Content uploaded by Helen Kowalewska
Author content
All content in this area was uploaded by Helen Kowalewska on Jan 19, 2023
Content may be subject to copyright.
Article
Journal of European Social Policy
2023, Vol. 0(0) 1–18
© The Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/09589287221148336
journals.sagepub.com/home/esp
Gendered employment patterns:
Women’s labour market outcomes
across 24 countries
Helen Kowalewska
Department of Social & Policy Sciences, University of Bath, Bath, UK
Abstract
An accepted framework for ‘gendering’the analysis of welfare regimes compares countries by degrees of
‘defamilialization’or how far their family policies support or undermine women’s employment participation.
This article develops an alternative framework that explicitly spotlights women’s labour market outcomes
rather than policies. Using hierarchical clustering on principal components, it groups 24 industrialized
countries by their simultaneous performance across multiple gendered employment outcomes spanning
segregation and inequalities in employment participation, intensity, and pay, with further differences by class.
The three core ‘worlds’of welfare (social-democratic, corporatist, liberal) each displays a distinctive pattern
of gendered employment outcomes. Only France diverges from expectations, as large gender pay gaps across
the educational divide –likely due to fragmented wage-bargaining –place it with Anglophone countries.
Nevertheless, the outcome-based clustering fails to support the idea of a homogeneous Mediterranean
grouping or a singular Eastern European cluster. Furthermore, results underscore the complexity and id-
iosyncrasy of gender inequality: while certain groups of countries are ‘better’overall performers, all have
their flaws. Even the Nordics fall behind on some measures of segregation, despite narrow participatory and
pay gaps for lower- and high-skilled groups. Accordingly, separately monitoring multiple measures of gender
inequality, rather than relying on ‘headline’indicators or gender equality indices, matters.
Keywords
Cluster analysis, comparative family policy, comparative social policy, defamilialization, gender inequality,
gendered trade-offs, welfare state outcomes, welfare state paradox, welfare state typologies, women’s
employment
Introduction
A common approach to analysing and comparing
welfare states involves grouping them into ‘welfare
regimes’based on the similarities and differences in
their policy ‘logics’–that is, the intentions, origins,
and general direction of their social policies
Corresponding author:
Helen Kowalewska, Department of Social & Policy Sciences,
University of Bath, Claverton Down, Bath BA2 7AY, UK.
Email: hk775@bath.ac.uk
(Lewis, 1997;Zimmerman, 2013)–and the relative
role of the state versus the family or market in
welfare provision. Esping-Andersen’s (1990)
welfare regime typology, which distinguished a
‘social-democratic’regime, a ‘liberal’regime, and
a‘corporatist’regime, continues to dominate and
guide this literature. However, building on early
research on gender and the welfare state (for ex-
ample, Land, 1971), feminist scholars have high-
lighted the gender-blindness of Esping-Andersen’s
analysis and other ‘mainstream’welfare regime
analyses. Mainstream research underplays
women’s unpaid care work within the family and
the role of welfare provision in reducing gender as
well as class inequality. Additionally, the over-
riding focus on decommodification –reducing
citizens’dependence on paid work for achieving a
socially acceptable standard of living –understates
the value of having a paid job in the first place,
especially for women by increasing their financial
independence from male partners. It also skims
over gender differences in the capacity for em-
ployment due to women’s disproportionate care
and domestic labour (for example, Daly, 1994;
Lewis, 1992,1997;O’Connor, 1993;Orloff, 1993;
Sainsbury, 1999).
Feminists have extended Esping-Andersen’s
framework by adding in gender-sensitive dimen-
sions. Most notably, defamilialization is the ability to
uphold an acceptable standard of living indepen-
dently of personal relationships through state benefits
and/or paid employment (Lister, 1994;Saraceno,
1997). Other feminists have sought to replace Esp-
ing-Andersen’s framework altogether: for example,
Lewis’s (1992) male-breadwinner typology differ-
entiates ‘strong’(UK up to the 1980s/1990s),
‘modified’(France), and ‘weak’(Sweden) variants
(see also Lewis and Ostner, 1995). Ultimately,
feminist scholars have highlighted the importance of
childcare services, parental leaves, and other family
policies designed to support women’s employment
and autonomy and transform men’s domestic roles
for analyses of welfare states and regimes (for ex-
ample, Lewis, 1992;O’Connor, 1993;Orloff, 1993).
Subsequently, there has been an explosion of
studies that group countries into different family
policy regimes or ‘constellations’. Family policy
regimes mostly overlap with Esping-Andersen’s
(1990) three-fold welfare regime typology, albeit
imperfectly. In the social-democratic welfare regime,
resembled by Scandinavian countries, comprehen-
sive state social provision helps to foster greater
equality between men and women in the division of
paid employment and, more recently, unpaid care
work. By contrast, the liberal welfare regime, ap-
proximated by Anglophone countries, relies on
market-based services with limited intervention in
the family. Meanwhile, the corporatist welfare re-
gime, seen in certain Continental European coun-
tries, actively promotes the family’s (read: women’s)
care role (for example, Esping-Andersen, 1999;
Korpi, 2000;Korpi et al., 2013;Leitner, 2003;
Th´
evenon, 2011). Differently from Esping-Andersen
(1990), though, some analysts identify France (and
sometimes Belgium, too) as embodying a fourth
regime type, which supports women to ‘choose’
between working for pay or staying at home (Misra
et al., 2007). In addition, many scholars recognize a
distinctive Mediterranean welfare regime: here, a
residual welfare state, amid strong inter-generational
and gendered caregiving norms, leaves care work to
the family (women) ‘by default’(Saraceno and Keck,
2010). Finally, some studies add an (albeit hetero-
geneous) post-Soviet regime characterized by market
provision and traditionalism (for example, Saraceno
and Keck, 2010;Th´
evenon, 2011).
This article presents an alternative framework. It
compares 24 high-income countries by their effective
gendered employment outcomes using hierarchical
clustering on principal components. It analyses
countries strictly by their gendered outcomes in the
domain of paid work
1
rather than by the institutional
design of their policies or a conflation of policies and
outcomes.
2
Policies represent aims, whereas out-
comes are the welfare ‘reality’or tangible conse-
quences of policies, intended or not (Ferragina et al.,
2015). Of course, the outcomes observed here are not
solely a product of social policies, but of myriad
circumstances with which welfare state institutions
interact (for example, macroeconomic conditions or
cultural norms). Furthermore, the analysis is strictly
descriptive. Nevertheless, if the outcome-based ty-
pology maps perfectly onto established family policy
typologies, then it seems reasonable to conclude that
2Journal of European Social Policy 0(0)
family policies matter for gendered employment
outcomes even if other factors are also at play.
Conversely, a poor fit between the outcome-based
groupings and institutional typologies would suggest
that other factors are more significant.
A sizeable literature has examined the gendered
employment outcomes associated with different
family policies, regimes, and approaches (for ex-
ample, Gornick and Jacobs, 1998;Mandel, 2012;
Misra et al., 2007;Stier et al., 2001;Th´
evenon, 2016;
Valentova, 2016); yet studies usually analyse one or
just a few outcomes at a time, which ‘shrinks’down
and restricts the meaning of gender inequality
(Lombardo et al., 2009). Even when studies do
consider multiple gendered employment outcomes,
they typically examine different outcomes sequen-
tially and separately from each other, thereby
masking potential interrelations between them (al-
though see Mandel, 2009).
Alternatively, some studies and international or-
ganizations use indices for a ‘holistic’measure of
gender inequality. Examples include the World
Economic Forum’s Global Gender Gap Index and the
European Institute for Gender Equality’s Index.
Again, though, in bundling together multiple, distinct
outcomes for women’s employment into a single
number, indices gloss over the interrelationships
between different ‘types’of inequality (Van der
Vleuten and Verloo, 2012). Added to this, a strong
performance on certain indicators that feed into the
overall index can ‘compensate’for and mask a poor
performance on other indicators, making it hard to
judge and rank countries’performance in a com-
prehensive way. Indeed, workshops with experts
suggest that policymakers typically find data on
individual indicators more engaging and useful than
aggregated index scores (Schmid et al., 2022).
This article instead takes a configurational ap-
proach. It analyses countries as multidimensional
patterns of gender inequality based on their simul-
taneous performance across 10 indicators of
women’s economic labour market outcomes, which
cover participation, intensity, segregation, and pay.
In turn, it determines constellations and associations
that cannot always be established when comparing
countries by a single indicator at a time or measuring
relationships between policies and outcomes, which
usually involves stripping out the broader context. I
treat the outcomes analysed as comprising an inex-
tricable ‘package’of gender in/equality, whose
meaning derives from the interrelationships between
the different outcomes that comprise it and whose
whole is greater than the sum of its parts. I call this
the gendered employment pattern to denote the
macro-level pattern of gender stratification in paid
labour that characterizes a society. I also consider
employment outcomes for women from different
educational groups to give insight into classed di-
mensions (Cooke, 2011).
The period of analysis is 2010–19. Although
cross-sectional analyses cannot capture change over
time and are always prone to distortion by outliers,
pooling annual data from multiple years helps to
reduce the influence of year-by-year volatility and
measurement errors (Ferragina et al., 2015;Kuitto,
2011). Arguably, the 2010s were a critical decade for
women’s employment given the gendered effects of
the 2008 financial crisis, austerity, and the subse-
quent economic recovery (for example, Karamessini
and Rubery, 2013). Despite this, reducing gender
inequalities in employment fell down the list of
policymakers’priorities, at least at the European
level: for example, the Employment Strategy’s
2010–20 revision dropped any gender-disaggregated
data and targets (Fagan and Rubery, 2018;Smith and
Villa, 2010). Additionally, marking ‘how things
were’in the decade leading up to the COVID-19
pandemic is crucial, since the impacts of the pan-
demic on women’s employment outcomes depend
partly on the state of pre-pandemic inequalities.
The results reveal that each of the three core
‘worlds’of welfare is reflected in a specific and
distinct gendered employment pattern. Only France
deviates from expectations by sitting in the same
cluster as Anglophone countries. This mainly reflects
France’s wide gender pay gaps, which are arguably
linked to its fragmented wage-bargaining system,
especially when compared to Belgium. However, the
outcome-based clustering fails to support the idea of
a homogeneous Mediterranean grouping or a sin-
gular Eastern European cluster, thereby highlighting
the inadequacy of welfare regime theory for un-
derstanding women’s employment beyond the three
core ‘worlds’. What is more, the analysis underscores
Kowalewska 3
the complex and idiosyncratic meaning of gender
inequality in comparative perspective. Each group of
countries has its respective strengths and weak-
nesses, and the interrelationships between the dif-
ferent dimensions of inequality are not always
replicated or predictable across borders. Notably,
greater inclusion of lower-skilled and other women
in the labour market can coexist with both worse and
better performances on other gendered labour market
outcomes, including for high-skilled women.
Background
Family policy typologies and women’s
employment participation
An accepted framework for analysing the gendered
character of welfare regimes is based on the design
and logics of family policies –most often, parental
leaves and/or childcare policies. Table S1 in the
Supplemental Material summarises some major
family policy typologies and groupings, as well as
their placements of the 24 countries analysed here.
These labels should not be seen as fixed descriptions
of countries, since welfare states are dynamic and
often display characteristics of more than one regime
type –as exemplified by the divergent placements of
certain countries across different typologies (for
example, Finland, Switzerland; Table S1). Regard-
less, the regime approach offers value in providing a
general guide to important institutional differences
(for example, Hopkin and Wincott, 2006;Maˆ
ıtre
et al., 2005). Table S2 presents descriptive statis-
tics for the main indicators used in these typologies.
To begin, the ‘earner–carer’strategy exhibits
generous childcare services and parental leaves, in-
cluding incentives for men’s caregiving (for exam-
ple, Korpi et al., 2013;Misra et al., 2007). Nordic
countries are usually in this grouping, although not
always Finland, where certain policies are more
typical of familialism –that is, state support for care
within the family (for example, Leitner, 2003). The
Nordics have historically had high female employ-
ment rates, including for low-skilled women, with
most women employed full-time or ‘long’part-time
hours (30 + per week) (Korpi et al., 2013). Ac-
cordingly, comparative social-science researchers
and commentators often hail Scandinavian societies
as a gender equality ‘paradise’and as role models for
others (for example, Esping-Andersen, 2016).
By contrast, the ‘market-oriented’(Korpi, 2000)
regime type, of which the United States and other
Anglophone countries are the best examples, offers
limited state support for the family’s care role. As the
label suggests, care is provided mainly through the
market. Consequently, there is intra-national varia-
tion in the provision, quality, and costs of services
(for example, Korpi et al., 2013;Mandel, 2009).
Employment among low- and medium-skilled
women is also lower than in the dual-earner re-
gime (for example, Anxo et al., 2007;Ferragina,
2019;Hook, 2015). Even so, low-level, means-tested
benefits mean that many women have to work for pay
(for example, Mandel, 2009), although in Australia,
the UK, and, to a lesser extent, Canada, mothers are
often part-time employed (for example, Sayer and
Gornick, 2012). Note that some studies include
Switzerland in this regime type due to its similar
reliance on marketized care services (Korpi et al.,
2013).
In the ‘traditional-family’regime (Korpi et al.,
2013), familialism is more ‘explicit’, as evidenced by
such measures as cash-for-care benefits and derived
social rights for non-employed partners (Leitner,
2003). In turn, women’s employment rates are low
to medium by comparative standards (for example,
Anxo et al., 2007;Ferragina, 2019); and, when they
are employed, women often have part-time or ‘mini’
jobs (for example, Anxo et al., 2007;Korpi et al.,
2013). Even high-skilled women have lower labour
force participation than elsewhere, as flatter wage
structures have impeded the growth of market-based
alternatives to the lack of state care services (Cooke,
2011). Despite declining policy supports for the
traditional male-breadwinner family model, espe-
cially in Germany, Continental European countries
continue to approximate this regime type (for ex-
ample, Leitner, 2014).
Within Continental Europe, family policy typol-
ogies often single out France (and sometimes Bel-
gium) as representing a separate regime type,
labelled the ‘choice strategy’(Misra et al., 2007)or
‘optional familialism’(Leitner, 2003). Policies are
intended to provide a ‘choice’between staying at
4Journal of European Social Policy 0(0)
home versus being employed, as well as between
different care providers (Klammer and Letablier,
2007;Leitner, 2003;Misra et al., 2007). In reality,
though, the low-level, flat-rate nature of the benefit
available to stay-at-home parents encourages low-
income women to trade off a low-paying job for what
is essentially a caring allowance. Meanwhile, al-
lowances for hiring in-home private nannies or
childminders give choice only to high-income in-
dividuals (Klammer and Letablier, 2007;Leitner,
2014). Indeed, in France, low-skilled women’s em-
ployment rates lag behind those of high-skilled
women (Anxo et al., 2007;Hook, 2015).
Many scholars identify Southern European
countries as representing a distinct regime, where
support for women’s care work is more ‘implicit’
(Leitner, 2003). Underdeveloped care services, amid
strong inter-generational, gendered caregiving norms
but with inadequate state benefits to financially
support the family’s care function, leave care work to
the family (women) ‘by default’(Saraceno and Keck,
2010). Furthermore, part-time employment is rarely
an option for ‘reconciling’employment and care due
to the dearth of part-time jobs (for example, Anxo
et al., 2007;Hook, 2015;Th´
evenon, 2011). Thus,
Southern European countries have among the lowest
employment rates for lower-skilled women (for ex-
ample, Anxo et al., 2007;Hook, 2015;Th´
evenon,
2011). The exception is Portugal, where support for
and rates of women’s (full-time) employment have
been high ever since the exodus of men to fight in the
colonial wars in the 1960s and early 1970s (for
example, Tavora, 2012).
Finally, family policy regime typologies that in-
corporate post-Soviet countries often place them in a
single group even while acknowledging intra-
regional diversity (for example, Th´
evenon, 2011).
Yet, region-specific studies suggest that Eastern
European countries are too heterogeneous to be
considered representative of a single regime type,
despite disagreement about the labelling of countries.
For instance, Szelewa and Polakowski (2008) argue
that leave and childcare provisions are implicitly
familialist in Czechia, Slovakia, and Slovenia, ex-
plicitly familialist in Poland, supportive of optional
familialism in Hungary and Lithuania, and defam-
ilializing in Estonia and Latvia. Focusing on the
Baltics, Aidukaite (2021) instead differentiates
Latvia as more traditional than Estonia and Lith-
uania. Javornik (2014) broadly agrees that Lithua-
nian family policies are employment-enabling while
policies in Czechia and Poland support women’s
caregiving; yet Javornik also argues that Estonia,
Latvia, and Hungary promote women’s caregiving,
whereas Slovenia incentivizes maternal employment
and active fatherhood. Slovenia does indeed boast
high female labour force participation (Javornik,
2014); and, despite their differences, Baltic coun-
tries share medium-high (full-time) employment
rates for women. By contrast, the Visegr´
ad states of
Czechia, Hungary, Slovakia, and Poland fall far
behind (for example, Th´
evenon, 2011).
A welfare state paradox?
Another relevant literature forwards the idea of a
welfare state ‘paradox’(Mandel and Semyonov,
2006) or gendered ‘trade-offs’(Cooke, 2011;Pettit
and Hook, 2009). According to these arguments,
although large welfare states bring more women into
the workforce –including women with low earnings
potential –they can have adverse consequences for
women’s career and earnings progression, with high-
skilled women suffering the largest penalties. So,
while the leaner liberal welfare regime is less in-
clusive, it does a better job than its social-democratic
counterpart of promoting women’s access to male-
dominated sectors and occupations, including the top
jobs and wages (Mandel and Semyonov, 2006).
Added to this, the liberal tenets of non-intervention in
markets and treating women ‘the same’as men have
underpinned extensive anti-discrimination and
equal-opportunity legislation in Anglophone coun-
tries, which further supports high-skilled women’s
career development (Mandel, 2009).
Part of this ‘paradox’reflects a selection effect. It
is plausible that under more generous, employment-
supportive family policies, labour force participation
is higher among women with low earnings potential
–who may otherwise be unable to afford market-
based care services –and women who would prefer
to stay at home if policies supported the homemaker
role. Yet, these women are potentially more inclined
to select into less competitive, less lucrative, and
Kowalewska 5
female-dominated occupations and sectors that fit
around their family responsibilities and allow them to
work alongside others with similar employment
behaviours (Hegewisch and Gornick, 2011;Janus,
2013;Korpi et al., 2013;Pettit and Hook, 2009). The
oversupply of women’s labour to a limited range of
occupations further depresses wages (Bergmann,
1974). By contrast, in contexts where fewer
women are in paid work, those women and mothers
who are employed potentially signal strong com-
mitment and career-drive to employers, resulting in
better career and pay rewards than where women’s
employment continuity is the norm (Evertsson and
Grunow, 2012).
Enlarged welfare states also offer more oppor-
tunities for women in feminine-typed public-sector
jobs (Mandel and Semyonov, 2006). High public-
sector wage floors and a higher reservation wage
from more generous social benefits mean that lower-
educated women usually earn more than their peers
in less generous contexts. In the latter, leaner social
provision, weaker unions, and labour market de-
regulation have given rise to a large private service
sector, in which wages are far lower (for example,
Mandel and Shalev, 2009;Shalev, 2008). Never-
theless, limits on deviations from uniform pay scales
for public employees and the dominance of a single
employer (that is, the state) mean that high-level
public-sector jobs typically offer lower salaries
than their private-sector equivalents. Thus, gender
pay gaps at the top of the labour market should be
larger under more versus less generous welfare ar-
rangements (Mandel, 2012;Rubery, 2015;Shalev,
2008).
Institutionalized rights to employment interrup-
tions for motherhood and caregiving are also rele-
vant. Moderate-length leaves (<2 years) promote
mothers’job retention and return, especially if well-
paid (for example, Nieuwenhuis et al., 2017;
Th´
evenon and Solaz, 2013); but leaves can also harm
women’s long-run earnings and career mobility,
particularly when they exceed 12–16 months and are
taken multiple times (Boeckmann et al., 2015;
Evertsson and Duvander, 2011). It becomes harder to
accumulate human capital and experience, maintain
skills and networks, and access opportunities and
training (for example, Boeckmann et al., 2015). Long
leaves are also associated with mothers performing
more housework (Schober, 2013). Relatedly, Gangl
and Ziefle (2015) find that exposure to the full-time
caregiver role can trigger a shift in women’s priorities
away from their careers.
On the demand side, rights to family-related
leave, reduced employment hours, and flexible
working risk increasing employer discrimination
against women in hiring, promotion, and pay deci-
sions (Mandel and Semyonov, 2006). Employers
may view users of leaves and other work–family
policies as less motivated and committed, potentially
amplifying status-based discrimination against
mothers (for example, Petts et al., 2022). Moreover,
such policies risk intensifying statistical discrimi-
nation against mothers if employers generalize from
previous negative experiences with leave-users to all
leave-users. In turn, managers may favour employees
who do not take advantage of work–family policies –
that is, men –irrespective of actual productivity
(Glass, 2004). Even childless women may experi-
ence statistical discrimination if employers presume
that they might in the future have children and make
use of leave and other work–family policies (Jessen
et al., 2019).
High-skilled women may experience more
policy-induced discrimination, since these workers
are harder and often more expensive to replace (for
example, Est´
evez-Abe, 2006;Mandel, 2012;Shalev,
2008). Furthermore, expectations of commitment are
generally greater for high-skilled jobs (England et al.,
2016). High-skilled women in Northern and Conti-
nental Europe may be particularly adversely affected,
where stronger employment protection legislation
makes it harder to hire and fire workers and there is
greater emphasis on on-the-job training. Therefore,
the costs associated with training staff –and losing
said staff, even if only temporarily –are greater
(Est´
evez-Abe, 2006). On top of this, while more
equal wage structures and higher union density and
collective-bargaining coverage in Continental and
Northern European countries improve pay for low-
earning women, these features may limit the wages
that high-skilled educated women can attain (for
example, Mandel and Shalev, 2009;Soskice, 2005).
Conversely, weaker trade unions and more deregu-
lated labour markets in Anglophone countries –and,
6Journal of European Social Policy 0(0)
to a lesser extent, Mediterranean and Eastern Eu-
ropean countries –mean that individual-level traits
should matter more for determining pay (for ex-
ample, Mandel and Shalev, 2009).
It should be noted that the welfare state paradox
and trade-off arguments are based on findings using
data from the 1990s. Certain studies using more
recent data suggest limited evidence for trade-off
arguments (for example, Brady et al., 2020;
Gr¨
onlund and Magnusson, 2016;Korpi et al., 2013).
For example, Korpi et al. (2013) find that women’s
chances of reaching the top jobs and wages are no
different in Nordic countries than elsewhere. An
additional problem with existing literature is its
implicit bias towards Western welfare states. Patterns
for Southern European countries are often based on a
single representative. Likewise, those (few) studies
that include Eastern European countries usually
generalize patterns to this region based on a sample
of two to four post-Socialist countries –most often,
the Visegr´
ad states –with the Baltics and Slovenia
underrepresented (for example, Mandel, 2012;
Mandel and Semyonov, 2006).
Accordingly, as Hook and Li (2020) argue: ‘We
need more research that considers multiple labour
market outcomes, which is at the crux of the welfare
state paradox or trade-off arguments…Studies
should engage both employment [participation] and
other labour market outcomes’(p.260). This article
seeks to do just that. It examines how 24 countries –
representing different family policy approaches and
including three Mediterranean and eight post-Soviet
states –‘fit’together based on their simultaneous
performance across multiple gendered labour market
outcomes.
Measures and approach
The cluster analysis takes Mandel’s (2009) similar
analysis as its starting point. Mandel clusters 14
countries by various dimensions of gender inequality
through a cross-sectional analysis of data covering a
single year within the period of 1990–2002, de-
pending on the indicator. Mandel’s results largely
align with institutional typologies, identifying a
social-democratic, a liberal, and a conservative
cluster, the latter of which includes Italy and Spain.
The results also reflect the tensions highlighted in the
welfare state paradox literature.
In contrast to Mandel’s(2009)analysis, the present
study incorporates more countries (24 versus 14) with
data on all indicators for all countries (an unac-
knowledged problem with Mandel’sanalysisisthe
absence of data on occupational segregation in Den-
mark; Mandel ‘estimates’Danish levels based on
those for Sweden, with no information provided on the
bases of this estimation). I use data from 2010–19,
taking the means of all 10 years (or for as many years
for which data are available). The selection of clus-
tering variables is also more conceptually relevant and
robust. Mandel (2009) conceives of gender inequality
broadly within society and includes such (albeit im-
portant) variables as the lone parent poverty rate and
share of male-breadwinner couples. By contrast, the
present article is delimited to gender inequalities in
economic employment outcomes.Itisbasedonfive
key constructs that reflect theoretically relevant di-
mensions of women’s relative economic employment
position across advanced economies, as identified in
prior research (for a review, see Ferragina, 2020). To
tap into the classed dimensions, variables include
those relevant to both low-/medium- and high-skilled
women.
Table S3 in the Supplemental Material summa-
rizes the key constructs and data sources. To avoid
giving undue weight to certain constructs over
others, I measure each construct strictly by two
variables and only include variables that are not
significantly and strongly correlated with each other
(note that Mandel did not check for multi-
collinearity). As Table S4 (Supplemental Material)
illustrates, variables are either not significantly as-
sociated with each other or only weakly to moder-
ately associated (r< 0.70).
The first construct concerns women’s participa-
tion in employment. Variable (a) is women’s em-
ployment rate relative to that of men for those with
below university education, while (b) considers this
gap for tertiary-educated persons. The second con-
struct covers gender gaps in employment intensity
for (c) part-time employment (1–29 hours per week)
and (d) long-hours employment (40 + hours per
week). The third construct, ‘vertical segregation’,
concerns women’s under-representation and men’s
Kowalewska 7
Figure 1. Gendered employment patterns across 24 high-income countries (2010–19). Based on the following variables:
gender gaps in employment participation for (a) low and medium educated and (b) highly educated; gender gaps in (c)
part-time (1–29 hours per week) and (d) long-hours employment (40+ per week); shares of women in (e) management
and (f) board positions; gender gaps in (g) private-sector and (h) service employment; gender pay gaps for (i) low and
medium educated and (j) highly educated.
Sources: Own calculations using the OECD’s Family Database, Gender Data Portal, Employment Database, and Education at a Glance Database,
the International Labour Organisation’s modelled estimates, the International Social Survey Programme Work Orientations module, and Eurostat.
8Journal of European Social Policy 0(0)
over-representation in more ‘desirable’top-level
occupations, namely, (e) managerial occupations
and (f) corporate board positions. The fourth con-
struct, ‘horizontal segregation’, is women’s under-/
over-representation in certain occupations or sectors
that are not ordered by any criterion; hence, it is more
a measure of ‘difference’than ‘inequality’(Meulders
et al., 2010). Variable (g) is women’s private-sector
under-representation, while (h) is women’s concen-
tration in service jobs. Finally, the fifth construct
concerns gender pay gaps for full-time employees
with (i) low/medium education levels and (j) high
education levels.
Where appropriate, clustering variables are based
on ratios to capture women’s position relative to men
and limit the influence of cross-national variation in
economic conditions (Cho, 2014). Odds ratios are
less intuitive and more liable to producing extreme
values that exaggerate gender differences (Davies
et al., 1998). Similarly, segregation indices cannot
indicate whether ‘pockets’of high or low segregation
are skewing the overall index score, or whether these
pockets are concentrated in low-paying, low-status
occupations versus high-paying, high-status ones
(Chang, 2000). Of course, ratios are relative mea-
sures that are squeezed or expanded from two sides,
meaning a smaller ratio may reflect a poor situation
for women and men (Jones et al., 2018). Therefore, I
pay attention to the broader context of the ratios
when interpreting the results.
I clustered countries using standardized data and
hierarchical clustering on principal components. This
approach combines three methods to obtain a better
cluster solution: first, a principal component analysis
reduces the data to a smaller set of dimensions for a
more reliable clustering solution later; second, a
hierarchical agglomerative cluster analysis (Ward’s
method) is performed on the principal components;
and third, k-means clustering improves the previous
partition (Otero et al., 2021).
Results
The factor map (Figure 1A) shows each country’s
positioning on the principal components and their
spatial clustering. The dendrogram (Figure 1B) gives
the results of the hierarchical cluster analysis. Table 1
summarizes the final five clusters. Analysis of var-
iance tests verify the robustness of these five clusters,
as differences between them are statistically signif-
icant on all variables. The only exception is the
gender pay gap for low- and medium-educated
persons. This reflects Estonia’s much wider gap
compared with all other countries and even its own
cluster; thus, Table 1 presents the analysis of variance
results with this caveat in mind.
Note that wide gender employment participation
gaps mark Greece out as an outlier. As prior research
has shown, these gaps –which were large pre-2008
crisis and only preserved by severe austerity –persist
throughout the life course. In Greece, young women
face difficulties in integrating into a first job, espe-
cially low-skilled women due to a lack of labour
demand. Thereafter, mothers receive little structural
support to keep their jobs, while a low retirement age
promotes women’s early exit from employment
(Karamessini, 2013).
The first gendered employment pattern, labelled
part time, contains the Continental European coun-
tries (except France) (Figure 1 and Table 1).
Women’s outnumbering of men in part-time jobs by
more than 4:1 may help to explain high vertical
segregation in this cluster. Research indicates that
part-time workers must often take jobs for which they
are overqualified, since there are fewer part-time
opportunities in higher-level occupations (for ex-
ample, Connolly and Gregory, 2008). What is im-
portant, though, is that previous research has
highlighted class distinctions in women’s experi-
ences of part-time employment: ‘good’part-time
jobs are typically undertaken (temporarily) by
middle-class professionals, whereas working-class
women are more often in precarious and poorly
paid part-time jobs (Stovell and Besamusca, 2022).
In Germany, for instance, the odds of having a ‘mini-
job’are 80% higher for low-skilled versus high-
skilled women; yet mini-jobs are associated with
lower pay and social security (Pfau-Effinger and
Reimer, 2019). Furthermore, for low-educated
women, part-time employment is rarely a ‘step-
pingstone’to full-time employment (Fagan et al.,
2014). Potentially, then, this cluster is less advan-
tageous for lower-educated than for high-skilled
women.
Kowalewska 9
For full-timers, gender pay gaps in the part time
cluster are narrower than in most others (Tabl e 1).
Given that reduced-hours employment is widely a
tool for employment–family reconciliation in Con-
tinental Europe (Matteazzi et al., 2018), it is plausible
that those women who select into full-time em-
ployment are less likely to have care responsibilities
and/or conform more closely with the ‘ideal’(male)
unencumbered worker norm (Acker, 1990). Mothers
and women who are full-time employed in a context
of high female part-time employment rates also
signal strong commitment to their employer, which
may underpin better career and pay rewards
(Evertsson and Grunow, 2012).
Spain slots into the part time cluster, too (Figure
1), although its similarities with the Continental
European countries are likely underpinned by dif-
ferent processes and carry different meaning. In
Spain, employers have expanded part-time jobs
mainly to increase labour flexibility since 2008,
particularly in marginal, service-sector occupations.
Indeed, research suggests that women’s concentra-
tion in part-time employment in Spain is less re-
flective of its use as a tool for work–family
reconciliation, and more symptomatic of involuntary
underemployment –especially given the low quality
of part-time jobs in Spain by European standards (for
example, Buddelmeyer et al., 2005).
The Anglophone countries consist of a second
group labelled lower segregation. Countries in this
cluster share below-average levels of gender em-
ployment segregation but are more diverse when it
comes to gender gaps in participation and pay. De-
spite similar gaps at high education levels, the av-
erage gender employment participation gap at lower
education levels is larger in the US than in Australia,
Canada, and the UK. These differences may reflect
lower demand for part-time employees and a rela-
tively less generous mix of family policies and social
support in the US (for example, Sayer and Gornick,
2012). Additionally, Canada and the US have larger
gender pay gaps among full-time workers than
Australia and the UK. Again, this may reflect more
widespread part-time employment in Australia and
the UK, as well as lower wage dispersion (Gornick
and Jacobs, 1996). Even so, all four Anglophone
countries have in common narrow class differences
in gendered pay penalties, i.e., the gender pay gap
varies minimally by education level.
An unexpected finding is the placement of France
in this same group (Figure 1). Compared with other
Continental European countries, gender gaps in
employment intensity are smaller in France; yet
France also has lower representation of women in
services and private-sector employment. Most sig-
nificantly, large gender pay gaps in France, despite
lower vertical segregation, push it into the lower
segregation cluster.
The Nordic countries fall into the high public-
sector participation cluster (Figure 1 and Table 1).
This cluster is characterized by a highly inclusive
labour market (although Finland’s positioning may
be unduly influenced by comparatively low male
employment: for instance, the employment rate for
low-skilled men was, on average, 60% over 2010–19
versus 68–73% for the other Nordic countries).
However, such inclusivity is at the price of women’s
reliance on the public sector for employment and
underrepresentation in private-sector employment. In
addition, while women in the Nordic cluster are less
concentrated in part-time employment than in other
clusters, the long-hours gender gap is the largest,
with men around 2.5 times more likely to be em-
ployed 40 + hours per week than women (except in
Sweden, where the ratio is 1.5).
Still, the Nordic group is a relatively high per-
former on most outcomes (Table 1). Despite evi-
dence of a ‘glass ceiling’effect in this group –that is,
a larger gender wage gap at higher versus lower
educational levels –the high-skilled gender pay gap
is still smaller than in the lower segregation (An-
glophone) cluster and only marginally wider than in
the part time (Continental European) cluster. The
high public-sector participation group also has the
highest share of women on boards and a similar share
of women in management as the part time cluster,
even if female directors and managers in Nordic
countries are possibly more concentrated in the
public sector, where maximum pay is usually lower.
Finally, the eight Eastern European countries
bifurcate into two distinct gendered employment
clusters (Figure 1 and Table 1). The low participation
and progression cluster comprises the Visegr´
ad
countries plus Estonia and is generally an all-round
10 Journal of European Social Policy 0(0)
Table 1. Profile characteristics (unstandardized data) of the gendered employment patterns and significance of differences between them based on a
hierarchical clustering on principal components of 24 high-income countries (2010–19).
All
countries
I. Part time
conservative &
Spain
II. Lower
segregation
liberal & France
III. High public-
sector
participation
Nordic
IV. Low participation
and progression
Visegr´
ad (& Estonia)
V. Long hours
Latvia, Lithuania,
Portugal, &
Slovenia
Greece
(outlier)
ANOVA
F-test
(a) Employment
participation
gap: below
university
education
0.81 0.82 0.79 0.86 0.77 0.85 0.61 7.460**
(b) Employment
participation
gap: university
educated
0.92 0.92 0.91 0.96 0.88 0.97 0.87 10.726**
(c) Part-time
employment
gap
2.65 4.15 2.69 1.90 2.38 1.89 2.37 17.446**
(d) Long-hours
employment
gap
0.68 0.52 0.56 0.48 0.89 0.95 0.83 18.735**
(e) Women in
management
34.44% 31.06% 36.62% 32.63% 34.63% 39.30% 27.34% 3.420*
(f) Women on
boards
21.59% 20.84% 27.05% 31.52% 13.53% 19.08% 8.75% 7.982**
(g) Private-sector
employment
gap
0.82 0.90 0.87 0.67 0.80 0.81 0.94 21.361**
(h) Service
employment
gap
1.43 1.40 1.34 1.39 1.58 1.46 1.23 11.609**
(i) Gender pay gap:
below
university
education
22.45% 19.61% 25.84% 18.47% 21.51% (38.35%)* 22.43% 23.62% 4.042*
(continued)
Kowalewska 11
Table 1. (continued)
All
countries
I. Part time
conservative &
Spain
II. Lower
segregation
liberal & France
III. High public-
sector
participation
Nordic
IV. Low participation
and progression
Visegr´
ad (& Estonia)
V. Long hours
Latvia, Lithuania,
Portugal, &
Slovenia
Greece
(outlier)
ANOVA
F-test
(j) Gender pay gap:
university
educated
25.26% 22.44% 26.44% 23.43% 30.63% 23.32% 21.64% 3.628*
Additional variables
Gender gap in
employment
(all education
levels)
0.88 0.87 0.88 0.95 0.83 0.94 0.70 13.101**
Gender gap in
industry
employment
0.35 0.32 0.29 0.27 0.44 0.44 0.36 19.601**
Gender gap in
public-sector
employment
1.86 1.46 1.84 2.65 1.82 1.83 1.24 14.678**
Maternal
employment
rates
70.32% 70.73% 69.48% 79.41% 62.58% 77.43% 55.36% 8.165**
Notes: *The figure in brackets is for Estonia only to reflect its difference from the Visegr ´
ad countries on this indicator; in addition, the ANOVA test in this row is based on treating
Estonia as an outlier. If Estonia were not treated as an outlier, the ANOVA F-test value would be 1.330 (p> 0.50). For variables a-d, g, and h, <1 means there are fewer women than
men, 1 indicates equal numbers, and >1 indicates that women outnumber men; for example, a value of 2.65 means that for every male employee, there are 2.65 female employees.
(a) Share of all lower and medium-educated (ISCED 0–4) persons in the workforce, ratio of women to men. (b) Share of all highly educated (ISCED 5–8) persons in the workforce,
ratio of women to men. (c) Share of employed persons working part time (1–30 hours per week), ratio of women to men. (d) Share of employed persons working 40+ hours per
week, ratio of women to men. (e) Share of managerial positions filled by women. (f) Share of board positions filled by women. (g) Ratio of employed women to employed men in
private-sector jobs. (h) Ratio of employed women to employed men in service jobs. (i) Difference in men’s and women’s average pay as a percentage of men’s pay for lower and
medium educated (ISCED 0–4) full-time employees. (j) Difference in men’s and women’s average pay as a percentage of men’s pay for highly educated (ISCED 5–8) full-time
employees. ‘Additional variables’are those that were not included in the cluster analysis, but on which differences between the clusters are statistically significant. *p< 0.5; **p<
0.001.
Sources: Own calculations using the OECD’s family Database, gender data portal, employment Database, and education at a glance Database, the international labour Or-
ganisation’s modelled estimates, the international social survey programme work orientations module, and Eurostat.
12 Journal of European Social Policy 0(0)
poor performer. By contrast, Latvia, Lithuania, and
Slovenia are in the long hours group, which features
high female labour force participation rates and less
intense gender inequalities and segregation. As the
name suggests, most (85%) employed women are in
40-hour jobs or longer (versus 90% of employed
men). Although both clusters display similar gender
pay gaps for lower-skilled groups, the long
hours cluster has a narrower high-skilled gender pay
gap. Again, though, context matters: it is plausible
that part of the explanation for the long hours
cluster’s better performance is the asymmetric effects
of the 2008 global financial crisis on male-dominated
sectors and wages, which hit this group of countries
particularly hard, rather than ‘true’equality per se
(Perugini and Selezneva, 2015).
Portugal also sits in the long hours cluster (Figure
1). Its positioning here likely reflects the afore-
mentioned cultural and policy legacies that have
underpinned high female (full-time) employment in
Portugal since the latter half of the twentieth century
(for example, Tavora, 2012). Once more, though,
these narrower gender gaps should be interpreted
within their context. Low wages and limited part-
time jobs create strong financial imperatives for
women in Portugal to take full-time jobs, despite
stark inequality in the division of labour within the
home (Tavora, 2012;Table S2 in the Supplemental
Material) and low job quality (for example, Esser and
Olsen, 2011). What is more, although Portugal
(slightly) outperforms Spain on women’s managerial
representation (34% versus 31%), Portugal lags
behind Spain on other measures, namely, women’s
representation in board positions (13% versus 17%)
and private-sector employment (0.78 versus 0.93) as
well as the high-skilled gender pay gap (29% versus
17%).
Discussion
Women are disadvantaged relative to men across all
employment outcomes in the 24 countries considered
here. Still, the analysis has highlighted nuances
across borders. By and large, the patterning of
gendered employment outcomes across Nordic,
Anglophone, and Continental European countries
mirrors the three-fold institutional typology
identified by Esping-Andersen (1990). This suggests
that women’s employment outcomes are a relatively
good proxy of welfare state and family policy ap-
proaches and logics in these contexts.
Against existing institutional typologies, though,
this article suggests that France is more similar to
Anglophone countries than to other Continental
countries or Belgium when it comes to gendered
employment outcomes. This result mainly reflects
large gender pay gaps across the educational divide
in France. Although gender pay disparities are
shaped by a multitude of circumstances, one insti-
tutional feature that stands out is collective bar-
gaining. In France, wage bargaining happens
primarily at the industry and company levels and so
is less centralized than in other Continental European
countries, especially Belgium (for example, Berson
and Jousselin, 2018). It is often argued that high-
skilled women benefit from the tighter link between
individual human capital endowments and pay under
decentralisation, even if low-skilled women do not
(for example, Soskice, 2005). Equally, though, an
absence of transparent and fair systems in decen-
tralized wage-bargaining contexts may allow for
increased managerial discretion and discrimination
in pay-setting under the guise of market forces;
hence, even high-skilled women can face difficulties
in attaining similar wages as men versus if wage
determination is centralized (Rubery and Johnson,
2019). These differences may help to understand
why gender pay gaps in France exceed those in
Belgium among both the university educated (22%
versus 16%) and below university educated (28%
versus 18%), despite these two countries’similar
family policy logics.
In addition, the results indicate that institutional
typologies are less instructive for understanding
gendered employment outcomes across Mediterra-
nean and Eastern European countries. While Spain is
in the part time cluster with Continental European
countries, Portugal’s legacy of high employment
among women but limited part-time opportunities
place it in the long hours group alongside Latvia,
Lithuania, and Slovenia. Meanwhile, gaping em-
ployment participation gaps plus high segregation
mark Greece out as an outlier. These diverse patterns
likely reflect certain shared particularities of the
Kowalewska 13
Mediterranean context, such as low wages, high (and
gendered) labour market segmentation, and tradi-
tional gender roles in the household (Tavora, 2012).
Still, grouping Southern European nations together
would clearly obfuscate significant differences in
women’s employment, thereby weakening empirical
research and theory-building. The same holds for the
eight Eastern European countries analysed here,
which bifurcate into a low participation and pro-
gression cluster (Visegr´
ad plus Estonia) versus a long
hours, somewhat better-performing cluster (Latvia,
Lithuania, Slovenia). Mainstream comparative ana-
lyses of gender, welfare states, and employment
commonly draw generalizations about the Eastern
European region based on a small sample of post-
Soviet countries –most often, the Visegr´
ad states.
Yet, the patterns presented here suggest this approach
may mislead.
Of course, some of these differences from insti-
tutional typologies may be because certain countries
have adapted their policies and experienced labour
market changes since many policy typologies were
devised. Still, in bringing to light marked differences
in gendered employment outcomes across countries
with similar institutional and cultural contexts, this
study offers a framework for the comparative anal-
ysis of gender and employment that is conducive to
hypothesis-testing and theory construction.
In addition to highlighting certain shortcomings
of institutional typologies for analysing women’s
employment outcomes, the results cast doubt on
welfare state paradox arguments. Comparatively
stronger scores on measures of women’s labour
market inclusion do not always coincide with
comparatively worse scores on other labour market
outcomes, including for high-skilled women. The
high public-sector participation group, comprising
Nordic countries, is more inclusive with narrower
pay gaps among low- and medium-educated groups
than the part time cluster comprising Continental
European countries; however, the former performs
similarly to if not better than the latter on women’s
share of managerial and board positions and high-
skilled gender pay gaps. The high public-sector
participation group also has narrower gender em-
ployment participation gaps than the lower segre-
gation group, to which the Anglophone countries and
France belong, while still displaying narrower gender
pay gaps at tertiary education levels. True, women’s
managerial representation in the high public-sector
participation cluster falls behind that of the lower
segregation one; even so, women’s share of board
seats is higher in the former, reflecting more decisive
legislation to correct for women’s historic absence
from the boardroom (Kowalewska, 2021).
Gendered employment patterns across Eastern
Europe cast further doubt on welfare state paradox
arguments. The long hours cluster (Latvia, Lithuania,
Slovenia) is more inclusive of lower-educated
women and has higher representation of women in
top jobs as well as narrower gender pay gaps among
the high-skilled than the low participation and
progression cluster (Visegr´
ad plus Estonia). Addi-
tionally, women in the long hours group are less
concentrated in services or reduced-hours employ-
ment than in the low participation and progression
cluster. By illustrating how the interrelationships
between the different indicators in one cluster of
countries are not necessarily replicated in others,
these results support calls for more studies that il-
luminate the context-dependence of women’s em-
ployment outcomes (Hook and Li, 2020).
The results also underline the value of separately
‘counting’a range of dimensions and measures of
gender inequality. Clearly, boiling gender equality
down to labour force participation rates or another
single measure or index has the potential to misrep-
resent a country’s performance. Relatedly, the results
suggest that it does not always make sense to talk of
certain countries or regime types as more or less
‘women-friendly’, since it depends on which indi-
cator(s) of gender equality we are consulting and for
whom. For instance, even in the high-participation
Nordic group, women are significantly underrepre-
sented in private-sector and long-hours positions,
which may underpin their low presence in certain
careers carrying high pay and influence (for example,
lawyer in private practice, Chief Executive of an
investment firm). This insight additionally under-
scores the need to always be explicit, transparent, and
reflective about how we are conceptualizing and
operationalizing gender inequality. Precision facili-
tates meaningful analysis and the accumulation of
debate and knowledge-building by helping to make
14 Journal of European Social Policy 0(0)
sure we are not reaching conflicting conclusions based
simply on our different underlying research choices
(Kantola and Verloo, 2018). It can also aid more
targeted policymaking.
Overall,the results give a picture of pre-COVID-19
gender inequalities in employment, which can provide
a reference point for research on the longer-term
gendered employment impacts of the pandemic as
they continue to unfurl and new data emerge. Nev-
ertheless, cross-national data that break down em-
ployment outcomes by gender intersected with other
factors are required to account for the heterogeneity of
women. This is important for better understanding
how employment outcomes are shaped not only by
gender, but simultaneously by its intersection with
other dimensions beyond those that could be explored
here, such as ethnicity, migration status, and sexual
orientation.
ORCID iD
Helen Kowalewska https://orcid.org/0000-0002-7991-5371
Acknowledgements
The author is grateful to two anonymous referees, the journal
editors, and colleagues from the Department of Social Policy
and Intervention at the University of Oxford for their in-
valuable comments on earlier versions of this work.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest
with respect to the research, authorship, and/or publication
of this article.
Funding
The author(s) disclosed receipt of the following financial
support for the research, authorship, and/or publication
of this article: This research was supported by the
Economic and Social Research Council (grant no. ES/
S016058/1).
Supplemental Material
Supplementary material for this article is available online.
Notes
1. Although important, unpaid work is beyond this arti-
cle’s scope.
2. For example, welfare state support for women’sem-
ployment is commonly operationalized by female labour
force participation rates (for example, Bambra, 2007).
References
Aidukaite, J (2021) Support to families with children in the
Baltic states. In: B Greve (ed). Handbook on Aus-
terity, Populism and the State. Cheltenham: Edward
Elgar, 221–242.
Acker, J (1990) Hierarchies, jobs, and bodies. Gender and
Society 4(2): 139–158.
Anxo, D, Fagan, C, Cebrian, I, et al. (2007) Patterns of
labour market integration in Europe. Socio-Economic
Review 5(2): 233–260.
Bambra, C (2007) Defamilization and welfare state re-
gimes: a cluster analysis. International Journal of
Social Welfare 16(4): 326–338.
Bergmann, BR (1974) Occupational segregation, wages,
and profits when employers discriminate by race or
sex. Eastern Economic Journal 1(2): 103–110.
Berson, C and Jousselin, ´
E (2018) Wage bargaining in
Europe. Banque de France Bulletin 217(2): 1–9.
Boeckmann, I, Misra, J and Budig, MJ (2015) Cultural and
institutional factors shaping mothers’employment
and working hours in postindustrial countries. Social
Forces 93(4): 1301–1333.
Brady, D, Blome, A and Kmec, JA (2020) Work–family
reconciliation policies and women’s and mothers’
labour market outcomes in rich democracies. Socio-
Economic Review 18(1): 125–161.
Buddelmeyer, H, Mourre, G and Ward-Warmedinger, M
(2005) Part-Time Work in EU Countries. Bonn: IZA.
Chang, ML (2000) The evolution of sex segregation regimes.
American Journal of Sociology 105(6): 1658–1701.
Cho, EY-N (2014) Defamilization typology re-examined:
re-measuring the economic independence of women
in welfare states. Journal of European Social Policy
24(5): 442–454.
Connolly, S and Gregory, M (2008) Women’s part-time
work and occupational change in Britain 1991–2001.
Economic Journal 118(526): F52–F76.
Kowalewska 15
Cooke, LP (2011) Gender-Class Equality in Political
Economies. London: Routledge.
Daly, M (1994) Comparing welfare states: towards a
gender-friendly approach. In: D Sainsbury (ed)
Gendering Welfare States, 101–117. London: Sage.
Davies, HT, Crombie, IK and Tavakoli, M (1998) When
can odds ratios mislead? BMJ 316(7136): 989–991.
England, P, Bearak, B, Budig, MJ, et al. (2016) Do highly
paid, highly skilled women experience the largest
motherhood penalty? American Sociological Review
81(6): 1161–1189.
Esping-Andersen, G (1990) The Three Worlds of Welfare
Capitalism. Cambridge: Polity Press.
Esping-Andersen, G (1999) Social Foundations of Postin-
dustrial Economies. Oxford: Oxford University Press.
Esping-Andersen, G (2016) Families in the 21st Century.
Stockholm: SNS F¨
orlag.
Esser, I and Olsen, KM (2011) Perceived job quality: au-
tonomy and job security within a multi-level frame-
work. European Sociological Review 28(4): 443–454.
Est´
evez-Abe, M (2006) Gendering the varieties of capitalism:
a study of occupational segregation by sex in advanced
industrial societies. World Politics 59(1): 142–175.
Evertsson, M and Duvander, A.-Z (2011) Parental leave –
possibility or trap? European Sociological Review
27(4): 435–450.
Evertsson, M and Grunow, D (2012) Women’s work in-
terruptions and career prospects in Germany and
Sweden. International Journal of Sociology and
Social Policy 32(9/10): 561–575.
Fagan, C and Rubery, J (2018) Advancing gender equality
through European employment policy: the impact of
the UK’s EU membership and the risks of Brexit.
Social Policy and Society 17(2): 297–317.
Fagan, C, Norman, H, Smith, Met al. (2014) In Search of
Good Quality Part-Time Employment. Geneva: In-
ternational Labour Office.
Ferragina, E (2019) Does family policy influence women’s
employment? Reviewing the evidence in the field.
Political Studies Review 17(1): 65–80.
Ferragina, E (2020) Family policy and women’s em-
ployment outcomes in 45 high-income countries.
Social Policy and Administration 54(7): 1016–1066.
Ferragina, E, Seeleib-Kaiser, M and Spreckelsen, T (2015)
The four worlds of ‘welfare reality’–social risks and
outcomes in Europe. Social Policy and Society 14(2):
287–307.
Gangl, M and Ziefle, A (2015) The making of a good
woman: extended parental leave entitlements and
mothers’work commitment in Germany. American
Journal of Sociology 121(2): 511–563.
Glass, J (2004) Blessing or curse? Work–family policies
and mother’s wage growth over time. Work and
Occupations 31(3): 367–394.
Gornick, JC and Jacobs, JA (1996) A cross-national
analysis of the wages of part-time workers. Work,
Employment and Society 10(1): 1–27.
Gornick, JC and Jacobs, JA (1998) Gender, the welfare
state, and public employment. American Sociological
Review 63(5): 688–710.
Gr¨
onlund, A and Magnusson, C (2016) Family-friendly
policies and women’s wages –is there a trade-off?
European Societies 18(1): 91–113.
Hegewisch, A and Gornick, JC (2011) The impact of
work–family policies on women’s employment.
Community, Work and Family 14(2): 119–138.
Hook, J (2015) Incorporating ‘class’into work–family
arrangements. Journal of European Social Policy
25(1): 14–31.
Hook, JL and Li, M (2020) Gendered tradeoffs. In: R
Nieuwenhuis and W van Lancker (eds). The Palgrave
Handbook of Family Policy. Cham: Palgrave Mac-
millan, 249–266.
Hopkin, J and Wincott, D (2006) New Labour, economic
reform and the European social model. British
Journal of Politics and International Relations 8(1):
50–68.
Janus, AL (2013) The implications of family policy re-
gimes for mothers’autonomy. Research in Social
Stratification and Mobility 34: 96–110.
Javornik, J (2014) Measuring state de-familialism: con-
testing post-socialist exceptionalism. Journal of Eu-
ropean Social Policy 24(3): 240–257.
Jessen, J, Jessen, R and Kluve, J (2019) Punishing potential
mothers? Evidence for statistical employer discrimi-
nation from a natural experiment. Labour Economics
59: 164–172.
Jones, M, Makepeace, G and Wass, V (2018) The UK
gender pay gap 1997–2015: what is the role of the
public sector? Industrial Relations 57(2):
296–319.
Kantola, J and Verloo, M (2018) Revisiting gender equality
at times of recession. European Journal of Politics
and Gender 1(1–2): 205–222.
16 Journal of European Social Policy 0(0)
Karamessini, M (2013) Structural crisis and adjustment in
Greece. In: M Karamessini and J Rubery (eds).
Women and Austerity. Abingdon: Routledge,
165–185.
Karamessini, M and Rubery, J (2013) Women and Aus-
terity: The Economic Crisis and the Future for
Gender Equality. Abingdon: Routledge.
Klammer, U and Letablier, M.-T (2007) Family policies in
Germany and France: the role of enterprises and social
partners. Social Policy and Administration 41(6):
672–692.
Korpi, W (2000) Faces of inequality: gender, class, and
patterns of inequalities in different types of welfare
states. Social Politics 7(2): 127–191.
Korpi, W, Ferrarini, T and Englund, S (2013) Women’s
opportunities under different family policy constel-
lations. Social Politics 20(1): 1–40.
Kowalewska, H (2021) Bringing women on board? Family
policies, quotas and gender diversity in top jobs.
Work, Employment and Society 35(4): 735–752.
Kuitto, K (2011) More than just money: patterns of dis-
aggregated welfare expenditure in the enlarged Eu-
rope. Journal of European Social Policy 21(4):
348–364.
Land, H (1971) Women, work and social security. Social
Policy and Administration 5(3): 183–192.
Leitner, S (2003) Varieties of familialism: the caring
function of the family in comparative perspective.
European Societies 5(4): 353–375.
Leitner, S (2014) Varieties of familialism: developing care
policies in conservative welfare states. In: P Sandermann
(ed) The End of Welfare as We Know It? Leverkusen-
Opladen: Verlag Barbara Budrich, pp. 37–52.
Lewis, J (1992) Gender and the development of welfare
regimes. Journal of European Social Policy 2(3):
159–173.
Lewis, J (1997) Gender and welfare regimes: further
thoughts. Social Politics 4(2): 160–177.
Lewis, J and Ostner, I (1995) Gender and the evolution of
European social policy. In: P Pierson and S Leibfried
(eds) European Social Policy. Washington: Brookings
Institution, pp. 159–193.
Lister, R (1994) ‘She has other duties’: women, citizen-
ship, and social security. In: S Baldwin and J
Falkingham (eds). Social Security and Social
Change. Hemel Hempstead: Harvester Wheatsheaf,
31–44.
Lombardo, E, Meier, P and Verloo, M (2009) The Dis-
cursive Politics of Gender Equality. London:
Routledge.
Maˆ
ıtre, B, Nolan, B and Whelan, CT (2005) Welfare re-
gimes and household income packaging in the Eu-
ropean Union. Journal of European Social Policy
15(2): 157–171.
Mandel, H (2009) Configurations of gender inequality: the
consequences of ideology and public policy. British
Journal of Sociology 60(4): 693–719.
Mandel, H (2012) Winners and losers: the consequences of
welfare state policy for gender wage inequality. Eu-
ropean Sociological Review 28(2): 241–262.
Mandel, H and Semyonov, M (2006) A welfare state
paradox: state interventions and women’s employ-
ment opportunities in 22 countries. American Journal
of Sociology 111(6): 1910–1949.
Mandel, H and Shalev, M (2009) Gender, class and va-
rieties of capitalism. Social Politics 16(2): 161–181.
Matteazzi, E, Pailh´
e, A and Solaz, A (2018) Part-time
employment, the gender wage gap and the role of
wage-setting institutions. European Journal of In-
dustrial Relations 24(3): 221–241.
Meulders, D, Plasman, R, Rigo, A, et al (2010) Horizontal
and Vertical Segregation. Brussels: Universite Libre
de Bruxelles.
Misra, J, Budig, MJ and Moller, S (2007) Reconciliation
policies and the effects of motherhood on employ-
ment, earnings and poverty. Journal of Comparative
Policy Analysis 9(2): 135–155.
Nieuwenhuis, R, Need, A and Van der Kolk, H (2017) Is there
such a thing as too long childcare leave? International
Journal of Sociology and Social Policy 37(1/2): 2–15.
O’Connor, J (1993) Gender, class and citizenship in the
comparative analysis of welfare state regimes. British
Journal of Sociology 44(3): 501–518.
Orloff, AS (1993) Gender and the social rights of citi-
zenship. American Sociological Review 58(3):
303–328.
Otero, G, Carranza, R and Contreras, D (2021) Spatial
divisions of poverty and wealth: does segregation
affect educational achievement? Socio-Economic
Review. Online First. DOI: 10.1093/ser/mwab022.
Perugini, C and Selezneva, E (2015) Labour market in-
stitutions, crisis and gender earnings gap in Eastern
Europe. Economics of Transition and Institutional
Change 23(3): 517–564.
Kowalewska 17
Pettit, B and Hook, JL (2009) Gendered Tradeoffs. New
York: Russell Sage.
Petts, RJ, Mize, TD and Kaufman, G (2022) Organisational
policies, workplace culture, and perceived job com-
mitment of mothers and fathers who take parental
leave. Social Science Research 103(2022): 102651.
Pfau-Effinger, B and Reimer, T (2019) The interplay of
welfare state policies with supply- and demand-side
factors in the production of marginalised part-time
employment among women in Germany. In: H
Nicolaisen, H Kavli and RS Jensen (eds). Dualisation
of Part-Time Work. Bristol: Policy Press, 245–264.
Rubery, J (2015) Austerity, the public sector and the threat
to gender equality. The Economic and Social Review
46(1): 1–27.
Rubery, J and Johnson, M (2019) Closing the Gender Pay
Gap: What Role for Trade Unions? Geneva: ILO.
Sainsbury, D (1999) Gender and Welfare State Regimes.
Oxford: Oxford University Press.
Saraceno, C (1997) Family Change, Family Policies and
the Restructuring of Welfare. Paris: OECD.
Saraceno, C and Keck, W (2010) Can we identify inter-
generational policy regimes in Europe? European
Societies 12(5): 675–696.
Sayer, LC and Gornick, JC (2012) Cross-national variation
in the influence of employment hours on child care
time. European Sociological Review 28(4): 421–442.
Schmid, C, Cook, R and Jones, L (2022) Measuring gender
inequality in Great Britain: proposal for a subnational
gender inequality index. Social Politics Online
First(23). doi: 10.1093/sp/jxac005.
Schober, P (2013) Maternal labour market return and
domestic work after childbirth in Britain and Ger-
many. Community, Work and Family 16(3): 307–326.
Shalev, M (2008) Class divisions among women. Politics
and Society 36(3): 421–444.
Smith, M and Villa, P (2010) The ever-declining role of
gender equality in the European Employment Strat-
egy. Industrial Relations Journal 41(6): 526–543.
Soskice, D (2005) Varieties of capitalism and cross-
national gender differences. Social Politics 12(2):
170–179.
Stier, H, Lewin-Epstein, N and Braun, M (2001) Welfare
regimes, family-supportive policies, and women’s
employment along the life-course. American Journal
of Sociology 106(6): 1731–1760.
Stovell, C and Besamusca, J (2022) Full-time hours, part-
time work: questioning the sufficiency of working
hours as a measure of employment status. Community,
Work and Family 25(1): 63–83.
Szelewa, D and Polakowski, MP (2008) Who cares?
Changing patterns of childcare in Central and Eastern
Europe. Journal of European Social Policy 18(2):
115–131.
Tavora, I (2012) The Southern European social model:
familialism and the high rates of female employment
in Portugal. Journal of European Social Policy 22(1):
63–76.
Th´
evenon, O (2011) Family policies in OECD coun-
tries. Population and Development Review 37(1):
57–87.
Th´
evenon, O (2016) Do ‘institutional complementarities’
foster female labour force participation? Journal of
Institutional Economics 12(2): 471–497.
Th´
evenon, O and Solaz, A (2013) Labour Market Effects of
Parental Leave Policies in OECD Countries. Paris:
OECD.
Valentova, M (2016) Generation and the propensity of long
career interruptions due to childcare under different
family policy regimes. International Sociology 31(6):
701–725.
Van der Vleuten, A and Verloo, M (2012) Ranking and
benchmarking: the political logic of new regulatory
instruments in the fields of gender equality and anti-
corruption. Policy and Politics 40(1): 73–88.
Zimmerman, MK (2013) Theorising inequality: compar-
ative policy regimes, gender, and everyday lives. The
Sociological Quarterly 54(1): 66–80.
18 Journal of European Social Policy 0(0)
