Introduction

Link for the screen cast on youtube : https://www.youtube.com/watch?v=9i75RdWIVTQ

Based on a figure provided by Eurostat, the European Union’s statistical agency, which states that “French women earn 15.4% less than men”, meaning that they work on a voluntary basis from 5 November at 16:47 until 31 December. Shocking right ?

As french woman that are going to enter the world of work in the years to come, we decided in this report to focus on the topic of gender equality in the working place throughout Europe. Therefore this study will be a comparison of data collected in each European country from 2006 to 2018 on 20-64 year old population. The data used can be found on Eurostat open source database (see references in the source section). The period studied has been chosen depending on the available data and allow to provides a pre/post financial crisis analysis. The age group studied is the most appropriate considering that most citizens enter the world around the age of 20 and leave it at around 64. The objective is to see if the working place is becoming more egalitarian over the years. Who are today the good students and the bad students in terms of gender equality at work? We may have in mind that the best countries on gender equality matter are nordic countries, but are the results consistent with the assumptions we have? This analysis is exploring the difference between men and women on: activity rate, unemployment rate, part-time employment, temporary employment and gap paid (divided into 6 parts), in order to discover which country is the best to work in Europe when you are a woman.

Methodology

The first step was to import the dataframe from Eurostat open source database with the Eurostat package. Then, the following step was to tidy the dataframe in order to only have the data needed for the analysis and classify them to make it easier for r to make graphs. It means suppressing the missing value, filtering only the interesting data for the analysis (countries, period, age) and changing wide dataframe to long dataframe. The third step was to analyse and make the data speak for themselves by creating plots. The last one was to try to find explanatory variables to those results by comparing them with socio-demographic variables through correlation analysis.

Part 1: Activity rate difference between men and women across european countries and over time

The activity rate increase between 2006 and 2018 for both women and men overall. Generally, women always have an activity rate lower than the men. We can notice that 2011 is a turning point: Before 2011, the activity rate for men was stagnating around 83% while activity rate for women was increasing. After 2011, the activity rate for men start to slightly increase to reach 84.4 in 2018. Concerning, women activity rate after 2011, it keep increasing a bit faster than before 2011: Indeed, The Beta of the trend before 2011 was 2.524 while the Beta after 2011 was 0.2. Women activity rate finally reach 72.4 in 2018. Despite, a positive evolution of the activity rate for women for the concerned period, the gap between men and women activity rate remain almost unchanged. In 2006, the difference in percentage between men and women activity rate was 15.1 while in 2018 this gap in percentage was 12.

Countries in red are those with the lowest gap of activity rate between woman and men. On the map above, we can see that the countries with the lowest gap of activity rate are mostly countries from the north of Europe : The scandinavian peninsula countries (Norway, Finland, Sweden, Denmark and Iceland), two other countries from the baltic region (Latvia and Lithuania) and Portugal and Romania. The results concerning the Scandinavian countries are consistent with the expectations because scandinavian countries are known to be equalitarian on gender matter. However, it is a little more surprising for the other countries. Even more surprising the country with the lowest activity rate gap is Lithuania with 3.8 % of différence between men and women for activity rate. This can be explain by the regulation there. Article 7 of the LET (Lithunia on Equal Treatment) which prohibits discrimination in the job market on the grounds of gender, race, nationality, language and so on. The Office of Equal Opportunities Ombudsperson monitors and controls the implementation of this regulation by employers for exemple.

The countries with the higher gap of activity rate are depicted in dark and light blue. The one with the higher gender gap is Turkey with 43%. This is consistent because Turkey is a country where the major religion is Islam and usually islamic woman are not allowed to work (with some exceptions). But we will not used Turkey for further analysis because it is not within the European Union. The higher gender gap within european countries can be found in Italy with 20.6 %.

In 2007 a National reform plan was undertaken by the Italian government called the “Libro Verde” in order to increase women participation in the labour market such as the improvement of the quality of childcare services. This is important to underline the fact that women are less present in the labour force particularly because of the double burden of having to conciliate work life and family life. Despite these reforms, the figures speak for themeselves, Italian women are still not enough present in the labour market. But why? The origins can be found in Childhood, women are taught to know how to keep a house (cooking, cleaning, taking care of the children and of their husband) meanwhile men are responsible for providing the family needs by working and earning money. Once again, why is it like this? We can aslo find answers in Religion. Italy more than other countries in Europe is deeply influenced by the Bible, majority of Italians are Christians, 88% of them belong to the oman Catholic Church.

The correlation between the activity rate and the gap activity rate for the years 2006,2010,2018 is -0.7814506, this is negative and only tells us that the two variables, activity rate and gap activity rate are evolving in differents ways.

Let’s now see the correlation between gap activity rate and the GDP in 2018. The correlation is 0.2629032 , this underlines a small correlation between gap activity rate and GDP in EU for 2018. Commonly, the GDP is a indicator of the wealth of a country, since the correlation is not that strong we can’t really say there’s a link. 

# Find the country with the lowest gap activity rate between woman & men  
mingap2018=min(PC_ActGap$gap) # LT lituania 3.8
PC_ActGap%>%filter(gap==mingap2018)%>%select(CNTR_CODE)
# Find the country with the higher woman activity rate  
maxgap2018=max(PC_ActGap$gap) # Turquey 43%
PC_ActGap%>%filter(gap==maxgap2018)%>%select(CNTR_CODE)
# Find the country with the higher woman activity  in Europe  
gap2018EU= PC_ActGap %>% 
  filter(! CNTR_CODE %in% c("EA17","EA18","EA19","EU15","EU28","TR","MK","MT"))
max2018UE=max(gap2018EU$gap) #20.6 Italy
gap2018EU%>%filter(gap==max2018UE)%>%select(CNTR_CODE)
# find the country  with the median gap activity rate
gapmoy=median(gap2018EU$gap) #9.5
gap2018EU%>% filter(gap==gapmoy)%>%select(CNTR_CODE) # BE,Belgium
# text for plot
li06=PC_ActGapTime%>%filter(time==2006,country=='LT')%>%select(gap)
li06=round(li06,2)
Lit <- grobTree(textGrob(c(li06,mingap2018), x=c(0.01,0.9) , y=c(0.2,0.12), hjust=0,
  gp=gpar(col="Blue", fontsize=10)))
it06=PC_ActGapTime%>%filter(time==2006,country=='IT')%>%select(gap)
Ita <- grobTree(textGrob(c(it06,max2018UE), x=c(0.01,0.9) , y=c(0.9,0.7), hjust=0,
  gp=gpar(col="Green", fontsize=10)))
be06=PC_ActGapTime%>%filter(time==2006,country=='BE')%>%select(gap)
Bel<- grobTree(textGrob(c(be06,gapmoy), x=c(0.01,0.9) , y=c(0.6,0.38), hjust=0,
  gp=gpar(col="Red", fontsize=10)))
# graph of the evolution of 3 country over time for gap activity rate
PC_ActGapTime %>% 
  filter(country %in%c("IT","BE","LT")) %>% group_by(country)%>% ggplot(.,) +
  geom_line(aes(x=time,y=gap,colour=country,group=country),size=1) +
  labs(title="Evolution of gap activity rate over time  ") + xlab("time") +
  ylab("gap paid") +
  scale_color_discrete(label= c("Belgium"," Italy", "Lithuania")) + theme_light() +
  theme(axis.text.x = element_text(angle=38))+ annotation_custom(Bel)+annotation_custom(Bel)+annotation_custom(Ita)+annotation_custom(Lit)

We are focusing on 3 countries to see how they have evolved over time. We focused on the country with the highest activity gap (Italy), the lowest activity gap (Lithuania) and the median gap (Belgium) with 9.5% . We have already discussed why we had those results for the first two, now let’s see why does Belgium has an activity gap rate that is higher than Lithuania. The reason behind a relatively important activity rate gap in Belgium is not legislation, it’s people state of mind. When asking to women why they think there’s such a gap between men and woman in the labour market, ⅓ of them argue it is because it would have a negative impact on their careers to be a women, once again the double burden of being a woman in the labour force plays a role here. According to ⅓ Belgium men, equality in employment opportunity for both gender is not yet a reality. The general tendency for the three countries is that gap activity rate is reducing in the period between 2006 and 2018. Going from 25.4% to 20.6 for Italy , from 15,1% to 9.5 for Belgium and from 7.6% to 3.8% for Lithuania.

Part 2: Unemployment rate difference between men and women across european countries and over time

From 2006 and 2018 the unemployment rate for both men and women decreased slightly from 7.7% for men to less than 7.5% and from 10% for women to 7.5% . At the beginning we had a gap of over 2.5 % between men and women for unemployment, as both unemployment rate are decreasing the gap is still not closing before 2008. From 2008 the gap may have reduced, unemployment is raising for both sex until 2013. There are two main reasons behind it. First, the Financial crisis is touching Europe from 2008 a year after the US, one year after that Europe has to face its own debt crisis which started in Greece. Companies, unable to refinance their governments are striked with the consequences of both crisis. A lot of them stopped hiring people and even firing some people because they can’t keep up. After 2013, unemployment is decreasing for both men and women and in the meantime, it seems like the gap is also trying to come back to how it was before the crisis . We can argue that the reason why the unemployment rate was closing after the crisis is because in desperate times it might not make a difference that you’re a man or woman.

Part 3: Part-time employment difference between men and women across european countries and over time

We have 4 categories of Part-time employment rate for women throughout Europe in 2017. Surprisingly, countries of eastern Europe have the lowest percentage of women having part time employment. It goes from 1.8% to 10.8% , let’s also add that the country with the lowest women in part time employment is within this group, it’s Bulgaria. Countries with a “medium” rate of part time employment for women are in Western Europe for the most part, it goes from 10.8% to 19.8% . Countries from Northern Europe have more women doing part time jobs (19.8% to 28.8%) . The country with the highest rate of women in part time employment is Netherlands. Now we will try to find explanations to these results. 

##### III b) 3 graphs 
# Gap H - F
TabF=PC_Partime %>% filter(sex=="F") %>% rename(PC_PF= PC_parttime_onemp)%>% select(-"sex")
TabH=PC_Partime %>% filter(sex=="M") %>% rename(PC_PH= PC_parttime_onemp)%>% select(-"sex")
PartTime=left_join(TabF,TabH,by=c('country','time'))
PartTime$gap=PartTime$PC_PF-PartTime$PC_PH
# choose 3 country
PartTime2018=PartTime%>% filter(time=="2018")%>% filter(! country %in% c("EA17","EA18","EA19","EU15","EU28","ME","MK","TR")) 
# negative for Montenegro so Men are more precare than woman
# Country with the lowest precarity gap
minpt2018=min(PartTime2018$gap)# 0.1 
mincountrypt2018=PartTime2018%>%filter(gap==minpt2018)%>%select(country)
mincountrypt2018 # BG, Bulgaria # 0.4
# Country with the higher precarity  gap
maxpt2018=max(PartTime2018$gap)# 58.3
maxcountrypt2018=PartTime2018%>%filter(gap==maxpt2018)%>%select(country)
maxcountrypt2018 #NL Netherland 
# Country with the median gender paid gap
#medpt2018=median(PartTime2018$gap) #20.3 
#medcountpt=PartTime2018 %>% filter(gap==20.3)%>%select(country)
# medcountpt # Norway
# European gender paid gap
EUpt=PartTime%>%filter(country=="EU28",time=="2018")%>%select(gap)
EUpt# 22.8

# text for plot
bu06=PartTime%>%filter(time==2006,country=='BG')%>%select(gap)
bu06=round(bu06,2)
bu <- grobTree(textGrob(c(bu06,minpt2018), x=c(0.01,0.9) , y=c(0.08,0.12), hjust=0,
  gp=gpar(col="Blue", fontsize=10)))
nl06=PartTime%>%filter(time==2006,country=='NL')%>%select(gap)
nl <- grobTree(textGrob(c(nl06,maxpt2018), x=c(0.01,0.9) , y=c(0.9,0.8), hjust=0,
  gp=gpar(col="Red", fontsize=10)))
no06=PartTime%>%filter(time==2006,country=='NO')%>%select(gap)
no18=20.3
no<- grobTree(textGrob(c(no06,no18), x=c(0.01,0.9) , y=c(0.65,0.5), hjust=0,
  gp=gpar(col="Green", fontsize=10)))
# graph of the evolution of 3 country over time forpart time activity rate
PartTime %>% 
  filter(country %in%c("NO","BG","NL")) %>% group_by(country)%>% ggplot(.,) +
  geom_line(aes(x=time,y=gap,colour=country,group=country),size=1) +
  labs(title="Evolution of part-time gap activity rate over years  ") + xlab("time") +
  ylab("gap paid") +
  scale_color_discrete(label= c("Netherland","Norway", " Bulgaria")) + theme_light() +
  theme(axis.text.x = element_text(angle=38))+
  annotation_custom(bu)+annotation_custom(no)+annotation_custom(nl)

Bulgaria is the country in Europe with the lowest rate of women in part time jobs, in 2006 it was 0.4% , it went to 0.1% in 12 years. An article from the World Bank published in March 2019 called “ Bulgaria Emerges as champion in Women Legal Rights Affecting Work “ help us understand this figure. Bulgaria is one of the 6 countries that removed all job restrictions on women over 10 years. The government wants to ensure gender equality in labour market, they believe that a world where women who have the skills to work in what they want to work would not only be fairer but also could benefit all. Women in Bulgaria occupy important high posts in politics, the President of the Parliament is a woman, so as the capital city mayor and 2 deputies ministers.

There are way more women doing part time job in Netherlands, they are 58.3 percent. From 1980 to 2016 women have been more present in the labour force, the percentage of women working in the Netherlands went from 35% to 70% , but 60% of them are actually doing part time jobs. Part time jobs offers job satisfaction, free time for children care and also free time for leisure activities. In Netherlands, the average father works five days a week whereas the average mother works 3 days a week, this can be explained by the fact that twice as many women as men look after the household and children.

Once again we have various situtations in Europe when we try to understand the relationship between the births and women in part time work from 2006 to 2018. If we focus on France in purple it seems like there’s a small change in the percentage of women in part time job depending on the number of births. When there’s more births there’s little less women in part time working, when the births decrease the number of women in part time job goes the other way.This might be because even when women are having part time job it is still hard for them to find time to take care of the children, once again double burden.

Part 4: Temporary employment difference between men and women across european countries and over time

In 2018, Slovenia was the country with the highest rate of women who didn’t want to find a permanent job, the percentage was 46.5% . Let’s try to find out why. We know that childcare is usually one of the main obstacle for women in the labour force, while in Slovenia they offered a good child care leave both for men and women this doesn’t explain why we have a little than 50% of women who don’t want to find a permanent job.

When looking to the evolution of difficulties for women and men to find a permanent job overtime, it’s more or less constant. We went from no gap in 2006 to a negative gap in 2018 almost minus 1, this tells us that women have more difficulties today than before to find a job. Before trying to give an explanation for the latter, we will look more closely to how the trend has evolved. Between 2007 and 2013, during the financial and the debt crisis, we have a negative gap for men and women when it comes to find a permanent job, we have a very small increase from 2013 to 2014 when there is no gap again before the gap is negative until today. If we compare the amount of women in the labour force at the beginning of the time frame to what it is today, we can see that it went from 44.5% to 46.2%. So maybe since we have more values in 2018 this highlight better the fact that women are struggling more to find permanent job than men are, the labour force today might be more representative.

The map we have of 2018 representing the gap between men and women to find a permanent job shows once the disparity in the EU for gender equality in the labour market. The country with women having less difficulties than men to find a permanent job is Lithuania with a gap of minus 9,5%. We already discussed the legislation there and how it could be a reason of why they have good conditions for women in the labour market. Now searching for the country with women having more difficulties than men to find a permanent job, we have Malta with a gap of minus 9,7%. This can be explained by the fact that there is a higher percentage of men working in Malta than women. Another factor explaining why women have more difficulties than men in finding permanent jobs in Malta is the mentality, once again we attribute the role of a woman to householding , motherhood and so on, it is indeed harder for them to find permanent jobs with the double buren already invoked before in this study.

Part 5: The gender pay gap across european countries and over time

Between 2010 and 2017, the gap paid between men and women decreased by 1.1% from 17,10% to 16% . Between 2011 and 2012 we had a slight increase up to 17.4% for the gap, right after this period we have a progressive decrease. What can explain this tendency? Since 2009, the Academy of European Law host annual seminars to discuss european law on equality between women and men. From 2014 to 2020, series of seminars in the framework of the European Commission’s Right, Equality and Citizenship Programme have been held and will be host to fight inequality.

There’s a lot of different situations in Europe regarding the gap paid between men and women in 2017. Some countries from Eastern Europe have a huge gap in wages from 15.6% to 20%, nordic countries score quite well with a gap going from 4.7% to 13. For countries from Western Europe it’s a mix of both situation. We will be focusing on 3 countries : Estonia with the highest gap, Finland with the median gap and finally Roumania with the lowest gap. However it is important to say that the gender pay gap measures inequality but not necessarily discrimination. 

# filter PC_Gapaid2 for work on max min medium
PC_Gapaid22017=PC_Gapaid2%>% filter(time=="2017")
# Country with the lowest gender paid gap
minGP2017=min(PC_Gapaid22017$mean_gappaid_percountry)
mincountGP=PC_Gapaid22017 %>% filter(mean_gappaid_percountry==minGP2017)%>%select(country)
mincountGP # RO, Roumania # 4.73% ( puis Marlte Belgique)
# Country with the hogher gender paid gap
maxGP2017=max(PC_Gapaid22017$mean_gappaid_percountry)
maxcountGP=PC_Gapaid22017 %>% filter(mean_gappaid_percountry==maxGP2017)%>%select(country)
maxcountGP # EE,Estonia # 20%, Republique Tcheque Autriche 
# Country with the median gender paid gap
median(PC_Gapaid22017$mean_gappaid_percountry) #13.52 we don't find this in our data 
medcountGP=13.5
medcountGP=PC_Gapaid22017 %>% filter(mean_gappaid_percountry==13.5)%>%select(country)
medcountGP # Finland
# European gender paid gap
EUGP=PC_Gapaid22017%>%filter(country=="EU28")
EUGP # 16%
##### V c)  graph of the evolution of 3 gap wages over time

# text for plot
ro07=PC_Gapaid2%>%filter(time==2007,country=='RO')%>%select(mean_gappaid_percountry)
ro07=7.47
ro <- grobTree(textGrob(c(ro07,minGP2017), x=c(0.01,0.9) , y=c(0.18,0.22), hjust=0,
  gp=gpar(col="Blue", fontsize=10)))

fi07=PC_Gapaid2%>%filter(time==2007,country=='FI')%>%select(mean_gappaid_percountry)
fi07=19.9
fi <- grobTree(textGrob(c(fi07,medcountGP), x=c(0.01,0.9) , y=c(0.5,0.5), hjust=0,
  gp=gpar(col="Green", fontsize=10)))

ee07=PC_Gapaid2%>%filter(time==2007,country=='EE')%>%select(mean_gappaid_percountry)
ee07=30.3
maxGP2017=round(maxGP2017,2)
ee<- grobTree(textGrob(c(ee07,maxGP2017), x=c(0.01,0.9) , y=c(0.85,0.7), hjust=0,
  gp=gpar(col="Red", fontsize=10)))

#plot
PC_Gapaid2 %>% 
  filter(country %in%c("FI","EE","RO")) %>% group_by(country)%>% ggplot(.,) +
  geom_line(aes(x=time,y=mean_gappaid_percountry ,colour=country,group=country),size=1) +
  labs(title="Evolution of gap paid over time  ") + xlab("time") + ylab("gap paid") +
  scale_color_discrete(label= c(" Estonia", "Finland", "Roumania")) + theme_light() +
  theme(axis.text.x = element_text(angle=38))+
  annotation_custom(ro)+annotation_custom(fi)+annotation_custom(ee)

The country with the highest paid gap in Europe is Estonia, it went from 30% to 20% between 2007 and 2017. Meanwhile productivity among employees should be the only thing explaining wage gap, discrimination is still held responsible for wage gap. In Estonia, lot of men are still dominant in leading positions compared to women, this can explain why we have such gap difference. It seems that when they occupy the same position, women work is still undervalued. The country with the median gap paid is Finland with 20% in 2010 and 13% in 2017. Finnish women are more implicated in women-dominated work such as in the care industry, those works have lower wages. They also have more precarious jobs such as part time or fixed term jobs. We still have a decrease in the gap in the frametime we are working with, this might be because of the Equal Pay Programme of the Finnish government with the Act on Equality between men and women including reconciling work and family life. The country with the lowest gap paid is Roumania it went from 7% to 5% between 2010 and 2017. This can be explained by the political regime before 1989. The communist regime promoted an official policy of gender equality for more than 40 years in education, employment, and wages differentiation based on gender. Romanian women do well in male -dominated jobs. During the past 20 years improvement in the healthcare sector have been made because it is believed that well-performing healthcare system increase productivity for women because it allows them to take more intensively part in the labour market.

Conclusion

At the beginning of this project, we already had our mind set on the countries who are doing great in reducing gender inequality in the labour market in Europe. We thought that Western and Northern countries would do well, actually way better than the eastern countries because western europe have been more liberal and modern starting from the end of World War II whereas eastern europe had been under communism with more traditionals values regarding women role in society. Working on this project underlined two things : eastern europe countries seemed to be ahead when it comes to gender equality in the labour market and we still have some changes to do in the way we view women in our society in order to have more gender equality in general. In the countries of eastern europe, such as Roumania, Bulgaria and Lithuania huge efforts have been made in their legislation to help women to have a more active part in the labour market and be at the same level as men. Usually, these legislations aimed at reducing the double burden on women, they have to conciliate family life and work life. This double burden seemed to be one of the main obstacle in gender equality regarding labour market. Let’s underline another reason for gender inequality in this area. The role of women in our societies is still influenced by traditions, in many people min a woman has to take care of the house and children while the man provides for his family needs by working. This is taught from a young age to women like in Italy. Also religion seems to play a part in this. We can say that inequalities in overall are reducing over years. But what does gender inequality look like if we focus on the very top of the income distribution? Do we find any evidence of the famous ‘glass ceiling’ preventing women from reaching the top? How did this change over time? There are still few women at the head of big companies and same for internationale organizations with eventually few exceptions.

THE END

Reference

Open source data from Eurostat

Complementary informations