Family Policies and Fertility - Examining the Link between Family
Policy Institutions and Fertility Rates in 33 Countries 1995-2010
Corresponding author:
PhD student Katharina Wesolowski
Sociology/Baltic and East European Graduate School (BEEGS)
Södertörn University
SE-141 89 Huddinge
Sweden
E-mail: katharina.wesolowski@sh.se
Telephone: +46 8 608 50 95
Associate Professor of Sociology Tommy Ferrarini
Swedish Institute for Social Research
Stockholm University
SE-106 91 Stockholm
Sweden
1
Abstract
In what ways are family policies related to fertility? Previous studies of OECD countries have
arrived at mixed results when analysing the effects of family policy expenditures or formal benefit
rates. This study draws on new institutional family policy data from a wider set of 33 countries in
a multidimensional analysis of the link between family policy institutions and fertility for the years
1995 to 2010. Pooled time-series regressions show that more extensive gender-egalitarian family
policies, i.e. earner-carer support, are linked to higher fertility, while policies supporting more
traditional family patterns as well as the degree of economic development show no statistically
significant effects. Analyses of the interaction between earner-carer support and female labour
force participation indicate that the impact of introducing more gender-egalitarian policies is
stronger in countries with lower levels of female labour force participation. Regressions with
differenced data sustain ideas of earner-carer support being linked to total fertility increase.
Keywords: fertility, family policies, female labour force participation, gender-egalitarian,
gender-traditional
2
Introduction
In the last few decades, total fertility rates have remained below the replacement rate of 2.1
children per woman of fertile age in most affluent countries, causing debate among policymakers,
as well as scholars, about the best ways to reverse, or at least slow down fertility decline. Family
policy legislation has entered the spotlight here. In research on welfare states and family change a
much-debated issue concerns the degree to which family policies impact on fertility. Several
studies indicate that some family policies may result in increases in fertility rates. However, the
empirical evidence has at times been inconclusive, to some extent because of the different ways of
conceptualising and measuring the contents of family policies (see Gauthier 2007).
Many previous studies of fertility outcomes have used family policy expenditures or the
formal pre-tax benefit levels given by family policies as explanatory factors. Although such
approaches have contributed important empirical insights, they are often insufficiently detailed
and fail to differentiate between the theoretically central institutional characteristics of policies.
Here, the institutional, or social-rights approach, built on the legislative structures of social-policy
transfers is proposed as a more precise method to analyse the causes and consequences of welfarestate organisation (see Korpi and Palme 1998). When this approach was extended to family policy
analysis the need to analyse policy in a multidimensional framework was highlighted (see Ferrarini
2003; Korpi 2000; Pettit and Hook 2009; Sainsbury 1999). Korpi´s (2000) approach, for example,
differentiates the gender-egalitarian aspects of family policy from the features supporting a more
traditional gender-division of paid and unpaid work. This approach allows for an analysis of
whether different types of family policy orientations affect childbearing decisions differently. This
is relevant, particularly against the background of the argument that family policies that help both
3
partners to combine work and family life, i.e. more gender-egalitarian family policies, are the best
way to raise fertility levels in low-fertility countries (e.g. McDonald 2006).
Family policies can be linked to childbearing through several pathways. One might expect
that the policies would impact on childbearing behaviour directly by increasing the size of
household budgets, thus decreasing the relative size of the direct costs of children. However,
family policies are also likely to have indirect effects on behaviour. Thus, they could reduce the
opportunity costs of childbearing by making the combination of paid work and family life easier
(see Gauthier and Hatzius 1997). In this context, gender-egalitarian and traditional family policies
are likely to have divergent effects on women’s employment. Gender-egalitarian family policies,
assisting with the combination of paid work and care, are particularly likely to increase female
labour force participation both before and after childbirth (see Ferrarini 2003; Gornick and Meyers
2008; Korpi 2000).
This paper aims to analyse the link between different family policy institutions and
fertility rates 1995-2010 in 33 countries, including both longstanding and newer members of the
EU as well as other post-communist countries. The study thus widens the analyses of recent family
policy development and fertility to also include post-communist countries in Eastern Europe,
where fertility decline has often been substantial. More precisely, the analyses aim to investigate
whether and how gender-egalitarian or traditional family policies are connected to fertility rates.
This is done by employing the multidimensional approach to family policy analysis originally
developed by Korpi (2000).
The following section of this article discusses results of previous studies analysing the
links between family policy and fertility. The third section elaborates upon the theoretical
underpinnings of the family policy dimensions employed in the analysis. The fourth section is
4
devoted to a discussion of data and variables used, while the fifth section describes estimation
methods. The sixth section presents empirical evidence, and the final section discusses the main
results.
Family Policy and Fertility – Previous Research
In what ways can family policy be expected to influence fertility in industrialised countries? There
are several explanations for the long-term fertility decrease. A general rise in income and an
increase in women’s labour force participation and education were for long assumed to introduce
a trade-off between the number of children and the investment in the child’s education. Moreover,
women’s increasing educational attainment and earnings implied that they would be more prone
to choose paid work over childbearing (Barro and Becker 1989; Blossfeld 1995). However, the
links on the country level between economic development and fertility as well as female
employment and fertility appear to have turned from a clearly negative correlation in the 1970s
and the 1980s to a positive one during the most recent decades. The main reason why female
employment and economic development are now positively related to fertility on the country level
has been sought in the proposition that family policies can actively assist with one’s ability to
combine paid work and family life (see Ahn and Mira 2002; d´Addio and d´Ercole 2005).
It has been pointed out that countries where family policies are specifically designed to
support the reconciliation of paid work and family life are the ones that have managed best to
counter the fertility decline (Castles 2004). Moreover, economic and social development may well
be a reason for the “fertility rebound” observed in rich countries. This can be seen in the results of
Myrskylä et al. (2009) showing that the rank correlation between the Human Development Index
5
(HDI) and fertility rates turned into a positive one in those countries where the HDI increased
above 0.9 between 1975 and 2005.
In this context, Luci and Thévenon (2010) argue that the female employment may be the
key factor behind the fertility rebound in rich countries. In an analysis of 30 OECD countries for
the period 1960-2007 they identified female labour market participation as a central factor behind
the increase in fertility rates in some countries. The proposed mechanism is that economic
development increases both the possibility for women to engage in paid work and the possibility
for parents to reconcile paid work and family life (Luci and Thévenon 2010).
Returning to the discussion of family policies, evidence from comparative macro-level
analyses supports the idea that they may influence fertility (Castles 2003; Ferrarini 2003; Gauthier
and Hatzius 1997; Rovny 2011; Ruhm and Teague 1995; Winegarden and Bracy 1995). Gauthier
and Hatzius (1997), for example, find a positive relationship between family allowances on fertility
rates in their study on 22 industrialised countries 1970-1990, even though the magnitude of the
correlation is not high.
Ferrarini (2003) found a positive correlation of both gender-egalitarian and gendertraditional family policies with fertility rates when analysing the impact of family policies on
fertility rates in 18 OECD countries in the period 1970-1995. Another one of his findings is that
gender-egalitarian family policies were connected to higher female labour force participation,
while gender-traditional family policies were connected to lower female labour force participation.
This is interpreted to indicate that gender-egalitarian family policies lower the opportunity costs
for women to be in paid employment, in contrast to traditional family policies that increase these
opportunity costs (Ferrarini 2003).
6
In a review of previous findings, Gauthier (2007) comes to the conclusion that there is
evidence supporting the argument that family policies actually may increase fertility even though
the effects do not appear to be large. However, previous empirical findings have at times been
contradictory, partly due to lack of available data on different types of public family policies and
other measurement and modelling issues. McDonald (2006) argues that policies facilitating the
reconciliation of paid work and child-rearing would be the most viable way to raise fertility, and
also maintains that already small impacts could raise the total fertility rate (TFR) above the lowestlow fertility levels.
A recent study by Luci-Greulich and Thévenon (2013) on family policy in 18 OECD
countries in the period 1982-2007 demonstrates that family policies may increase fertility rates.
Different family policy measures were analysed and the results indicate that each policy instrument
has a positive effect. Spending on cash benefits, on parental leave benefits, on maternity grants
related to childbirth, and enrolment in childcare for children below the age of three were all
positively correlated with fertility rates. An overall conclusion drawn by the authors is that a
combination of different family policies facilitates childbirth although their influence differs
depending on the family policy context in each country. The authors, however, do not go into detail
regarding which particular type of combination would be the most favourable (Luci-Greulich and
Thévenon 2013).
The study of Luci-Greulich and Thévenon (2013) is one of the first studies to include data
on family policy that stretches into the most recent decade. However, their study does not cover
Eastern European countries. Several researchers have discussed the tendency of family policies to
change towards a more familialist or male-breadwinner model in post-communist countries (Ciccia
and Verloo 2012; Saxonberg and Szelewa 2007). Examples of post-communist countries currently
7
applying a male-breadwinner model given by Ciccia and Verloo (2012) are the Czech Republic,
Estonia, Hungary, Latvia, Poland, and Slovakia. In her analysis of 13 post-communist countries,
Rostgaard (2004) also comes to the conclusion that family policies have developed in a more
gender-traditional direction in several of these countries, making the reconciliation of work and
child-rearing harder, especially for women.
However, the development of family policies in Eastern Europe has also been shown to
be quite diverse. It is not necessarily oriented towards a refamilialisation, but in some instances
also emphasises more gender equality (Aidukaite 2006; Billingsley and Ferrarini 2014). For
example, Slovenia´s family policy has in several studies been shown to have clear genderegalitarian features (Billingsley and Ferrarini 2014; Ciccia and Verloo 2012).
Much of the discussion about recent fertility decline in Europe points to the potential role
that can be played by family policy. Against the background of the above, it seems urgent to
expand the analysis of the link between family policies and fertility to post-communist countries.
Do results from longstanding welfare states hold when the systematic comparative analysis is
extended to include also the post-communist countries?
As discussed above, family policies may in several ways impact on fertility as well as on
the potentially important intermediate factor of female employment. One obvious direct effect of
family policy transfers is that they increase the size of the household budget and thus make it easier
to meet the direct costs of children (costs for household goods, education, housing etc.). Here, it is
important to note that family policy legislation also may have indirect effects on childbearing
decisions. On the one hand, they could support paid work (and care) of both parents and thus lower
the opportunity costs for giving birth, especially for women. On the other hand, they could sustain
8
gendered divisions of labour, where women’s main responsibility for care work is traded against
less involvement in paid work (Korpi 2000; Sainsbury 1996).
A Multidimensional Perspective on Family Policy Institutions
Family policy legislation became central in comparative welfare state analysis when gender
perspectives challenged the dominating class-based or structural-economic explanations to
differences between welfare states (Orloff 2009). Feminist critique in particular came to target
Esping-Andersen’s (1990) typology of the “three worlds of welfare capitalism” for neglecting
women’s unpaid work (O´Connor et al. 1999; Orloff 1993). One response was to develop new
gender-regime typologies, based on the structure of family policies as well as their gender-related
outcomes (Crompton 1998; Lewis 1992; Pfau-Effinger 1998; Siaroff 1994). These efforts
contributed considerably to welfare-state analyses by highlighting the gender aspects of welfare
states. However, they also shared some shortcomings with other regime typologies concerning
causal analysis. One reason for this is that regime typologies often mix welfare institutions and
their effects in their very basis, restricting their explanatory potential concerning the same effects
(Korpi and Palme 1998).
Basing the analysis solely on institutional family policy indicators was an approach
developed in order to improve the explanatory potential in comparative empirical analyses. While
some of the early studies used family policy indicators to evaluate the “family-friendliness” of
welfare states along a single scale, other researchers pointed to the multidimensional structures of
family policy institutions (Korpi 2000; Sainsbury 1996). However, family policies were not
necessarily “women-friendly” but could support different gender divisions of labour. Korpi (2000)
used a multidimensional approach to distinguish between gender-egalitarian policies that support
9
gender equality in paid and unpaid work, on the one hand, and traditional policies supporting
marked gender divisions of labour on the other. Later studies have fruitfully used this approach in
empirical analyses of gender inequalities of paid and unpaid work, as well as childbearing
(Billingsley and Ferrarini 2014; Ferrarini 2006).
Korpi et al. (2013) later added a “dual-carer” dimension that also takes into account the
degree to which fathers are supported in their care role. However, empirically speaking, the dualearner and dual-carer dimensions appear to have been developed in tandem and their occurrence
is highly positively correlated, while both of them show a negative relationship with the traditionalfamily support dimension (Korpi et al. 2013). Therefore, dual-carer and dual-earner support forms
can be combined analytically into a so-called “earner-carer” support dimension.
As countries´ family policies often contain varying amounts of both earner-carer and
traditional-family support forms, the institutional approach allows them to vary along both
dimensions simultaneously. It also permits the countries to have contradictory elements in their
family policies – for example, both gender-egalitarian and gender-traditional policies can be highly
developed simultaneously. Such policy constellations have often been shaped by class and gender
interest formation, changing political power relations, and policy inertia producing a particular
layering of different family policies with contradictory elements (Ferrarini 2006). The use of
family policy dimensions that are allowed to vary in degree also facilitates an analysis of policy
change that cannot be captured by only attaching static regime labels to countries.
Data
The development of institutional social-rights data based on legislative structures emanated from
the challenge posed by well-known validity problems associated with expenditure data in welfare-
10
state analyses (see Bolzendahl 2011; Esping-Andersen 1990; Gilbert 2009; Goodin et al. 1999;
Kangas and Palme 2007). Although several recent studies using expenditures have provided
important empirical results (see Kalwij 2010; Luci and Thévenon 2010), data on expenditures still
often have insufficient detail to separate theoretically central institutional characteristics of
policies.1 Institutional social-rights data are also less affected than expenditure data by the
outcomes of the policies that are subject to study.2 Moreover, they are less sensitive to changes in
the gross domestic product (GDP), which is used as the denominator when constructing
expenditure-based indicators.
Institutional analyses of family policy are not entirely new to the study of fertility
outcomes. Central studies have for example used legislated replacement rates in per cent 3 (see
Castles 2003; Gauthier and Hatzius 1997). However, there are some major drawbacks to the use
of formally legislated rates. First, as taxation of benefits is not considered, bias is introduced in the
comparison between taxable and non-taxable benefits. Second, legislated benefit ceilings are not
taken into account. This means that benefits with seemingly high formal replacement rates may
have factual replacement levels that are considerably lower because the earnings-ceilings of
benefits are often set at a fairly low income.
1
Parental leave benefit expenditures are sometimes available as an aggregate indicator. Yet, it should be noted that an earnings-
related parental leave benefit with shorter duration and a flat-rate parental leave benefit with longer duration may have similar
expenditures, but completely opposite effects on the gender distributions of paid and unpaid work – which in turn are likely to be
related to fertility.
2
One obvious example of an outcome that affects family policy expenditures is the number of children born in a country. To study
the link between family policy and fertility by the use of such a predictor is likely to produce biased results.
3
Replacement rates denote how much in per cent of a wage is replaced when being on parental leave for example.
11
In order to address the above-mentioned problems, the core independent variables used
in this study are the net replacement rates of family benefits, which express the size of benefits
after income taxation as proportion of an average production worker’s after-tax wage. Data for the
countries are mainly taken from the Social Citizenship Indicator Program (SCIP) and the Social
Policy Indicator database (SPIN), developed at Stockholm University and covering 33 countries
every fifth year between 1995 and 2010.4 This replacement-rate data is based on the calculation of
entitlements for a model family according to the rules stated in national legislation. The benefits
are the annual after-tax replacement rates for a family with two adults (one working full-time and
one on leave) and two children (of which one is an infant) expressed as a percentage of an average
production worker´s net wage.
[Table 1 here]
Table 1 shows the two dimensions of family support and their constitutive family policy benefits.
The traditional-family dimension is measured by a set of benefits that are typically not related to
previous work record and are paid in low flat-rate amounts or as lump-sum payments. Included in
this dimension are thus childcare leave allowances, which in many European countries are paid in
low flat-rate amounts for extended leave, lump-sum maternity grants that are paid in connection
to childbirth, child benefits paid in cash and via the tax system, and tax deductions for the main
4
The following countries are included: Australia, Austria, Belgium, Bulgaria, Canada, the Czech Republic, Denmark, Estonia,
Finland, France, Germany, Greece, Hungary, Ireland, Italy, Japan, Latvia, Lithuania, the Netherlands, New Zealand, Norway,
Poland, Portugal, Romania, Russia, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Ukraine, the United Kingdom and
the United States.
12
earner with an economically inactive (or less active) spouse (marriage subsidies). The variable
measures the average of the annual generosity5 of all those benefits as a percentage of an average
production worker´s net wage. The component of childcare leave included in the variable takes
into account the generosity of the benefit during the first year after the termination of earningsrelated parental leave.
The earner-carer dimension is measured by the annual generosity of earnings-related postnatal leave benefits paid to mothers and fathers during the first year after childbirth as a percentage
of an average production worker’s after-tax wage. To capture the degree of earnings-relatedness,
the parent on leave is assumed to have worked two years before childbirth, earning an average
wage, before spending a leave period with the infant.
The availability of public day-care is another factor that is likely to be central to
childbearing decisions. However, as welfare-state analysts are aware, longitudinal and
comparative institutional data on public day-care are difficult to find, and for the Eastern European
countries even valid cross-sectional data are hard to come by. Nevertheless, it should be pointed
out that previous studies of longstanding welfare states have shown that earnings-related parental
leave benefits and the extent of public childcare for the youngest children are highly correlated.
This is due to the fact that these policies have often developed together, pushed by similar driving
forces. Indicators based on the legislative structure of family policy transfers have here been shown
to function as proxies for the broader orientation of family policy (Ferrarini 2006).
The other variables included in the analyses are the total fertility rate (TFR), female labour
force participation, unemployment, and the Gross Domestic Product (GDP).
5
‘Annual generosity’ denotes the fact that the variables capture how much of an annual average production worker´s net wage is
replaced by the benefit/s.
13
The total fertility rate (TFR) is the outcome variable in the analyses. This measure is the
sum of age-specific fertility rates for women aged 15-49 years in a given year, and it indicates the
number of live births a woman would have if throughout her reproductive period she experienced
the age-specific fertility rates of the observation year. Although fertility rates that are adjusted for
timing effects should be preferred over the measure used here (see discussion in Bongaarts and
Feeney 1998), it is hard to obtain the data needed for the adjustment for all the countries included
in the analysis. Therefore, the TFR is used; however, keeping in mind that it is a measure sensitive
to postponement or advancement of childbirths.
Female labour force participation is the proportion of women aged 15 to 64 in the labour
force of a country. Here a more refined measure would have been preferred, for example, the
labour force participation of women of fertile age. Again, it was hard to find the data needed for
all the countries included in the analysis, and thus the less refined measure was utilised in the
study. Female labour force participation is included as the most important control variable, as
indicated by the results of previous studies (see the section on previous research) that see female
labour force participation as a vital component of fertility change.
In line with earlier studies, the analyses also include unemployment and GDP as
indicators of the general macro-economic situation in a country (see Ferrarini 2003; Gauthier and
Hatzius 1997). Unemployment is measured in per cent unemployed of the labour force in each
country. The Gross Domestic Product (GDP) data from the World Bank are measured in gross
domestic product converted to thousands of US Dollars according to the purchasing power parity
(PPP) rates, per capita.
14
Method
Total fertility rates are regressed on the two policy-dimension scores, earner-carer support and
traditional-family support, using pooled time-series analysis, and controlling for female labour
force participation, unemployment rates, and GDP. Because the number of countries exceeds the
number of time points substantially, certain analytical restrictions have to be considered. The error
terms from OLS-regression equations on pooled data have been shown to be temporally
autoregressive, cross-sectionally heteroskedastic, and cross-sectionally correlated (Hicks 1994).
Under such circumstances, standard errors are likely to be severely underestimated. Therefore, the
models will be estimated with panel-corrected standard errors (see Beck and Katz 1995).
The main models are estimated with country-fixed effects and corrections for first-order
auto-regressivity which have been used in previous comparative analyses with relatively few time
points (see Huber and Stephens 2000). However, as the total sample is restricted due to the number
of observations for which comparative rule-based data is available, alternative specifications will
be carried out to test the robustness of the results. In particular, change models based on differenced
data, where change in policy is related to change in TFR, will be estimated. Differencing generally
produces much more conservative estimates on links between independent variables and
outcomes, but it should be pointed out that the use of first differences also reduces the sample size
with one temporal point.
Results
In this section, first, descriptive results will be presented, followed by empirical evidence from the
pooled time-series regressions.
[Table 2 here]
15
Table 2 lists the variables included in the regressions, their mean values, standard deviation (Std.
Dev), and their coefficients of variation (CV). As can be seen, the dependent variable (TFR) shows
small but non-trivial variation. The family policy variables show larger variation, not least when
it comes to earner-carer support. Unemployment and GDP also show substantial variation, while
female labour force participation has the lowest relative variation as measured by the coefficient
of variation.
[Table 3 here]
Table 3 introduces a series of pooled regression models, each including country-fixed effects (not
reported). Model 1-3 includes the two types of family support separately first, and then together.
These regressions show that earner-carer support has a positive and statistically significant link to
TFR, while traditional-family support does not come out with a statistically significant correlation.
Model 4 also introduces female labour force participation alongside the two policy variables, and
shows that both earner-carer support and female labour force participation are positively and
significantly linked to TFR. The coefficient for earner-carer support is slightly weakened as
compared to Model 3, which is in line with ideas that some of the impacts of such policies are
mediated through higher female employment, as they explicitly support female employment.
Model 5 introduces a multiplicative interaction term between female labour force
participation and earner-carer support. Both variables show significant correlations with fertility
rates and the interaction term shows a negative but not statistically significant correlation. In the
full model (Model 6), the interaction effect between earner-carer support and female labour force
participation is significant and negative. Earner-carer support and female labour force participation
16
both show positive and significant effects on fertility rates. Model 6 also includes the control
variables of unemployment level and GDP. However, neither of them has a significant effect on
fertility rates.
[Fig.1 here]
To facilitate the interpretation of the interaction effect introduced in Model 5, Figure 1 graphically
illustrates predicted fertility rates at different levels of earner-carer support and at different levels
of female employment. Here the observed range of female labour force participation was used,
and each line indicates different levels of earner-carer support (20, 40, 60 and 80%, respectively).
The negative interaction term from Table 3 manifests itself in the decreasing slope of higher
earner-carer support on TFR at higher levels of female employment. At the highest levels of female
labour force participation, it appears as if more extensive earner-carer support would decrease
TFR. However, the differences between the slopes are only statistically significant for female
labour force participation rates below 70%.
It can also be noted from the decreasing distances between the different slopes in Figure
1 that an increase in earner-carer support should have a higher impact on fertility rates at lower
levels of female labour force participation. In other words, the results lend support to the idea that
there is a positive effect of earner-carer support on total fertility, but decreasing returns from
earner-carer support with rising female employment.
[Table 4 here]
17
Table 4 shows the results from a series of alternative regression analyses using first differences. It
should be noted that the fact that one time point is lost by differencing means that the results are
not strictly comparable to the results of the pooled time-series analyses in Table 3. Here, each
independent variable is first introduced alone. All independent variables have the expected
directions, but earner-carer support is the only one that is statistically significant. In a full model
(Model 6), with controls for all the independent variables, traditional-family support also has a
significant and negative effect. Using this more conservative estimation technique thus suggests
that changes in family policies can be linked to changes in fertility rates. Here again, an increase
in gender-egalitarian policies is linked to increases in fertility rates. In contrast, the results of
Model 6 indicate that increases in gender-traditional policies seem to be linked to decreases in
fertility rates.
Discussion
Can family policy institutions be anticipated to influence fertility change in industrialised
countries? The results of this study provide affirmative evidence to this question. Using new
institutional data and performing pooled time-series regressions, the link between family policy
institutions and fertility in 33 countries was investigated. As described in the theoretical and
methodological sections, the multidimensional approach employed separates institutional features
of family policies that build on diverging ideas about the gendered division of paid and unpaid
work. Earner-carer support eases the reconciliation of paid work and child-rearing, while
traditional-family support maintains a gendered division of the same, with a male breadwinner and
a stay-at-home spouse. The indicators used also try to avoid the validity problems of expenditure
data and the formal replacement rates, which do not take into account the tax effects of benefits
and benefit ceilings.
18
The results of the analyses show that different family policy orientations have different
relationships to fertility and that an important part of the puzzle lies in whether policies support
more gender-egalitarian behaviours or not. More gender-egalitarian family policies are associated
with higher fertility in both sets of analyses. Policies supporting traditional family patterns show
no statistically significant results in the first set of analyses or even the tendency to be correlated
with lower fertility in the analyses with differenced data.
Family policies could influence fertility by decreasing the direct costs of children through
cash benefits and by lowering the opportunity costs, especially of women, by making the
reconciliation of paid work and child-rearing easier. Moreover, Luci and Thévenon (2010)
demonstrated the importance of female labour force participation as an influence on fertility levels.
Ferrarini (2003) found that gender-egalitarian family policies were correlated with higher female
labour force participation. This could be part of the explanation for why earner-carer policies and
female labour force participation had positive links to fertility levels in the analyses in this study
and why they interact in their influence. Earner-carer policies partly seem to influence fertility
through female labour force participation as earnings-related benefits give incentives to enter and
stay in paid work, while also making the combination of paid work and child-rearing easier.
Regarding the results of the economic control variables, unemployment has a nonsignificant effect on fertility rates. In addition, GDP does not show a statistically significant
relationship with fertility. This result is in line with the arguments made by Luci-Greulich and
Thévenon (2013) according to which female labour force participation may be more important
than the degree of economic development as an influence on changes in fertility in recent decades,
and family policy legislation could play a major role in this development.
19
It is also interesting to note that tendencies manifested in earlier studies on Western
countries (see Castles 2004; Ferrarini 2003, 2006) hold and are further specified when former
communist countries are included in the analyses. Previous studies had already concluded that a
multidimensional perspective on family policy appeared to be a fruitful way of analysing family
policies in both Western and Eastern European countries (Ferrarini and Sjöberg 2010), although
fertility was not directly discussed. The present study shows that the expansion of the analysis to
include post-communist countries and also more recent time periods leads to interesting results
that also identify which type of family policy is connected with higher fertility levels, i.e. more
gender-egalitarian family policies. In this context, it may be questioned whether trends towards
expanding gender-traditional family policies in several post-communist countries will be an
effective way to raise fertility levels in the long run.
The results give more weight to the arguments that policies assisting the combination of
paid work and child-rearing, i.e. more gender-egalitarian family policies, are connected with
higher fertility levels (see McDonald 2006). On the other hand, taking the decreasing returns of
gender-egalitarian family policies at higher levels of female labour force participation manifested
in the interaction effect into account, the effects on fertility levels of introducing more genderegalitarian family policies are likely to differ depending on the level of female labour force
participation. Here, countries with lower female employment would be more likely to benefit from
increasing gender equality in family policies.
The study has concentrated on cash and fiscal family policy transfers. However, the
results are probably not confined to this set of policies, as it has previously been shown that family
policy transfers to some extent also function as proxies for the broader family policy matrix. In
particular, the countries where transfers support more gender-egalitarian divisions of paid and
20
unpaid work also tend to have highly developed public day-care for the youngest children.
Therefore, in future analyses, the institutional structure of public childcare and probably also elder
care needs to be considered. Broadly comparative data on public services that is longitudinal and
covers a greater number of countries is still severely lacking. However, it is likely that a
multidimensional institutional framework could be fruitfully used with other parts of family policy
legislation – including not only family policy services but also other pieces of family law, such as
joint-custody legislation – in the evaluation of central socioeconomic and demographic outcomes.
21
References
Ahn, N., & Mira, P. (2002). A note on the changing relationship between fertility and female
employment rates in developed countries. Journal of Population Economics, 15(4), 667–
682. doi:10.1007/s001480100078
Aidukaite, J. (2006). Reforming family policy in the Baltic States: The views of the elites.
Communist and Post-Communist Studies, 39(1), 1–23.
doi:10.1016/j.postcomstud.2005.09.004
Barro, R. J., & Becker, G. S. (1989). Fertility choice in a model of economic growth.
Econometrica, 57(2), 481–501.
Beck, N., & Katz, J. N. (1995). What to do (and not to do) with time-series cross-section data.
American Political Science Review, 89(3), 634–647. doi:10.2307/2082979
Billingsley, S., & Ferrarini, T. (2014). Family policy and fertility intentions in 21 European
countries. Journal of Marriage and Family, 76(2), 428–445. doi:10.1111/jomf.12097
Blossfeld, H. (1995). The new role of women: Family formation in modern societies. Boulder,
Colorado: Westview Press.
Bolzendahl, C. (2011). Beyond the big picture: Gender influences on disaggregated and domainspecific measures of social spending, 1980–1999. Politics & Gender, 7(1), 35–70.
doi:10.1017/S1743923X10000553
Bongaarts, J., & Feeney, G. (1998). On the quantum and tempo of fertility. Population and
Development Review, 24(2), 271–291. doi:10.2307/2807974
Castles, F. G. (2003). The world turned upside down: Below replacement fertility, changing
preferences and family-friendly public policy in 21 OECD countries. Journal of
European Social Policy, 13(3), 209–227. doi:10.1177/09589287030133001
22
Castles, F. G. (2004). The future of the welfare state. Oxford: Oxford University Press.
http://www.oxfordscholarship.com/view/10.1093/0199270171.001.0001/acprof9780199270170. Accessed 11 April 2014
Ciccia, R., & Verloo, M. (2012). Parental leave regulations and the persistence of the male
breadwinner model: Using fuzzy-set ideal type analysis to assess gender equality in an
enlarged Europe. Journal of European Social Policy, 22(5), 507–528.
doi:10.1177/0958928712456576
Crompton, R. (1998). Women’s employment and state policies. Innovation: The European
Journal of Social Science Research, 11(2), 129–146.
doi:10.1080/13511610.1998.9968558
d´Addio, A. C., & d´Ercole, M. (2005). Trends and determinants of fertility rates in OECD
countries: The role of policies (Working Paper 27). Geneva: Organization for Economic
Co-operation and Development (OECD).
Esping-Andersen, G. (1990). The three worlds of welfare capitalism. Cambridge: Polity Press.
Ferrarini, T. (2003). Parental leave institutions in eighteen post-war welfare states. Stockholm:
Stockholm University.
Ferrarini, T. (2006). Families, states and labour markets - Institutions, causes and consequences
of family policy in post-war welfare states. Cheltenhamn: Edward Elgar Publishing.
Ferrarini, T., & Sjöberg, O. (2010). Social policy and health: Transition countries in a
comparative perspective. International Journal of Social Welfare, 19(Issue Supplement
s1), S60–S88. doi:10.1111/j.1468-2397.2010.00729.x
23
Gauthier, A. H. (2007). The impact of family policies on fertility in industrialized countries: A
review of the literature. Population Research and Policy Review, 26(3), 323–346.
doi:10.1007/s11113-007-9033-x
Gauthier, A. H., & Hatzius, J. (1997). Family benefits and fertility: An econometric analysis.
Population Studies, 51(3), 295–306. doi:10.1080/0032472031000150066
Gilbert, N. (2009). The least generous welfare state? A case of blind empiricism. Journal of
Comparative Policy Analysis: Research and Practice, 11(3), 355–367.
doi:10.1080/13876980903221122
Goodin, R. E., Headey, B., Muffels, R., & Dirven, H.-J. (1999). The real worlds of welfare
capitalism. Cambridge: Cambridge University Press.
Gornick, J. C., & Meyers, M. K. (2008). Creating gender egalitarian societies: An agenda for
reform. Politics & Society, 36(3), 313–349. doi:10.1177/0032329208320562
Hicks, A. (1994). Introduction to pooling. In A. Hicks & T. Janoski (Eds.), The comparative
political economy of the welfare state (pp. 169–188). New York: Cambridge University
Press.
Huber, E., & Stephens, J. D. (2000). Partisan governance, women’s employment, and the social
democratic service state. American Sociological Review, 65(3), 323–342.
doi:10.2307/2657460
Kalwij, A. (2010). The impact of family policy expenditure on fertility in Western Europe.
Demography, 47(2), 503–519. doi:10.1353/dem.0.0104
Kangas, O., & Palme, J. (2007). Social rights, structural needs and social expenditure: A
comparative study of 18 OECD countries 1960–2000. In J. Clasen & N. Siegel (Eds.),
24
Investigating welfare state change. Cheltenhamn: Edward Elgar Publishing.
http://www.elgaronline.com/view/9781845427399.00015.xml. Accessed 10 April 2014
Korpi, W. (2000). Faces of inequality: Gender, class, and patterns of inequalities in different
types of welfare states. Social Politics: International Studies in Gender, State & Society,
7(2), 127–191. doi:10.1093/sp/7.2.127
Korpi, W., Ferrarini, T., & Englund, S. (2013). Women’s opportunities under different family
policy constellations: Gender, class, and inequality tradeoffs in Western countries reexamined. Social Politics: International Studies in Gender, State & Society, 20(1), 1–40.
doi:10.1093/sp/jxs028
Korpi, W., & Palme, J. (1998). The paradox of redistribution and strategies of equality: Welfare
state institutions, inequality, and poverty in the Western countries. American Sociological
Review, 63(5), 661–687. doi:10.2307/2657333
Lewis, J. (1992). Gender and the development of welfare regimes. Journal of European Social
Policy, 2(3), 159–173.
Luci, A., & Thévenon, O. (2010). Does economic development drive the fertility rebound in
OECD countries? (Report 167). Paris: Institut National d’Etudes Démographiques
(INED).
Luci-Greulich, A., & Thévenon, O. (2013). The impact of family policies on fertility trends in
developed countries. European Journal of Population / Revue européenne de
Démographie, 29(4), 387–416. doi:10.1007/s10680-013-9295-4
McDonald, P. (2006). Low fertility and the state: The efficacy of policy. Population and
Development Review, 32(3), 485–510.
25
Myrskylä, M., Kohler, H.-P., & Billari, F. C. (2009). Advances in development reverse fertility
declines. Nature, 460(7256), 741–743. doi:10.1038/nature08230
O´Connor, J. S., Orloff, A. S., & Shaver, S. (1999). States, markets, families: Gender, liberalism
and social policy in Australia, Canada, Great Britain and the United States. Cambridge:
Cambridge University Press.
Orloff, A. S. (1993). Gender and the social rights of citizenship: The comparative analysis of
gender relations and welfare states. American Sociological Review, 58(3), 303–328.
doi:10.2307/2095903
Orloff, A. S. (2009). Gendering the comparative analysis of welfare states: An unfinished
agenda. Sociological Theory, 27(3), 317–343. doi:10.1111/j.1467-9558.2009.01350.x
Pettit, B., & Hook, J. L. (2009). Gendered tradeoffs: Women, family, and workplace inequality in
twenty-one countries. New York: Russell Sage Foundation.
Pfau-Effinger, B. (1998). Gender cultures and the gender arrangement - A theoretical framework
for cross‐national gender research. Innovation: The European Journal of Social Science
Research, 11(2), 147–166. doi:10.1080/13511610.1998.9968559
Rostgaard, T. (2004). Family support policy in Central and Eastern Europe - A decade and a
half of transition (UNESCO Early Childhood and Family Policy Series Volume 8). Paris:
United Nations Educational, Scientific and Cultural Organization (UNESCO).
Rovny, A. E. (2011). Welfare state policy determinants of fertility level: A comparative analysis.
Journal of European Social Policy, 21(4), 335–347. doi:10.1177/0958928711412221
Ruhm, C. J., & Teague, J. L. (1995). Parental leave policies in Europe and North America
(Working Paper No. 5065). Cambridge, Massachusetts: National Bureau of Economic
Research.
26
Sainsbury, D. (1996). Gender equality and welfare states. Cambridge: Cambridge University
Press.
Sainsbury, D. (Ed.). (1999). Gender and welfare state regimes. Oxford: Oxford University Press.
Accessed 15 December 2014
Saxonberg, S., & Szelewa, D. (2007). The continuing legacy of the communist legacy? The
development of family policies in Poland and the Czech Republic. Social Politics:
International Studies in Gender, State & Society, 14(3), 351–379. doi:10.1093/sp/jxm014
Siaroff, A. (1994). Work, welfare and gender equality: A new typology. In D. Sainsbury (Ed.),
Gendering welfare states (pp. 82–100). London: SAGE Publications Ltd.
Winegarden, C. R., & Bracy, P. M. (1995). Demographic consequences of maternal-leave
programs in industrial countries: Evidence from fixed-effects models. Southern Economic
Journal, 61(4), 1020–1035.
27
Table 1. Family policy dimensions and included family policy transfer types
Family policy dimension
Family policy transfer
Traditional-family support
Childcare leave
Maternity grants
Cash and fiscal child allowances
Marriage subsidies
Earner-carer support
Maternity leave insurance
Dual parental leave insurance
Paternity leave insurance
28
Table 2. The regression variables, their means, standard deviations and coefficients of variation
for the 33 countries, 1995-2010
Variable
Definition
Mean
Std. Dev
CV
TFR
Total fertility rate
1.54
0.26
0.17
Earner-carer
Earner-carer support
38.34
30.14
0.79
Traditional
Traditional-family
support
20.15
11.31
0.56
Femlab
Female labour force
participation
64.35
7.36
0.11
Unemployment
Unemployment rate
8.76
4.20
0.48
GDP
Gross Domestic Product,
per capita, 1000 PPPconverted US dollars
23.14
10.97
0.47
29
Table 3. Pooled time-series cross-section regression of fertility rates on different determinants in 33 countries 1995-2010 with country
fixed effects (N=132). Prais-Winsten regression, correlated panels corrected standard errors (PCSEs).a
TFR
Model 1
Earner-Carer
0.001***
(0.0002)
Traditional
Model 2
0.0009
(0.001)
Model 3
Model 4
Model 5
Model 6
0.001***
(0.0002)
0.0006**
(0.0002)
0.009*
(0.005)
0.009*
(0.004)
0.0009
(0.001)
0.0009
(0.001)
0.0006
(0.001)
-0.00007
(0.0007)
0.010***
(0.002)
0.014***
(0.002)
0.010**
(0.004)
-0.0001
(0.00007)
-0.0001*
(0.00006)
Femlab
Femlab x Earner-Carer
(Interaction term)
a
Unemployment
-0.001
(0.003)
GDP
0.005
(0.004)
Constant
1.260***
(0.061)
1.237***
(0.096)
1.211***
(0.095)
0.588**
(0.171)
0.326
(0.177)
0.621**
(0.198)
Common rho
-0.136
-0.116
-0.124
-0.147
-0.152
-0.142
Country fixed effects not shown, panel-corrected standard errors in parentheses, *p<0.05, **p<0.01, ***p<0.001
30
Table 4. Pooled time-series cross-section regression of fertility rates on different determinants in 33 countries 1995-2010, with
differenced data (N=99). Prais-Winsten regression, correlated panels corrected standard errors (PCSEs).b
∆ TFR
Model 1
∆ Earner-Carer
0.0008**
(0.0003)
∆ Traditional
Model 2
Model 3
Model 4
Model 6
0.0007**
(0.0002)
-0.0009
(0.001)
∆ Femlab
-0.0007***
(0.0001)
0.008
(0.005)
∆ GDP
0.005
(0.006)
0.005
(0.009)
∆ Unemployment
b
Model 5
-0.002
(0.016)
-0.002
(0.004)
-0.001
(0.014)
Constant
0.029
(0.036)
0.036
(0.034)
0.021
(0.040)
0.010
(0.058)
0.033
(0.038)
0.052
(0.096)
Common rho
0.054
0.026
-0.077
0.046
0.034
-0.349
Country fixed effects in Model 6 not shown, panel-corrected standard errors in parentheses, *p<0.05, **p<0.01, ***p<0.001
31
Fig. 1. Predictive probabilities TFR at different levels of female labour force participation and
replacement rates of earner-carer support (based on Model 5, Table 3).
1.5
1.4
1.3
Predicted TFR
1.6
1.7
Predictive Margins
50
55
60
65
70
femlab
earner_carer=20
earner_carer=60
32
earner_carer=40
earner_carer=80
75