Gender Unemployment Gaps: Evidence from the New EU Member States
ABSTRACT Using EU LFS data, we analyze gender unemployment gaps in eight new EU member states – the Czech Republic, Hungary, Slovakia, Poland, the three Baltic states and Slovenia – over the last decade. While there are substantial unemployment gaps in the four central European countries and, more recently, also in Slovenia, there is no statistical difference between female and male unemployment rates in the three Baltic states. The estimated cost of having children, in terms of the higher probability of unemployment and lower unemployment to employment transition rate, is the highest in countries with the longest and most substantial drop in the labor force participation of women after childbirth. We show that country differences in family leave policies can explain much of the cross-country variation in the gender unemployment gaps.
- SourceAvailable from: Claudia Olivetti[Show abstract] [Hide abstract]
ABSTRACT: There is evidence of a negative cross-country correlation between gender wage and employment gaps. We arguethat non-random selection of women into work explains an important part of such correlation and thus of theobserved variation in wage gaps. The idea is that, if women who are employed tend to have relatively high-wagecharacteristics, low female employment rates may become consistent with low gender wage gaps simplybecause low-wage women would not feature in the observed wage distribution. We explore this idea across theUS and EU countries estimating gender gaps in potential wages. We recover information on wages for those notin work in a given year using alternative imputation techniques. Imputation is based on (i) wage observationsfrom nearest available waves in the sample, (ii) observable characteristics of the nonemployed and (iii) astatistical repeated-sampling model. We then estimate median wage gaps on the resulting imputed wagedistributions, thus simply requiring assumptions on the position of the imputed wage observations with respectto the median, but not on their level. We obtain higher median wage gaps on imputed rather than actual wagedistributions for most countries in the sample. However, this difference is small in the US, the UK and mostcentral and northern EU countries, and becomes sizeable in Ireland, France and southern EU, all countries inwhich gender employment gaps are high. In particular, correction for employment selection explains more thana half of the observed correlation between wage and employment gaps.01/2006;
- [Show abstract] [Hide abstract]
ABSTRACT: This paper discusses the implication of recent results on the structure of gender wage gaps in transition economies for the literature on gender segregation. Differences in employment rates of low-wage women driven by initial transition policies may be responsible for different wage penalties to predominantly female occupations. New evidence presented here also suggests that the introduction of Western-type anti-discrimination policies has had little immediate effect on the structure of female-male wage differences. (JEL: J3, J7, P3) Copyright (c) 2005 The European Economic Association.Journal of the European Economic Association 02/2005; 3(2-3):598-607. · 1.36 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The labour market status of many nonworking persons is at the boundary between unemployment and inactivity. Like the unemployed, they seek and are available for work; unlike them, their last search action was not recent enough to meet the International Labour Office definition of unemployment. In this paper we examine by nonparametric tests how the transition probabilities of these out-of-the-labour-force job seekers differ from those of the unemployed as well as the other nonparticipants. First, using data from the European Community Household Panel, we show that in most EU countries these job seekers constitute a distinct labour market state. Second, we rely on information available only in the Italian Labour Force Survey to derive a measure of search intensity that we use to break down the out-of-the-labour-force job seekers. On the basis of their transition probabilities, the most active are indistinguishable from the unemployed. (JEL: J64, J22, R23) Copyright (c) 2006 by the European Economic Association.Journal of the European Economic Association 02/2006; 4(1):153-179. · 1.36 Impact Factor
Gender Unemployment Gaps:
Evidence from the New EU Member States
Alena Biˇ c´ akov´ a∗
March 3, 2009
Inverse relationship between gender unemployment gaps and female
labor force participation has been documented for the Old EU member
states. No such pattern is observed among the New EU member states,
where countries like Czech Republic or Slovakia show both high female
participation rates and sizable gender differences in unemployment.
Using EU LFS data, we analyze the determinants of the gender un-
employment gaps among the New EU member states over the last
decade. Parametric and flexible Oaxaca-Blinder decompositions point
at the family characteristics as the key factor behind the unemploy-
ment gaps. Countries with the biggest unemployment gaps show the
highest labor market cost of having children, in terms of both higher
probability of unemployment and lower unemployment to employment
transition rate. It turns out that this is due to a substantial reduction
of labor force participation of women in the years following the child-
birth. Subsequent gender differences in work experience and employers’
expectations about the labor force behavior of women after childbirth
seem to contribute to the high gender unemployment gaps.
JEL classification: J13, J71
Keywords: Gender Discrimination; Gender Unemployment Gap; Female Labor Force Par-
ticipation; Family Gap
∗CERGE–EI, Praha, Czech Republic.
Email: email@example.com, web:
Gender discrimination at the labor market can have different forms, ranging
from the inequality in pay, job segregation, and job quality, to unequal
chances of employment.
While the main focus of the previous research has been on the gender
differences in pay per short period of time, such as hourly or monthly wage,
probability of holding a job directly affects the life-time earnings. More-
over, higher job separation rate or lower probability of finding a job may
both lower the wages offered at the labor market and the reservation wage,
resulting in lower observed accepted wages (see Black JOLE 1995). The
gender differences in unemployment are therefore likely to also affect the
gender differences in wages and vice versa.
While there is now a wealth of empirical evidence on the gender wage
gaps in most countries across the world (see .... for an overview), the research
focused on the gender differentials in unemployment rates is still scarce.
Azmat et al. (2006), who analyze the gender unemployment gaps in West
European countries and the US using ECHP and CPS data over the second
half of 1990s, and Stefanova-Lauerova and Terrell (2007), the only paper that
focuses on the gender unemployment gaps in Central or Eastern Europe, but
includes micro data analysis only for the Czech Republic for early 1990s,
extending the results to East Germany, Poland and Russia using secondary
data sources, are the only exceptions.
This paper complements the existing literature by documenting and an-
alyzing the gender unemployment gaps over the last decade (1996-2007) in
the eight New EU Member States, using the European Union Labor Force
While documenting the gender unemployment gaps and their discrim-
inatory part of the Oaxaca-Blinder decomposition in countries where the
research is lacking has value in its own right, the New EU member states
represent a particularly interesting group of countries for the research at
hand. As most of the previous research on gender discrimination has been
predominately carried on the West European countries or the US, the anal-
ysis of the cross-country differences and its outcomes have been typically
driven by the so-called North and South divide of Europe. Namely, the gen-
der discrimination outcomes negatively correlate with the extent of female
labor force participation, which span from 64 % and 3.1% gender unem-
ployment gap (Italy) in the South European countries to 86 % and 0.1 %
(Sweden) in Nordic countries.
Figure 1 plots the gender unemployment gaps and female labor force
participation in the Old and the New EU member states. It is apparent that
while the relationship in the OLD EU member states is negative (correlation
of -0.56 % significant at 2 % significance level), there is no correlation among
the new EU member states, or, if any, the correlation is positive (0.12 and
New EU Member states share features, determined by their Communist
past, which distinguish them from the Old EU Member states. As work was
compulsory and state created vacancies in order to assure work for every-
body, the unemployment rate was artificially maintained at zero rate. With
zero unemployment rate, the initial gender unemployment gap in these coun-
tries at the beginning of the transition towards market-based economies, did
not exist. In a decade starting ten years after the change in the regime, we
observe gender difference in unemployment rates exceeding 3 % in Czech
Republic, Slovakia, and Poland on one hand, and zero or negative gender
unemployment gap in the Baltic states on the other. In Hungary and Slove-
nia, the positive and significant gender unemployment gap emerges only in
the last two years.
Figure 2 shows the evolution of gender unemployment gap measured as
the difference between the female and male unemployment rate in the New
EU member states over the past decade. The confidence interval suggests
whether the gender difference in unemployment rates is significant.
The Figure suggests that the gender unemployment gaps have been
rather persistent, either stagnant or rising in size. The only exception is
Poland with a decreasing gender gap.
Part of the compulsory work and zero unemployment policy enforced by
the Communist regime was the norm of working women, also encouraged by
widely available public child care as well as the intra-family child care in the
multi-generational living arrangements. As a consequence, most of the post-
Communist countries entered transition with very high female labor force
participation rates, similar to the highest rates observed in the Old European
states. However, freedom to choose one’s labor market status, reduction in
the child care provision, but also sharp increases in unemployment during
the transition periods resulted in women leaving the labor force. As shown
in Figure 3, in sharp contrast with the ever-increasing in female labor force
participation in Old EU Member states, the participation of women in six
of the New EU member states has been either stagnant or decreased over
the past decade, and had a steady increasing trend only in Hungary and
Slovenia. However, with the exception of Hungary and Poland, the female
participation rate still remains above the 80 % level.
This paper explores this new evidence from the eight New EU Member
states to help determine which factors stand behind the gender unemploy-
Using the EU LFS data, we conduct a parametric and flexible Oaxaca-
Blinder decomposition of the gender unemployment gaps and the gender
differences in the transition rate to and from unemployment. Our results
suggest that gender differences in human capital (measured by age and edu-
cation) can explain only very little from the unemployment differentials. As
women are on average more educated than men in six of the eight countries,
the “unexplained” part of the gender unemployment gap there exceeds the
raw gender unemployment gap. The key driving force for the high gender
unemployment gaps observed in several of these countries turns out to be the
family factors. As marital status and children have opposite effect on gen-
der labor market outcomes, reducing unemployment of men and increasing
unemployment of women, we document gender unemployment gaps of sin-
gle individuals without children and gender unemployment gaps of married
individuals with children separately.
Countries with the highest and most persistent gender unemployment
gaps show the highest labor market cost of having children, in terms of both
higher unemployment and lower unemployment to employment transition
rate. Having children increases the likelihood of being unemployed, pre-
dominantly driven by lower probability of moving from unemployment to
employment. It turns out that this is due to a substantial reduction of labor
force participation of women in the years following the child-birth. Sub-
sequent gender differences in work experience and employers’ expectations
about the labor force behavior of women after childbirth seem to contribute
to the high gender unemployment gaps.
We discuss the potential mechanisms, why in some countries the transi-
tion of women with children from unemployment to employment is substan-
tially lower than that of men. We link the observed cross-country variation
in gender unemployment gaps to the cross-country differences in family-
related policies. In contrast to the evidence from the Old EU member states,
suggesting an inverse relationship between gender unemployment gaps and
female labor force participation, the new EU Member states with higher
overall labor force participation of women, such as Czech Republic or Slo-
vakia tend to have higher gender unemployment gaps. However, family-
related policies, in particular the maternity and parental leaves, results in
temporary but substantial drop in labor force participation among women
with very young children in these countries, which help explain the high
gender unemployment gaps observed in these countries.
See Figure 2.
Azmat et al. (2006) analyzes the gender unemployment gaps in West Euro-
pean countries and the US using ECHP and CPS data over the second half
of 1990s. They document substantial unemployment gaps among Mediter-
ranean countries, followed by Benelux and Germanic countries, but no or
negative unemployment gaps in Nordic and Anglo-saxon countries. The
high gender unemployment gaps, which occur mostly in countries with high
overall unemployment rate, are caused by gender differences in both employ-
ment to unemployment and unemployment to employment transitions. The
authors suggest that it is the interaction of the gender differences in human
capital accumulation and labor market institutions that drive the results.
The only paper that focuses on the gender unemployment gaps in Cen-
tral or Eastern Europe is Stefanova-Lauerova and Terrell (2007). While the
primary analysis concerns only the Czech Republic, the results are extended
also to East Germany, Poland and Russia, based on secondary data sources.
They find that gender unemployment gaps in these countries in early tran-
sition (first half of 1990s) are mostly driven by gender differences in the
transition from unemployment to employment.
TO BE FINISHED
We first document the variation in the gender unemployment gaps across
the New Member states of European Union and their significance. We next
estimate a full model of individual unemployment probabilities, as functions
of standard human capital characteristics, by gender, country and year. We
then compare whether the observed cross-country differences in gender un-
employment gaps are due to the differences in the characteristics of their
populations (e.g. in some countries women and men are more equally ed-
ucated than in others) or differences in the way these characteristics effect
the labor market outcomes for the two genders.
For each country and year, we estimate a model of a probability that a
given individual is unemployed:
Prob(Ui= 1) = f(Xiβ)
where Prob(Ui= 1) is the probability that individual i is unemployed, Xi
are the characteristics of an individual i.
We decompose the gender unemployment, in the spirit of Oaxaca-Blinder
methodology adopted for a probability model suggested by (Fairlie 2006)
and (Yun 2000), into the part due to differences in the characteristics of
women and men and the part due to the different contributions of these
characteristics towards the estimated probabilities.
We complement the results with a more flexible decomposition based
on re-weighting of the gender unemployment gaps across narrowly defined
socioeconomic groups with the overall gender-neutral weights to separate
the impact of the variation in personal characteristics.1
subgroups based on discrete versions of individual characteristics X. The
overall gender unemployment gap Ugapdefined as the difference between the
female uFand male uMunemployment rate can be then written in terms of
the J sub-groups as follows:
We construct J
1This is a simplified version of the Nopo (2008) for discrete explanatory variables .....