare
should predict engagement in non- professional self- employment more
strongly than in professional self- employment.
However, women’s self- employment is often low- earning and does not
provide paid leave for caregiving (Carr, 1996). Self- employed women
may thus require a nancial safety cushion that might be provided by
their husband’s earnings. If this is the case, being married or cohabitating
should positively a ect women’s likelihood of self- employment, and this
e ect should be stronger for women in lower- paying non- professional
self- employment activities. Second, marriage/cohabitation might posi-
tively a ect women’s self- employment if women are more likely to join
their partner’s business. Recent research nds that having a self- employed
husband dramatically increases the likelihood of becoming self- employed
for a woman (see, for example, Arum and Müller, 2004; Budig, 2006;
Firkin et al., 2003; Greene, 2000; Taniguchi, 2002). It seems more likely
M2401 - BRUSH TEXT.indd 138M2401 - BRUSH TEXT.indd 138 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 139
that a woman is likely to join the husband’s self- employment activity to
support her husband’s career rather than to advance her own. We thus
expect that having a self- employed spouse will a ect the likelihood of
non- professional self- employment the strongest. Yet, not all women may
become self- employed to balance work and family demands. Single and
childless women may not experience work and family con ict. Still others
may become self- employed to earn a higher amount of money, advance
their careers and/or socio- economic class positions (Budig, 2006). Since
professional occupations are more likely to have good job characteris-
tics, workers transitioning between wage and self- employed professional
work may rather be ‘pulled’ by the attractiveness of self- employment
than ‘pushed’ by the unattractiveness of wage employment. This second
group of workers should be more likely to be professionals than non-
professionals prior to entering professional self- employment. They should
be more likely to enter self- employment from wage employment, rather
than after a labor force absence (Budig, 2006). From these arguments, we
derive our rst two hypotheses:
Hypothesis 1a: Motherhood positively a ects women’s entry into non-
professional self- employment (but not professional self- employment) as a
solution to reconcile child- care and work.
Hypothesis 1b: The partner’s occupational status (for example, self-
employment) is a predictor for women’s entry into non- professional self-
employment (but not women’s professional self- employment).
INSTITUTIONAL ENVIRONMENT OF WOMEN’S
ENTREPRENEURSHIP
Little is known about the factors that are responsible for international
variations in women’s self- employment rates and their occupational
pro le. Past research on women’s self- employment concentrates on di er-
ences in psychological characteristics (for example, risk preferences, fair of
failure) (Koellinger et al., 2007; Langowitz and Minniti, 2007; Minniti and
Nardone, 2007; Wagner, 2007) and resources (for example, human capital,
nancial capital and social capital) (Verheul and Thurik, 2000). However,
variation in self- employment rates across countries indicates that country-
level factors, such as institutional environments and policies, may matter.
While some recent work considers the institutional embeddedness of
entrepreneurship and self- employment (see, for example, Bjornskov and
Foss, 2008; Carree et al., 2002; Nyström, 2008; Wennekers et al., 2002),
M2401 - BRUSH TEXT.indd 139M2401 - BRUSH TEXT.indd 139 2/8/10 16:12:082/8/10 16:12:08
140 Women entrepreneurs and the global environment for growth
most prior work did not analyse the association between the country-
speci c institutions and women’s entrepreneurship. Given the evidence
that American women turn to self- employment to balance family and
work responsibilities (Brush, 1992; Budig, 2006), cross- national vari-
ation in women’s self- employment rates and occupational pro le may
be linked to welfare state policies that attempt to reconcile unpaid- care
responsibilities and paid work.
We propose that policies that support women’s employment and child-
care are likely to in uence their self- employment participation, particu-
larly in non- professional occupations. Subsequently, the e ect of children
on women’s entry into non- professional self- employment will di er among
countries with di erent institutions regulating women’s reconciliation of
family and work duties. Welfare states di er in the supply of publicly sup-
ported measures (for example, childcare coverage, maternity entitlement,
voluntary family leaves, exible arrangements of working time and social
tax policies) that in uence women’s participation in the labor market
(Gornick et al., 1997; Korpi, 2000; OECD, 2001: 129–66; Sainsbury,
1994b). For example, liberal and laissez- faire welfare states such as the
US or the UK maintain policies oriented to individual solutions. Public
involvement in family policies is weak. Childcare facilities are provided on
a private basis and no public support exists for the provision of childcare
facilities and/or other family support orientated services, compared to
other types of welfare regimes (Gusta son and Wetzels, 1997). Moreover,
parental leave is short, with very low income compensation (Esping-
Andersen, 1990; Gusta son and Wetzels, 1997). This constellation of poli-
cies di ers markedly from conservative welfare states such as Germany,
the Netherlands and Austria, which largely support a traditional male
breadwinner model. In such regimes, priority is given to a general family
support that reduces the family costs of caregiving, for example, cash
child allowances, family tax bene ts for minor children, and tax bene ts
for mothers of young children staying at home. Public day- care services
exist predominantly for children above 3 years old, but not minor children
(under 3 years old). Full- time childcare is rather limited in such countries.
For example, in Germany only 9 per cent of children of kindergarten age
attend all- day- institutions (Gustafsson and Wetzels, 1997; Lohmann and
Luber, 2004; OECD, 2001). As a rule, such countries are characterized
by a traditional division of labor, with men being responsible for earning
incomes and women being responsible for childcare. Women’s (especially
young mothers’) participation in the labor market is rather restricted,
resulting in higher part- time rates of female employment.
Despite similarities among several countries in Continental Europe
which follow the so- called ‘conservative’/‘corporatist’ welfare state model,
M2401 - BRUSH TEXT.indd 140M2401 - BRUSH TEXT.indd 140 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 141
there are clear di erences with regard to speci c policy measures which
in uence women’s compatibility of family and paid work. For instance,
although Belgium, France and Italy are characterized as conservative
welfare states, their policies actively support women’s employment
(Lohmann, 2004). Further, France is the best performer on policies for
mothers with children under 6 and policies for mothers with children
aged 3 to school- age among 14 investigated welfare states. This re ects
the strong French commitment to nearly universal enrollment of children
in écoles maternelles beginning at age 3. Moreover, France scores very
well with regard to the policy support for infants (Gornick et al., 1997,
Lohmann and Luber, 2004). Similarly, Belgium has highly developed
packaging policies that support employment for mothers with children
under age 6. This can be exempli ed by the fact that 95 per cent of children
in Belgium and France are in public day- care or preschools, compared
to only 14 and 38 per cent of children in publicly funded childcare, for
example, in the US and the UK, respectively, that is, countries which rank
at the lowest end of the spectrum. By the same token, Italy scores well with
regard to policies supporting employment for mothers with children under
6, and it provides generous maternity policies and widespread preschool
coverage for children aged over 3, although it has only moderate support
for mothers with infants (Gornick et al., 1997; OECD, 2001).
The preceding discussion leads us to predict that the impact of chil-
dren on the entry into (non- professional) self- employment will di er
among countries with di erent welfare state institutions that in uence the
compatibility of women’s family and work duties. Thus, we develop the
following hypotheses:
Hypothesis 2a: Children will positively in uence low- quali ed women’s
decision to become self- employed in liberal welfare states such as the UK
and US. Moreover, the association between children and non- professional
self- employment should be the strongest in the UK and US compared to
any other type of welfare regimes of Western industrialized societies.
Hypothesis 2b: Children will positively in uence low- quali ed women’s
decision to become self- employed in most corporatist conservative welfare
states of Western Europe but not France, Belgium and Italy. Given
extended welfare state provisions for mothers in these countries, low-
quali ed women will not be pushed into self- employment to reconcile
family and work duties.
Beyond these broad predictions for country di erences, we also con-
sider whether speci c policies may be correlated with self- employment
M2401 - BRUSH TEXT.indd 141M2401 - BRUSH TEXT.indd 141 2/8/10 16:12:082/8/10 16:12:08
142 Women entrepreneurs and the global environment for growth
outcomes cross- nationally. Following prior work (Gornick et al., 1997;
Korpi, 2000; Misra et al., 2007; OECD, 2001; Sainsbury, 1994a), institu-
tions which may in uence mothers’ ability to combine family and work
duties can be classi ed into three main groups, namely, (1) institutions
which support women’s uninterrupted labor force attachments around
the time of childbirth (‘leave policies’), (2) institutions which increase the
supply or reduce the cost of non- maternal child care (‘childcare’) and (3)
institutions which provide childcare via public school schedules (‘school
scheduling’) that are compatible with parental employment. Next, we
derive hypotheses about the way these institutions may impact women’s
decision to work as self- employed as well as their choice of professional
status of self- employment.
Leave Policies – Paid Maternity Leave
Theoretical predictions concerning the impact of maternity leave on
women’s attachment to the labor force are ambivalent. On the one hand,
maternity leaves that allow women to return to the same job should
increase their attachment to paid work. This should decrease the likeli-
hood of self- employment, particularly in non- professional work, by
making wage work more attractive. On the other hand, policies that allow
for long leave periods may limit certain career- enhancing opportunities
that require continuity at work (for example, opportunities for training
and promotion) or result in human capital depreciation. The scarce empir-
ical evidence on this topic is controversial (for a review see, for example,
Gornick et al., 1997; Korpi, 2000).
Given women’s lower endowments with resources in non- professional
work to reconcile family and work duties, we expect that maternity leave
policies will impact non- professional self- employment more strongly than
professional self- employment. Speci cally, we expect that the absence
of maternity leave will push women out of the paid work force and into
non- professional self- employment. But, given the higher earnings and
potentially greater employment bene ts garnered by women in profes-
sional work, we do not expect the same positive e ect of children on self-
employment in countries with no provisions for maternity leave. At the
same time, very long leaves that negatively impact women’s attachment to
their employer might increase both non- professional and professional self-
employment, as women seek these entrepreneurial activities as alternatives
to being left out of the workforce following extended care leaves. We thus
assume that there will be a positive association between the length of the
country- speci c policy measures that impact paid maternity leave and
women’s non- professional self- employment.
M2401 - BRUSH TEXT.indd 142M2401 - BRUSH TEXT.indd 142 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 143
Hypothesis 3: Long length of maternity leave should increase women’s
share of self- employment overall, and women’s share of non- professional
self- employment. Additionally, high replacement wages should increase
women’s attachment to paid employment and lower the necessity of self-
employment, particularly among less- skilled women.
Hypothesis 4: The higher the country- speci c maternity leave wage
replacement, the higher the opportunity costs of becoming self- employed
relative to wage employment and thus the lower women’s share of overall,
professional, and non- professional self- employment.
Public School Schedules
From a labor supply perspective, it can be suggested that school schedules
are highly likely to in uence mothers’ employment decisions, because
public schools provide childcare for mothers of school- aged children.
Improvements in public school schedules would imply having children
enrolled in public school for longer hours per day and more weeks per
year, as well as schedules that are more consistent with typical employ-
ment schedules (Gornick et al., 1997; Korpi, 2000; OECD, 2001). As
a consequence, improvements in women’s childcare options should be
associated with increases in maternal labor supply.
However, is there a link between public school schedules and (the profes-
sional status of) women’s self- employment? Assuming that low- quali ed
women primarily choose non- professional self- employment because the
latter is compatible with self- provided childcare, the hypothesis can be put
forward that longer school days would decrease low- quali ed women’s
likelihood of becoming self- employed, since they would lower their need
for child care. This argument leads to the following hypothesis:
Hypothesis 5: The longer school- hours per week, the lower women’s share
of non- professional self- employment.
DATA, VARIABLES AND METHODS
This study draws on the European Labor Force Survey (ELFS, 2004–05)
and the National Longitudinal Survey of Youth (NLSY) for American
data. The ELFS includes 14 645 female founders and 657 909 female
employees (the reference category) from 22 Western and Eastern European
countries in 2004. The NLSY is a multi- stage strati ed national probabil-
ity sample that includes 6283 female respondents interviewed annually
M2401 - BRUSH TEXT.indd 143M2401 - BRUSH TEXT.indd 143 2/8/10 16:12:082/8/10 16:12:08
144 Women entrepreneurs and the global environment for growth
from 1979 to 1998, of which 1377 became self- employed during the 21-
year period. The NLSY over- sampled various populations (that is, racial
minority groups). Descriptive analyses use sample weights, but due to the
spell structure of the event history data, sample weights are not used in
regressions. In both sources of data, we analyse female respondents aged
between 22 and 44 years.
Dependent Variable
For the ELFS data, we use the International Standard Classi cation of
Occupations (ISCO- 88) to capture the dependent variable, which meas-
ures women’s entry into professional self- employment versus unskilled/
low- skilled self- employment. The di erentiation between professional
and non- professional self- employment roughly resembles the collapsed
EGP (Erikson, Goldthorpe and Portocarero) class scheme (Erikson
and Goldthorpe, 1992). Category I, which describes ‘professional self-
employment’, is comprised of professional, semi- professional, and higher
managerial occupations (equivalent to EGP I and II). Category II, which
describes ‘low- skilled self- employment’, consists of traditional skilled,
petty bourgeois self- employment including shopkeepers and restaurant
owners (mostly EGP IIIa, V and VI) as well as unskilled self- employment
(EGP IIIb and VII). In the NLSY, we de ne the respondent as self-
employed if she answered ‘yes’ to the survey question: ‘Are you self-
employed in this job?’ Analyses distinguish between two subgroups of
self- employment based on the three- digit Census codes for professional
and managerial occupations. Professional and managerial occupations
(coded 1–199) are classi ed as ‘professional’ occupations, while all other
occupations are classi ed as ‘non- professional’ occupations.
We use a discrete time event history model, Cox’s proportional hazards
regression, to analyse the transition rates into self- employment for the US
NLSY data. Only the rst transition into self- employment is modeled.
The competing risks model takes the form of a multinomial logistic regres-
sion model with three competing outcomes (for elaboration see Allison,
1984). The competing hazards model predicts self- employment in profes-
sional versus non- professional occupations. Event history analysis allows
us to handle the ubiquitous problems of sample selectivity, non- random
sorting, censoring, time- varying explanatory variables, and unobserved
heterogeneity (Wooldridge, 2002). Cox proportional hazards model is
designed to explain why some individuals are at greater risk to experience
an event than others, and to estimate the e ect of time- varying covariates
(such as marital status and motherhood) on the change in the hazard of an
event occurring (such as entry into self- employment).
M2401 - BRUSH TEXT.indd 144M2401 - BRUSH TEXT.indd 144 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 145
If we do not account for the fact that female entrepreneurs are not a
random sample of individuals, but have characteristics which made them
select into the labor market in the rst place, a simple regression model of
women’s choice of professional self- employment versus non- professional
self- employment on a set of covariates can generate misleading estima-
tions of the regression parameters. Ignoring the ‘sample selection bias’
would lead us to the overestimation of the predicted values (Heckman,
1979). To correct for the sample selection bias, we use the Heckman probit
model with a sample selection equation in the ELFS. The Heckman cor-
rection has a two- stage structure. In the rst stage, the selection equation
is used as a basis for predicting women’s labor market participation in
the sample. In the selection equation, we introduce an ‘extra’ exogenous
variable, which has an e ect on the selection equation, predicting women’s
decision either to participate in the labor market or not, but not on the
main equation, which determines women’s choice of professional self-
employment versus non- professional self- employment. In the second
stage, transformed predictions are included as an additional variable into
the main equation model. For technical details, see Greene (2003).
Independent Variables
We harmonized all available variables from the NLSY and ELFS that
could be consistently measured. In the ELFS data, variables are measured
in the year of observation. In the NLSY data, respondents who never
become self- employed are treated as right- censored observations. Here,
time- varying family characteristics include marital/cohabitation status,
the partner’s self- employment status and the number of children (within
child age categories). Unpartnered women (neither married nor cohabitat-
ing) and those with either unemployed or retired partners are the reference
category for the partner’s status variable. We include two dummy variables
indicating partner’s status. The partner self- employed dummy variable
is coded 1 if the woman has a partner who is self- employed. The partner
employed dummy variable is coded 1 if the woman has a partner in the wage
and salaried work. The reference category is de ned as having no partner or
an unemployed partner. We use three categorical variables to measure the
number of children aged 0 to 4 years (preschoolers), the number of children
aged 5 to 9 years, and the number of children aged 10 to 18 years.
Time- varying measures of human capital are education, age (proxy for
experience) and employment status prior to self- employment. Given the
high level of inconsistency in the coding of education across the countries in
our analysis, we were only able to capture high education (post- secondary
quali cations) versus low education with a dummy variable indicating high
M2401 - BRUSH TEXT.indd 145M2401 - BRUSH TEXT.indd 145 2/8/10 16:12:082/8/10 16:12:08
146 Women entrepreneurs and the global environment for growth
education. We use four age categories and include three dummy variables
indicating that the respondent is aged 30 to 34 years, 35 to 40 years and 40
to 45 years (with 18 to 29 years as the reference category). Thus, we also
control for employment status prior to becoming self- employed by using a
dummy which is ’1’ if the respondent was not employed prior to founding
her own business or becoming self- employed. Time- varying demographic
characteristics include whether the respondent lives in a rural, suburban or
urban area (rural being the reference category). We include these measures
since self- employment opportunities may be greater in areas with higher
population density.
Moreover, we utilize macro- level data on family policy indicators which
revolve around (1) leave policies, (2) early childhood education and care,
and (3) school scheduling for 22 countries. These data are provided by
Misra et al. (2007).
FINDINGS
The share of women’s professional self- employment has experienced
an increase during 1992 and 2004 in most Western European countries
(Germany, Sweden, the Netherlands, Belgium, the UK, Italy, France,
Austria, Greece, Portugal and Spain). However, countries di er in the scope
of the growth of women’s professional self- employment. At the aggregated
level, there is a clear indication for the impact of the country’s institu-
tional make- up on women’s share of professional self- employment. More
exactly, the probability of entering a low- skilled occupation is the highest in
Central- Eastern Europe (Poland, Slovenia, Slovakia, Estonia, Hungary and
Lithuania). In contrast, there is a positive association between the share of
women’s professional self- employment and highly regulated types of labor
markets in Belgium, the Netherlands, Luxembourg, Germany, Austria
and France. In the US, the proportion of women who are self- employed
has grown in both professional and non- professional self- employment.
However, the growth has been much larger for non- professional self-
employment.
Demographic characteristics show that self- employed women are
younger than their employed counterparts across all countries (with
self- employed professionals being the youngest, on average), and more
highly educated (again, with self- employed professionals having the
highest average education). In terms of family structure, we nd great
heterogeneity among the self- employed, with a disproportionate number
of non- professionals being mothers and partnered with men who are also
self- employed. The majority of female professional founders have no
M2401 - BRUSH TEXT.indd 146M2401 - BRUSH TEXT.indd 146 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 147
children (59 per cent), compared to the share of female non- professional
founders (43 per cent). Also, female professional founders live without a
partner more often (50.5 per cent) than their non- professional counter-
parts (39.4 per cent). In contrast, couple entrepreneurs (that is, businesses
which have been set up by couples) are more prevalent among female non-
professionals (21 per cent) than female professionals (15.8 per cent).
Summing up, female non- professional founders seem to be much more
strongly embedded in the family context than their professional founders
in almost all investigated countries.1 But, can the bivariate association
between self- employment in professional/non- professional occupations
and women’s individual characteristics be reproduced in the multivariate
estimations?
First, the e ect of family composition on women’s inclination to
become self- employed di ers substantially across professional and
non- professional occupations. On the one hand, responsibilities for
caring children (up to 4 years old and 5–9 years old) positively a ect
women’s decision to engage in non- professional occupations and
strongly signi cantly across the majority of countries analysed (Austria,
Germany, Ireland, the Netherlands, Luxembourg, Greece, Portugal,
Spain, Hungary, Poland, the Czech Republic). This strongly sup-
ports hypotheses 1a and 2b that motherhood is a strong predictor for
women’s entry into non- professional self- employment in corporatist and
conservative welfare states in Western Europe. Moreover, children are
strongly positively associated with mothers’ entry into non- professional
self- employment in the US and the UK as liberal welfare states with
missing childcare infrastructure. The results for the UK and US thus
strongly con rm hypothesis 2a. Furthermore, looking at the regression
coe cients for the impact of minor children (up to 4 years and 5–9 years
old) on women’s non- professional self- employment in the UK and the
US, one nds that they are far stronger than the respective coe cients
for other welfare states in Western Europe (for example, Austria or
Germany). In contrast, children do not in uence mothers’ entry into
non- professional self- employment in France, Belgium and Italy. This
supports hypothesis 2b that low- quali ed women in France, Belgium
and Italy are not apparently pushed into self- employment to combine
family and work duties, given extended welfare states provisions for
mothers to reconcile family and work duties.
Furthermore, children either do not have any e ect on the entry
into professional self- employment or impact women’s professional self-
employment strongly negatively in most Western European countries, the
US and some Eastern European countries (Latvia, Poland and Slovenia).
This is clear evidence for hypothesis 1a, which states that balancing family
M2401 - BRUSH TEXT.indd 147M2401 - BRUSH TEXT.indd 147 2/8/10 16:12:082/8/10 16:12:08
148
Table 8.1 Determinants of entry into professional and non- professional self- employment (Heckman probit regression)
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Austria Belgium Germany France
Children
Child up to
4 years
0.198** 0.277** 0.048 20.11 0.054 0.062 0.055 20.042
(0.068) (0.077) (0.071) (0.072) (0.051) (0.073) (0.053) (0.059)
Child 5–9 years 0.158** 0.174** 0.146 20.282** 0.115** 20.049 0.069 20.073
(0.052) (0.056) (0.075) (0.104) (0.033) (0.043) (0.046) (0.061)
Child
10–19 years
0.149** 20.176** 20.062 20.122 20.018 20.324** 0.027 20.358**
(0.047) (0.065) (0.077) (0.088) (0.032) (0.047) (0.046) (0.077)
Partner
Partner self-
empl.
0.240** 0.165* 0.319** 0.155 0.207** 0.097 0.473** 0.256**
(0.054) (0.065) (0.075) (0.088) (0.049) (0.065) (0.044) (0.063)
Partner
employee
20.033 0.004 20.140* 20.207** 20.175** 20.142** 20.226** 20.120**
(0.041) (0.049) (0.057) (0.064) (0.03) (0.038) (0.034) (0.042)
M2401 - BRUSH TEXT.indd 148M2401 - BRUSH TEXT.indd 148 2/8/10 16:12:082/8/10 16:12:08
149
Ireland Luxembourg Netherlands Cyprus
Children
Child up to
4 years
0.056 0.018 20.156 20.262 0.013 0.126* 0.061 20.276
(0.055) (0.075) (0.107) (0.152) (0.038) (0.057) (0.083) (0.214)
Child 5–9 years 0.151** 20.153* 20.083 20.404* 0.122** 0.049 20.141 20.269
(0.046) (0.067) (0.102) (0.166) (0.039) (0.06) (0.104) (0.187)
Child
10–19 years
0.03 20.127* 20.015 24.964 0.071 20.113 20.194* 20.564**
(0.047) (0.065) (0.106) (4191.535) (0.041) (0.073) (0.094) (0.211)
Partner
Partner self-
empl.
0.491** 0.390** 1.152** 0.440** 0.600** 0.193** 0.099 0.058
(0.043) (0.056) (0.111) (0.155) (0.036) (0.057) (0.099) (0.167)
Partner
employee
0.079* 0.148** 0.173 0.051 20.154** 20.216** 0.143 20.184
(0.038) (0.047) (0.094) (0.108) (0.032) (0.045) (0.074) (0.139)
M2401 - BRUSH TEXT.indd 149M2401 - BRUSH TEXT.indd 149 2/8/10 16:12:082/8/10 16:12:08
150
Table 8.2 Determinants of entry into professional and non- professional self- employment (Heckman probit regression)
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
Greece Italy Portugal Spain
Children
Child up to
4 years
0.023 20.009 20.050* 20.041 0.122** 20.156 0.03 0.217**
(0.038) (0.069) (0.021) (0.033) (0.039) (0.086) (0.023) (0.038)
Child 5–9 years 0.111** 20.172* 0.014 20.097* 0.107** 20.144 0.137** 0.096*
(0.039) (0.076) (0.023) (0.039) (0.041) (0.092) (0.023) (0.041)
Child
10–19 years
0.175** 20.1 0.062** 20.128** 20.031 20.058 0.149** 0.014
(0.032) (0.068) (0.019) (0.034) (0.04) (0.07) (0.02) (0.034)
Partner
Partner self-
empl.
20.05 0.312** 0.116** 0.125** 0.339** 0.269** 0.405** 0.212**
(0.034) (0.059) (0.022) (0.034) (0.04) (0.089) (0.027) (0.038)
Partner
employee
20.284** 0.1 20.146** 20.05 20.073* 20.003 20.112** 20.071*
(0.032) (0.063) (0.018) (0.029) (0.032) (0.065) (0.018) (0.035)
Czech Republic Estonia Hungary Latvia
Children
Child up to
4 years
0.421** 0.784** 0.148 1.137** 0.546** 0.847** 20.051 0.164
(0.121) (0.209) (0.439) (0.074) (0.072) (0.112) (0.136) (0.211)
Child 5–9 years 0.164** 20.044 0.216 0.018 0.116** 20.137 0.116 20.104
(0.042) (0.068) (0.148) (0.049) (0.04) (0.072) (0.125) (0.207)
Child
10–19 years
20.031 20.268** 0.226 0.013 0.078* 20.1 0.067 20.743**
(0.04) (0.07) (0.138) (0.074) (0.037) (0.058) (0.118) (0.264)
M2401 - BRUSH TEXT.indd 150M2401 - BRUSH TEXT.indd 150 2/8/10 16:12:082/8/10 16:12:08
151
Partner
Partner self-
empl.
0.322** 0.205** 20.137 20.114 0.333** 0.187** 0.713** 0.602**
(0.039) (0.068) (0.195) (0.099) (0.042) (0.058) (0.113) (0.18)
Partner
employee
20.161** 20.041 20.011 20.114 20.098** 20.245** 0.101 20.385*
(0.033) (0.054) (0.1) (0.068) (0.031) (0.057) (0.084) (0.182)
Lithuania Poland Slovenia Slovak Republic
Children
Child up to
4 years
0.152 0.824** 0.310** 20.018 0.148 0.275** 20.036 1.091**
(0.312) (0.04) (0.071) (0.161) (0.094) (0.059) (0.128) (0.044)
Child 5–9 years 0.14 0.011 0.216** 20.056 20.163 0.043 0.082 0.017
(0.11) (0.012) (0.05) (0.109) (0.12) (0.062) (0.065) (0.025)
Child
10–19 years
0.068 20.018 0.067 20.342** 0.131 0.015 0.051 20.024
(0.096) (0.009) (0.048) (0.12) (0.083) (0.057) (0.059) (0.026)
Partner
Partner self-
empl.
0.028 0.036 0.425** 0.330** 0.067 20.867 0.528** 0.172**
(0.145) (0.023) (0.049) (0.109) (0.123) (32.955) (0.066) (0.042)
Partner
employee
20.15 20.007 20.125** 20.068 20.118 20.084* 0.170** 0.070**
(0.089) (0.021) (0.04) (0.088) (0.079) (0.042) (0.053) (0.025)
M2401 - BRUSH TEXT.indd 151M2401 - BRUSH TEXT.indd 151 2/8/10 16:12:082/8/10 16:12:08
152 Women entrepreneurs and the global environment for growth
and work duties is not a motive to become self- employed for highly quali-
ed women. Among Western industrialized states, the outliers from this
pattern are Austria, the Netherlands and Spain. In these countries, women
with children (up to 4 years) are highly likely to work as self- employed
professionals – a pattern which is also surprisingly reproduced for most
post- Socialist countries in Eastern Europe, namely, the Czech Republic,
Estonia, Hungary, Lithuania, Slovenia and the Slovak Republic.
Besides, having a self- employed partner impacts the female founder’s
decision to work in non- professional occupations positively and highly
signi cantly in Europe (but Belgium where the e ect is only signi cant at
a 10 per cent level) – a nding which is remarkably stable across 18 out of
22 investigated countries with di erent institutional regimes. This gives
clear support for hypothesis 1b. But, spousal self- employment activity also
in uences women’s professional self- employment likelihood positively
and strongly signi cantly in most countries, supporting the contention
that ‘self- employment often comes in twos’ (Arum and Müller, 2004: 446).
Table 8.3 Determinants of entry into professional and non- professional
self- employment (heckman probit regression )
Model 1
non- profess.
Model 2
profess.
Model 1
non- profess.
Model 2
profess.
UK USA
Children
Child up to
4 years
0.344** 0.158 0.315** 0.059
(0.047) (0.088) (0.067) (0.184)
Child
5–9 years
0.125** 20.199** 0.419** 0.148
(0.038) (0.059) (0.087) (0.251)
Child
10–19 years
20.105* 20.289** 0.420** 20.502
(0.041) (0.062) (0.129) (0.480)
Partner
Partner self-
empl.
0.340** 0.275** 1.385** 0.742*
(0.039) (0.058) (0.109) (0.304)
Partner
employee
20.075** 20.062 0.550** 0.152
(0.029) (0.041) (0.061) (0.169)
Notes:
** p , .01 , * p , 0.05, two- tailed tests. Heckman selectivity correction includes dummies
for marital status, age categories, child up to 4 years and nationality.
All models include control variables (age and education).
Source: European Labor Force (ELFS, 2004–05), own calculations.
M2401 - BRUSH TEXT.indd 152M2401 - BRUSH TEXT.indd 152 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 153
Indeed, the pattern of cross- national variation in e ects of the spousal self-
employment on women’s professional self- employment is largely similar to
the variation found for the e ect of spousal self- employment on women’s
non- professional self- employment. This contradicts hypothesis 1b.
Summarizing, spousal self- employment is a positive and robust predictor
(at a 1 per cent level) for the wife’s entry into professional self- employment
in 17 out of 22 European countries and the US. However, the mecha-
nisms which underlie the positive relationship between spousal involve-
ment in self- employment and women’s non- professional and professional
self- employment might be di erent. Low- quali ed women entering non-
professional self- employment may join their husband’s self- employment
activity to support their husband’s business and career (Budig, 2006;
Firkin et al., 2003). This can be the case for family- run businesses in craft,
retail, catering and lodging. As indicated by prior work, the spouse of
self- employed males usually performs ancillary roles, for example, she is
doing accounts, administration, answering phones, and so on (Firkin et
al., 2003). On the other hand, the positive association between spousal
self- employment and high quali ed women’s choice of self- employment
suggests that marriage, generally, is a valuable asset (Becker, 1965) that pro-
vides a suitable background for self- employment. Collectively, a married
couple can put up more resources ( nancial capital, self- employment
relevant know- how and skills, networks, and so on, and emotional support
(Brüderl and Preisendörfer, 1998) for starting their own businesses. With
the support of the self- employed husband, a married woman thus would
be more willing to take the risk. Finally, self- employed husbands may also
serve as ‘role models’ (Bandura, 1977), thus supporting the wife’s perceived
feasibility of becoming of becoming self- employed (‘If he can do it, so can
I’) (Tonoyan et al., 2005). Indeed, having a self- employed partner in the US
increases women’s likelihood of non- professional self- employment signi -
cantly, more exactly, by a factor of 4. However, it has a far smaller e ect on
the likelihood of transition into professional self- employment.
On the other hand, we observe a strong negative association between
the spousal involvement in dependent employment and women’s entry
into both professional and non- professional self- employment in almost all
investigated European countries. Put di erently, women with husbands
in wage and salaried work are signi cantly less likely to become self-
employed (both in professional and non- professional occupations) than
their counterparts who either do not have partners/husbands or live with
an unemployed partner/husband.
Next, we turn to a set of correlation analyses. We run correlations between
family policy indicators and women’s total share of self- employment, pro-
fessional self- employment and non- professional self- employment. This
M2401 - BRUSH TEXT.indd 153M2401 - BRUSH TEXT.indd 153 2/8/10 16:12:082/8/10 16:12:08
154 Women entrepreneurs and the global environment for growth
analysis gives us a sense of how work- family policies may impact women’s
engagement in entrepreneurship.
First, we observe a positive correlation between paid maternity leave
and women’s non- professional self- employment (Pearson correlation
being 0.40). This means that the longer the paid maternity leave, the longer
women’s detachment from the labor force. To the extent women’s non-
professional self- employment is a result of poor labor market character-
istics, it seems to be intuitive that longer labor force detachments increase
the probability of becoming self- employed for low- quali ed women.
This supports hypothesis 3. However, there is almost no correlation
between paid maternity leave and women’s professional self- employment.
Second, we notice a negative relationship between maternity leave wage
replacement rates and both women’s non- professional and professional
self- employment, although the correlation values are not high, namely
20.22 and 0.29, respectively. This holds true for both types of women’s
entrepreneurship and substantiates hypothesis 4.
Table 8.4 Correlation between family policy indicators* and female
entrepreneurship rates in 22 European countries
Share of
women’s non-
Professional
self-
employment
Share of
women’s
professional
self-
employment
Share of
women’s self-
employment
(total)
Number
of
countries
A: Leave policies
1. weeks of maternity
leave
0.40 20.09 0.28 22
2. maternity leave
wage replacement
rate
20.22 20.29 20.26 22
B: Early childhood education and care
3. % of 0–2- years- old
children in public
childcare
20.28 0.08 20.20 20
4. % of 3–6- years- old
children in public
childcare
0.10 0.38 0.20 19
C: School scheduling
5. School hours per
week
20.22 0.24 20.11 18
Source: Misra et al. (2006).
M2401 - BRUSH TEXT.indd 154M2401 - BRUSH TEXT.indd 154 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 155
Furthermore, one notices a negative relationship between the percent-
age of children aged 0–4 years in public childcare and women’s share of
non- professional self- employment. This gives support for the hypothesis 5
that the lower the country- speci c publicly provided child care, the higher
women’s share of non- professional self- employment. This may result from
two factors. The rst is that the demand for private childcare in these
countries may create entrepreneurial opportunities for women to provide
childcare and other household services in the non- professional sector. It
may also result from the necessity of women in non- professional employ-
ment to turn to self- employment when childcare is unavailable in order to
balance work and family responsibilities.
At the same time, there is almost no relationship between the percentage
of children aged 0–2 years in public childcare and women’s share of pro-
fessional self- employment. The positive association between the percent-
age of 3–6- year- old children and women’s professional self- employment
does not hold. Again, these results strongly support hypothesis 5.
Finally, we observe a negative association between school hours per
week and women’s share of non- professional self- employment (the cor-
relation value being 20.22), and a positive relationship between school
hours per week and women’s share of professional self- employment (the
correlation value being 0.24). While the former nding supports hypothesis
6, no theoretical justi cation can be given for the latter.
SUMMARY AND DISCUSSION
This study makes several contributions to prior research. First, we show
profound heterogeneity in women’s self- employment in a cross- national
comparison among 23 European countries and the US. Second, we high-
light di erences in the individual determinants of women’s entry into
low- skilled versus high- skilled occupations in self- employment. Third, we
examine the impact of country- speci c institutions on women’s entry into
self- employment, showing the association between welfare state policies
which regulate women’s reconciliation of family and work and entry into
(non- professional) self- employment.
Self- employed women represent a strongly heterogeneous group. Women
working in low- skilled occupations seem to pursue self- employment
to combine childcare and work responsibilities (we call them ‘family-
oriented’ female self- employed). It could be shown that the motherhood
is a strong predictor for the entry into entrepreneurship for low quali ed
women – a result which is strikingly similar for almost all investigated
countries. Albeit self- employment is a job solution for highly quali ed but
M2401 - BRUSH TEXT.indd 155M2401 - BRUSH TEXT.indd 155 2/8/10 16:12:082/8/10 16:12:08
156 Women entrepreneurs and the global environment for growth
less ‘family embedded’ women who enter professional occupations, but
they are not pushed into self- employment to reconcile family and work
duties. Rather, they are pulled into it by the prospect of better earnings
and career advancement, thus resembling the pattern found for self-
employed men (Budig, 2006).
In contrast to motherhood, which is only a predictor for women’s
entry into non- professional self- employment, spousal self- employment is
strongly positively associated with both professional and non- professional
self- employment. Put di erently, women’s low- skilled and high- skilled
entrepreneurship are embedded in matrimonial relationships, thus sup-
porting the contention that ‘self- employment often comes in twos’ (Arum
and Müller, 2004: 446). Still, the mechanisms which underlie the positive
association between spousal self- employment and women’s entry into
non- professional versus professional self- employment may be di erent.
For instance, less quali ed women may join husbands’ self- employment
activity to support their husbands’ work and career. This is usually the
case for low- skilled or skilled self- employment in crafts, retail, catering,
and lodging, that is, typical family- run businesses (cf. Aldrich and Cli ,
2003; Firkin et al., 2003). Also, they might draw on strong ties with own
husbands to compensate for the shortage of nancial and human capital
(Brüderl and Preisendörfer, 1998). In contrast, the positive e ect of the
spousal self- employment on women’s entry into high- skilled occupations
could be traced back to other mechanisms, for example, self- employed
husbands could play a role by serving as positive ‘role models’ for highly
skilled women.
Moreover, our ndings suggest that institutional variations between
welfare states are responsible for cross- country variations of the occu-
pational status of women’s self- employment. This complements recent
work on female entrepreneurship which recommends addressing the
heterogeneity among female entrepreneurs when examining institutional
environments in which the entrepreneur is embedded (De Bruin et al.,
2007; Jennings and McDougald, 2007). As a general pattern, countries
with a higher level of women’s entrepreneurship usually have lower levels
of women’s professional self- employment. In contrast, the highest share
of women’s professional self- employment is found in highly regulated and
conservative/corporatist welfare states of Western Europe.
Furthermore, the di erences in welfare states regimes which regulate
women’s reconciliation of family and work duties moderate the e ect of
motherhood on women’s entry into entrepreneurship. For instance, the
e ect of motherhood on becoming self- employed in non- professional
occupations is the highest in liberal welfare states such as the UK and the
US, where the publicly funded infrastructure for women’s reconciliation
M2401 - BRUSH TEXT.indd 156M2401 - BRUSH TEXT.indd 156 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 157
of family and work is largely missing. On the aggregated level, the weaker
public institutions supporting mother’s employment are (for example,
weak public childcare, school scheduling which do not support mothers’
full- time employment), the higher the country’s share of non- professional
self- employed women. Besides, countries with longer paid maternity
leave and thus women’s longer detachment from the labor force seem to
have higher share of women’s non- professional self- employment. Finally,
countries with higher maternity leave wage replacement (and thus higher
opportunity costs of becoming self- employed) have lower rates of women’s
professional and non- professional self- employment.
CAVEATS AND IMPLICATIONS FOR PUBLIC
POLICY AND RESEARCH
Several implications for future research can be derived from our analysis.
First, we have discovered that some but not all post- Socialist countries
in Central- Eastern Europe resemble the pattern found for the corporat-
ist welfare states in Western Europe quite strongly. This calls for future
research on the role of the institutional environments of Central- Eastern
European countries for women’s entrepreneurship, considering that
research on this topic is still in infancy (De Bruin et al., 2007), despite some
recent contributions on it (see, for example, Welter et al., 2006).
A limitation of our study is the use of di erent data sources and
methods, namely, the panel data and survival analysis for the US sample
and cross- sectional data and Heckman correction model for the European
sample. While the most important results could be corroborated in both
samples, it would be useful to employ panel data for the European sample
too to account for unobserved heterogeneity.
Furthermore, it would also be rewarding to study country- speci c
informal institutions, that is, traditions, customs, societal norms, ‘shared
mental models’, unwritten codes of conduct, ideologies and templates
(Baumol, 1990; Denzau and North, 1994; North, 1990), and thus the
cultural aspects of the environment in which (women’s) entrepreneurship
takes place (instead of only focusing on formal political and economic
rules and regulations). If the society predominantly de nes women
through roles associated with family and household responsibilities (see,
for example, Welter, 2006, for Germany), it will design formal rules and
regulations which provide women with incentives to stay at home and care
for their children (Pfau- E nger, 2004), thus subsequently in uencing the
occupational character of self- employment work.
Fruitful research and policy agenda could also focus on examining
M2401 - BRUSH TEXT.indd 157M2401 - BRUSH TEXT.indd 157 2/8/10 16:12:082/8/10 16:12:08
158 Women entrepreneurs and the global environment for growth
working conditions and outcomes (for example, earnings, job stability)
that result from professional versus non- professional self- employment.
Supposing that highly skilled and well- compensated professions in self-
employment provide higher returns to self- employment and have higher
survival and growth chances (for the US evidence see Budig, 2006),
it would probably be more rewarding to create institutional environ-
ments that are more favorable for women’s entry into professional self-
employment.
NOTE
1. The respective tables are available upon request.
REFERENCES
Aldrich, H.E. and Cli , J. (2003), ‘The pervasive e ects of family on entrepre-
neurship: towards a family embeddedness perspective’, Journal of Business
Venturing, 18, 573–96.
Allison, P. (1984), Event History Analysis, Newbury Park, CA: Sage.
Arum, R. (1997), ‘Trends in male and female self- employment: growth in a new
middle class or increasing marginalization of the labor force?’, Research in
Strati cation and Mobility, 15, 209–38.
Arum, R. and Müller, W. (2004), The Reemergence of Self- Employment. A
Comparative Study of Self- employment Dynamics and Social Inequality,
Princeton, NJ and Oxford: Princeton University Press.
Bandura, A. (1977), Social Learning Theory, Englewood Cli s, NJ: Prentice Hall.
Baumol, W.J. (1990), ‘Entrepreneurship: productive, unproductive, and destruc-
tive’, Journal of Political Economy, 98 (5), 893–921.
Becker, G.S. (1965), ‘A theory of the allocation of time’, Economic Journal, 75,
493–517.
Bjornskov, C. and Foss, N.J. (2008), ‘Economic freedom and entrepreneurial
activity: some cross- country evidence’, Public Choice, 134 (314), 307–28.
Brüderl, J. and Preisendörfer, P. (1998), ‘Network support and the success of
newly founded businesses’, Small Business Economics, 10 (3), 213–225.
Brush, C. (1992), ‘Research on women business owners: past trends, a new per-
spective and future directions’, Entrepreneurship Theory and Practice, 16 (4),
5–30.
Budig, M.J. (2006). ‘Intersections on the road to self- employment: gender, family,
and occupational class’, Social Forces, 84 (4), 2223–39.
Carr, D. (1996), ‘Two paths to self- employment? Women’s and men’s self-
employment in the United States’, Work and Occupations, 23, 26–53.
Carree, M., van Steel, A., Thurik, R. and Wennekers, S. (2002), ‘Economic devel-
opment and business ownership: an analysis using data of 23 OECD countries
in the period 1976–1996’, Small Business Economics, 19, 271–90.
M2401 - BRUSH TEXT.indd 158M2401 - BRUSH TEXT.indd 158 2/8/10 16:12:082/8/10 16:12:08
The impact of family structure and family policies in Europe and the US 159
De Bruin, A., Brush, C. and Welter, F. (2007), ‘Advancing a framework for
coherent research on women’s entrepreneurship’, Entrepreneurship: Theory and
Practice, 5, 323–39.
Denzau, A.T. and North, D.C. (1994), ‘Shared mental models: ideologies and
institutions’, Kyklos, 47 (1), 3–31.
ELFS (2004–25), ‘EUROSTAT LFS Users Guide: Labor Force Survey: ano-
nymized data sets’, European Commission EUROSTAT. Erikson, R. and J.H.
Goldthorpe (1972), The Constant Flux, Oxford: Clarendon.
Esping- Anderson, G. (1990), The Three Worlds of Welfare Capitalism, Cambridge:
Polity Press.
Firkin, P., Dupuis, A. and De Bruin, A. (2003), ‘Familial entrepreneurship’, in P.
Spoonley, A. Dupuis, and A. De Bruin (eds), Work and Working in 21st Century
New Zealand, Palmerston North, New Zealand: Dunmore Press, pp. 154–79.
Gornick, J.C., Meyers, M.K. and Ross, K.E. (1997), ‘Supporting the employment
of mothers: policy variation across fourteen welfare states’, Journal of European
Social Policy, 7 (1), 45–70.
Greene, P. (2000), ‘Self- employment as an economic behavior: an analysis of self-
employed women’s human and social capital’, National Journal of Sociology,
12, 1–55.
Greene, W.H. (2003), Econometric Analysis, Englewood Cli s, NJ: Prentice
Hall.
Gusta son, S. and Wetzels, C. (1997), ‘Family policies and women’s labor
force transitions in connection with childbirth’, Vierteljahreshefte zur
Wirtschaftsforschung, 3 (4), 118–24.
Heckman, J. (1979), ‘Sample selection bias as a speci cation error’, Econometrica,
47, 153–61.
Jennings, J.E. and McDougald, M.S. (2007), ‘Work–family interface experiences
and coping strategies: implications for entrepreneurship research and practice’,
Academy of Management Review, 32, 747–60.
Koellinger, P., Minniti, M. and Schade, C. (2007), ‘I think I can, I think I can –
overcon dence and entrepreneurial behavior’, Journal of Economic Psychology,
28 (4), 502–07.
Korpi, W. (2000), ‘Faces of inequality: gender, class, and patterns of inequalities in
di erent types of welfare states’, Social Politics, 7 (2), 127–91.
Langowitz, N. and Minitti, M. (2007), ‘The entrepreneurial propensity of women’,
Entrepreneurship Theory and Practice, 31 (3), 341–64.
Lohmann, H. (2004), Beru iche Selbständigkeit von Fransen und Männern
im internationalen Vergleich – welche Rolle spielt die Vereinbarkeit von
Familie und Erwerbstätigkeit’, in G. Schmid. M. Gangl and P. Kupta (eds),
Arbeitsmarktpolitik und Struktarwandel: Empirische Analysen, BeitrAB 286,
Nuremberg: Institute für Arbeitsmarkt- und Berufsforschung, pp. 205–26.
Lohmann, H. and Luber, S. (2004), ‘Trends in self- employment in Germany: dif-
ferent types, di erent developments’, in R. Arum and W. Müller (eds), The Re-
Emergence of Self- Employment Dynamics and Social Inequality, Princeton, NJ:
University Press, pp. 36–75.
McManus, P. (2001), ‘Women’s participation in self- employment in Western
industrialized societies’, International Journal of Sociology, 31, 70–97.
Minniti, M. and Nardone, C. (2007), ‘Being in someone else’s shoes: the role of
gender in nascent entrepreneurship’, Small Business Economics, 28, 223–38.
Misra, J., Moller, S. and Budig, M.J. (2007), ‘Work–family policies and poverty
M2401 - BRUSH TEXT.indd 159M2401 - BRUSH TEXT.indd 159 2/8/10 16:12:082/8/10 16:12:08
160 Women entrepreneurs and the global environment for growth
for partnered and single women in Europe and North America’, Gender and
Society, 21 (6), 804–27.
North, D.C. (1990), Institutions, Institutional Change and Economic Performance,
Cambridge: Cambridge University Press.
Nyström, K. (2008), ‘The institutions of economic freedom and entrepreneurship:
evidence from panel data’, Public Choice, 136 (3), 269–82.
Organisation for Economic Co- operation and Development (OECD) (2001),
‘Balancing work and family life: helping parents into paid employment’,
Employment outlook, 1, 129–64.
Pfau- E nger, B. (2004), Development of Culture, Welfare States and Women’s
Employment in Europe, Hamburg: Ashgate.
Presser, H. (1995), ‘Job, family, and gender: determinants of nonstandard work
schedules among employed Americans in 1991’, Demography, 32, 577–98.
Sainsbury, D. (ed.). (1994a), Gendering Welfare States, London: Sage
Publications.
Sainsbury D. (1994b), ‘Women’s and men’s social rights: gendering dimensions of
welfare states’, in D. Sainsbury (ed.), Gendering Welfare States, London: Sage
Publications, pp. 150–70.
Taniguchi, H. (2002), ‘Determinants of women’s entry into self- employment’,
Social Science Quarterly, 83, 875–93.
Thomas, L. and Ganster, D. (1995), ‘Impact of family- supportive work variables
on Work–family con ict and strain’, Journal of Applied Psychology, 80, 6–15.
Tonoyan, V., Strohmeyer, R. and Wittmann, W.W. (2005), ‘Gendered and cross-
country di erences in the perceived di culty of becoming self- employed: the
impact of individual resources and institutional restrictions’, in Frontiers of
Entrepreneurship Research, Wellesley, MA: Babson College, pp. 57–72.
Verheul, I. and Thurik, A.R. (2001), ‘Start- up capital: does gender matter?’, Small
Business Economics, 16 (4), 329–45.
Wagner, J. (2007), ‘What a di erence a Y makes – female and male nascent entre-
preneurs in Germany’, Small Business Economics, 28 (1), 1–21.
Welter, F. (2006), ‘Women’s entrepreneurship in Germany: progress in a still
traditional environment’, in C.G. Brush, N. Carter, P.G. Greene and M. Hart
(eds), Growth Oriented Women Entrepreneurs and Their Businesses: A Global
Research Perspective, Cheltenham, UK and Northampton, MA, USA: Edward
Elgar, pp. 128–53.
Welter, F., Smallbone, D. and Isakova, N. (eds) (2006), Enterprising Women in
Transition Economies, Aldershot: Ashgate.
Wennekers, A.R.M., Noorderhaven, N.G., Hofstede, G. and Thurik, A.R. (2002),
‘Cultural and economic determinants of business ownership across coun-
tries’, Frontiers of Entrepreneurship Research, Babson Park, MA: Center for
Entrepreneurial Studies, pp. 179–90.
Wooldridge, J.M. (2002), Econometric Analysis of Cross- Section and Panel Data,
Cambridge, MA: MIT Press.
M2401 - BRUSH TEXT.indd 160M2401 - BRUSH TEXT.indd 160 2/8/10 16:12:082/8/10 16:12:08
