Content uploaded by Iga Magda
Author content
All content in this area was uploaded by Iga Magda on Oct 30, 2020
Content may be subject to copyright.
Firms and wage inequality in Central and Eastern Europe∗
Iga Magda†
, Jan Gromadzki‡
, Simone Moriconi§
June 2020
Abstract
We use large linked employer-employee data to analyze wage inequality patterns in
Central and Eastern European (CEE) countries between 2002 and 2014. We show that,
unlike in many other advanced economies, wage inequality levels have decreased in almost
all CEE countries. These reductions in wage inequality resulted from disproportionately
large increases in wages at the bottom of the wage distribution, and from decreases in
between-firm wage inequality. We further find that the declines in wage inequality were
driven by large wage structure effects that compensated for changes in the composition
of workers.
Keywords: wages, wage inequality, RIF regression, quantile decomposition, linked
employer-employee data
JEL Classification: D22, J31, J40
∗This paper has benefited from the financial support provided by the National Science Center, Poland (DEC-
2013/10/E/HS4/00445) and by the World Bank Group (FY2016 DGF Network for Jobs and Development-DGF
File: 502916-05). This paper is also part of the project “Migration And Labor Supply When Culture Matters”,
financed by French National Research Agency (ANR-18-CE26-0002 , AAPG2018). We acknowledge French
ANR for financial support. We would like to thank Peter Orazem, Carl Singleton, the participants of the 2018
EALE, 2018 IZA World Labour conference, the 2018 HSE/IZA workshop, and the 2019 IAAEU seminar for
their comments and remarks. We also gratefully acknowledge use of the Python/Stata template provided by
von Gaudecker (2014). This paper uses Eurostat data. Eurostat has no responsibility for the results and the
conclusions, which are those of the authors.
†SGH Warsaw School of Economics; Institute for Structural Research (IBS), Warsaw, Poland; IZA, Bonn,
Germany. E-mail: iga.magda@ibs.org.pl.
‡SGH Warsaw School of Economics; Institute for Structural Research (IBS), Warsaw, Poland. E-mail:
jan.gromadzki@ibs.org.pl.
§IESEG School of Management; LEM-CNRS 9221. E-mail: s.moriconi@ieseg.fr.
1
1 Introduction
The issue of increasing income inequality is being publicly debated in most OECD countries.
Many of the questions raised in these discussions center around the extent to which changes
in wage inequality levels are driving income differentials. Much of the existing literature on
this topic has focused on firm-level determinants, and has recognized the important role of
inter-industry and firm-level wage differentials (Abowd, Kramarz, & Margolis, 1999; Du Caju
et al., 2010; Krueger & Summers, 1988; Martins, 2004). We know far less about how between-
firm wage inequality levels change over time, and whether firm-level factors have contributed
to the increases in the wage inequality levels observed in many OECD countries. This paper
contributes to this debate by investigating the workplace features that are likely to drive wage
inequality, and how it changes at different points of the wage distribution.
While there is extensive research on recent determinants of wage inequality in the US, Ger-
many, and many other advanced countries, this paper focuses on Central and Eastern Euro-
pean (CEE) countries. CEE countries are interesting not only because there is little compar-
ative evidence on recent changes in their wage structures, but also because the countries in
this geographical region have experienced similar wage inequality trends. While the transition
from a socialist to a market economy was associated with increasing wage dispersion (Brzezin-
ski, 2018), we show that the wage patterns continued to change in the period that followed.
Specifically, we find that whereas wage inequality levels further increased in many advanced
countries in the 2000s, they stabilized or declined in Central and Eastern Europe during this
period.
There have been several institutional and economic changes in the region that likely con-
tributed to the observed changes in wage inequality. These changes, which we discuss in
detail in section 3, involved reforms of labor codes that increased workers’ bargaining power,
increases in minimum wages, and large migration outflows to Western European countries.
We argue that all of these developments had the potential to reduce wage inequality. There
is a broad consensus that an increase in the minimum wage reduces wage inequality, as both
its direct and its spillover effects are concentrated at the lower end of the wage distribution
2
(Autor, Manning, & Smith, 2016; Cengiz et al., 2019; DiNardo, Fortin, & Lemieux, 1996).
According to wage bargaining models, workers’ bargaining power and workers’ outside options
determine wages (Pissarides, 2000). Therefore, the increase in workers’ outside options due to
the opening of Western labor markets should have led to an increase in workers’ wages in the
CEE countries. These increases were likely concentrated at the bottom of wage distribution,
as the demand for migrant workers was largely limited to low-skilled jobs (Black et al., 2010).
This paper has three main objectives. First, we aim to present a clear picture of the changes
in the wage dispersion patterns in post-transition CEE countries between 2002 and 2014 using
harmonized, comparative data from a large, linked employer-employee dataset of the Euro-
pean Structure of Earnings Survey (ESES). We study both the variance of wages and the
quantiles of the wage distribution. Second, we intend to analyze the role of firms in determin-
ing wage inequality, and to examine how much of this inequality is due to wage differentials
arising between firms, and to within-firm wage inequality. Third, we investigate the potential
micro-level factors associated with higher or lower levels of wage inequality, and particularly
the drivers of the observed decrease in wage inequality during the 2006-2014 period.
Our results suggest that during the study period, wage inequality levels decreased in most
CEE countries, especially in the Baltic states and Romania, where the initial wage inequality
levels were the highest in the region. In these countries, the largest increases in real wages
occurred at the bottom of the wage distribution. This may be related to the fact that those
countries experienced the largest migration outflows and minimum wage increases during the
study period, as argued above. Czechia, where the wage inequality level remains the lowest in
the region, was the only CEE country that experienced a (slight) increase in wage inequality,
which was observed both at the lower and the upper part of the wage distribution. Czechia
also had the smallest migration outflows in CEE, and its minimum wage expressed as a frac-
tion of the average wage decreased considerably.
We further show that the differences in the variance of wages across the CEE countries were
primarily driven by the differences in the between-firm component of wage inequality (and,
3
to a lesser extent, by wage inequality within firms). We gain additional insight into the
determinants of wage inequality by applying recentered influence function (RIF) regressions
following Firpo, Fortin, and Lemieux (2018). We show that workplace characteristics played
an important role in wage inequalities; and that of these workplace characteristics, the levels
of education and the ages of an employee’s co-workers were as crucial as her/his occupational
and sectoral affiliation. Decomposition of the changes shows that reductions in wage inequal-
ity in the region between 2006 and 2014 were largely attributable to wage structure effects
(changes in the wage premia paid for individual- and firm-level characteristics, as well as in
the intercept), rather than to composition effects (changes in covariates). In line with Firpo,
Fortin, and Lemieux (2018), we find that composition effects increased inequality, as the gains
were greater at the top of the wage distribution. However, unlike in the US, we do not see
polarization in wage growth, as the changes in workers’ returns to wages were concentrated
at the bottom of the wage distribution, and thus led to decreases in inequality.
2 Literature review
Our paper is related to two main strands of literature. The first strand is comprised of studies
on changes in wage inequality and their determinants. Some of the most important works on
this topic include Autor, Katz, and Kearney (2006); Autor, Katz, and Kearney (2008) for the
US; Fortin et al. (2012) for Canada; Dustmann, Ludsteck, and Schönberg (2009) for Germany;
and Machin (2016) for the UK. This literature has looked at the macro-level drivers of wage
inequality, and has examined how trade and labor market frictions, technological change, and
migration have contributed to wage inequality (Acemoglu & Autor, 2011; Akerman et al.,
2013; Autor, Manning, & Smith, 2016; Ge & Yang, 2014; Goldschmidt & Schmieder, 2017;
Helpman et al., 2017; Krishna, Poole, & Senses, 2012). Some studies (Autor, Katz, & Kearney,
2008; Lemieux, 2006) have taken a micro perspective, and have shown that the rise in wage
inequality has been highly heterogeneous across worker characteristics, including education,
age, and type of occupation. A striking feature of the steady rise in wage inequality that took
place in the US from the 1970s onward is that earnings increased more at higher percentiles
of the earnings distribution, even for the same skill levels. The literature on this trend has
grown considerably in recent decades, and has focused mainly on developed economies (the
4
US and Western European countries) and some emerging economies (e.g., Brazil, China, see
Alvarez et al. (2018); Appleton, Song, and Xia (2014); Messina and Silva (2017)). Only a
few studies have dealt explicitly with recent developments in wage inequality in the CEE
countries, which experienced a strong increase in wage dispersion during the transition to a
market economy (Aristei & Perugini, 2014; Milanovic & Ersado, 2012). This phase seems to
have been followed by a period in which wage inequality was slowly decreasing (Tyrowicz &
Smyk, 2019); although the patterns varied across countries (Aristei & Perugini, 2012). Pryor
(2014) emphasized that even after the surge in wage inequality levels during the transition, the
degree of wage dispersion remained lower (around the 2000s) in the CEE countries than it was
in most OECD countries. A more recent study by Mysíková and Večerník (2018) compared
the developments in wage inequality in Poland and Czechia with those in Austria just before
and after the Great Recession (2007). They found that in the two CEE countries, income
polarization did not increase, and levels of wage inequality remained low along the gender,
skill, and occupational dimensions. Our paper contributes to this literature by showing that
wage inequality decreased in nine CEE countries during the 2000-2014 period.
The second strand of literature we want to contribute to focuses on firm-level drivers of wage
inequality. The overall level of wage inequality can be decomposed into a within-firm compo-
nent (wage differentials that arise within firms) and a between-firm component (differences in
the average wages of firms). Establishment effects matter, as employers are affected differently
by the various factors that shape changes in the wage distribution, such as skill-biased tech-
nological change or changes in labor market institutions; whereas workers are sorted among
employers. Card et al. (2018) developed a theoretical model of wage setting that assumes
that workers have idiosyncratic tastes for different workplaces; and that an increase in firm
productivity will lead to an increase in individual wages because firms do not observe workers’
preference shocks. Thus, according to this model, an increase in the dispersion of produc-
tivity across firms will lead to an increase in levels of between-firm wage inequality, and the
propagation of productivity increases to wages depends directly on rent-sharing elasticity. For
the UK, Bell, Bukowski, and Machin (2018) found that rent-sharing elasticity has decreased
sharply since the 1980s, which has resulted in a reduction in the impact of increasing pro-
ductivity differentials on wage inequality. Hence, it seems that increases in the dispersion of
5
firm productivity can explain only a portion of the observed increases in levels of between-firm
wage inequality.
The empirical studies on the contribution of the between-firm component were summarized
by Card et al. (2018). Barth et al. (2016) has shown that the increased variance of average
earnings across establishments can explain about half of the rise in US wage inequality during
the 1970-2000 period. Handwerker and Spletzer (2016) showed that the growing contribution
of establishment effects to the widening of the distribution of wages is only partially explained
by changes in the distribution of occupations among workplaces. Song et al. (2019) used
linked employer-employee data to analyze the contributions of firms to the rise in earnings
inequality in the United States from 1978 to 2013. They showed that about two-thirds of the
increase in the variance of (log) earnings occurred between firms. They pointed out that the
heterogeneity of the composition of the workforce among firms played a major role in this
development. In a similar vein, Antonczyk, Fitzenberger, and Sommerfeld (2010) found that
workplace effects contributed substantially to the increase in wage inequality in Germany.
Card, Heining, and Kline (2013) also looked at West Germany (between 1985 and 2009), and
confirmed that increasing firm-level heterogeneity explained a large share of the rise in wage
inequality. By contrast, the role of the between-firm component was found to be relatively
small in Italy (Devicienti, Fanfani, & Maida, 2019), Sweden (Akerman et al., 2013), and the
United Kingdom (Schaefer & Singleton, 2019).
Very few studies have touched upon the potential role of firms in shaping wage inequality
in the CEE countries, though a recent World Bank study (Kelly et al., 2017) has suggested
that in Bulgaria, Estonia, and Latvia, differences in wages across firms explain more than half
of wage inequality, while differences in educational attainment levels or occupations across
workers explain only a third or less of wage inequality.
We add four contributions to the previous literature. First, we provide evidence based on
harmonized data on recent reductions in wage inequality in most CEE countries, and discuss
these changes in the light of institutional and economic developments in the CEE. Second,
6
we investigate the contributions of the within-firm and the between-firm component to the
levels of and the changes in the overall wage inequality in nine CEE countries. Third, we
conduct a detailed analysis of the micro-level determinants of wage inequality, and of how
wage inequality has changed over time. Finally, while the literature on drivers of increases in
wage inequality is abundant, we provide novel evidence on the determinants of wage inequality
decreases as opposed to wage inequality increases.
3 Institutional and Economic Background
In the early 2000s, the CEE countries had not yet completed the transition from having a
centrally planned economy with artificially low levels of wage inequality to having a capitalist,
market-based economy.1The countries in the region continued to reform their legal and eco-
nomic frameworks to meet the EU requirements. As part of this process, the CEE countries
made changes to their national labor codes and minimum wages. After the EU enlargement
in 2004, the CEE countries continued to introduce structural adjustments in line with the
agenda of the EU’s Cohesion Policy (Sedelmeier, 2008). These changes were accompanied by
high levels GDP growth. On average, GDP growth was twice as high in the CEE countries as
it was in the EU incumbent countries between 2004 and 2014. This strong economic growth
in the CEE countries was driven by high rates of investment in the region (the GDP share of
investment was, on average, 20-25% higher in the CEE countries than it was in the Western
European countries), FDI inflows, re-industrialization, and increased participation in trade
and Global Value Chains (Carstensen & Toubal, 2004; Parteka & Wolszczak-Derlacz, 2013).
The post-enlargement period was also marked by a wave of emigration, as labor markets in
Western Europe were opened to CEE workers (Kaczmarczyk & Okólski, 2008; Kahanec & Zim-
mermann, 2016; Zaiceva & Zimmermann, 2008). Between 2006 and 2014 the number of CEE
migrants in the EU almost tripled. Nevertheless, there was substantial variation in migration
patterns across the CEE countries. In 2014, nearly 12% of the Romanian population were
1Still, they were at different stages in this process. Czechia, Poland, Hungary, Slovakia, and the Baltic
countries were at more advanced stages in this transition. Romania and Bulgaria were at earlier stages in this
process (Carstensen & Toubal, 2004), which led to their later entry into the EU.
7
living in other EU countries, compared to less than 2% of the Czech population (see Table D.3).
Rapid economic growth in the CEE countries contributed to an overall increase in the demand
for labor. This rising labor demand, coupled with high rates of emigration, put pressure on
wages (Kohl, 2009). Thus, wages in the CEE countries grew considerably during the 2006-
2014 period. Between 2006 and 2014, average real hourly wages increased by an average of
40% in the region, compared to by an average of roughly 23% in the EU28.2
It is also likely that many of the institutional adjustments that these countries made con-
tributed to their high levels of wage growth and todecreases in wage inequality. In the years
immediately after the EU accession, the CEE countries needed to consolidate the system of
labor market institutions that had resulted from the institutional restructuring during the
transition. Meeting EU employment law standards was a necessary step for successful inte-
gration into the EU. Among the first changes the CEE countries made were major reforms
of their national labor codes, and the introduction of multiple measures that reinforced their
adherence to the European standards, which led to increases in workers’ bargaining power.3
A second direction of reform was in the implementation of wage floors through the enforce-
ment of minimum wages and/or extension mechanisms. After the EU accession, most CEE
countries introduced new wage floors or reinforced existing ones. Currently, all CEE countries
have a wage floor, which was generally the result of the introduction of a national minimum
wage, and/or of bi/tripartite collective negotiations with extension mechanisms (as in the
case of Romania and Bulgaria, see Kohl, 2009). In most CEE countries, the minimum wage
2During this period, the Great Recession hit the CEE countries hard (particularly the Baltic States, where
GDP decreased by more than 14%). However, after the downturn, these countries quickly resumed their high
rates of growth. At the same time, there was a clear pattern of dualization in the labor markets associated
with changing contractual arrangements. Numbers of temporary jobs and of self-employed workers increased
while full-time employment rates declined, in particular in the countries that were most affected by the Great
Recession, with mixed effects on wage inequality (Brzezinski, 2018; Hoelscher, Perugini, and Pompei, 2011).
3These included actions aimed at strengthening the social partners’ ability to bargain, and at engaging in
the European Social dialogue and the procedural regulation of labor markets (e.g., through the implementation
and reinforcement of work councils). Labor market institutions were also reformed through involvement in the
European "Open Method of Coordination" in the fields of labor and social policies. These measures ultimately
increased trade unions’ bargaining power, directly or indirectly, by reinforcing the procedural legitimacy of
collective agreements. Thus, the power and the effectiveness of collective bargaining institutions were enhanced,
even though trade union densities and coverage declined in the region after 2002. See Magda, Marsden, and
Moriconi, 2016 for details.
8
increased substantially, both in real terms and as a fraction of average wages (see D.1 and
D.2). Bulgaria and Czechia were the only CEE countries where the increases in the minimum
wage were smaller than the increases in average wages. In general, the countries that had the
lowest minimum wages in 2006 (Bulgaria, Romania, and Latvia) also had relatively high wage
inequality. Those countries experienced the highest levels of minimum wage growth between
2006 and 2014. In our analysis, we argue that these increases in the minimum wage likely
contributed to the reduction in wage inequality, as wages at the bottom of the wage distribu-
tion increased significantly in those countries.
Furthermore, the wage distributions in all countries may have been affected by technological
change. Keister and Lewandowski (2017) have shown that in most CEE countries, the intensity
of routine cognitive tasks has increased. The authors argued that sustained demand for routine
work prevented increases in wage inequality in the CEE countries. This pattern contrasts with
that in Western Europe, where decreased demand for routine work led to increases in wage
inequality.
4 Data
We use repeated cross-sectional data from the European Structure of Earnings Survey (ESES)
for the years 2002, 2006, 2010, and 2014. The ESES is a large matched employer–employee
dataset provided by Eurostat. It includes information on workers’ earnings, and on their
individual-, job-, and firm-level characteristics. We use data for the following nine CEE coun-
tries: Czechia, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, Slovakia, and Bulgaria.
We additionally draw on ESES data for the Netherlands, Norway, Sweden, and Portugal in
order to compare some of our results with those for Western European countries.
While the ESES data are characterized by a high degree of cross-country comparability, we
had to carry out a number of cleaning steps to guarantee that the national samples and our
analyses were fully harmonized across countries. In particular, we dropped observations that
referred to workers in the smallest firms (fewer than 10 workers), because comparable data
were available for only some of the countries. We also dropped observations from the top
9
and the bottom 0.1% of the hourly wage distribution to avoid outliers. In the 2002 wave of
the survey, the inclusion of observations from the non-market services sector was optional.
Because the 2002 data for Estonia, Latvia, and Hungary are incomplete, we were not able to
obtain comparable datasets for all countries for that year. For this reason, we have chosen to
analyze the 2002 data only for countries with datasets that included all sectors, and only for
descriptive statistics. We provide detailed analyses for the 2006-2014 period only. The sizes
of the final samples are large: they range from 32,000 observations in Lithuania in 2010 to
more than two million observations in Czechia in 20144. Such large samples reduce the risk
of any potential sample biases. Summary statistics across countries and years are presented
in Table 1. To check for potential data inconsistency issues, we compare the ESES wage
data with an alternative data source. We show that our descriptive statistics based on the
ESES are in line with the statistics provided by the OECD (see Appendix E for further details).
Table 1: Summary statistics
(a) Number of observations
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 151 090 1 023 598 135 978 629 101 229 423 417 173
2006 162 838 1 914 027 114 656 676 050 114 892 271 872 639 784 247 433 670 603
2010 175 575 1 948 513 108 903 781 240 32 773 198 862 668 022 262 983 767 368
2014 168 345 2 148 818 112 569 770 148 38 483 153 540 707 999 270 582 863 864
(b) Number of firms
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 2 454 2 289 5 915 13 392 8 870 1 391
2006 4 596 11 673 2 628 13 916 5 305 7 641 13 978 10 778 2 971
2010 5 187 11 193 2 502 13 681 2 690 5 261 14 423 12 161 4 739
2014 4 904 12 159 2 348 12 638 3 089 3 688 14 608 12 075 5 695
(c) Mean of hourly earnings (PPS)
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 1.96 5.40 3.76 5.73 2.52 4.46
2006 2.93 6.92 5.60 5.89 5.18 4.66 7.09 3.69 4.53
2010 4.60 7.45 7.14 7.25 5.94 4.35 8.69 5.09 7.05
2014 4.92 8.35 7.66 7.76 6.30 6.37 10.10 5.20 7.92
Data: European Structure of Earnings Survey.
4The reduction in the number of observations in Lithuania between 2006 and 2010 was caused by a change
in the sampling procedure. This should not affect our results, because we use in all of our calculations the
sample weights provided by Eurostat, which take these changes into account. Furthermore, we show that the
descriptive statistics are in line with an external source of data, the OECD (see Appendix E).
10
Panel C in Table 1summarizes the changes in average hourly gross wages in the CEE coun-
tries between 2002 and 2014. We can see that real wages were lowest in the late EU entrants,
Romania and Bulgaria; and were, on average, twice as high in Czechia and Poland. In all
of the CEE countries, there were substantial increases in average earnings over the analyzed
period.
Our baseline measure of wages is real hourly gross wage, expressed in PPS (Purchasing Power
Standard)5. This measure includes earnings, earnings related to overtime, special payments
for shift work, social security contributions, and taxes; but it does not include annual bonuses
and allowances that were not paid in each period. We use the variance of log hourly wages
as our measure of wage inequality. This is a common statistical measure of dispersion, and,
unlike other popular measures of inequality, such as the Gini coefficient or the 90-10 wage gap,
the variance is additively decomposable into the between-firm component and the within-firm
component, which we draw on in our analysis. We use log wages because the variance of log
wages is a mean independent measure (unlike the variance of wages, see Atkinson (1970)).
The downside of using variance as a measure of inequality is that it may be masking changes
at the tails of the wage distribution. Therefore, we supplement our study with an analysis
of changes in quantiles of the wage distribution. Moreover, we show that the trends in wage
inequality based on alternative measures (Gini index, Atkinson index, Theil index, and decile
dispersion ratios) are very similar to the trends in changes in the variance of wages (see Tables
A.1-A.6 in the Appendix).
5 Methodological approach
Our analysis is carried out in two main steps. First, we analyze levels of and changes in
wage inequality in each country over time, discussing changes at the mean and at the tails
of the wage distribution. We also determine the respective contributions of the within-firm
component and the between-firm component to total wage inequality. In the second step, we
investigate the determinants of the levels of wage inequality, as well as the changes in wage
5PPS is an artificial currency unit derived by the Eurostat that accounts for cross-country differences in
the prices of goods. Thus, one PPS can buy the same amount of goods and services in each country.
11
inequality, over time.
We start the first part of our analysis by normalizing wages for each year and country, such
that individual wages are defined as ˆwit =log100 ∗wit
¯wt, where wit denotes the individual
hourly wage and ¯wtis the average hourly wage in a given year t. We then calculate the
variance of log wages for each country and each year and other measures of wage inequality,
presented in Tables A.1-A.6 in the Appendix.
For each country, we analyze to what extent the level of overall wage inequality and its
changes are determined by the within-firm and the between-firm wage inequality, following
the methodology applied by Lazear and Shaw (2009) and Barth et al. (2016). We decompose
the overall variance of log wages (V ar( ˆwit)) into the within-firm component (V ar(within))
and the between-firm component (V ar(between)). Thus, the variance decomposition of log
wages, V ar( ˆwit ) = V ar(within) + V ar(between), is given by the following equation:
V ar( ˆwit ) = 1
NtX
i
( ˆwit −ˆ
¯wt)2=1
NtX
j
X
i∈j
( ˆwit −ˆ
¯wjt )2+1
NtX
j
Njt (ˆ
¯wjt −ˆ
¯wt)(1)
where ˆ
¯wtis the average log wage in year tin a given country, ˆ
¯wjt denotes the average log
wage for workers in firm jin year t,Ntis the number of all workers in year t, and Njt is the
number of workers in firm jin year t.
We also repeat the above analysis, but while looking at residual wage inequality; that is,
the wage inequality that remains after the workers’ and workplaces’ observable characteristics
are accounted for. First, for each year and country, we estimate a standard Mincerian wage
equation of the following form:
ˆwi=β0+β1Xi+β2Xj+i(2)
where Xiis a set of individual and job characteristics, such as age, gender, education, occu-
pation, type of contract; and Xjis a set of firm characteristics, such as the enterprise’s sector
and forms of economic and financial control. We also account for peer effects (share of female
12
workers, share of workers with tertiary education, share of workers aged 50 or older, and share
of workers with a tenure of less than two years) in order to capture more firm heterogeneity
(Card & De La Rica, 2006). Next, we calculate the residuals from the estimated model, and
analyze the variance of the obtained residuals. In other words, the residual variance is the
variance of the unexplained component of wages.
While the exercises above provide us with a broad picture of the aggregate wage dispersion
trends, they give us little insight into the determinants of these trends. Several recent stud-
ies have tried to distinguish the individual determinants of wage inequality (associated with
gender, age, job experience) from job and firm characteristics (Barth et al., 2016; Handw-
erker & Spletzer, 2016). To add to this line of research, we estimate in the second step the
variance of wages as a function of worker and firm characteristics (the same characteristics
as in the Mincerian equation above). To this end, we use the recentered influence function
regression, which calculates the partial effect of a small change in the distribution of covariates
on the distributional statistic of interest (Firpo, Fortin, & Lemieux, 2018), which in our case
is the variance. In other words, we calculate the recentered influence function value for each
observation according to the following formula:
RI F ( ˆwit) = ( ˆwit −ˆ
¯wt)2(3)
Next, we estimate the following model for each country and each year:
RI F ( ˆwit) = β0+β1Xit +β2Xjt +it (4)
The notation is the same as in Equation (2). We obtain the estimated partial effects of small
changes in the distribution of selected variables on the variance of log wages for each country
and for each year. Thus, we can observe differences in the magnitude of the effects over time.
Furthermore, to gain a better understanding of the determinants of changes in inequality
over time, we decompose the changes in the variance of log wages into the composition effect
(changes in the covariates Xit and Xjt) and the wage structure effect (changes in the coeffi-
13
cients from the RIF regression β0,β1,β2) following Firpo, Fortin, and Lemieux (2018). The
decomposition is given by the following equation:
V ar( ˆwi,2014)−V ar( ˆwi,2006)=(E[X|Y= 2014] −E[X|Y= 2006])0β2006
+E[X|Y= 2006](β2014 −β2006)
(5)
The first term reflects changes in the variance driven by changes in the covariates, assuming
that the coefficients remained at the 2006 level (composition effect). The second term captures
the unexplained part of the changes in wage inequality; namely, the change in the coefficients
(β2014 −β2006), assuming that the covariates remained at the 2006 level (wage structure effect).
Finally, to better understand the heterogeneity of the sources of changes in the wage distribu-
tion, we decompose the changes at different deciles using an unconditional quantile regression
(Firpo, Fortin, & Lemieux, 2018).
6 Results
6.1 Overall wage dispersion and its changes
Table 2: Variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.34 0.19 0.37 0.34 0.42 0.25
2006 0.33 0.21 0.28 0.29 0.36 0.46 0.36 0.42 0.24
2010 0.33 0.23 0.28 0.30 0.33 0.34 0.31 0.38 0.23
2014 0.33 0.23 0.27 0.29 0.29 0.31 0.32 0.36 0.23
Data: European Structure of Earnings Survey.
The results show that levels of wage inequality varied substantially across the CEE countries
(Table 2). In 2014, the lowest wage inequality levels were observed in Czechia and Slovakia
(where the variance of log wages amounted to 0.23), while the highest wage inequality level
was observed in Romania (0.36). When we compare the wage inequality levels in the CEE
countries to those in the more advanced European countries (Table F.2), we see that the
levels in Czechia and Slovakia were similar to the level in the Netherlands, and that the high
variance of wages in Romania corresponded to the level of wage inequality in Portugal (where
14
wages were the most dispersed among EU countries, if measured with the D9/D1 decile dis-
persion (Eurostat, 2014)). The average level of the variance of log wages observed in the CEE
countries was around three times higher than it was in the two Scandinavian countries in our
study sample (Norway and Sweden). All in all, we find that wages were, on average, more
unequal in the CEE countries than in the older EU member states; a result that is confirmed
by the Eurostat D9/D1 dispersion statistics.
Figure 1: Overall variance of log wages: 2002-2014
Note: Figure shows variance of log of normalised gross hourly wages. Tables A.7-A.9 show variance of log of
normalised gross hourly wages by sector.
Data: European Structure of Earnings Survey.
There were substantial changes in the wage inequality patterns in the CEE countries between
the early to mid-2000s and 2014 (Table 2). These changes included a slight increase in the
level of wage inequality in Czechia, the CEE country that had the lowest initial level; there,
the variance of log wages increased from 0.19 in 2002 to 0.23 in 2014. Over the same period,
the levels of wage dispersion decreased in the CEE countries that had high initial wage in-
equality levels. The variance of log wages decreased the most in Latvia (from 0.46 in 2006
15
to 0.31 in 2014), Romania (from 0.42 in 2006 to 0.36 in 2014), and Lithuania (from 0.37 in
2002 to 0.29 in 2014). Wage inequality levels remained stable in Bulgaria, Estonia, Hungary,
and Slovakia. The data suggest that the sharpest declines in wage inequality levels occurred
after 2006 (between 2006 and 2010, in particular). When we look at the 2002-2006 sub-period
(during which seven of the nine CEE countries we analyze entered the European Union), we
observe hardly any changes in the overall wage dispersion patterns – although it should be
noted that we have information for only a few of the CEE countries in this period. In sum, the
differences in the levels of wage dispersion among the CEE countries narrowed considerably
in the 2000s and the early 2010s (see Figure 1).
The narrowing of wage inequality occurred mostly at the lower tails of the wage distribution
(see Tables A.5-A.6). Between 2006 and 2014, the D50/D10 ratio decreased in Hungary,
Lithuania, Latvia, Poland, and Romania; it increased in Czechia; and it remained stable in
the remaining countries. The Baltic states were the only CEE countries where the D90/D50
ratio decreased between 2006 and 2014, reflecting a decrease in wage inequality in the upper
part of the wage distribution. None of the other CEE countries experienced large changes
in the D90/D50 ratio. To shed more light on these changes, Figure 2shows the cumulative
distributions of real hourly wages in 2006 and 2014. We see a substantial increase in real wages
at the very bottom of the distribution in Hungary, Latvia, Lithuania, Poland, and Romania
(compared to the other points of the wage distribution). These changes were likely facilitated
by the institutional and economic changes discussed in section 2 (i.e., changes related to
minimum wage policies, sustained economic growth, and emigration), as there was greater
density around the minimum wage in 2014 than in 2006 (see Figure A.1 for kernel density
estimates) in the countries with narrowed lower-end wage distributions. It appears that in
Bulgaria, Estonia, and Slovakia, real wages increased equally across the wage distribution.
Finally, wages in Czechia increased slightly throughout most of the wage distribution, although
less at the very bottom and more at the very top.
16
Table 3: Contribution of the within component to level and change in variance of log wages
Level 2006 Change 2006-2014
(percent) (percent)
Estonia 60 70
Czechia 55 16
Slovakia 50 19
Lithuania 49 58
Hungary 48 25
Latvia 47 46
Poland 44 35
Romania 36 56
Bulgaria 29 51
Note: the first column shows the contribution of the within-firm component to the
level of the variance of log wages in 2006 (V ar(within2006 )
V ar( ˆwi,2006)). The unreported be-
tween component is 100% minus the reported within component. The second col-
umn shows the contribution of the within component to the change of the variance
(|∆V ar(within)|
(|∆V ar(within)|+|∆V ar(between)|)).
Data: European Structure of Earnings Survey.
Table 4: Variance decomposition
(a) Within-firm variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.10 0.11 0.18 0.15 0.16 0.12
2006 0.09 0.12 0.17 0.14 0.18 0.22 0.16 0.15 0.12
2010 0.10 0.11 0.15 0.14 0.16 0.16 0.15 0.14 0.11
2014 0.11 0.11 0.14 0.14 0.13 0.15 0.14 0.12 0.12
Data: European Structure of Earnings Survey.
(b) Between-firm variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.24 0.09 0.19 0.19 0.26 0.13
2006 0.23 0.09 0.11 0.15 0.18 0.25 0.20 0.26 0.12
2010 0.22 0.12 0.13 0.16 0.18 0.18 0.17 0.24 0.11
2014 0.22 0.12 0.12 0.15 0.15 0.17 0.18 0.24 0.11
Data: European Structure of Earnings Survey.
17
Figure 2: Cumulative Distributions of Hourly Wages: 2006 and 2014
(a) Bulgaria (b) Czechia (c) Estonia
(d) Hungary (e) Latvia (f) Lithuania
(g) Poland (h) Romania (i) Slovakia
Notes: For each percentile, the statistics are based on the minimum hourly wages among individuals in that
percentile of earnings in each year. The vertical axis is log-scaled. Changes in log hourly wages (2006-2014)
are shown in Figure A.2.
Data: European Structure of Earnings Survey.
6.2 The role of between- and within-firm wage inequality
The overall wage inequality at the country level arises from the dispersion in average wages
between firms, and from the inequality in wages that exists within firms. Thus, as we discussed
in the methodological section, we can decompose overall wage inequality into two components:
within-firm and between-firm wage inequality. Tables 3and 4summarize the results of such
18
an exercise.
The CEE countries differed primarily with respect to between-firm wage inequality, as this
component explained most of the existing differences in the total wage inequality levels between
countries (see Table 4). In 2014, within-firm wage inequality varied from 0.11 in Bulgaria and
Czechia to 0.15 in Latvia; while between-firm wage inequality ranged from 0.1 in Slovakia to
0.24 in Romania. Thus, between-firm wage inequality was the main contributor to differences
in the levels of total wage inequality among the CEE countries. The countries with high levels
of overall wage inequality (Romania, Bulgaria) had much higher levels of between-firm wage
inequality than the countries with low levels of overall wage inequality (Czechia, Slovakia),
whereas the levels of within-firm wage inequality in these two groups of countries were more
similar. The share of within-firm wage inequality in overall wage inequality varied from 33%
in Bulgaria to 54% in Estonia (in 2014). These patterns appear to be similar to those ob-
served in the four Western European countries to which we compare our results for the CEE
countries: i.e., in the Netherlands, Norway, Portugal, and Sweden, the levels of between-firm
wage inequality varied more than the levels of within-firm wage inequality.
In the CEE countries, between-firm wage inequality was both higher and more dispersed than
within-firm wage inequality in the early to mid-2000s as well. Among the CEE countries for
which 2002 data are available, within-firm wage inequality varied in 2002 from 0.10 in Bul-
garia to 0.18 in Lithuania, while the variance of wages between firms in 2002 ranged from a
low of 0.09 in Czechia to 0.26 in Romania. Thus, even in the early 2000s, between-firm wage
inequality accounted for the majority of the total wage inequality in all of the CEE countries
except for Czechia. It is important to note, however, that there was no single pattern of
changes over time. For instance, Romania saw a decrease in both within-firm and between-
firm wage inequality, but the decline was greater in the former than in the latter component.
By contrast, in Czechia, the increase in wage inequality was driven by the increase in the
between-firm variance of wages. In most of the CEE countries, both within-firm and between-
firm wage inequality decreased over the study period. Table 3shows that the between-firm
component was the main driver of the changes in wage inequality levels between 2006 and 2014
in five CEE countries (Czechia, Hungary, Latvia, Poland, and Slovakia). In Bulgaria, Estonia,
19
Lithuania, and Romania, the decrease in within-firm inequality contributed the most to the
changes in overall wage inequality. Interestingly, it appears that the decrease in between-firm
wage inequality was mainly attributable to the higher rates of growth in average wages in the
low- than in the high-paying firms, except in Czechia, Slovakia, and Estonia (cf. Figure A.3
in the Appendix).
In terms of both the absolute level and the share of total wage inequality, between-firm wage
inequality was generally higher in countries with higher levels of the overall variance of wages.
Interestingly, this was also the case in the Western European countries (see Table F.2 in the
Appendix, Card, Heining, and Kline (2013) for Germany and Barth et al. (2016) for the US).
In both Bulgaria and Portugal, between-firm wage inequality explained around two-thirds
of total wage inequality. This component played a smaller role in the Netherlands, where
the share of between-firm wage inequality was similar to the average level observed among
the CEE countries; and it played an even smaller role in Sweden, where between-firm wage
inequality accounted for only one-third of total wage inequality.
6.3 Residual variance
We now check whether our findings on the role of within-firm and between-firm wage inequal-
ity are robust once we account for observed worker and firm characteristics. To this end, we
estimate the Mincerian wage equation (equation (2)), calculate the variance of the residuals,
and then decompose this residual variance into within-firm and between-firm components.
Our findings indicate that residual wage inequality accounted for around 40-60% of overall
wage inequality (see Table 5). This means that while the observable characteristics of work-
ers and firms explained around one-half of the wage inequality in the CEE countries, the
remaining inequality was related to unobserved factors. These factors include both work-
ers’ characteristics that impact their productivity and wages, and firm-specific wage premia
(associated with, for instance, rent-sharing, union bargaining power, managerial inputs, or
efficiency wages). Moreover, when we look at residual wage inequality rather than at total
wage inequality, we see that the share of the within-firm variance is higher. This means that
20
Table 5: Residual variance decomposition
(a) Total residual variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.18 0.10 0.22 0.15 0.21 0.14
2006 0.17 0.10 0.14 0.14 0.22 0.31 0.15 0.21 0.12
2010 0.16 0.11 0.12 0.13 0.19 0.20 0.14 0.19 0.11
2014 0.16 0.10 0.13 0.13 0.17 0.18 0.14 0.18 0.12
Note: Table shows the decomposition of residual variance of normalised log gross hourly wages. The residuals
are calculated from the estimated Mincerian wage equation that includes worker and firm characteristics.
Data: European Structure of Earnings Survey.
(b) Within-firm residual variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.06 0.05 0.11 0.07 0.09 0.07
2006 0.05 0.06 0.09 0.07 0.11 0.14 0.08 0.08 0.07
2010 0.06 0.06 0.07 0.07 0.09 0.10 0.08 0.07 0.06
2014 0.06 0.06 0.09 0.08 0.08 0.10 0.08 0.07 0.07
Note: Table shows the decomposition of residual variance of normalised log gross hourly wages. The residuals
are calculated from the estimated Mincerian wage equation that includes worker and firm characteristics.
Data: European Structure of Earnings Survey.
(c) Between-firm residual variance of log wages
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.12 0.04 0.11 0.08 0.12 0.07
2006 0.11 0.04 0.05 0.07 0.11 0.17 0.07 0.12 0.06
2010 0.10 0.05 0.05 0.06 0.10 0.10 0.06 0.11 0.05
2014 0.10 0.04 0.05 0.05 0.09 0.09 0.06 0.11 0.05
Note: Table shows the decomposition of residual variance of normalised log gross hourly wages. The residuals
are calculated from the estimated Mincerian wage equation that includes worker and firm characteristics.
Data: European Structure of Earnings Survey.
a non-negligible share of wage inequality between firms is driven by their sectoral structure
and the heterogeneity of workers’ sorting into firms. Within-firm residual wage inequality ex-
plained almost 40% of total residual wage inequality in Bulgaria and Romania, around 47% in
Lithuania, and 50-70% in most other CEE countries. The share of within-firm residual wage
inequality was also higher in countries with lower levels of overall wage inequality; and it was
lower in high-inequality countries like Bulgaria and Romania, where between-firm (residual)
wage inequality was relatively high. These patterns are in line with those observed for the
overall wage levels. Thus, while a large share of wage inequality was attributable to observable
heterogeneity among workers within firms, workers’ sorting and inter-industry and firm wage
differentials drove between-firm wage inequality, as well as the differences in the role and the
size of this component across the CEE.
21
6.4 Microeconomic determinants of levels and changes in wage inequality
We continue our analysis by investigating the role of micro-level factors in shaping wage in-
equality in the CEE countries. We aim to capture the potential contributions of a set of
individual and firm characteristics to the observed wage inequality levels, and changes in
these contributions over time. First, we estimate RIF regressions that explain wage inequality
levels. Next, we display the results of the decomposition of changes in wage inequality over
time in order to show how the role of these characteristics changed over time along the wage
distribution and across countries. These results shed additional light on the role of firms, as we
find that firm-level factors are crucial in determining the levels and changes in wage inequality.
RIF regression results offer interesting insights into the contributions of micro factors to the
observed wage inequality levels in the CEE countries (see Tables B.1-B.5). We find that firm-
level characteristics were crucial in explaining the levels of wage inequality. First, we show
that sectoral affiliation was an important determinant of wage inequality, with financial and
insurance services contributing the most to increased levels in all countries. Second, we find
that peer effects played a large role: i.e., in all of the countries and years analyzed, workplaces
with large shares of tertiary-educated workers contributed substantially to increases in wage
inequality, while workplaces with large shares of older workers contributed to decreases in
wage inequality, all other things being equal. Third, we find that public sector workplaces
had lower levels of wage inequality. Finally, our results indicate that occupation was strongly
related to the level of wage inequality.
In order to analyze the contribution of micro-level factors to changes in wage inequality over
time, we decompose the above estimates using the approach by Firpo, Fortin, and Lemieux
(2018), as we discussed in the methodology section. This approach allows us to distinguish
between the composition effect (i.e., changes in individual characteristics and firm character-
istics) and the wage structure effect (i.e., returns to these characteristics) on the change in
the variance of log wages between 2006 and 2014. The analysis is performed for each country
separately. We find that composition effects (the changing structure of workers’ and firms’
characteristics) contributed to increases in wage inequality, while wage structure effects (which
22
Figure 3: Decomposition of overall change in variance of log wages into composition and wage
structure effects
Note: Figure shows the results of the decomposition of changes in the variance of log of normalized gross hourly
wages between 2006 and 2014 based on RIF regressions following Firpo, Fortin, and Lemieux (2018). Compo-
sition effects capture changes in log wages driven by changes in the covariates, assuming that the coefficients
remained at the 2006 level. The wage structure effects reflect the unexplained share of changes in log wages
due to changes in returns to covariates.
Data: European Structure of Earnings Survey.
23
reflect how much employers were willing to pay for these characteristics) contributed to de-
creases in inequality (see Figure 3). Thus, the overall observed pattern of decreasing wage
inequality resulted from larger changes in returns to covariates, rather than from changes in
covariates. Czechia was the only country where changes in covariates led to a (slight) increase
in wage inequality. In Bulgaria, Hungary, and Slovakia, inequality-increasing changes in co-
variates were offset by changes in returns, which resulted in stable wage inequality levels. The
largest composition effect was observed in Poland, and this effect would have led to increased
inequality had it not been offset by substantial wage structure effects. The largest inequality-
decreasing wage structure effect was found in Latvia, where the variance of wages decreased
substantially.
Changes in variance may mask developments at the tails of the wage distribution. Hence,
we additionally estimate unconditional quantile regressions, and decompose the changes in
real wage growth between 2006 and 2014 into composition and wage structure effects for each
decile of the wage distribution. This analysis helps us better understand where the changes
in wage inequality came from. The results show that in most countries, the largest wage
increases were in the lowest deciles, which led to a reduction in overall wage inequality (see
Figure 4). We find that these decreases in wage inequality were driven by large wage structure
effects (changes in the returns to covariates), which were concentrated at the bottom of the
distribution. In the absence of wage structure effects, wage inequality would have increased
because the wage gains driven by the composition effects were concentrated at the top of the
wage distribution.
We observe the largest changes in returns to wages (wage structure effects) at the bottom
of the wage distribution in the countries that also experienced the largest minimum wage
increases (Latvia, Poland, Romania, see Figure D.2). The only two countries in which the
wage structure effects were not the largest for the bottom decile were the countries where
the minimum wage increases were smaller than the increases in average earnings (Bulgaria
and Czechia). These findings reinforce our view that the institutional adjustments and the
variation in these adjustments across countries likely contributed to the narrowing of the wage
distribution. In addition, we observe the largest wage structure effects in the countries with
24
the greatest migration outflows (Latvia, Lithuania, Romania, see Figure D.3).
Figure 4: Decomposition of total wage change into composition and wage structure effects:
deciles
(a) Bulgaria (b) Czechia (c) Estonia
(d) Hungary (e) Latvia (f) Lithuania
(g) Poland (h) Romania (i) Slovakia
Notes: Figure shows the results of the decomposition of changes in log real hourly wages (euro) between
2006 and 2014 based on unconditional quantile regressions using RIF following Firpo, Fortin, and Lemieux
(2018). Composition effects capture changes in log wages driven by changes in the covariates, assuming that
the coefficients remained at the 2006 level. The wage structure effects reflect the unexplained share of changes
in log wages due to changes in returns to covariates.
Data: European Structure of Earnings Survey.
The detailed results of the decomposition provide us with interesting insights into the micro-
determinants of changes in inequality (Tables B.6-B.17 in the Appendix). The analysis of both
25
the changes in the variance and the quantiles show a coherent picture. First, we see that the
decline in returns to tertiary education was an important factor associated with decreasing
wage inequality. In most of the CEE countries, we observe that the decreases in returns to
tertiary education were largest at the top of the wage distribution. We see the opposite pat-
tern for the composition effects related to tertiary education, but find that these composition
effects were smaller than the wage structure effects 6. Second, we observe that the returns to
age became more inequality-increasing in all of the countries except for Poland. The largest
increases in the returns to age were found for the top decile of the wage distribution.
Finally, the decomposition of changes in quantiles of the wage distribution offers interesting
insights into the drivers of between-firm inequality. In particular, we see that the firm-level
share of tertiary educated workers increased more at the top than at the bottom of the wage
distribution. In several countries (Bulgaria, Latvia, Lithuania, and Hungary), these composi-
tion effects were compensated for by decreases in returns to firm characteristics, which again
were larger at the top than at the bottom of the wage distribution, contributing to its narrow-
ing. Finally, in all of the countries that experienced significant decreases in wage inequality
(except Lithuania), we observe that changes in the intercept contributed substantially to this
trend. These changes were likely linked to institutional adjustments, and may have affected
both within- and between-firm wage inequality7.
6.5 Robustness tests
We run two additional sets of analyses as robustness checks. First, since public sector employ-
ment constituted an important share of employment in the CEE countries during the study
period, and thus contributed to the observed decreases in wage inequality, we decided to run
an additional analysis that included private sector employees only. The results show that in
6There are several plausible explanations for the observed decline in returns to tertiary education. First, the
substantial increases in the supply of tertiary educated workforce could lower the price of high-skilled labor.
Minimum wages could have also contributed to the converging returns to covariates, as they mechanically
reduce the gap between high- and low-skilled workers. An increased outside option due to migration could
also decrease returns to tertiary education, as most emigrants performed low-skilled jobs.
7We find similar results for the role of firm-level factors in changes in the between-firm wage inequality when
we decompose changes in the variance of average wages at the firm level (Tables B.18 -B.20 in the Appendix).
26
the private sector, as in the total economy, changes in coefficients contributed to decreases in
inequality, and changes in endowments contributed to increases in inequality in all of the CEE
countries except Czechia. In other respects, however, there were no significant differences in
the results of the two RIF regressions (see Tables C.1-C.5 in the Appendix) and the decom-
position into wage structure and composition effects (see Tables C.6-C.8).
Second, in our analysis, we excluded observations with the top and the bottom 0.1% of wages
to account for potential errors in reporting wages. As this could bias our results, we replicate
the results using the full sample, and show that trimming of the sample does not affect the
results of the analysis (see Appendix C.1).
We acknowledge that we are likely not observing all of the changes at the top of the wage
distribution. First, high earners might have moved out of paid employment into other forms of
employment not captured by our data (self-employment, managerial contracts, etc.). Second,
we observe monetary benefits only, even though some of the changes in wage inequality might
have been driven by changes in non-monetary compensation, especially at the top of the labor
income distribution. We believe that these reservations do not alter our results.
7 Conclusions
Using harmonized linked employer-employee data, we analyzed changes in wage distributions
in Central and Eastern European countries. We found that in all but one of these countries,
wage inequality decreased between the early 2000s and the mid-2010s. Czechia, which still has
the lowest levels of wage inequality in the region, was the only CEE country that saw a slight
increase in wage inequality during this period. The reduction in wage inequality occurred
mainly in the lower part of the wage distribution, although the Baltic states saw a decrease
in wage dispersion in the upper part of the wage distribution as well.
27
Our aim was to shed more light on the factors associated with decreases in wage inequality
in the CEE. To do so, we first decomposed the variance of log wages into the within- and
between-firm components, and found that in both the early 2000s and 2014, wage inequality
in the CEE was greater between than within firms. Our results show that the contributions of
the between-firm component explained most of the cross-country differences in levels of wage
inequality, and in how these levels changed over time. Thus, it appears that wage inequality in
these countries was largely driven by where an individual was working, and with whom s/he
was working. The role of the between-firm component in wage inequality in the CEE countries
puts these countries closer to the US than to Sweden or Italy. High-inequality CEE countries
(Bulgaria and Romania in particular) continue to have much larger shares of between-firm
wage inequality than low-inequality CEE countries, mainly due to the greater heterogeneity
of their firms.
We further analyzed the micro determinants of changes in wage inequality by decomposing
changes in both the variance of log wages and the quantiles of wage distribution into wage
structure and composition effects. We found that the decline in wage inequality was mainly
driven by wage structure effects; i.e., by changes in the returns to individual and firm char-
acteristics. These changes were largest at the bottom of the wage distribution, and decreased
along the wage distribution. This pattern stands in contrast to the U-shaped wage struc-
ture effects observed in the US (Firpo, Fortin, & Lemieux, 2018), where wage premia have
increased both for low-paid and high-paid workers. We also found that like in the US, the
composition effects (changes in the structure of workers with respect to their individual and
workplace characteristics) in the CEE contributed to increases in wage inequality, as they were
concentrated at the top of the distribution. However, as the composition effects were much
smaller than the wage structure effects, there was a substantial reduction in wage inequality
in the CEE that was absent in most Western countries.
Among the micro factors that had an impact on changes in the wage distribution, educational
attainment stands out. The increase in the number of tertiary-educated workers was an impor-
tant inequality-increasing factor (in line with what Lemieux (2006) found for the US). In the
CEE countries, workers’ rising educational attainment levels were accompanied by decreasing
28
returns to tertiary education. We also found that firm-level characteristics played a key role
in determining both the levels and the changes in wage inequality, which further confirmed
the importance of the between-firm component of wage inequality. Finally, a large share of
changes remains unexplained by micro-level covariates, as this share is captured by changes
in the constant of the wage equation.
While the formal analysis of the potential impact of institutional changes on the wage dis-
tribution is beyond the scope of our paper, we argue that the substantial increases in the
minimum wages in the CEE countries we observed likely contributed to the narrowing of the
bottom levels of their wage distributions. Migration flows and cross-border commuting likely
also had an effect by increasing workers’ outside options. In line with these expectations, we
found that inequality decreased the most in the CEE countries that experienced the largest
minimum wage increases and very large migration outflows. The problem of how the effects
of particular institutional changes on wage inequality could be disentangled, while accounting
for potential reverse causality, would benefit from further research.
29
References
Abowd, J. M., Kramarz, F., & Margolis, D. N. (1999). High wage workers and high wage
firms. Econometrica,67 (2), 251–333. https://doi.org/10.1111/1468-0262.00020
Acemoglu, D., & Autor, D. H. (2011). Chapter 12 - skills, tasks and technologies: Implications
for employment and earnings (D. Card & O. Ashenfelter, Eds.). In D. Card & O.
Ashenfelter (Eds.). Elsevier. https://doi.org/10.1016/S0169-7218(11)02410-5
Akerman, A., Helpman, E., Itskhoki, O., Muendler, M.-A., & Redding, S. (2013). Sources of
wage inequality. American Economic Review,103(3), 214–219. https:// doi . org /10.
1257/aer.103.3.214
Alvarez, J., Benguria, F., Engbom, N., & Moser, C. (2018). Firms and the decline in earnings
inequality in Brazil. American Economic Journal: Macroeconomics,10 (1), 149–189.
https://doi.org/10.1257/mac.20150355
Antonczyk, D., Fitzenberger, B., & Sommerfeld, K. (2010). Rising wage inequality, the decline
of collective bargaining, and the gender wage gap. Labour Economics,17 (5), 835–847.
https://doi.org/10.1016/j.labeco.2010.04.008
Appleton, S., Song, L., & Xia, Q. (2014). Understanding urban wage inequality in China
1988–2008: Evidence from quantile analysis. World Development,62, 1–13. https://
doi.org/10.1016/j.worlddev.2014.04.005
Aristei, D., & Perugini, C. (2012). Inequality and reforms in transition countries. Economic
Systems,36 (1), 2–10. https://doi.org/10.1016/j.ecosys.2011.09.001
Aristei, D., & Perugini, C. (2014). Speed and sequencing of transition reforms and income
inequality: A panel data analysis. Review of Income and Wealth,60 (3), 542–570. https:
//doi.org/10.1111/roiw.12090
Atkinson, A. B. (1970). On the measurement of inequality. Journal of Economic Theory,2(3),
244–263.
Autor, D. H., Katz, L. F., & Kearney, M. S. (2006). The polarization of the U.S. labor
market. American Economic Review,96 (2), 189–194. https : / / doi . org / 10 . 1257 /
000282806777212620
Autor, D. H., Katz, L. F., & Kearney, M. S. (2008). Trends in U.S. wage inequality: Revising
the revisionists. Review of Economics and Statistics,90 (2), 300–323. https://doi.org/
10.1162/rest.90.2.300
Autor, D. H., Manning, A., & Smith, C. L. (2016). The contribution of the minimum wage to
US wage inequality over three decades: A reassessment. American Economic Journal:
Applied Economics,8(1), 58–99. https://doi.org/10.1257/app.20140073
Barth, E., Bryson, A., Davis, J. C., & Freeman, R. (2016). It’s where you work: Increases in
the dispersion of earnings across establishments and individuals in the United States.
Journal of Labor Economics,34 (S2), 244–263. https://doi.org/10.1086/684045
30
Bell, B., Bukowski, P., & Machin, S. (2018). Rent Sharing and Inclusive Growth (CEP Dis-
cussion Papers dp1584). Centre for Economic Performance, LSE. https://ideas.repec.
org/p/cep/cepdps/dp1584.html
Black, R., Pantıru, C., Okólski, M., & Engbersen, G. (2010). A continent moving west?:
EU enlargement and labour migration from Central and Eastern Europe. Amsterdam
University Press.
Brzezinski, M. (2018). Income inequality and the Great Recession in Central and Eastern
Europe. Economic Systems,42 (2), 219–247. https://doi.org/10.1016/j.ecosys.2017.
07.003
Card, D., & De La Rica, S. (2006). Firm-level contracting and the structure of wages in Spain.
Industrial and Labor Relations Review,59 (4), 573–592. https: // doi. org/ 10. 1177 /
001979390605900403
Card, D., Heining, J., & Kline, P. (2013). Workplace heterogeneity and the rise of West German
wage inequality. The Quarterly Journal of Economics,128 (3), 967–1015. https://doi.
org/10.1093/qje/qjt006
Card, D., Cardoso, A. R., Heining, J., & Kline, P. (2018). Firms and labor market inequality:
Evidence and some theory. Journal of Labor Economics,36 (S1), 13–70. https://doi.
org/10.1086/694153
Carstensen, K., & Toubal, F. (2004). Foreign direct investment in Central and Eastern Euro-
pean countries: A dynamic panel analysis. Journal of Comparative Economics,32 (1),
3–22. https://doi.org/10.1016/j.jce.2003.11.001
Cengiz, D., Dube, A., Lindner, A., & Zipperer, B. (2019). The Effect of Minimum Wages
on Low-Wage Jobs. The Quarterly Journal of Economics,134 (3), 1405–1454. https:
//doi.org/10.1093/qje/qjz014
Devicienti, F., Fanfani, B., & Maida, A. (2019). Collective bargaining and the evolution of
wage inequality in Italy. British Journal of Industrial Relations,57 (2), 377–407. https:
//doi.org/10.1111/bjir.12444
DiNardo, J., Fortin, N. M., & Lemieux, T. (1996). Labor market institutions and the distri-
bution of wages, 1973-1992: A semiparametric approach. Econometrica,64 (5), 1001–
1044. https://doi.org/10.2307/2171954
Du Caju, P., Kátay, G., Lamo, A., Nicolitsas, D., & Poelhekke, S. (2010). Inter-industry
wage differentials in EU countries: What do cross-country time varying data add to
the picture? Journal of the European Economic Association,8(2-3), 478–486. https:
//doi.org/10.1111/j.1542-4774.2010.tb00518.x
Dustmann, C., Ludsteck, J., & Schönberg, U. (2009). Revisiting the German wage structure.
The Quarterly Journal of Economics,124 (2), 843–881. https://doi.org/10.1162/qjec.
2009.124.2.843
Eurostat. (2014). D9 D1 dispersion ratio in the EU Member States. Retrieved September 21,
2018, from https://ec.europa.eu/eurostat/statistics-explained/index.php?title=File:
D9_D1_dispersion_ratio_in_the_EU_Member_States,_2014_update.png
31
Firpo, S., Fortin, N., & Lemieux, T. (2018). Decomposing wage distributions using recen-
tered influence function regressions. Econometrics,6.https : / / doi . org / 10 . 3390 /
econometrics6020028
Fortin, N., Green, D. A., Lemieux, T., Milligan, K., & Riddell, W. C. (2012). Canadian
inequality: Recent developments and policy options. Canadian Public Policy,38 (2),
121–145. https://doi.org/10.3138/cpp.38.2.121
Gaudecker, H.-M. von. (2014). Templates for reproducible research projects in economics [https:
//github.com/hmgaudecker/econ-project-templates]. https://github.com/hmgaudecker/
econ-project-templates.
Ge, S., & Yang, D. T. (2014). Changes In China’s Wage Structure. Journal of the European
Economic Association,12 (2), 300–336. https://doi.org/10.1111/jeea.12072
Goldschmidt, D., & Schmieder, J. F. (2017). The rise of domestic outsourcing and the evolution
of the German wage structure. The Quarterly Journal of Economics,132 (3), 1165–
1217. https://doi.org/10.1093/qje/qjx008
Handwerker, E. W., & Spletzer, J. R. (2016). The role of establishments and the concentration
of occupations in wage inequality. In Inequality: Causes and consequences (pp. 167–
193). https://doi.org/10.1108/S0147-912120160000043013
Helpman, E., Itskhoki, O., Muendler, M.-A., & Redding, S. (2017). Trade and inequality:
From theory to estimation. The Review of Economic Studies,84 (1), 357–405. https:
//doi.org/10.1093/restud/rdw025
Hoelscher, J., Perugini, C., & Pompei, F. (2011). Wage inequality, labour market flexibility
and duality in Eastern and Western Europe. Post-Communist Economies,23 (3), 271–
310. https://doi.org/10.1080/14631377.2011.595119
Kaczmarczyk, P., & Okólski, M. (2008). Demographic and labour-market impacts of migration
on Poland. Oxford Review of Economic Policy,24 (3), 599–624. https://doi.org/10.
1093/oxrep/grn029
Kahanec, M., & Zimmermann, K. F. (2016). EU post-enlargement migration and the Great
Recession: Lessons and policy implications, In Labor Migration, EU Enlargement, and
the Great Recession.https://doi.org/10.1007/978-3-662-45320-9_17
Keister, R., & Lewandowski, P. (2017). A routine transition in the digital era? The rise of
routine work in Central and Eastern Europe. Transfer: European Review of Labour
and Research,23 (3), 263–279. https://doi.org/10.1177/1024258917703557
Kelly, T., Liaplina, A., Tan, S. T., & Winkler, H. (2017). Reaping digital dividends: Leveraging
the internet for development in Europe and Central Asia. World Bank. https://doi.
org/10.1596/978-1-4648-1025-1
Kohl, H. (2009). Freedom of Association, Employees’ Rights and Social Dialogue in Central
and Eastern Europe and the Western Balkans (tech. rep.). Friedrich-Ebert-Stiftung
Division for International Dialogue.
32
Krishna, P., Poole, J. P., & Senses, M. Z. (2012). Trade, labor market frictions, and residual
wage inequality across worker groups. American Economic Review,102 (3), 417–423.
https://doi.org/10.1257/aer.102.3.417
Krueger, A. B., & Summers, L. H. (1988). Efficiency wages and the inter-industry wage struc-
ture. Econometrica,56 (2), 259–293. https://doi.org/10.2307/1911072
Lazear, E. P., & Shaw, K. L. (2009). The structure of wages: An international comparison.
University of Chicago Press. https://doi.org/10.7208/chicago/9780226470511.001.0001
Lemieux, T. (2006). Increasing residual wage inequality: Composition effects, noisy data, or
rising demand for skill? American Economic Review,96 (3), 461–498. https://doi.org/
10.1257/aer.96.3.461
Machin, S. (2016). Rising wage inequality, real wage stagnation and unions. In Inequal-
ity: Causes and consequences (pp. 329–354). https : / / doi . org / 10 . 1108 / S0147 -
912120160000043017
Magda, I., Marsden, D., & Moriconi, S. (2016). Lower coverage but stronger unions? Insti-
tutional changes and union wage premia in central europe. Journal of Comparative
Economics,44 (3), 638–656.
Martins, P. S. (2004). Industry wage premia: evidence from the wage distribution. Economics
Letters,83 (2), 157–163. https://doi.org/10.1016/j.econlet.2003.11.002
Messina, J., & Silva, J. (2017). Wage inequality in Latin America: Understanding the past to
prepare for the future. The World Bank. https://doi.org/10.1596/978-1-4648-1039-8
Milanovic, B., & Ersado, L. (2012). Reform and inequality during the transition: An analysis
using panel household survey data, 1990–2005, In Economies in transition. Springer.
https://doi.org/10.1057/9780230361836_4
Mysíková, M., & Večerník, J. (2018). Personal earnings inequality and polarization: The Czech
Republic in comparison with Austria and Poland. Eastern European Economics,56(1),
57–80. https://doi.org/10.1080/00128775.2017.1402685
Parteka, A., & Wolszczak-Derlacz, J. (2013). The impact of trade integration with the Euro-
pean Union on productivity in a posttransition economy: The case of Polish manufac-
turing sectors. Emerging markets Finance and trade,49 (2), 84–104. https://doi.org/
10.2753/REE1540-496X490206
Pissarides, C. A. (2000). Equilibrium unemployment theory. MIT press.
Pryor, F. L. (2014). A note on income inequality in East and Central Europe. Comparative
Economic Studies,56 (1), 42–51. https://doi.org/10.1057/ces.2013.31
Schaefer, D., & Singleton, C. (2019). Recent changes in British wage inequality: Evidence from
large firms and occupations. Scottish Journal of Political Economy.https://doi.org/
10.1111/sjpe.12225
Sedelmeier, U. (2008). After conditionality: Post-accession compliance with EU law in East
Central Europe. Journal of European Public Policy,15(6), 806–825. https://doi.org/
10.1080/13501760802196549
33
Song, J., Price, D. J., Guvenen, F., Bloom, N., & Wachter, T. von. (2019). Firming up in-
equality. The Quarterly Journal of Economics,134 (1), 1–50. https://doi.org/10.1093/
qje/qjy025
Tyrowicz, J., & Smyk, M. (2019). Wage inequality and structural change. Social Indicators
Research,141 (2), 503–538. https://doi.org/10.1007/s11205-018-1846-y
Zaiceva, A., & Zimmermann, K. F. (2008). Scale, diversity, and determinants of labour mi-
gration in Europe. Oxford Review of Economic Policy,24 (3), 427–451. https://doi.
org/10.1093/oxrep/grn028
34
Appendix A Descriptive statistics
A.1 Wage distribution
Figure A.1: Kernel density estimates
(a) Bulgaria
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(b) Czechia
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(c) Estonia
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(d) Hungary
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(e) Latvia
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(f) Lithuania
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(g) Poland
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(h) Romania
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
(i) Slovakia
0 .005 .01 .015 .02 .025
Density
0 100 200 300 400 500
Hourly wage (mean wage=100)
2006 2014
Notes: Graphs show kernel density estimates of hourly wages in 2006 and 2014. Dashed lines show the results
for 2006 and solid lines show the results for 2014. Wages are indexed so that mean wage equals 100. In order
to make the figures clear and comparable, we show results for wages not exceeding five times the mean wage.
Data: European Structure of Earnings Survey.
35
Figure A.2: Cumulative distributions of individual hourly wages: 2006-2014 change
(a) Bulgaria (b) Czechia (c) Estonia
(d) Hungary (e) Latvia (f) Lithuania
(g) Poland (h) Romania (i) Slovakia
Notes: Figure shows changes in percentiles of log real hourly wages between 2006 and 2014.
Data: European Structure of Earnings Survey.
36
Figure A.3: Cumulative distributions of average firm-level hourly wages: 2006-2014 change
(a) Bulgaria (b) Czechia (c) Estonia
(d) Hungary (e) Latvia (f) Lithuania
(g) Poland (h) Romania (i) Slovakia
Notes: Figure shows changes in percentiles of log average firm-level hourly wages between 2006 and 2014.
Data: European Structure of Earnings Survey.
37
A.2 Alternative measures of wage inequality
Table A.1: Gini coefficient
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.34 0.26 0.36 0.34 0.38 0.30
2006 0.35 0.27 0.31 0.32 0.35 0.39 0.35 0.38 0.30
2010 0.35 0.28 0.30 0.33 0.34 0.34 0.33 0.37 0.28
2014 0.36 0.28 0.30 0.32 0.32 0.33 0.34 0.37 0.29
Data: European Structure of Earnings Survey.
Table A.2: Atkinson index (= 2)
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.29 0.18 0.31 0.29 0.34 0.23
2006 0.28 0.19 0.25 0.26 0.30 0.37 0.30 0.34 0.22
2010 0.28 0.21 0.24 0.26 0.28 0.29 0.27 0.32 0.21
2014 0.28 0.21 0.24 0.25 0.25 0.27 0.27 0.31 0.21
Data: European Structure of Earnings Survey.
Table A.3: Theil T index
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.20 0.12 0.22 0.21 0.26 0.19
2006 0.23 0.14 0.16 0.19 0.21 0.26 0.21 0.26 0.17
2010 0.23 0.14 0.16 0.20 0.19 0.20 0.19 0.26 0.15
2014 0.24 0.15 0.15 0.19 0.17 0.20 0.20 0.25 0.15
Data: European Structure of Earnings Survey.
Table A.4: Decile dispersion ratio (90-10)
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 4.78 2.93 4.87 4.71 5.43 3.43
2006 4.10 3.14 3.95 4.03 4.86 6.12 5.00 5.57 3.34
2010 4.10 3.33 3.98 4.02 4.54 4.54 4.55 4.60 3.29
2014 4.18 3.36 3.84 3.63 4.00 4.01 4.58 4.61 3.38
Data: European Structure of Earnings Survey.
38
Table A.5: Decile dispersion ratio (50-10)
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 2.24 1.71 2.10 1.96 2.26 1.75
2006 1.68 1.71 1.93 1.84 2.20 2.49 2.07 2.24 1.72
2010 1.76 1.79 2.01 1.75 2.03 2.04 1.94 1.86 1.73
2014 1.69 1.81 1.97 1.65 1.88 1.84 1.90 1.82 1.74
Data: European Structure of Earnings Survey.
Table A.6: Decile dispersion ratio (90-50)
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 2.13 1.72 2.32 2.40 2.41 1.95
2006 2.44 1.83 2.05 2.19 2.21 2.46 2.41 2.49 1.95
2010 2.33 1.86 1.98 2.30 2.24 2.23 2.35 2.47 1.90
2014 2.47 1.86 1.95 2.20 2.13 2.17 2.41 2.53 1.94
Data: European Structure of Earnings Survey.
A.3 Variance of log wages: sectors
Table A.7: Variance of log wages: manufacturing and construction
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.38 0.19 0.34 0.30 0.36 0.22
2006 0.31 0.18 0.25 0.31 0.35 0.45 0.31 0.33 0.23
2010 0.29 0.19 0.24 0.29 0.29 0.31 0.25 0.32 0.20
2014 0.30 0.19 0.22 0.25 0.27 0.29 0.26 0.28 0.20
Data: European Structure of Earnings Survey.
Table A.8: Variance of log wages: market services
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.43 0.23 0.44 0.37 0.58 0.31
2006 0.40 0.29 0.33 0.39 0.39 0.52 0.38 0.52 0.31
2010 0.40 0.31 0.33 0.36 0.36 0.40 0.35 0.48 0.29
2014 0.40 0.31 0.32 0.33 0.32 0.38 0.36 0.46 0.28
Data: European Structure of Earnings Survey.
39
Table A.9: Variance of log wages: non-market services
year Bulgaria Czechia Estonia Hungary Lithuania Latvia Poland Romania Slovakia
2002 0.20 0.14 0.35 0.34 0.36 0.18
2006 0.26 0.16 0.28 0.21 0.33 0.42 0.32 0.41 0.18
2010 0.24 0.16 0.25 0.26 0.32 0.29 0.30 0.36 0.17
2014 0.24 0.17 0.25 0.27 0.27 0.26 0.29 0.35 0.19
Data: European Structure of Earnings Survey.
40
Appendix B Detailed results: RIF regressions and Decomposi-
tion into Composition and Wage Structure Effects
Table B.1: Results of RIF regression: Bulgaria and Romania
Bulgaria Romania
2002 2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.055*** 0.028*** -0.004 -0.015** 0.297*** 0.025*** 0.029*** -0.018***
secondary education -0.003 -0.025*** -0.043*** -0.021*** -0.026*** 0.004 -0.032*** -0.013***
reference: under 30 years old
30-49 years old -0.000 0.018*** 0.066*** 0.091*** -0.002 0.034*** 0.051*** 0.080***
50 years old or more 0.022*** 0.026*** 0.067*** 0.091*** 0.084*** 0.112*** 0.078*** 0.099***
reference: male
female -0.064*** -0.069*** -0.071*** -0.081*** -0.031*** -0.025*** -0.025*** -0.051***
reference: tenure of less than a year
tenure: 1-4 years -0.023*** 0.015*** -0.004 -0.009** -0.003 -0.013*** -0.001 -0.007**
tenure: 5-9 years -0.013*** 0.040*** 0.013*** 0.010** -0.017*** -0.012*** -0.004 -0.010**
tenure: 10 years or more 0.013*** 0.088*** 0.037*** 0.031*** 0.013** 0.018*** 0.016*** 0.038***
reference: ISCO 5
ISCO 1 0.411*** 0.553*** 0.558*** 0.650*** 0.480*** 0.991*** 0.635*** 0.673***
ISCO 2 0.069*** 0.183*** 0.145*** 0.215*** -0.216*** 0.280*** -0.035*** 0.109***
ISCO 3 -0.045*** -0.055*** -0.021*** -0.048*** -0.164*** -0.058*** -0.157*** -0.126***
ISCO 4 -0.092*** -0.102*** -0.122*** -0.117*** -0.321*** -0.177*** -0.251*** -0.186***
ISCO 6 -0.050** -0.044* 0.011 0.740*** -0.133*** -0.049** -0.109*** -0.036
ISCO 7 -0.041*** -0.050*** -0.080*** -0.089*** -0.191*** -0.092*** -0.175*** -0.122***
ISCO 8 -0.062*** -0.083*** -0.100*** -0.134*** -0.242*** -0.109*** -0.199*** -0.148***
ISCO 9 0.002 -0.000 0.022*** 0.014*** -0.081*** 0.040*** -0.018*** -0.034***
reference: permanent contract
fixed contract 0.003 0.068*** 0.021*** 0.034*** 0.024*** -0.046*** -0.016** -0.051***
Firm effects
reference: NACE C
NACE B 0.209*** 0.268*** 0.197*** 0.280*** 0.330*** 0.307*** 0.343*** 0.613***
NACE D+E 0.194*** 0.229*** 0.163*** 0.203*** 0.206*** 0.074*** 0.098*** 0.087***
NACE F -0.082*** -0.111*** -0.059*** -0.039*** -0.025*** 0.027*** -0.002 -0.029***
NACE G 0.004 -0.047*** -0.095*** -0.110*** 0.084*** 0.050*** -0.016*** -0.024***
NACE H+J -0.007 0.008 0.129*** 0.167*** 0.208*** 0.081*** 0.118*** 0.122***
NACE I 0.038*** -0.039*** -0.075*** -0.128*** 0.049*** 0.064*** -0.016** 0.024***
NACE K 0.267*** 0.216*** 0.071*** 0.004 0.607*** 0.703*** 0.560*** 0.344***
NACE L+M+N -0.019*** 0.035*** 0.106*** 0.068*** 0.002 0.115*** 0.015*** 0.003
NACE O -0.204*** -0.144*** -0.187*** -0.143*** 0.029*** 0.197*** 0.029*** 0.039***
NACE P -0.310*** -0.396*** -0.298*** -0.268*** -0.321*** -0.187*** -0.360*** -0.322***
NACE Q -0.208*** -0.111*** -0.151*** -0.107*** -0.066*** -0.067*** -0.116*** -0.103***
NACE R+S -0.093*** -0.012* -0.123*** -0.154*** 0.032*** -0.004 -0.149*** -0.131***
reference: private ownership of a firm
public ownership of a firm -0.067*** -0.078*** -0.110*** -0.112*** -0.072*** -0.024*** -0.017*** -0.061***
tenure: less than 2 years (share) 0.117*** 0.018*** 0.071*** 0.065*** 0.138*** 0.073*** 0.028*** 0.101***
age: 50 years or more (share) -0.486*** -0.375*** -0.374*** -0.445*** -0.361*** -0.284*** -0.160*** -0.198***
tertiary education (share) 0.250*** 0.378*** 0.404*** 0.325*** 0.497*** 0.245*** 0.488*** 0.533***
female (share) -0.058*** -0.046*** 0.003 -0.029*** 0.073*** 0.045*** -0.009 -0.027***
constant 0.488*** 0.390*** 0.326*** 0.345*** 0.429*** 0.255*** 0.333*** 0.219***
Observations 150,392 162,838 175,575 168,345 220,284 241,708 262,983 270,582
R-squared 0.175 0.187 0.198 0.217 0.221 0.260 0.227 0.250
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The co-
efficients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log
hourly wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages.
Dummy variables indicating 1-digit level occupational groups from International Standard Classification of Occupations (ISCO) are in-
cluded. There was no inconsistency in 1-digit level occupational groups between ISCO-88 and ISCO-08. ISCO 1 - managers, ISCO 2 -
Professional, ISCO 3 - Technicians and associate professionals, ISCO 4 - Clerical support workers, ISCO 5 - Service and sales workers,
ISCO 6 - Skilled agricultural, forestry and fishery workers, ISCO 7 - Craft and related trades workers, ISCO 8 - Plant and machine
operators, and assemblers, ISCO 9 - Elementary occupations. Dummy variables indicating NACE Level 1 sectors were included (NACE
Rev.2). Few Level 1 sectors were pooled for the reason of inconsistencies between NACE Rev.1 and NACE Rev.2. NACE B - Mining and
Quarrying, NACE C - Manufacturing, NACE D+E - Electricity, Gas, Steam and Air Conditioning Supply, Water Supply; Sewerage, Waste
Management and Remediation Activities, NACE F - Construction, NACE G - Wholesale and Retail Trade; Repair of Motor Vehicles and
Motorcycles, NACE H+J - Transportation and Storage, Information and Communication, NACE I - Accommodation and Food Service
Activities, NACE K - Financial and Insurance Activities, NACE L+M+N - Real Estate Activities, Professional, Scientific and Technical
Activities, Administrative and Support Service Activities, NACE O - Public Administration and Defence; Compulsory Social Security,
NACE P - Education, NACE Q - Human Health and Social Work Activities, NACE R+S - Arts, Entertainment and Recreation, Other
Service Activities.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
41
Table B.2: Results of RIF regression: Czechia and Slovakia
Czechia Slovakia
2002 2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.163*** 0.155*** 0.141*** 0.084*** 0.066*** 0.059*** 0.049*** 0.008***
secondary education -0.043*** -0.054*** -0.054*** -0.066*** -0.070*** -0.111*** -0.091*** -0.094***
reference: under 30 years old
30-49 years old 0.039*** 0.078*** 0.089*** 0.098*** 0.056*** 0.069*** 0.089*** 0.098***
50 years old or more 0.050*** 0.079*** 0.093*** 0.103*** 0.068*** 0.065*** 0.094*** 0.101***
reference: male
female -0.039*** -0.052*** -0.049*** -0.057*** -0.061*** -0.055*** -0.056*** -0.055***
reference: tenure of less than a year
tenure: 1-4 years -0.008*** -0.015*** -0.017*** -0.031*** -0.008*** 0.004** -0.003* -0.018***
tenure: 5-9 years -0.004*** 0.001 -0.017*** -0.033*** 0.008** 0.018*** 0.008*** -0.016***
tenure: 10 years or more -0.018*** -0.001 0.003*** -0.028*** -0.018*** 0.020*** 0.004** -0.017***
reference: ISCO 5
ISCO 1 0.274*** 0.312*** 0.366*** 0.447*** 0.350*** 0.465*** 0.474*** 0.411***
ISCO 2 -0.139*** -0.124*** -0.052*** -0.029*** -0.126*** -0.081*** -0.051*** -0.030***
ISCO 3 -0.098*** -0.101*** -0.107*** -0.129*** -0.158*** -0.127*** -0.112*** -0.095***
ISCO 4 -0.097*** -0.135*** -0.164*** -0.173*** -0.054*** -0.125*** -0.133*** -0.108***
ISCO 6 -0.009 -0.002 -0.069*** -0.066*** 0.038 0.018 -0.023** 0.031***
ISCO 7 -0.123*** -0.152*** -0.147*** -0.172*** -0.196*** -0.145*** -0.130*** -0.101***
ISCO 8 -0.128*** -0.161*** -0.157*** -0.171*** -0.193*** -0.163*** -0.153*** -0.114***
ISCO 9 0.005*** 0.044*** 0.058*** 0.068*** -0.033*** 0.005** -0.002 0.061***
reference: permanent contract
fixed contract 0.018*** 0.023*** 0.008*** -0.015*** 0.031*** 0.001 0.022*** 0.004***
Firm effects
reference: NACE C
NACE B -0.000 0.033*** 0.036*** 0.046*** -0.005 -0.061*** 0.030*** 0.061***
NACE D+E 0.002 0.094*** 0.064*** 0.055*** 0.153*** 0.157*** 0.098*** 0.098***
NACE F -0.007*** -0.012*** -0.005*** -0.047*** -0.038*** -0.028*** -0.011*** 0.008***
NACE G -0.016*** 0.013*** -0.012*** -0.003*** 0.054*** -0.025*** -0.022*** -0.025***
NACE H+J -0.009*** 0.069*** 0.103*** 0.088*** 0.039*** 0.029*** 0.073*** 0.092***
NACE I 0.028*** 0.017*** 0.156*** 0.072*** -0.009 0.015*** 0.010*** -0.006*
NACE K 0.053*** 0.265*** 0.200*** 0.169*** 0.077*** 0.123*** 0.078*** 0.066***
NACE L+M+N -0.013*** 0.013*** 0.046*** 0.030*** 0.122*** 0.046*** 0.020*** 0.041***
NACE O -0.103*** -0.024*** -0.031*** -0.071*** -0.033*** -0.041*** -0.031*** -0.038***
NACE P -0.155*** -0.108*** -0.158*** -0.187*** -0.198*** -0.179*** -0.172*** -0.112***
NACE Q -0.060*** -0.032*** -0.030*** -0.020*** 0.006 -0.038*** -0.017*** 0.012***
NACE R+S -0.055*** -0.007*** -0.037*** -0.052*** -0.087*** -0.048*** -0.100*** -0.031***
reference: private ownership of a firm
public ownership of a firm -0.037*** -0.086*** -0.093*** -0.082*** -0.089*** -0.095*** -0.109*** -0.114***
tenure: less than 2 years (share) 0.053*** 0.027*** 0.090*** 0.063*** 0.010** -0.053*** 0.020*** 0.022***
age: 50 years or more (share) -0.157*** -0.203*** -0.117*** -0.111*** -0.369*** -0.274*** -0.183*** -0.142***
tertiary education (share) 0.137*** 0.176*** 0.116*** 0.192*** 0.283*** 0.266*** 0.240*** 0.193***
female (share) 0.036*** 0.041*** 0.021*** 0.001 -0.014*** -0.021*** 0.020*** 0.036***
constant 0.286*** 0.290*** 0.255*** 0.307*** 0.440*** 0.441*** 0.313*** 0.276***
Observations 978,110 1,914,027 1,948,513 2,148,818 391,714 670,603 767,368 863,864
R-squared 0.183 0.201 0.207 0.219 0.130 0.200 0.216 0.191
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the
detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
42
Table B.3: Results of RIF regression: Estonia and Poland
Estonia Poland
2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.079*** 0.055*** 0.078*** 0.230*** 0.215*** 0.155*** 0.120***
secondary education -0.026*** -0.034*** -0.018*** -0.012*** -0.006*** -0.016*** -0.018***
reference: under 30 years old
30-49 years old 0.077*** 0.085*** 0.090*** 0.072*** 0.114*** 0.109*** 0.103***
50 years old or more 0.065*** 0.084*** 0.084*** 0.117*** 0.159*** 0.133*** 0.125***
reference: male
female -0.080*** -0.072*** -0.089*** -0.037*** -0.046*** -0.055*** -0.068***
reference: tenure of less than a year
tenure: 1-4 years -0.005 -0.008** 0.000 -0.047*** -0.004* -0.015*** -0.003*
tenure: 5-9 years 0.029*** -0.003 0.005 -0.043*** -0.010*** -0.020*** -0.015***
tenure: 10 years or more -0.004 -0.004 -0.005 -0.050*** -0.015*** 0.006*** 0.026***
reference: ISCO 5
ISCO 1 0.366*** 0.370*** 0.286*** 0.383*** 0.263*** 0.285*** 0.309***
ISCO 2 -0.018*** -0.045*** -0.048*** -0.021*** -0.099*** -0.027*** -0.023***
ISCO 3 -0.077*** -0.145*** -0.094*** -0.127*** -0.210*** -0.139*** -0.142***
ISCO 4 -0.123*** -0.166*** -0.147*** -0.184*** -0.220*** -0.180*** -0.170***
ISCO 6 0.257*** 0.025 -0.064* -0.061*** -0.142*** 0.026 -0.062***
ISCO 7 -0.060*** -0.125*** -0.087*** -0.092*** -0.128*** -0.078*** -0.087***
ISCO 8 -0.096*** -0.129*** -0.132*** -0.150*** -0.190*** -0.121*** -0.122***
ISCO 9 0.209*** 0.129*** 0.097*** -0.006** -0.048*** 0.009*** -0.001
reference: permanent contract
fixed contract 0.050*** 0.045*** 0.043***
Firm effects
reference: NACE C
NACE B -0.001 0.118*** 0.077*** 0.233*** 0.207*** 0.103*** 0.213***
NACE D+E -0.076*** 0.020** 0.039*** 0.028*** 0.017*** -0.010*** -0.010***
NACE F 0.044*** 0.038*** 0.019*** -0.015*** -0.022*** 0.009*** 0.015***
NACE G 0.037*** 0.002 0.028*** -0.003 -0.023*** -0.008*** 0.025***
NACE H+J 0.048*** 0.112*** 0.132*** 0.010*** 0.016*** 0.041*** 0.045***
NACE I 0.019** -0.007 -0.015** 0.040*** 0.013** -0.006 0.028***
NACE K 0.240*** 0.193*** 0.235*** -0.026*** 0.056*** 0.050*** 0.019***
NACE L+M+N 0.070*** 0.014*** 0.064*** 0.008*** 0.038*** 0.056*** 0.044***
NACE O -0.092*** -0.044*** -0.030*** -0.104*** -0.076*** -0.138*** -0.113***
NACE P -0.035*** -0.093*** -0.034*** -0.017*** 0.010*** 0.042*** 0.070***
NACE Q 0.101*** 0.090*** 0.075*** -0.077*** -0.129*** -0.080*** -0.083***
NACE R+S -0.004 -0.010 0.013* -0.082*** -0.048*** -0.073*** -0.078***
reference: private ownership of a firm
public ownership of a firm -0.037*** -0.060*** -0.033*** -0.131*** -0.118*** -0.084*** -0.069***
tenure: less than 2 years (share) -0.016** 0.036*** 0.029*** 0.109*** 0.135*** 0.081*** 0.083***
age: 50 years or more (share) -0.112*** -0.115*** -0.084*** -0.179*** -0.153*** -0.159*** -0.082***
tertiary education (share) 0.146*** 0.152*** 0.094*** 0.304*** 0.196*** 0.166*** 0.153***
female (share) 0.002 0.075*** 0.067*** 0.083*** 0.084*** 0.065*** 0.027***
constant 0.251*** 0.249*** 0.211*** 0.322*** 0.285*** 0.212*** 0.194***
Observations 114,656 108,903 112,569 629,101 639,784 667,963 707,999
R-squared 0.161 0.183 0.134 0.199 0.183 0.185 0.170
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the
detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
43
Table B.4: Results of RIF regression: Lithuania and Latvia
Lithuania Latvia
2002 2006 2010 2014 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.139*** 0.132*** 0.131*** 0.035*** 0.077*** 0.033*** 0.000
secondary education -0.010** -0.001 0.014 -0.012 -0.028*** -0.029*** -0.032***
reference: under 30 years old
30-49 years old 0.035*** 0.066*** 0.062*** 0.087*** 0.079*** 0.084*** 0.103***
50 years old or more 0.046*** 0.074*** 0.079*** 0.097*** 0.070*** 0.067*** 0.095***
reference: male
female -0.058*** -0.069*** -0.081*** -0.088*** -0.073*** -0.079*** -0.067***
reference: tenure of less than a year
tenure: 1-4 years 0.002 0.016*** 0.005 -0.013** 0.018*** -0.001 0.018***
tenure: 5-9 years 0.018*** 0.054*** 0.014* -0.008 0.052*** 0.015*** 0.021***
tenure: 10 years or more 0.028*** 0.048*** 0.052*** 0.008 0.038*** 0.006* 0.004
reference: ISCO 5
ISCO 1 0.310*** 0.274*** 0.235*** 0.329*** 0.320*** 0.331*** 0.322***
ISCO 2 -0.029*** -0.060*** -0.047*** -0.027*** 0.007 -0.007* 0.017***
ISCO 3 -0.041*** -0.060*** -0.100*** -0.088*** -0.100*** -0.100*** -0.104***
ISCO 4 -0.110*** -0.129*** -0.145*** -0.122*** -0.121*** -0.121*** -0.152***
ISCO 6 0.028 0.159*** 0.026 -0.165 0.103*** 0.038 0.081***
ISCO 7 -0.026*** -0.013** -0.046*** -0.060*** -0.034*** -0.059*** -0.047***
ISCO 8 -0.031*** -0.058*** -0.115*** -0.122*** -0.038*** -0.063*** -0.067***
ISCO 9 0.085*** 0.118*** 0.083*** 0.038*** 0.124*** 0.086*** 0.086***
reference: permanent contract
fixed contract -0.029*** 0.052*** 0.041*** -0.009 0.187*** 0.064*** 0.047***
Firm effects
reference: NACE C
NACE B 0.067*** 0.020 -0.026 -0.036 -0.100*** -0.111*** -0.027
NACE D+E 0.081*** 0.072*** 0.018 0.026* 0.120*** -0.013* 0.003
NACE F 0.012** 0.074*** -0.061*** -0.049*** -0.003 -0.032*** -0.050***
NACE G 0.012** 0.010** 0.011 -0.039*** 0.029*** -0.029*** -0.027***
NACE H+J 0.086*** 0.084*** 0.065*** 0.052*** 0.072*** 0.072*** 0.092***
NACE I 0.072*** 0.064*** 0.049*** 0.016 0.067*** 0.014* -0.034***
NACE K 0.260*** 0.300*** 0.244*** 0.247*** 0.286*** 0.281*** 0.322***
NACE L+M+N -0.022*** -0.007 0.006 0.007 0.068*** -0.023*** 0.008
NACE O 0.026*** 0.059*** -0.021 -0.035*** -0.081*** -0.186*** -0.191***
NACE P -0.012* 0.010 0.090*** 0.007 0.002 -0.113*** -0.146***
NACE Q -0.052*** 0.074*** 0.041*** 0.065*** 0.068*** -0.008 0.036***
NACE R+S -0.055*** -0.022*** -0.053*** -0.063*** 0.002 -0.101*** -0.089***
reference: private ownership of a firm
public ownership of a firm -0.083*** -0.119*** -0.100*** -0.089*** -0.147*** -0.061*** -0.075***
tenure: less than 2 years (share) 0.052*** 0.063*** 0.040*** 0.004 0.077*** 0.059*** 0.020***
age: 50 years or more (share) -0.205*** -0.096*** -0.080*** -0.067*** -0.183*** -0.171*** -0.158***
tertiary education (share) 0.359*** 0.231*** 0.161*** 0.144*** 0.349*** 0.367*** 0.402***
female (share) -0.026*** 0.014** -0.014 0.020** -0.022*** -0.019*** -0.031***
constant 0.284*** 0.195*** 0.237*** 0.240*** 0.341*** 0.277*** 0.226***
Observations 135,978 114,892 32,773 38,483 271,872 198,862 153,540
R-squared 0.159 0.132 0.138 0.146 0.117 0.166 0.157
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the
detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
44
Table B.5: Results of RIF regression: Hungary
Hungary
2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.217*** 0.205*** 0.095***
secondary education -0.028*** -0.032*** -0.085***
reference: under 30 years old
30-49 years old 0.080*** 0.091*** 0.096***
50 years old or more 0.106*** 0.106*** 0.120***
reference: male
female -0.064*** -0.077*** -0.069***
reference: tenure of less than a year
tenure: 1-4 years -0.014*** -0.048*** -0.113***
tenure: 5-9 years -0.005*** -0.049*** -0.091***
tenure: 10 years or more -0.021*** -0.041*** -0.103***
reference: ISCO 5
ISCO 1 0.341*** 0.351*** 0.428***
ISCO 2 -0.051*** -0.044*** 0.010***
ISCO 3 -0.101*** -0.112*** -0.080***
ISCO 4 -0.124*** -0.117*** -0.089***
ISCO 6 0.045*** 0.034*** 0.117***
ISCO 7 -0.109*** -0.122*** -0.090***
ISCO 8 -0.151*** -0.133*** -0.152***
ISCO 9 0.025*** 0.143*** 0.082***
reference: permanent contract
fixed contract 0.015*** -0.024*** -0.056***
Firm effects
reference: NACE C
NACE B 0.037*** 0.011 -0.012
NACE D+E 0.074*** 0.020*** -0.016***
NACE F 0.028*** -0.070*** -0.108***
NACE G 0.024*** -0.084*** -0.035***
NACE H+J 0.048*** 0.046*** 0.044***
NACE I -0.017*** -0.101*** -0.100***
NACE K 0.223*** 0.271*** 0.267***
NACE L+M+N 0.001 -0.024*** -0.071***
NACE O -0.027*** -0.081*** 0.044***
NACE P -0.320*** -0.379*** -0.223***
NACE Q -0.094*** -0.123*** -0.049***
NACE R+S -0.087*** -0.188*** -0.114***
reference: private ownership of a firm
public ownership of a firm -0.085*** -0.058*** -0.049***
tenure: less than 2 years (share) 0.079*** 0.100*** 0.148***
age: 50 years or more (share) -0.163*** -0.164*** -0.194***
tertiary education (share) 0.316*** 0.362*** 0.220***
female (share) -0.054*** -0.026*** -0.099***
constant 0.322*** 0.312*** 0.379***
Observations 676,050 781,240 770,148
R-squared 0.252 0.244 0.248
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo,
Fortin, & Lemieux, 2018). The coefficients measure the impact of an infinitesimal shift to the
right in the distribution of the regressors on variance of normalized log hourly wages in a given
country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1%
hourly wages. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
45
Table B.6: Decomposition of overall change in variance of log wages into composition and
wage structure effects: Bulgaria, Czechia and Estonia
Bulgaria Czechia Estonia
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.002*** -0.016*** 0.006*** -0.014*** -0.002*** -0.000
secondary education 0.000*** 0.002 0.001*** -0.009*** -0.000*** 0.005
reference: under 30 years old
30-49 years old -0.000*** 0.037*** 0.002*** 0.011*** -0.001*** 0.006**
50 years old or more 0.001*** 0.023*** -0.001*** 0.007*** 0.003*** 0.007***
reference: male
female -0.001*** -0.006** -0.001*** -0.003* 0.001*** -0.005
reference: tenure of less than a year
tenure: 1-4 years -0.000*** -0.008*** 0.000*** -0.005*** 0.001 0.002
tenure: 5-9 years 0.002*** -0.006*** 0.000 -0.008*** -0.000 -0.005***
tenure: 10 years or more -0.001*** -0.012*** -0.000 -0.008*** -0.000 -0.000
reference: ISCO 5
ISCO 1 0.003*** 0.005*** -0.006*** 0.006*** -0.001* -0.005***
ISCO 2 0.011*** 0.006*** -0.003*** 0.014*** -0.001* -0.006***
ISCO 3 0.000*** 0.001 0.002*** -0.006*** 0.001*** -0.003**
ISCO 4 0.001*** -0.001*** -0.001*** -0.003*** 0.000*** -0.002**
ISCO 6 -0.000*** 0.002*** -0.000 -0.000** 0.000** -0.000***
ISCO 7 0.001*** -0.005*** 0.006*** -0.003*** 0.001*** -0.003***
ISCO 8 0.002*** -0.007*** -0.001*** -0.002** 0.003*** -0.004***
ISCO 9 0.000 0.002*** -0.000*** 0.002*** 0.001*** -0.012***
reference: permanent contract
fixed contract -0.003*** -0.003*** 0.001*** -0.008*** -0.001*** -0.000
Firm effects
reference: NACE C
NACE B -0.001*** 0.000 -0.000*** 0.000** 0.000 0.001***
NACE D+E 0.001*** -0.001* 0.001*** -0.001*** 0.000*** 0.002***
NACE F 0.002*** 0.005*** 0.000*** -0.002*** 0.000*** -0.001**
NACE G -0.001*** -0.008*** -0.000** -0.002** 0.001*** -0.001
NACE H+J 0.000 0.015*** 0.001*** 0.002*** 0.001*** 0.008***
NACE I -0.000*** -0.004*** 0.000** 0.001*** 0.000* -0.001**
NACE K 0.002*** -0.006*** 0.001*** -0.002*** -0.002*** -0.000
NACE L+M+N 0.001*** 0.003*** 0.000** 0.002** 0.000*** -0.001
NACE O 0.001*** 0.000 -0.000*** -0.004*** 0.001*** 0.005***
NACE P 0.003*** 0.011*** -0.000 -0.006*** -0.000*** 0.000
NACE Q -0.001*** 0.000 -0.000*** 0.001* -0.000** -0.002**
NACE R+S 0.000 -0.003*** 0.000 -0.001*** 0.000 0.000
reference: private ownership of a firm
public ownership of a firm 0.005*** -0.009*** -0.001*** 0.001 0.001*** 0.001
tenure: less than 2 years (share) -0.001*** 0.020*** -0.000*** 0.012*** -0.000 0.017***
age: 50 years or more (share) -0.016*** -0.024*** 0.002*** 0.026*** -0.005*** 0.010*
tertiary education (share) 0.027*** -0.020*** 0.006*** 0.003 -0.010*** -0.014***
female (share) -0.001*** 0.009* 0.001*** -0.018*** 0.000 0.036***
constant -0.044*** 0.017** -0.040**
total 0.043*** -0.043*** 0.015*** 0.002 -0.008*** -0.005**
Observations 331,183 4,062,845 227,225
Notes: Table represent the results of the decomposition of changes in variance of normalized log hourly wages between 2006 and 2014 into
composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the top 0.1% and
the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
46
Table B.7: Decomposition of overall change in variance of log wages into composition and
wage structure effects: Latvia, Lithuania and Hungary
Latvia Lithuania Hungary
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.007*** -0.032*** 0.020*** -0.046*** 0.011*** -0.037***
secondary education 0.003*** -0.002 0.000 -0.005 0.001*** -0.032***
reference: under 30 years old
30-49 years old -0.002*** 0.011*** -0.005*** 0.010** 0.002*** 0.009***
50 years old or more 0.003*** 0.009*** 0.005*** 0.008** -0.001*** 0.004***
reference: male
female -0.001*** 0.003 -0.001*** -0.010** 0.002*** -0.002
reference: tenure of less than a year
tenure: 1-4 years -0.001*** 0.000 -0.000*** -0.009*** -0.000 -0.032***
tenure: 5-9 years 0.000*** -0.006*** 0.002*** -0.012*** 0.000 -0.016***
tenure: 10 years or more 0.002*** -0.009*** 0.001*** -0.011*** 0.001*** -0.022***
reference: ISCO 5
ISCO 1 -0.003*** 0.000 -0.006*** 0.005** -0.005*** 0.005***
ISCO 2 0.000 0.002 -0.003*** 0.010*** -0.001*** 0.012***
ISCO 3 0.001*** -0.001 0.000 -0.003** -0.001*** 0.004***
ISCO 4 0.002*** -0.002*** 0.001*** 0.000 0.004*** 0.002***
ISCO 6 -0.000*** -0.000 -0.000*** -0.000*** -0.000 0.000***
ISCO 7 0.001*** -0.002 0.001** -0.007*** -0.001*** 0.003***
ISCO 8 0.000*** -0.003*** 0.000** -0.007*** -0.002*** -0.000
ISCO 9 0.001*** -0.005*** -0.001*** -0.009*** 0.001*** 0.009***
reference: permanent contract
fixed contract 0.001*** -0.008*** 0.001*** -0.003*** -0.000*** -0.003***
Firm effects
reference: NACE C
NACE B -0.000*** 0.000*** -0.000 -0.000* 0.000** -0.000*
NACE D+E 0.001*** -0.003*** 0.000*** -0.001** -0.000*** -0.002***
NACE F 0.000 -0.003*** -0.002*** -0.009*** -0.000*** -0.005***
NACE G -0.000*** -0.007*** 0.000 -0.007*** 0.000 -0.006***
NACE H+J 0.002*** 0.002** 0.002*** -0.003*** 0.001*** -0.000
NACE I 0.000 -0.003*** 0.000*** -0.001*** 0.000* -0.002***
NACE K -0.001*** 0.001 -0.001*** -0.001 0.002*** 0.001**
NACE L+M+N -0.000*** -0.004*** -0.000 0.001 0.000 -0.005***
NACE O 0.001*** -0.010*** 0.000 -0.007*** -0.000*** 0.011***
NACE P 0.000 -0.029*** -0.000 -0.000 0.006*** 0.013***
NACE Q 0.001*** -0.003** 0.001*** -0.001 0.000*** 0.004***
NACE R+S -0.000 -0.002*** 0.000** -0.001** 0.002*** -0.001***
reference: private ownership of a firm
public ownership of a firm -0.004*** 0.033*** 0.002*** 0.011** 0.003*** 0.016***
tenure: less than 2 years (share) -0.005*** -0.023*** -0.004*** -0.023*** 0.006*** 0.029***
age: 50 years or more (share) -0.009*** 0.009 -0.007*** 0.010 0.001*** -0.009***
tertiary education (share) 0.030*** 0.022*** 0.035*** -0.040*** 0.015*** -0.031***
female (share) -0.000*** -0.005 0.000* 0.003 0.001*** -0.023***
constant -0.115*** 0.045** 0.057***
total 0.030*** -0.183*** 0.043*** -0.115*** 0.046*** -0.050***
Observations 425,412 153,375 1,446,198
Notes: Table represent the results of the decomposition of changes in variance of normalized log hourly wages between 2006 and 2014 into
composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the top 0.1% and
the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
47
Table B.8: Decomposition of overall change in variance of log wages into composition and
wage structure effects: Poland, Romania and Slovakia
Poland Romania Slovakia
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.020*** -0.037*** 0.000*** -0.014*** 0.005*** -0.016***
secondary education 0.000*** -0.007*** 0.000 -0.011*** 0.009*** 0.010***
reference: under 30 years old
30-49 years old -0.001*** -0.006*** -0.000 0.027*** -0.001*** 0.015***
50 years old or more 0.007*** -0.009*** 0.004*** -0.003** 0.003*** 0.011***
reference: male
female -0.000*** -0.011*** -0.000** -0.012*** -0.000*** 0.000
reference: tenure of less than a year
tenure: 1-4 years 0.000* 0.000 0.001*** 0.002 -0.000 -0.006***
tenure: 5-9 years -0.000*** -0.001 -0.001** 0.001 0.001*** -0.008***
tenure: 10 years or more -0.000** 0.015*** 0.000*** 0.006** 0.001*** -0.012***
reference: ISCO 5
ISCO 1 0.004*** 0.004*** 0.012*** -0.018*** 0.003*** -0.003***
ISCO 2 -0.003*** 0.020*** 0.023*** -0.039*** -0.005*** 0.010***
ISCO 3 0.003*** 0.008*** 0.002*** -0.006*** 0.007*** 0.005***
ISCO 4 0.002*** 0.004*** 0.001*** -0.001 -0.002*** 0.001***
ISCO 6 -0.000*** 0.000*** 0.000** 0.000 -0.000 0.000
ISCO 7 0.003*** 0.006*** 0.004*** -0.005*** 0.010*** 0.005***
ISCO 8 0.000*** 0.008*** 0.005*** -0.005*** -0.000*** 0.008***
ISCO 9 0.001*** 0.004*** 0.001*** -0.011*** -0.000 0.004***
Firm effects
reference: NACE C
NACE B -0.000*** 0.000 -0.004*** 0.003*** 0.000*** 0.001***
NACE D+E 0.000*** -0.001*** 0.000*** 0.001 0.001*** -0.002***
NACE F -0.000*** 0.002*** -0.000*** -0.004*** 0.001*** 0.001***
NACE G -0.000*** 0.007*** 0.001*** -0.009*** -0.000*** 0.000
NACE H+J 0.000*** 0.003*** 0.002*** 0.004*** 0.001*** 0.006***
NACE I 0.000** 0.000** 0.001*** -0.001*** -0.000 -0.000*
NACE K 0.000*** -0.001*** 0.001*** -0.005*** 0.001*** -0.001***
NACE L+M+N 0.000*** 0.000 0.004*** -0.011*** 0.001*** -0.000
NACE O -0.000*** -0.002*** -0.001*** -0.010*** -0.001*** 0.000
NACE P -0.000** 0.008*** 0.001*** -0.012*** -0.003*** 0.008***
NACE Q 0.002*** 0.003*** -0.000*** -0.003*** 0.000*** 0.004***
NACE R+S 0.000*** -0.000*** -0.000 -0.003*** 0.001*** 0.000*
reference: private ownership of a firm
public ownership of a firm 0.008*** 0.017*** 0.001*** -0.011*** -0.001*** -0.006***
tenure: less than 2 years (share) -0.001*** -0.015*** -0.005*** 0.009*** 0.004*** 0.022***
age: 50 years or more (share) -0.007*** 0.019*** -0.011*** 0.022*** -0.011*** 0.042***
tertiary education (share) 0.018*** -0.017*** 0.002*** 0.093*** 0.024*** -0.022***
female (share) 0.000*** -0.028*** 0.000*** -0.034*** -0.000*** 0.029***
reference: permanent contract
fixed contract -0.001*** -0.000 0.000 0.001
constant -0.091*** -0.035*** -0.165***
total 0.059*** -0.097*** 0.041*** -0.096*** 0.047*** -0.058***
Observations 1,347,783 512,290 1,534,467
Notes: Table represent the results of the decomposition of changes in variance of normalized log hourly wages between 2006 and 2014 into
composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the top 0.1% and
the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
48
Table B.9: Decomposition of total wage change into composition and wage structure effects:
Bulgaria, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education -0.001*** -0.001 0.008*** 0.015*** 0.005*** -0.019**
secondary education 0.000 0.001 0.000 0.043*** 0.001*** 0.032***
reference: under 30 years old
30-49 years old -0.002*** -0.042*** -0.002*** -0.043*** -0.002*** 0.044***
50 years old or more 0.005*** -0.031*** 0.005*** -0.029*** 0.006*** 0.017***
reference: male
female 0.000*** 0.002 -0.001*** 0.002 -0.002*** -0.018***
reference: tenure of less than a year
tenure: 1-4 years -0.001*** 0.002 -0.001*** -0.004 -0.002*** -0.009**
tenure: 5-9 years 0.003*** 0.006*** 0.009*** -0.005** 0.008*** -0.004
tenure: 10 years or more -0.000*** 0.010*** -0.002*** -0.020*** -0.002*** -0.024***
reference: ISCO 5
ISCO 1 0.000*** 0.005*** 0.007*** -0.014*** 0.009*** 0.024***
ISCO 2 0.004*** 0.015*** 0.059*** -0.070*** 0.039*** 0.047***
ISCO 3 -0.001*** 0.010*** -0.007*** -0.029*** -0.001*** 0.013***
ISCO 4 -0.000*** 0.005*** -0.003*** -0.011*** 0.001*** -0.002
ISCO 6 -0.000 0.000*** -0.000 0.001*** -0.000** 0.006***
ISCO 7 -0.001*** 0.013*** -0.009*** -0.008*** -0.001*** -0.008***
ISCO 8 -0.002*** 0.012*** -0.011*** -0.024*** 0.001*** -0.014***
ISCO 9 0.002*** -0.002** 0.002*** -0.004*** 0.002*** -0.001
reference: permanent contract
fixed contract -0.000 0.002*** -0.001*** 0.002*** -0.006*** -0.004***
Firm effects
reference: NACE C
NACE B -0.000*** 0.000*** -0.001*** -0.001*** -0.003*** 0.004***
NACE D+E 0.000** 0.001*** 0.003*** -0.011*** 0.004*** -0.001
NACE F -0.002*** -0.009*** -0.000 -0.002*** 0.004*** 0.008***
NACE G 0.000** 0.011*** -0.001** -0.001 -0.003*** -0.005**
NACE H+J -0.000 0.000 0.000 -0.002 -0.002*** 0.038***
NACE I 0.001*** 0.003*** -0.000 -0.001** -0.001*** -0.004***
NACE K 0.000*** -0.001*** 0.003*** -0.004*** 0.007*** -0.016***
NACE L+M+N -0.004*** 0.004*** -0.004*** 0.002 -0.002*** 0.012***
NACE O 0.000*** -0.001*** 0.000*** -0.011*** 0.003*** -0.005**
NACE P 0.001*** 0.001** 0.003*** 0.023*** 0.008*** 0.027***
NACE Q -0.001*** 0.003*** -0.005*** 0.022*** -0.004*** 0.002
NACE R+S 0.002*** -0.001*** 0.005*** -0.005*** 0.004*** -0.009***
reference: private ownership of a firm
public ownership of a firm -0.003*** 0.010*** -0.012*** -0.022*** 0.005*** -0.010
tenure: less than 2 years (share) 0.004*** -0.022*** 0.023*** 0.078*** 0.013*** 0.082***
age: 50 years or more (share) 0.002*** -0.046*** 0.001** -0.161*** -0.028*** -0.224***
tertiary education (share) 0.011*** 0.001 0.035*** -0.009 0.073*** -0.002
female (share) -0.000*** 0.023*** -0.002*** -0.042*** -0.003*** 0.042***
constant 0.489*** 0.738*** 0.359***
total 0.018*** 0.472*** 0.102*** 0.395*** 0.133*** 0.375***
Observations 331,183 331,183 331,183
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
49
Table B.10: Decomposition of total wage change into composition and wage structure effects:
Czechia, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.005*** 0.012*** 0.007*** -0.006*** 0.018*** -0.020***
secondary education -0.002*** 0.040*** -0.001*** -0.001 0.001*** 0.019***
reference: under 30 years old
30-49 years old -0.000*** -0.023*** 0.002*** 0.001 0.006*** 0.023***
50 years old or more 0.000** -0.012*** -0.001*** -0.003** -0.002*** 0.014***
reference: male
female -0.002*** 0.021*** -0.003*** 0.010*** -0.005*** 0.006**
reference: tenure of less than a year
tenure: 1-4 years -0.003*** -0.001 -0.003*** -0.003** -0.002*** -0.014***
tenure: 5-9 years 0.001*** 0.001 0.001*** -0.005*** 0.001*** -0.019***
tenure: 10 years or more 0.002*** 0.007** 0.002*** 0.000 0.002*** -0.016***
reference: ISCO 5
ISCO 1 -0.009*** -0.001** -0.009*** 0.000 -0.020*** 0.011***
ISCO 2 0.014*** -0.014*** 0.012*** -0.001 0.006*** 0.035***
ISCO 3 -0.009*** -0.011*** -0.007*** -0.006*** -0.004*** -0.024***
ISCO 4 0.003*** 0.001 0.001*** 0.005*** -0.001*** -0.005***
ISCO 6 0.000 0.000 -0.000 -0.000 0.000 -0.001
ISCO 7 -0.013*** 0.002 -0.003*** -0.001 0.009*** -0.009***
ISCO 8 0.003*** -0.008*** 0.000*** -0.007*** -0.002*** -0.016***
ISCO 9 0.001*** -0.005*** 0.000*** -0.001*** 0.000*** -0.003***
reference: permanent contract
fixed contract -0.004*** 0.018*** -0.003*** -0.004*** -0.001** -0.009***
Firm effects
reference: NACE C
NACE B -0.000*** 0.000 -0.001*** 0.000*** -0.001*** 0.000
NACE D+E -0.000*** 0.000 0.001*** -0.002*** 0.002*** -0.003***
NACE F 0.001*** 0.001* 0.001*** -0.004*** 0.001*** -0.004***
NACE G 0.001*** 0.002* 0.000*** -0.004*** 0.000 -0.008***
NACE H+J -0.000*** -0.017*** 0.001*** -0.007*** 0.003*** -0.007***
NACE I -0.001*** -0.010*** -0.000*** -0.003*** -0.000*** -0.004***
NACE K -0.000*** -0.002*** 0.000*** -0.002*** 0.002*** -0.007***
NACE L+M+N -0.004*** -0.002** -0.001*** -0.000 -0.001*** -0.005***
NACE O -0.001*** -0.014*** -0.000 -0.011*** 0.000** -0.033***
NACE P -0.000 -0.009*** -0.000 -0.005*** -0.000 -0.024***
NACE Q 0.000 -0.009*** -0.000 -0.006*** -0.001*** -0.008***
NACE R+S 0.000*** -0.001 0.000*** -0.002*** 0.000*** -0.004***
reference: private ownership of a firm
public ownership of a firm 0.001*** 0.021*** 0.001*** 0.004** -0.001*** 0.017***
tenure: less than 2 years (share) 0.002*** -0.069*** 0.001*** -0.041*** -0.000* -0.028***
age: 50 years or more (share) 0.003*** -0.079*** 0.004*** 0.010** 0.007*** 0.022***
tertiary education (share) 0.011*** -0.000 0.010*** -0.001 0.024*** -0.005
female (share) -0.004*** 0.092*** -0.005*** 0.023*** -0.003*** 0.035***
constant 0.216*** 0.254*** 0.248***
total -0.007*** 0.146*** 0.008*** 0.182*** 0.039*** 0.156***
Observations 4,062,845 4,062,845 4,062,845
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
50
Table B.11: Decomposition of total wage change into composition and wage structure effects:
Estonia, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education -0.003*** -0.008 -0.005*** -0.008** -0.007*** 0.003
secondary education 0.001*** -0.020 0.001*** -0.006 0.000 0.004
reference: under 30 years old
30-49 years old 0.001*** -0.013** 0.000 0.022*** -0.003*** 0.025***
50 years old or more -0.006*** -0.009 -0.003*** 0.009** 0.002** 0.027***
reference: male
female 0.001*** 0.031*** 0.002*** 0.012** 0.003*** 0.006
reference: tenure of less than a year
tenure: 1-4 years -0.006*** -0.010 -0.013*** -0.014*** -0.007*** -0.010*
tenure: 5-9 years -0.000 0.000 -0.000 -0.016*** -0.000 -0.018***
tenure: 10 years or more 0.001*** -0.001 0.002*** -0.014*** 0.001*** -0.010
reference: ISCO 5
ISCO 1 -0.001* -0.002 -0.001* -0.003*** -0.002* -0.015***
ISCO 2 0.019*** -0.001 0.021*** -0.016*** 0.019*** -0.021***
ISCO 3 -0.004*** -0.011*** -0.004*** -0.001 -0.002*** -0.014***
ISCO 4 -0.001*** -0.001 -0.001*** 0.003** 0.000*** -0.002
ISCO 6 -0.000 0.001* 0.000 -0.000 0.000*** -0.000***
ISCO 7 -0.001*** -0.002 -0.002*** -0.000 -0.000 -0.009***
ISCO 8 -0.006*** -0.001 -0.005*** -0.005*** 0.002*** -0.009***
ISCO 9 -0.003*** 0.033*** -0.000*** 0.003* 0.000 -0.004**
reference: permanent contract
fixed contract -0.001** -0.004*** -0.000 -0.002*** -0.002*** -0.003*
Firm effects
reference: NACE C
NACE B -0.000*** -0.001*** -0.000** 0.002*** -0.001*** 0.002**
NACE D+E -0.000 -0.003*** 0.000** 0.001** 0.001*** 0.004***
NACE F -0.000*** -0.004*** 0.000*** -0.004*** 0.000*** -0.008***
NACE G -0.002*** -0.005 0.001*** -0.011*** 0.002*** -0.012***
NACE H+J -0.001*** -0.019*** 0.002*** -0.013*** 0.001*** 0.006**
NACE I -0.001 -0.000 0.001*** -0.005*** 0.000 -0.002**
NACE K 0.001*** -0.002*** -0.001*** -0.002*** -0.004*** -0.003**
NACE L+M+N -0.000** -0.010*** -0.000 -0.002 0.000*** -0.006**
NACE O 0.001*** -0.008*** 0.001*** 0.010*** 0.003*** 0.015***
NACE P -0.003*** -0.011*** -0.001*** 0.001 -0.003*** 0.003
NACE Q 0.000 -0.002 -0.000** -0.003* -0.001** -0.007***
NACE R+S 0.001*** -0.005*** 0.001*** -0.004*** 0.001*** -0.002
reference: private ownership of a firm
public ownership of a firm 0.002*** 0.020*** 0.001** -0.003 0.004*** 0.004
tenure: less than 2 years (share) 0.002*** -0.097*** 0.002*** -0.061*** 0.001 -0.024*
age: 50 years or more (share) -0.016*** -0.034** -0.016*** -0.005 -0.029*** 0.022
tertiary education (share) -0.011*** 0.010 -0.021*** 0.032*** -0.034*** -0.023**
female (share) -0.000 -0.023 -0.000 -0.017 -0.000 0.085***
constant 0.571*** 0.503*** 0.349***
total -0.036*** 0.360*** -0.040*** 0.383*** -0.054*** 0.351***
Observations 227,225 227,225 227,225
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
51
Table B.12: Decomposition of total wage change into composition and wage structure effects:
Latvia, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.004*** -0.000 0.027*** -0.043*** 0.021*** -0.072***
secondary education -0.002** -0.001 -0.006*** -0.024*** 0.001 -0.025***
reference: under 30 years old
30-49 years old 0.001*** 0.013*** 0.003*** 0.023*** -0.002*** 0.052***
50 years old or more -0.003*** 0.013*** -0.009*** 0.035*** 0.001** 0.042***
reference: male
female -0.000*** 0.000 -0.001*** -0.004 -0.002*** -0.021**
reference: tenure of less than a year
tenure: 1-4 years -0.001*** -0.012*** -0.003*** 0.003 -0.003*** -0.001
tenure: 5-9 years 0.001*** -0.008*** 0.002*** 0.008*** 0.002*** -0.013***
tenure: 10 years or more 0.006*** -0.013*** 0.019*** -0.021*** 0.012*** -0.029***
reference: ISCO 5
ISCO 1 -0.001*** -0.009*** -0.007*** -0.009*** -0.008*** 0.016***
ISCO 2 0.007*** -0.009*** 0.037*** -0.042*** 0.016*** 0.024***
ISCO 3 -0.002*** -0.010*** -0.007*** -0.010*** -0.001*** -0.013***
ISCO 4 -0.002*** -0.006*** -0.005*** -0.002* 0.001*** -0.008***
ISCO 6 0.000*** 0.000** 0.000 0.000** -0.000* 0.001**
ISCO 7 -0.001*** -0.005*** -0.007*** 0.001 0.000** -0.001
ISCO 8 -0.000 -0.004*** -0.001*** -0.003 0.000*** -0.009***
ISCO 9 -0.001*** -0.008*** -0.000*** 0.004* 0.000** 0.002
reference: permanent contract
fixed contract -0.000*** 0.005*** 0.000*** 0.004*** 0.002*** -0.003**
Firm effects
reference: NACE C
NACE B 0.000*** -0.000*** 0.000*** -0.001*** 0.000*** -0.001**
NACE D+E -0.001*** 0.001*** 0.002*** -0.003*** 0.003*** -0.008***
NACE F -0.000*** 0.000 0.000*** 0.002 -0.000*** -0.011***
NACE G 0.001*** 0.010*** -0.000 -0.001 0.000 -0.001
NACE H+J -0.002*** 0.003*** 0.001*** 0.001 0.002*** 0.012***
NACE I -0.000 0.005*** -0.000 0.002*** -0.000 -0.001*
NACE K -0.000** 0.000 -0.001*** -0.001*** -0.002*** 0.002
NACE L+M+N 0.000*** 0.006*** 0.001*** 0.013*** -0.000* -0.001
NACE O 0.002*** 0.012*** 0.004*** 0.018*** 0.004*** -0.020***
NACE P -0.010*** 0.007*** -0.023*** 0.016*** -0.016*** -0.068***
NACE Q -0.000*** -0.000 -0.002*** 0.002 0.002*** -0.013***
NACE R+S 0.004*** 0.003*** 0.009*** 0.003*** 0.006*** -0.004***
reference: private ownership of a firm
public ownership of a firm 0.004*** -0.051*** 0.003*** 0.007 -0.004*** 0.012
tenure: less than 2 years (share) 0.003*** 0.029*** 0.008*** -0.082*** -0.002*** -0.076***
age: 50 years or more (share) 0.001* -0.032*** -0.002*** -0.206*** -0.018*** -0.083***
tertiary education (share) 0.010*** -0.071*** 0.045*** -0.071*** 0.068*** 0.083***
female (share) -0.001*** 0.019*** -0.002*** -0.113*** -0.002*** -0.045***
constant 0.736*** 0.748*** 0.436***
total 0.018*** 0.623*** 0.084*** 0.256*** 0.081*** 0.153***
Observations 425,412 425,412 425,412
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
52
Table B.13: Decomposition of total wage change into composition and wage structure effects:
Lithuania, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.009*** 0.002 0.057*** -0.064*** 0.059*** -0.114***
secondary education -0.000 0.004 -0.006*** 0.003 -0.003*** -0.017**
reference: under 30 years old
30-49 years old 0.002*** -0.004 0.003*** 0.012 -0.005*** 0.025***
50 years old or more -0.003*** 0.004 -0.007*** 0.017** 0.005*** 0.028***
reference: male
female -0.000*** 0.004 -0.002*** 0.014* -0.003*** -0.020*
reference: tenure of less than a year
tenure: 1-4 years -0.000*** -0.003 -0.002*** 0.003 -0.002*** -0.013**
tenure: 5-9 years 0.001*** 0.000 0.006*** -0.000 0.006*** -0.016***
tenure: 10 years or more 0.002*** -0.005 0.009*** -0.014** 0.007*** -0.026***
reference: ISCO 5
ISCO 1 -0.001*** 0.004** -0.012*** 0.013*** -0.017*** 0.012***
ISCO 2 0.004*** 0.027*** 0.028*** 0.048*** 0.005*** 0.054***
ISCO 3 -0.000 0.004** -0.000 0.010*** 0.000 -0.003
ISCO 4 -0.000*** 0.000 -0.001*** -0.001 0.001*** 0.000
ISCO 6 0.000** 0.000** 0.000** 0.000 -0.000 -0.000**
ISCO 7 -0.000 -0.001 -0.006*** 0.015*** -0.002*** -0.017***
ISCO 8 -0.000** 0.004** -0.001** 0.006* -0.000 -0.011***
ISCO 9 0.001*** 0.003 0.001*** 0.008*** -0.000** -0.002
reference: permanent contract
fixed contract 0.001*** 0.002*** 0.001*** -0.002 0.003*** -0.010***
Firm effects
reference: NACE C
NACE B -0.000* -0.000 -0.000* 0.000 -0.000* -0.000***
NACE D+E -0.000** 0.001*** 0.000*** -0.004*** 0.001*** -0.003**
NACE F -0.001*** -0.009*** -0.007*** -0.038*** -0.007*** -0.037***
NACE G -0.000*** 0.003 -0.001*** -0.009** -0.000** -0.013***
NACE H+J -0.002*** 0.001 -0.007*** -0.001 -0.001 0.004
NACE I -0.000** 0.001 -0.001*** -0.000 -0.000** -0.002**
NACE K -0.000** -0.001*** -0.001*** -0.002*** -0.001*** -0.003***
NACE L+M+N -0.000*** 0.000 -0.004*** 0.006*** -0.002*** 0.003
NACE O -0.000 0.001 -0.000 -0.003 -0.000 -0.008**
NACE P 0.001*** -0.005 0.002*** 0.002 0.002*** 0.015**
NACE Q -0.000 -0.003* -0.003*** -0.016*** 0.001** 0.000
NACE R+S 0.001*** -0.002*** 0.004*** -0.006*** 0.002*** -0.003***
reference: private ownership of a firm
public ownership of a firm -0.001*** -0.004 -0.002*** 0.033*** 0.001*** -0.011
tenure: less than 2 years (share) 0.006*** 0.036*** 0.011*** 0.012 0.002** -0.005
age: 50 years or more (share) -0.001 -0.030*** -0.012*** -0.060*** -0.017*** -0.024
tertiary education (share) 0.003** -0.031*** 0.053*** -0.081*** 0.089*** -0.096***
female (share) -0.000** 0.009 -0.001*** -0.031* -0.000*** -0.008
constant 0.332*** 0.240*** 0.363***
total 0.017*** 0.347*** 0.099*** 0.110*** 0.122*** 0.041***
Observations 153,375 153,375 153,375
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
53
Table B.14: Decomposition of total wage change into composition and wage structure effects:
Hungary, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education -0.001*** 0.049*** 0.019*** -0.002 0.030*** -0.019***
secondary education 0.000*** 0.096*** -0.003*** -0.030*** 0.001*** -0.008**
reference: under 30 years old
30-49 years old 0.000 -0.005 0.001*** 0.027*** 0.005*** 0.058***
50 years old or more -0.000*** -0.012*** -0.001*** 0.014*** -0.003*** 0.016***
reference: male
female 0.001*** 0.011*** 0.002*** 0.002 0.006*** -0.013***
reference: tenure of less than a year
tenure: 1-4 years 0.000 0.021*** 0.000 -0.005*** 0.000 -0.033***
tenure: 5-9 years -0.002*** 0.005*** -0.002*** -0.009*** -0.001*** -0.016***
tenure: 10 years or more -0.004*** 0.007*** -0.006*** -0.010*** -0.001*** -0.031***
reference: ISCO 5
ISCO 1 -0.004*** -0.014*** -0.007*** 0.007*** -0.015*** 0.016***
ISCO 2 0.004*** -0.023*** 0.013*** 0.012*** 0.003*** 0.030***
ISCO 3 0.002*** -0.034*** 0.003*** 0.002 -0.000*** -0.016***
ISCO 4 -0.007*** -0.014*** -0.003*** 0.008*** 0.005*** -0.005***
ISCO 6 0.000 -0.000** 0.000 0.000*** 0.000 0.000
ISCO 7 0.001*** -0.016*** -0.000 0.015*** -0.002*** -0.013***
ISCO 8 0.003*** -0.034*** 0.001*** -0.002 -0.004*** -0.030***
ISCO 9 -0.004*** -0.043*** -0.007*** 0.019*** -0.005*** -0.017***
reference: permanent contract
fixed contract 0.000 0.003*** 0.001*** 0.000 -0.000 -0.003***
Firm effects
reference: NACE C
NACE B -0.000* 0.000* -0.000 -0.000 -0.000 -0.000
NACE D+E -0.000** -0.000 -0.000*** 0.001* -0.000*** -0.006***
NACE F 0.004*** 0.014*** 0.002*** -0.000 0.002*** -0.003***
NACE G -0.000 0.025*** -0.000 -0.002** -0.000 -0.004**
NACE H+J -0.001*** 0.002*** 0.000*** -0.008*** 0.000 0.002
NACE I 0.000*** 0.005*** 0.000*** -0.000 0.000*** -0.001***
NACE K 0.000*** -0.004*** 0.002*** -0.005*** 0.004*** 0.000
NACE L+M+N -0.002*** 0.011*** -0.001*** -0.001* -0.001*** -0.010***
NACE O -0.002*** -0.025*** -0.001*** -0.036*** -0.001*** -0.049***
NACE P 0.004*** 0.008*** 0.005*** -0.009*** 0.017*** -0.022***
NACE Q 0.000*** -0.003*** 0.000*** -0.023*** 0.001*** -0.004**
NACE R+S 0.002*** 0.001** 0.003*** -0.004*** 0.006*** -0.004***
reference: private ownership of a firm
public ownership of a firm -0.005*** -0.075*** -0.002*** -0.024*** 0.003*** -0.028***
tenure: less than 2 years (share) -0.026*** 0.077*** -0.013*** 0.048*** -0.004*** 0.073***
age: 50 years or more (share) -0.000*** -0.018*** 0.002*** 0.014*** 0.003*** -0.076***
tertiary education (share) 0.022*** -0.091*** 0.019*** 0.011*** 0.051*** -0.038***
female (share) -0.003*** -0.002 0.005*** 0.059*** 0.004*** 0.013
constant 0.478*** 0.176*** 0.409***
total -0.016*** 0.401*** 0.031*** 0.244*** 0.104*** 0.167***
Observations 1,446,198 1,446,198 1,446,198
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
54
Table B.15: Decomposition of total wage change into composition and wage structure effects:
Poland, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.000 -0.012*** 0.022*** -0.012*** 0.044*** -0.060***
secondary education -0.001** -0.022*** -0.005*** -0.014*** -0.002*** -0.040***
reference: under 30 years old
30-49 years old 0.000*** 0.013*** -0.001*** -0.003 -0.002*** -0.002
50 years old or more -0.003*** 0.015*** 0.007*** -0.008*** 0.014*** -0.008***
reference: male
female -0.000*** 0.025*** -0.001*** 0.002* -0.001*** -0.027***
reference: tenure of less than a year
tenure: 1-4 years -0.004*** -0.016*** -0.004*** -0.005*** -0.005*** -0.009***
tenure: 5-9 years 0.002*** -0.031*** 0.002*** -0.011*** 0.002*** -0.025***
tenure: 10 years or more 0.001*** -0.079*** 0.001*** -0.023*** 0.001*** 0.015***
reference: ISCO 5
ISCO 1 0.009*** -0.027*** 0.012*** -0.007*** 0.012*** 0.009***
ISCO 2 0.009*** -0.059*** 0.015*** -0.012*** 0.006*** 0.028***
ISCO 3 -0.006*** -0.029*** -0.005*** -0.008*** 0.001*** 0.003***
ISCO 4 -0.003*** -0.019*** -0.001*** -0.006*** 0.001*** -0.002***
ISCO 6 0.000*** -0.000*** -0.000 -0.000** -0.000** 0.000
ISCO 7 -0.004*** -0.015*** -0.002*** 0.003*** 0.002*** 0.006***
ISCO 8 -0.001*** -0.026*** -0.000*** -0.004*** 0.000*** 0.004***
ISCO 9 -0.002*** -0.006*** 0.003*** 0.002*** 0.002*** 0.002***
Firm effects
reference: NACE C
NACE B 0.000*** 0.001*** -0.001*** -0.000*** -0.001*** 0.002***
NACE D+E -0.000*** 0.002*** 0.001*** -0.003*** -0.000 -0.002***
NACE F -0.000*** -0.004*** -0.000*** -0.004*** -0.000*** -0.001
NACE G -0.000 -0.007*** -0.001*** -0.000 -0.001*** 0.007***
NACE H+J -0.001*** -0.003*** 0.000*** -0.009*** 0.001*** -0.004***
NACE I -0.001*** 0.001*** 0.000** -0.002*** -0.000*** 0.000*
NACE K -0.000*** -0.003*** 0.000*** -0.007*** 0.000 -0.005***
NACE L+M+N -0.001*** 0.007*** -0.000*** -0.000 -0.000*** -0.002***
NACE O -0.000*** 0.000 -0.000*** -0.013*** -0.000*** -0.016***
NACE P 0.001*** 0.002** 0.000*** -0.012*** -0.000*** 0.013***
NACE Q -0.000*** -0.012*** 0.002*** -0.013*** 0.003*** -0.010***
NACE R+S 0.001*** 0.000 0.001*** -0.002*** 0.001*** -0.003***
reference: private ownership of a firm
public ownership of a firm -0.017*** -0.046*** -0.006*** 0.025*** 0.005*** 0.037***
tenure: less than 2 years (share) 0.002*** 0.060*** 0.000*** -0.008*** -0.000*** -0.013***
age: 50 years or more (share) -0.006*** 0.037*** -0.008*** 0.010*** -0.021*** 0.058***
tertiary education (share) 0.062*** -0.104*** 0.054*** -0.017*** 0.069*** -0.036***
female (share) -0.001*** 0.112*** -0.002*** 0.027*** -0.000** 0.006
constant 0.646*** 0.401*** 0.291***
total 0.035*** 0.401*** 0.083*** 0.266*** 0.133*** 0.216***
Observations 1,347,783 1,347,783 1,347,783
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
55
Table B.16: Decomposition of total wage change into composition and wage structure effects:
Romania, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education -0.000** 0.005*** 0.001*** 0.007 0.001*** 0.011*
secondary education -0.000 0.002 0.002*** 0.013** 0.002*** -0.039***
reference: under 30 years old
30-49 years old -0.000 -0.007*** -0.000 -0.036*** -0.000 0.042***
50 years old or more 0.000 -0.001 0.004*** -0.020*** 0.009*** -0.009**
reference: male
female -0.000* 0.004*** -0.000** 0.012*** -0.000** -0.014***
reference: tenure of less than a year
tenure: 1-4 years -0.000* -0.003* -0.002*** 0.003 -0.001*** -0.005
tenure: 5-9 years 0.002*** -0.007*** 0.007*** -0.003 0.003*** -0.009**
tenure: 10 years or more 0.001*** -0.014*** 0.007*** -0.011*** 0.005*** -0.010*
reference: ISCO 5
ISCO 1 0.002*** -0.011*** 0.010*** -0.003*** 0.026*** -0.041***
ISCO 2 0.014*** -0.038*** 0.067*** -0.010*** 0.084*** -0.111***
ISCO 3 -0.005*** -0.011*** -0.025*** -0.008*** -0.007*** -0.014***
ISCO 4 -0.001*** -0.010*** -0.003*** 0.001 -0.000** -0.012***
ISCO 6 -0.000** -0.000** 0.000*** -0.000*** 0.000*** -0.000***
ISCO 7 -0.004*** -0.017*** -0.011*** 0.007*** -0.002*** -0.025***
ISCO 8 -0.005*** -0.014*** -0.011*** 0.003** -0.001*** -0.019***
ISCO 9 -0.001*** 0.006*** -0.001*** 0.002 0.001*** -0.020***
reference: permanent contract
fixed contract -0.000*** 0.001*** -0.001*** 0.003*** -0.002*** 0.003***
Firm effects
reference: NACE C
NACE B 0.000 0.000 -0.006*** -0.002*** -0.010*** 0.008***
NACE D+E -0.000*** 0.002*** 0.001*** -0.007*** 0.001*** 0.002
NACE F 0.000*** 0.004*** 0.000*** -0.015*** 0.000 -0.015***
NACE G -0.001*** 0.015*** -0.001*** -0.019*** -0.000*** -0.019***
NACE H+J -0.003*** 0.011*** -0.000 -0.014*** 0.001 0.009***
NACE I -0.002*** 0.004*** -0.000** -0.006*** 0.000 -0.005***
NACE K -0.000*** 0.001*** 0.000*** -0.006*** 0.002*** -0.012***
NACE L+M+N -0.003*** 0.008*** -0.006*** -0.008*** 0.001** -0.016***
NACE O 0.001*** 0.013*** 0.002*** -0.022*** 0.000 -0.019***
NACE P 0.001*** 0.013*** 0.002*** -0.036*** 0.003*** -0.052***
NACE Q -0.001*** 0.005*** -0.001*** -0.019*** -0.002*** -0.015***
NACE R+S -0.000** 0.001*** -0.000** -0.008*** -0.000** -0.011***
reference: private ownership of a firm
public ownership of a firm -0.004*** -0.033*** -0.010*** -0.003 -0.008*** -0.054***
tenure: less than 2 years (share) 0.010*** 0.043*** 0.014*** -0.006 -0.002** 0.005
age: 50 years or more (share) 0.002*** -0.013*** -0.004*** -0.008 -0.019*** 0.029***
tertiary education (share) 0.001*** -0.037*** 0.003*** 0.049*** 0.005*** 0.192***
female (share) -0.000 0.001 -0.001*** 0.085*** 0.000** -0.026***
constant 0.669*** 0.353*** 0.511***
total 0.003*** 0.591*** 0.038*** 0.268*** 0.090*** 0.239***
Observations 512,290 512,290 512,290
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
56
Table B.17: Decomposition of total wage change into composition and wage structure effects:
Slovakia, quantiles
10th Quantile 50th Quantile 90th Quantile
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.028*** -0.010*** 0.027*** 0.014*** 0.039*** -0.040***
secondary education -0.026*** -0.046*** -0.010*** 0.024*** 0.003*** -0.009***
reference: under 30 years old
30-49 years old -0.000** -0.024*** -0.001*** -0.000 -0.002*** 0.047***
50 years old or more 0.001*** -0.018*** 0.004*** -0.006*** 0.007*** 0.029***
reference: male
female -0.000*** 0.007*** -0.000*** 0.004** -0.001*** -0.002
reference: tenure of less than a year
tenure: 1-4 years -0.007*** 0.009*** -0.010*** -0.002 -0.010*** -0.014***
tenure: 5-9 years 0.004*** 0.009*** 0.006*** 0.000 0.007*** -0.020***
tenure: 10 years or more 0.007*** 0.015*** 0.010*** 0.004** 0.012*** -0.025***
reference: ISCO 5
ISCO 1 0.002*** -0.004*** 0.003*** -0.000 0.008*** -0.003
ISCO 2 0.024*** -0.005** 0.025*** -0.002 0.018*** 0.015***
ISCO 3 -0.020*** -0.013*** -0.017*** -0.004*** -0.004*** -0.000
ISCO 4 0.004*** 0.001 0.002*** -0.000 -0.001*** 0.003***
ISCO 6 0.000 -0.000 0.000*** -0.000 0.000*** 0.000
ISCO 7 -0.010*** -0.004*** -0.005*** 0.001 0.016*** 0.012***
ISCO 8 0.001*** -0.015*** 0.000*** -0.012*** -0.001*** 0.010***
ISCO 9 0.001*** -0.011*** 0.001*** -0.001 0.001*** 0.007***
reference: permanent contract
fixed contract -0.002*** 0.006*** -0.003*** 0.001 -0.003*** 0.011***
Firm effects
reference: NACE C
NACE B -0.000** -0.000*** -0.000*** -0.000 0.001*** 0.002***
NACE D+E -0.000*** 0.001*** 0.000*** -0.002*** 0.002*** -0.004***
NACE F 0.002*** -0.002*** 0.003*** -0.001** 0.007*** 0.005***
NACE G -0.001*** 0.001 -0.001*** 0.002* -0.002*** 0.010***
NACE H+J -0.001*** -0.016*** 0.001*** -0.018*** -0.001*** 0.009***
NACE I 0.001*** -0.002*** 0.000*** -0.001*** 0.000*** -0.001**
NACE K -0.000*** -0.002*** 0.001*** -0.000** 0.002*** -0.001**
NACE L+M+N -0.003*** -0.009*** -0.003*** 0.002** -0.002*** 0.001
NACE O -0.003*** -0.011*** -0.001*** -0.004*** -0.006*** 0.004**
NACE P -0.003*** -0.020*** -0.002*** -0.002* -0.011*** 0.007***
NACE Q 0.002*** -0.005*** 0.001*** 0.002*** 0.003*** 0.011***
NACE R+S 0.002*** -0.004*** 0.004*** -0.001*** 0.006*** 0.001**
reference: private ownership of a firm
public ownership of a firm 0.001*** 0.015*** 0.001*** -0.004** -0.001*** -0.010***
tenure: less than 2 years (share) 0.001* -0.050*** -0.002*** -0.029*** 0.016*** 0.052***
age: 50 years or more (share) -0.003*** -0.061*** -0.011*** -0.044*** -0.034*** 0.076***
tertiary education (share) 0.010*** 0.013*** 0.020*** 0.001 0.064*** -0.026***
female (share) -0.001*** 0.024*** -0.001*** -0.009** -0.002*** 0.143***
constant 0.780*** 0.622*** 0.135***
total 0.009*** 0.550*** 0.041*** 0.534*** 0.134*** 0.436***
Observations 1,534,467 1,534,467 1,534,467
Notes: Table represent the results of the decomposition of changes in log real hourly wages between 2006 and 2014 for the 10th, 50th and
90th percentile of wage distribution into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed
sample does not include the top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
57
Table B.18: Decomposition of overall change in variance of firm average log wages into com-
position and wage structure effects: Bulgaria, Czechia and Estonia
Bulgaria Czechia Estonia
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Firm effects
reference: NACE C
NACE B -0.001 -0.000 -0.000 0.000 0.000 0.001**
NACE D+E 0.001 -0.001 0.001** -0.000 0.000 0.002**
NACE F 0.002*** 0.004** -0.000 -0.000 0.000 -0.001
NACE G -0.002** -0.004 -0.001** -0.001 0.001 -0.001
NACE H+J -0.001 0.017*** 0.001*** 0.004 0.000 0.008*
NACE I -0.000 -0.002** 0.000* 0.002*** 0.002** -0.002
NACE K 0.002* -0.006** 0.000 -0.002* -0.001 0.005**
NACE L+M+N 0.001 0.006 0.001*** 0.000 0.000 0.000
NACE O 0.001 0.004 0.000 -0.003 0.000 0.005
NACE P 0.002 0.014** -0.000 -0.002 -0.000 0.002
NACE Q -0.002* 0.009* -0.000** 0.002 0.000 0.002
NACE R+S -0.000 -0.002** -0.000 -0.001* -0.000 0.001
reference: private ownership of a firm
public ownership of a firm 0.002 -0.007 -0.000 -0.004 0.002 0.009
tenure: less than 2 years (share) -0.000 0.019 -0.000 0.026*** -0.001 0.030*
age: 50 years or more (share) -0.014*** -0.008 0.001*** 0.037*** -0.002 0.012
tertiary education (share) 0.031*** -0.025 0.007*** 0.014 -0.010*** -0.018
female (share) -0.001* 0.004 0.001 -0.028** -0.000 0.024
constant -0.058 -0.033 -0.065*
total 0.022*** -0.036*** 0.011*** 0.011* -0.008* 0.015*
Observations 9,500 23,832 4,976
Notes: Table represent the results of the decomposition of changes in variance of firm average normalized log hourly wages between 2006
and 2014 into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the
top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
Table B.19: Decomposition of overall change in variance of firm-average log wages into com-
position and wage structure effects: Latvia, Lithuania and Hungary
Latvia Lithuania Hungary
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Firm effects
reference: NACE C
NACE B -0.000 0.000* -0.000 -0.000 0.000 -0.000
NACE D+E 0.002*** -0.004*** 0.000 -0.001 -0.000 -0.002
NACE F 0.000 0.000 -0.001** -0.005** -0.001*** -0.005***
NACE G -0.000 -0.004 0.000 0.001 0.000 -0.006***
NACE H+J 0.001** 0.003 0.002*** -0.003 0.001* -0.000
NACE I 0.000 -0.003* 0.000 -0.001 -0.000 -0.002***
NACE K -0.001 0.001 -0.000 -0.000 0.002*** 0.002
NACE L+M+N -0.000 -0.002 0.000 0.003 0.001 -0.003
NACE O 0.001 -0.005 0.000 -0.006*** -0.000 0.027***
NACE P -0.008*** 0.002 0.001 0.007 0.004*** 0.014***
NACE Q -0.001* 0.000 -0.001* 0.004 0.000 0.005**
NACE R+S 0.000 -0.001 0.000 0.000 0.001* 0.000
reference: private ownership of a firm
public ownership of a firm -0.004*** 0.042*** 0.001 0.003 0.002*** 0.016
tenure: less than 2 years (share) -0.003** -0.003 -0.003** -0.027*** 0.007*** 0.056***
age: 50 years or more (share) -0.010*** 0.032* -0.007*** 0.016 0.001*** 0.016
tertiary education (share) 0.024*** 0.004 0.032*** -0.035** 0.020*** -0.048***
female (share) -0.001* 0.003 -0.000 0.006 0.002*** -0.022
constant -0.148*** -0.017 -0.090***
total -0.000 -0.083*** 0.025*** -0.056*** 0.038*** -0.043***
Observations 11,329 8,394 26,554
Notes: Table represent the results of the decomposition of changes in variance of firm average normalized log hourly wages between 2006
and 2014 into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the
top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
58
Table B.20: Decomposition of overall change in variance of firm-average log wages into com-
position and wage structure effects: Poland, Romania and Slovakia
Poland Romania Slovakia
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Firm effects
reference: NACE C
NACE B -0.001 0.000 -0.004*** 0.003*** 0.000 0.001**
NACE D+E 0.000 -0.000 0.001* 0.000 0.001 -0.001
NACE F 0.000 0.002*** -0.000 -0.003 0.000 0.002*
NACE G 0.000 0.010*** 0.001** -0.006*** 0.000 0.001
NACE H+J 0.000 0.004** 0.001** 0.005 0.001 0.007**
NACE I 0.000 0.000 0.001* 0.001 -0.000 0.000
NACE K 0.000 -0.001 0.001 -0.005* 0.000 -0.001
NACE L+M+N 0.000 0.001 0.005*** -0.011*** 0.002*** 0.002
NACE O -0.000 0.001 -0.001* -0.003 -0.001 0.004
NACE P 0.000 0.007** 0.001 -0.011*** -0.002** 0.007**
NACE Q 0.001*** 0.005*** -0.001* 0.005* 0.000 0.001
NACE R+S 0.000 0.000 -0.000 -0.002** -0.000 0.000
reference: private ownership of a firm
public ownership of a firm 0.006*** 0.006 0.000 -0.014 -0.001 -0.009
tenure: less than 2 years (share) -0.001 -0.014** -0.004*** 0.006 0.005** 0.021*
age: 50 years or more (share) -0.002** 0.018** -0.010*** 0.027*** -0.008*** 0.035**
tertiary education (share) 0.024*** -0.035*** 0.003** 0.065*** 0.028*** -0.027*
female (share) 0.000 -0.041*** 0.000 -0.048*** -0.000 0.012
constant -0.019 -0.026 -0.091**
total 0.030*** -0.055*** -0.006* -0.017** 0.026*** -0.035***
Observations 28,586 22,572 8,666
Notes: Table represent the results of the decomposition of changes in variance of firm average normalized log hourly wages between 2006
and 2014 into composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the
top 0.1% and the bottom 0.1% hourly wages. For the detailed explanation of NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
59
Appendix C Sensitivity tests
C.1 Results without trimming
Figure C.1: Overall variance of log wages: 2002-2014, full sample
Note: Figure shows variance of log of normalised gross hourly wages. Sample includes top
0.1% and bottom 0.1%.
Data: European Structure of Earnings Survey.
60
Figure C.2: Decomposition of Change in Variance of Hourly Wages into Composition and
Wage Structure Effects, full sample
Note: Figure shows the results of the decomposition of change in variance of hourly wages between 2006 and
2014 into composition and wage structure effects based on RIF regressions. Sample includes top 0.1% and
bottom 0.1%.
Data: European Structure of Earnings Survey.
61
C.2 Results without public sector
Table C.1: Results of RIF regression: Bulgaria and Romania (excluding public sector)
Bulgaria Romania
2002 2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.106*** 0.057*** 0.034*** 0.000 0.406*** 0.018*** 0.043*** -0.065***
secondary education -0.001 -0.024*** -0.020*** -0.019*** -0.016*** -0.003 -0.002 -0.001
reference: under 30 years old
30-49 years old 0.009** 0.019*** 0.064*** 0.094*** 0.017*** 0.038*** 0.073*** 0.092***
50 years old or more 0.006 -0.003 0.042*** 0.077*** 0.084*** 0.068*** 0.074*** 0.084***
reference: male
female -0.062*** -0.063*** -0.067*** -0.075*** -0.041*** -0.038*** -0.033*** -0.060***
reference: tenure of less than a year
tenure: 1-4 years 0.004 0.024*** 0.010** 0.003 0.002 -0.000 0.009** -0.004
tenure: 5-9 years 0.033*** 0.066*** 0.031*** 0.016*** -0.015** 0.027*** 0.012** 0.011**
tenure: 10 years or more 0.079*** 0.170*** 0.093*** 0.059*** -0.009 0.022*** 0.026*** 0.048***
reference: ISCO 5
ISCO 1 0.679*** 0.715*** 0.788*** 0.879*** 0.710*** 1.136*** 0.847*** 0.844***
ISCO 2 0.226*** 0.339*** 0.369*** 0.421*** -0.140*** 0.444*** 0.197*** 0.324***
ISCO 3 0.091*** 0.028*** 0.054*** 0.004 -0.107*** 0.033*** -0.063*** -0.060***
ISCO 4 -0.052*** -0.095*** -0.109*** -0.101*** -0.206*** -0.060*** -0.161*** -0.130***
ISCO 6 0.063 0.028 0.076** 0.908*** -0.123*** 0.043 -0.065* -0.035
ISCO 7 0.063*** 0.019*** -0.019*** -0.034*** -0.152*** -0.065*** -0.098*** -0.088***
ISCO 8 0.026*** -0.029*** -0.056*** -0.095*** -0.168*** -0.083*** -0.129*** -0.127***
ISCO 9 0.001 0.002 0.024*** 0.029*** -0.100*** 0.001 0.018*** -0.042***
reference: permanent contract
fixed contract 0.012** 0.025*** -0.005 0.063*** -0.020 -0.029** 0.025*** -0.036***
Firm effects
reference: NACE C
NACE B 0.134*** 0.176*** 0.128*** 0.202*** 0.126*** 0.354*** 0.235*** 0.669***
NACE D+E 0.217*** 0.217*** 0.097*** 0.040*** -0.021 0.056*** 0.004 0.013
NACE F -0.042*** -0.088*** -0.026*** -0.043*** -0.041*** 0.025*** -0.011** -0.064***
NACE G -0.017*** -0.046*** -0.066*** -0.092*** -0.013** 0.013*** -0.038*** -0.062***
NACE H+J 0.078*** 0.065*** 0.194*** 0.193*** 0.221*** 0.104*** 0.055*** 0.090***
NACE I 0.092*** -0.002 -0.046*** -0.113*** 0.015 0.031*** -0.009 -0.012
NACE K 0.410*** 0.204*** 0.067*** 0.005 0.477*** 0.679*** 0.426*** 0.266***
NACE L+M+N 0.012 0.055*** 0.117*** 0.081*** -0.012 0.116*** 0.015*** -0.017***
NACE P 0.055 -0.097*** -0.063** -0.272*** -0.175*** -0.020 -0.275*** -0.285***
NACE Q -0.328*** -0.574*** -0.167*** -0.181*** -0.437*** -0.195*** -0.290*** -0.351***
NACE R+S -0.055*** -0.043*** -0.096*** -0.207*** 0.063*** -0.061*** -0.101*** -0.139***
tenure: less than 2 years (share) -0.035*** -0.044*** 0.055*** 0.072*** 0.004 0.004 0.017*** 0.064***
age: 50 years or more (share) -0.649*** -0.381*** -0.355*** -0.422*** -0.497*** -0.299*** -0.170*** -0.190***
tertiary education (share) 0.412*** 0.599*** 0.470*** 0.371*** 1.076*** 0.349*** 0.678*** 0.684***
female (share) -0.170*** -0.081*** 0.023*** -0.034*** -0.004 -0.013** 0.023*** -0.019***
constant 0.476*** 0.324*** 0.197*** 0.247*** 0.377*** 0.241*** 0.165*** 0.146***
Observations 84,017 106,996 123,992 124,450 144,604 173,531 168,987 175,087
R-squared 0.207 0.254 0.274 0.279 0.284 0.290 0.288 0.312
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. Sample
includes only private sector firms. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
62
Table C.2: Results of RIF regression: Czechia and Slovakia (excluding public sector)
Czechia Slovakia
2002 2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.247*** 0.238*** 0.229*** 0.158*** 0.089*** 0.127*** 0.123*** 0.056***
secondary education -0.046*** -0.050*** -0.048*** -0.059*** -0.056*** -0.111*** -0.087*** -0.090***
reference: under 30 years old
30-49 years old 0.042*** 0.086*** 0.105*** 0.108*** 0.044*** 0.076*** 0.099*** 0.104***
50 years old or more 0.048*** 0.079*** 0.092*** 0.101*** 0.054*** 0.058*** 0.090*** 0.091***
reference: male
female -0.053*** -0.053*** -0.057*** -0.060*** -0.043*** -0.057*** -0.063*** -0.055***
reference: tenure of less than a year
tenure: 1-4 years -0.007*** -0.009*** -0.002 -0.017*** 0.004 0.011*** -0.005*** -0.022***
tenure: 5-9 years -0.001 0.015*** 0.004*** -0.008*** 0.032*** 0.034*** 0.018*** -0.010***
tenure: 10 years or more -0.017*** 0.007*** 0.021*** -0.001 0.007 0.050*** 0.018*** 0.004**
reference: ISCO 5
ISCO 1 0.335*** 0.379*** 0.455*** 0.529*** 0.480*** 0.544*** 0.518*** 0.481***
ISCO 2 -0.172*** -0.074*** -0.008*** 0.044*** -0.049*** -0.038*** 0.039*** 0.028***
ISCO 3 -0.111*** -0.106*** -0.110*** -0.127*** -0.157*** -0.096*** -0.070*** -0.090***
ISCO 4 -0.114*** -0.146*** -0.175*** -0.183*** -0.142*** -0.102*** -0.120*** -0.134***
ISCO 6 -0.036*** 0.004 -0.092*** -0.105*** -0.058 0.036** 0.017 -0.056***
ISCO 7 -0.136*** -0.134*** -0.140*** -0.156*** -0.149*** -0.099*** -0.077*** -0.084***
ISCO 8 -0.151*** -0.153*** -0.150*** -0.154*** -0.146*** -0.122*** -0.105*** -0.099***
ISCO 9 -0.050*** -0.016*** 0.004** -0.008*** -0.024*** -0.003 0.008*** -0.001
reference: permanent contract
fixed contract 0.014*** 0.030*** 0.008*** -0.014*** 0.006 0.000 0.018*** 0.002
Firm effects
reference: NACE C
NACE B 0.035*** 0.057*** 0.048*** 0.054*** -0.013 -0.071*** 0.024*** 0.063***
NACE D+E -0.003 0.031*** 0.017*** -0.035*** 0.047*** 0.004 0.088*** 0.096***
NACE F -0.005*** -0.001 -0.003** -0.051*** 0.006 -0.015*** -0.009*** 0.014***
NACE G -0.029*** 0.005*** -0.016*** -0.005*** 0.069*** -0.027*** -0.012*** -0.024***
NACE H+J 0.040*** 0.101*** 0.119*** 0.082*** 0.033*** 0.039*** 0.094*** 0.105***
NACE I 0.022*** 0.012*** 0.148*** 0.082*** 0.040*** 0.049*** 0.031*** 0.008**
NACE K 0.032*** 0.201*** 0.177*** 0.123*** 0.019** 0.026*** 0.008** 0.026***
NACE L+M+N -0.006*** -0.011*** 0.044*** 0.033*** 0.189*** 0.032*** 0.024*** 0.050***
NACE P -0.042*** -0.252*** -0.304*** -0.353*** -0.331*** -0.308*** -0.401*** -0.255***
NACE Q -0.069*** -0.130*** -0.107*** -0.082*** 0.111*** -0.069*** -0.070*** -0.027***
NACE R+S -0.018*** 0.011*** 0.013*** -0.011*** -0.101*** -0.059*** -0.173*** -0.041***
tenure: less than 2 years (share) 0.018*** 0.039*** 0.069*** 0.071*** 0.024*** -0.031*** 0.018*** 0.030***
age: 50 years or more (share) -0.206*** -0.169*** -0.117*** -0.081*** -0.394*** -0.295*** -0.202*** -0.180***
tertiary education (share) 0.164*** 0.295*** 0.164*** 0.251*** 0.223*** 0.373*** 0.312*** 0.183***
female (share) 0.057*** 0.107*** 0.045*** 0.022*** 0.075*** 0.031*** 0.065*** 0.048***
constant 0.304*** 0.211*** 0.207*** 0.236*** 0.346*** 0.345*** 0.230*** 0.251***
Observations 600,224 1,007,549 1,152,883 1,242,217 247,517 441,569 503,585 572,365
R-squared 0.212 0.235 0.236 0.251 0.131 0.224 0.242 0.216
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. Sample
includes only private sector firms. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
63
Table C.3: Results of RIF regression: Estonia and Poland (excluding public sector)
Estonia Poland
2006 2010 2014 2002 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.129*** 0.084*** 0.112*** 0.301*** 0.267*** 0.172*** 0.126***
secondary education -0.017*** -0.030*** -0.013*** -0.012*** -0.005 -0.016*** -0.020***
reference: under 30 years old
30-49 years old 0.077*** 0.091*** 0.091*** 0.098*** 0.129*** 0.131*** 0.132***
50 years old or more 0.044*** 0.068*** 0.068*** 0.152*** 0.170*** 0.150*** 0.142***
reference: male
female -0.054*** -0.061*** -0.081*** -0.056*** -0.077*** -0.084*** -0.092***
reference: tenure of less than a year
tenure: 1-4 years -0.026*** -0.014*** -0.006 -0.028*** -0.008** -0.008*** 0.003
tenure: 5-9 years 0.006 -0.013** 0.005 -0.012*** 0.003 0.000 -0.001
tenure: 10 years or more -0.015** -0.016*** -0.009* -0.049*** -0.025*** 0.011*** 0.035***
reference: ISCO 5
ISCO 1 0.461*** 0.518*** 0.369*** 0.681*** 0.521*** 0.509*** 0.482***
ISCO 2 0.184*** 0.132*** 0.134*** 0.042*** -0.075*** -0.026*** -0.052***
ISCO 3 0.009 -0.071*** -0.025*** -0.083*** -0.164*** -0.120*** -0.133***
ISCO 4 -0.133*** -0.156*** -0.135*** -0.170*** -0.204*** -0.178*** -0.183***
ISCO 6 0.175 0.129** -0.097* -0.122*** -0.208*** 0.033 -0.073***
ISCO 7 -0.017*** -0.063*** -0.028*** -0.068*** -0.118*** -0.080*** -0.091***
ISCO 8 -0.047*** -0.075*** -0.073*** -0.126*** -0.178*** -0.124*** -0.140***
ISCO 9 0.124*** 0.061*** 0.051*** -0.031*** -0.057*** -0.011*** -0.010***
reference: permanent contract
fixed contract 0.048*** 0.070*** 0.083***
Firm effects
reference: NACE C
NACE B 0.011 0.103*** 0.094*** 0.257*** 0.423*** 0.178*** 0.278***
NACE D+E -0.081*** -0.034** -0.031** 0.070*** 0.040*** 0.018*** 0.071***
NACE F 0.034*** 0.017*** 0.006 -0.030*** -0.038*** -0.016*** -0.017***
NACE G 0.055*** 0.018*** 0.048*** -0.043*** -0.060*** -0.035*** -0.006**
NACE H+J 0.038*** 0.106*** 0.110*** 0.087*** 0.049*** 0.052*** 0.054***
NACE I 0.065*** 0.031*** 0.028*** 0.026*** -0.008 -0.019*** -0.016***
NACE K 0.172*** 0.159*** 0.187*** -0.122*** 0.062*** 0.020*** -0.013***
NACE L+M+N 0.113*** 0.037*** 0.085*** 0.006 0.017*** 0.032*** 0.024***
NACE P -0.218*** -0.248*** -0.081*** -0.096*** -0.393*** -0.336*** -0.261***
NACE Q 0.012 0.013 0.033*** -0.260*** -0.205*** -0.126*** -0.108***
NACE R+S -0.001 0.016 0.013 -0.031*** 0.013 0.040*** 0.044***
tenure: less than 2 years (share) -0.044*** 0.053*** 0.022*** 0.078*** 0.050*** 0.039*** 0.038***
age: 50 years or more (share) -0.116*** -0.106*** -0.058*** -0.297*** -0.166*** -0.151*** -0.089***
tertiary education (share) 0.089*** 0.149*** 0.070*** 0.758*** 0.511*** 0.434*** 0.367***
female (share) -0.032*** 0.051*** 0.054*** 0.105*** 0.040*** 0.032*** 0.000
constant 0.249*** 0.193*** 0.162*** 0.195*** 0.227*** 0.139*** 0.138***
Observations 76,863 66,752 69,999 293,325 316,821 336,871 404,022
R-squared 0.190 0.225 0.165 0.269 0.245 0.252 0.227
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. Sample
includes only private sector firms. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
64
Table C.4: Results of RIF regression: Lithuania and Latvia (excluding public sector)
Lithuania Latvia
2002 2006 2010 2014 2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.178*** 0.174*** 0.131*** 0.038** 0.103*** 0.055*** 0.054***
secondary education 0.005 0.003 0.009 -0.006 -0.029*** -0.044*** -0.024***
reference: under 30 years old
30-49 years old 0.017*** 0.059*** 0.045*** 0.090*** 0.079*** 0.108*** 0.117***
50 years old or more 0.004 0.033*** 0.036*** 0.085*** 0.038*** 0.067*** 0.091***
reference: male
female -0.078*** -0.079*** -0.089*** -0.099*** -0.091*** -0.093*** -0.073***
reference: tenure of less than a year
tenure: 1-4 years 0.007 0.023*** 0.044*** -0.013* 0.014*** 0.008** 0.017***
tenure: 5-9 years 0.051*** 0.083*** 0.049*** 0.010 0.075*** 0.030*** 0.032***
tenure: 10 years or more 0.034*** 0.099*** 0.127*** 0.037*** 0.083*** 0.033*** 0.005
reference: ISCO 5
ISCO 1 0.451*** 0.323*** 0.356*** 0.424*** 0.378*** 0.446*** 0.418***
ISCO 2 0.092*** -0.021** 0.076*** 0.093*** 0.142*** 0.155*** 0.213***
ISCO 3 0.024** -0.032*** -0.036** -0.030** -0.053*** -0.014** -0.049***
ISCO 4 -0.086*** -0.157*** -0.145*** -0.125*** -0.135*** -0.099*** -0.122***
ISCO 6 0.105 0.079 0.032 -0.143 0.055** 0.079** 0.097**
ISCO 7 0.013 -0.023*** -0.011 -0.033*** -0.012* 0.030*** 0.015*
ISCO 8 0.010 -0.067*** -0.065*** -0.097*** -0.024*** 0.021*** -0.002
ISCO 9 0.006 0.010 0.030* -0.010 0.015** 0.036*** 0.034***
reference: permanent contract
fixed contract -0.027*** 0.063*** 0.002 0.001 0.077*** 0.047*** 0.021**
Firm effects
reference: NACE C
NACE B 0.105*** 0.046* 0.010 -0.015 -0.028 -0.050*** -0.006
NACE D+E -0.023 -0.082*** 0.020 0.018 0.095*** -0.020 -0.054***
NACE F -0.017** 0.104*** -0.072*** -0.069*** -0.009 -0.039*** -0.040***
NACE G -0.007 -0.007 0.001 -0.053*** 0.027*** -0.015*** -0.008
NACE H+J 0.072*** 0.068*** 0.015 0.028*** 0.062*** 0.084*** 0.090***
NACE I 0.016 0.035*** 0.025 -0.004 0.063*** 0.038*** 0.001
NACE K 0.255*** 0.293*** 0.248*** 0.245*** 0.272*** 0.200*** 0.276***
NACE L+M+N -0.035*** -0.017* 0.009 -0.001 0.099*** 0.001 0.021**
NACE P -0.071** 0.057 -0.185** -0.207*** -0.111*** -0.208*** -0.293***
NACE Q -0.114*** -0.009 -0.065** 0.098*** 0.022 -0.005 0.087***
NACE R+S 0.030* -0.025* 0.010 -0.020 0.036*** -0.002 0.119***
tenure: less than 2 years (share) -0.047*** 0.040*** 0.037** -0.021* 0.007 0.075*** 0.036***
age: 50 years or more (share) -0.299*** -0.107*** -0.076*** -0.126*** -0.308*** -0.299*** -0.189***
tertiary education (share) 0.435*** 0.264*** 0.254*** 0.152*** 0.377*** 0.381*** 0.281***
female (share) -0.076*** -0.018* -0.037** 0.007 -0.055*** 0.009 -0.000
constant 0.318*** 0.212*** 0.170*** 0.233*** 0.413*** 0.205*** 0.166***
Observations 67,576 71,351 18,407 24,961 151,134 108,080 58,685
R-squared 0.193 0.152 0.170 0.185 0.135 0.203 0.180
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo, Fortin, & Lemieux, 2018). The coeffi-
cients measure the impact of an infinitesimal shift to the right in the distribution of the regressors on variance of normalized log hourly
wages in a given country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1% hourly wages. Sample
includes only private sector firms. For the detailed explanation of ISCO and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
65
Table C.5: Results of RIF regression: Hungary (excluding public sector)
Hungary
2006 2010 2014
Individual effects
reference: primary education
tertiary education 0.332*** 0.321*** 0.230***
secondary education -0.004 -0.012*** -0.028***
reference: under 30 years old
30-49 years old 0.100*** 0.108*** 0.097***
50 years old or more 0.086*** 0.094*** 0.101***
reference: male
female -0.075*** -0.072*** -0.063***
reference: tenure of less than a year
tenure: 1-4 years -0.004 0.004 -0.026***
tenure: 5-9 years 0.029*** 0.016*** -0.010**
tenure: 10 years or more 0.034*** 0.046*** 0.005
reference: ISCO 5
ISCO 1 0.364*** 0.373*** 0.458***
ISCO 2 0.100*** 0.128*** 0.127***
ISCO 3 -0.098*** -0.091*** -0.076***
ISCO 4 -0.150*** -0.117*** -0.117***
ISCO 6 -0.003 -0.016 0.000
ISCO 7 -0.104*** -0.081*** -0.054***
ISCO 8 -0.148*** -0.080*** -0.090***
ISCO 9 -0.004 0.065*** 0.013
reference: permanent contract
fixed contract 0.001 0.023*** 0.033***
Firm effects
reference: NACE C
NACE B 0.065*** 0.025 0.021
NACE D+E 0.054*** 0.028*** 0.006
NACE F 0.016*** -0.018*** -0.013**
NACE G -0.003 -0.063*** -0.021***
NACE H+J 0.107*** 0.050*** 0.031***
NACE I -0.031*** -0.038*** -0.022***
NACE K 0.209*** 0.206*** 0.148***
NACE L+M+N -0.017*** 0.019*** -0.016***
NACE P -0.528*** -0.497*** -0.426***
NACE Q -0.130*** -0.142*** -0.062***
NACE R+S -0.079*** -0.144*** -0.069***
tenure: less than 2 years (share) 0.044*** 0.049*** 0.047***
age: 50 years or more (share) -0.259*** -0.191*** -0.145***
tertiary education (share) 0.446*** 0.349*** 0.306***
female (share) -0.016** 0.048*** 0.009
constant 0.260*** 0.175*** 0.174***
Observations 124,960 122,372 136,216
R-squared 0.288 0.276 0.284
Notes: Table shows the coefficients estimated by Recentered Influence Function regression (Firpo,
Fortin, & Lemieux, 2018). The coefficients measure the impact of an infinitesimal shift to the
right in the distribution of the regressors on variance of normalized log hourly wages in a given
country in a given year. Trimmed sample does not include the top 0.1% and the bottom 0.1%
hourly wages. Sample includes only private sector firms. For the detailed explanation of ISCO
and NACE codes see Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
66
Table C.6: Decomposition of overall change in variance of log wages into composition and
wage structure effects: Bulgaria, Czechia and Estonia (excluding public sector)
Bulgaria Czechia Estonia
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.005*** -0.016*** 0.006*** -0.011*** -0.001*** -0.003
secondary education 0.001*** 0.003 0.000*** -0.007** 0.000** 0.003
reference: under 30 years old
30-49 years old 0.000 0.039*** 0.003*** 0.012*** -0.000** 0.007*
50 years old or more -0.000 0.023*** -0.001*** 0.006*** 0.002*** 0.008***
reference: male
female -0.001*** -0.006* -0.001*** -0.003 0.002*** -0.014***
reference: tenure of less than a year
tenure: 1-4 years -0.001*** -0.008*** 0.000** -0.003** 0.003*** 0.007**
tenure: 5-9 years 0.005*** -0.011*** 0.000*** -0.005*** -0.000 -0.000
tenure: 10 years or more 0.002*** -0.015*** 0.000 -0.002 -0.001 0.001
reference: ISCO 5
ISCO 1 0.004*** 0.008*** -0.009*** 0.007*** 0.002*** -0.006***
ISCO 2 0.014*** 0.009*** -0.001*** 0.011*** 0.004*** -0.004**
ISCO 3 0.000** -0.002* 0.001*** -0.004*** 0.000 -0.005***
ISCO 4 -0.001*** -0.001 -0.001*** -0.003*** 0.000** -0.000
ISCO 6 0.000** 0.003*** 0.000 -0.000** 0.000** -0.000***
ISCO 7 -0.001*** -0.008*** 0.007*** -0.004*** 0.000* -0.002
ISCO 8 0.001*** -0.010*** -0.003*** -0.000 0.002*** -0.004**
ISCO 9 -0.000 0.004*** -0.000*** 0.001 0.001*** -0.009***
reference: permanent contract
fixed contract -0.001*** 0.003*** 0.001*** -0.010*** -0.001*** 0.001**
Firm effects
reference: NACE C
NACE B -0.001*** 0.000 -0.000*** -0.000 0.000 0.001***
NACE D+E 0.002*** -0.004*** 0.000*** -0.001*** -0.000* 0.001**
NACE F 0.003*** 0.004*** 0.000 -0.003*** 0.000** -0.002**
NACE G -0.001*** -0.008*** -0.000 -0.002 0.002*** -0.001
NACE H+J 0.003*** 0.013*** 0.004*** -0.002*** 0.001*** 0.008***
NACE I -0.000 -0.006*** 0.000 0.002*** 0.001*** -0.002***
NACE K 0.003*** -0.008*** 0.001*** -0.002*** -0.002*** 0.000
NACE L+M+N 0.002*** 0.003* -0.000* 0.005*** -0.000 -0.003*
NACE P -0.000 -0.001*** 0.001*** -0.001*** 0.000 0.001***
NACE Q -0.008*** 0.007*** -0.001*** 0.001*** 0.000 0.001
NACE R+S -0.000 -0.004*** 0.000 -0.000* -0.000 0.000
tenure: less than 2 years (share) 0.004*** 0.058*** -0.001*** 0.013*** 0.000 0.028***
age: 50 years or more (share) -0.011*** -0.012*** 0.003*** 0.022*** -0.005*** 0.019***
tertiary education (share) 0.054*** -0.064*** 0.007*** -0.006* -0.004*** -0.003
female (share) -0.001*** 0.021*** 0.002*** -0.034*** 0.001*** 0.042***
constant -0.077*** 0.026** -0.087***
total 0.078*** -0.062*** 0.019*** 0.002 0.007*** -0.019***
Observations 231,446 2,249,766 146,862
Notes: Table represent the results of the decomposition of changes in variance of normalized log hourly wages between 2006 and 2014 into
composition and wage structure effect following Firpo, Fortin, and Lemieux (2018). Trimmed sample does not include the top 0.1% and
the bottom 0.1% hourly wages. Sample includes only private sector firms. For the detailed explanation of ISCO and NACE codes see
Table B.1.
* p<0.1, ** p<0.05, ***p<0.01
Data: European Structure of Earnings Survey.
67
Table C.7: Decomposition of overall change in variance of log wages into composition and
wage structure effects: Latvia, Lithuania and Hungary (excluding public sector)
Latvia Lithuania Hungary
Composition Wage Structure Composition Wage Structure Composition Wage Structure
Individual effects
reference: primary education
tertiary education 0.006*** -0.015*** 0.028*** -0.052*** 0.018*** -0.022***
secondary education 0.002*** 0.002 -0.000 -0.005 0.000 -0.015***
reference: under 30 years old
30-49 years old -0.001*** 0.018*** -0.003*** 0.015*** 0.004*** -0.002
50 years old or more 0.001*** 0.016*** 0.002***