Bachelor Thesis
The effects of the labor migration from EU10
countries on the Dutch labor market:
Evidence on the effects on wages and job opportunities for the
Dutch natives
Michelle Lim 10047573
Economics & Business Specialization: Economics Supervisor: I. Rozentale 07-07-2014
Table of Content
1. Introduction 3
2. Research approach 4
3. Theoretical and empirical literature review 6
3.1 Theoretical model 6
3.2 Effects of labor migration in previous studies 6 3.2.1 Effects of labor migration on wages 7 3.2.2 Effects of labor migration on job opportunities 8 3.2.3 Advantages and disadvantages of previous studies 9 3.4 Conclusion 10
4. Labor market in the Netherlands 11
4.1 The effects of labor migration on the Dutch labor market 11 4.1.1 Wage effects of the former migration flows 11 4.2 EU10 labor migration in the Netherlands 12
4.2.1 Permanent migration 12
4.2.2 Temporary migration 13
4.2.3 Effects of EU10 labor migration 13 5. Overall conclusion 16
1. Introduction
Since a few years the debate about whether countries should allow labor migrants has been fuelled again. Main reason for this debate is the enlargement of the European Union (EU) in 2004 (Heyma et al, 2008). In May 2004 ten countries joined the EU, Cyprus, Malta, Estonia, Latvia, Lithuania, Poland, Hungary, the Czech Republic, Slovakia and Slovenia (further in the text EU10). This expansion has been the most extensive one within the EU in one time. Bulgaria and Romania followed in 2007. As a member of the EU, citizens have the right to move and work freely amongst other EU countries. The Dutch government and some other governments of EU15 countries (the EU countries before the expansion) were afraid of more unemployment and lower wages for the natives, due to free labor mobility. Because of the different welfare states, politicians and citizens in the EU15 are afraid for huge influxes of migrants coming from EU10 countries, looking for a better standard of living by taking jobs away from natives. The UK, Sweden and Ireland opened their borders immediately after the expansion, but the Netherlands and some other countries, e.g. Germany and Austria, imposed a transition period. Dutch borders opened for the citizens of the EU10 in 2007 and for Romanian and Bulgarian citizens in 2014 (Constant, 2011). Since EU citizens do not need a working permit anymore to work in another EU country, it has become much easier to migrate and work somewhere else (Hartog and Zorlu, 2004).
Labor migration and their effects on the labor market have always been a topic with a lot of different opinions and visions. That is why a lot of research about the effects of labor migrants has been done so far. In recent years, the concentration of research has moved from the US to Europe, due to the expansion of the EU. Research in Europe is mainly done in Germany and the UK, but some studies have been conducted in other countries, such as Austria, the Netherlands and France.
In the Netherlands the amount of labor migrants arriving from Eastern Europe has risen by 137%, from 98.000 to 237.000 between 2007 and 2012 (Ooijevaar et al, 2013). This increasing number of migrants emphasizes the fear politicians and citizens had. Constant (2011) states that 36 percent of the Dutch citizens think that the EU expansion has contributed to job losses in their own country. In addition, Constant (2011) split the opinions of EU-citizens into different segments. 64 percent of the people with only a primary school degree think the EU expansion contributed to job losses. For people with a college degree or higher, this is 40%.
However, according to Constant (2011) the geographic mobility of workers can act as the equilibrating factor between demand and supply in labor markets, as labor migration can improve the efficiency, productivity and wellbeing of all societies.
Thus, did the EU expansion have a positive or negative effect on the Dutch labor market? It is interesting to see what actually happened in the Netherlands after the borders opened to the citizens of the EU10. This theoretical thesis will try to answer the following research question:
What is the effect of EU10 labor immigration on wages and job opportunities of Dutch native workers?
To get a well-based and objective answer to this question, sub-questions are formulated, namely:
1. What would be the effects of labor migration for natives according to the developed theories, with a closer look on wages and job opportunities? 2. What is known about the effects of immigration on wages and job
opportunities of native workers according to previous empirical studies? 3. Which effects of immigration on wages and job opportunities occurred on
the Dutch labor market since 1950?
4. What is the current situation in the Netherlands regarding labor migration with a particular emphasis on EU10 countries?
By answering the formulated sub-questions, we will retrieve the information required regarding labor migration, such that a clear, well-based answer on the main research question can be formulated.
These questions are interesting, as labor migration is always dynamic and at the moment it is a very current topic in the Netherlands. Despite the large amount of studies that have been performed so far, a strong debate is still ongoing between politicians and citizens in the Netherlands and other countries. I want to analyze if this debate is plausible. Also, the studies that are performed on the current EU labor migration into the Netherlands are mainly studies about the expected effects on the labor market. There are still no studies done that measure the effects of EU10 labor immigration on the Dutch labor market after the free labor mobility was legalized.
The remainder of the thesis is structured as follows. First, the research approach of this thesis is explained. In chapter three the theoretical models and empirical studies are considered. By explaining theoretical models and discussing some studies performed, the first two sub-questions can be answered. In chapter four the focus will be kept on the Dutch labor market. The effects of labor migration on the Dutch labor market from 1950 until now and the current situation in the Netherlands will be considered. After this, the findings will be summarized in order to answer the main research question in a conclusion.
2. Research approach
The research question is answered by carrying out a structured literature-based case study. By using data from Centraal Bureau voor de Statistiek (CBS) and secondary data from reports, relying on the theory and results of previously carried out empirical studies a hypothesis for the effects of labor migration from EU10 countries on wages and job opportunities of the Dutch labor market can be formulated.
The literature used in this thesis is searched for in Google Scholar. By using the search terms, labor migration, effects, labor market, wage, unemployment and EU a selection has been made. By reading the abstracts of different studies a second selection was made. When the topic of the abstract was related to the effects of labor migration on wages and job opportunities for natives in host countries, the article was included for review in this thesis.
For the first two sub-questions studies about former migration flows are used. For the last sub-questions mainly Dutch studies are used. These studies are searched for in Google Scholar by using the search terms ‘effecten, arbeidsmarkt, Nederland, EU-uitbreiding’. Most Dutch studies about labor migration from EU10 countries are related to the expected effects of the incoming labor migrants.
An objective conclusion can be drawn regarding the research question formulated by using the systematic (non exclusive) method described above.
3. Theoretical and empirical literature review
This chapter serves as a background for this thesis and will first discuss the theories about the impact of labor migration on wages and job opportunities. Next, previous studies will be mentioned and discussed to see what the effects of labor migration have been. This way, the first two sub-questions will be addressed:
1. What would be the effects of labor migration for natives according to the developed theories, with a closer look on wages and job opportunities?
2. What is known about the effects of immigration on wages and job opportunities of native workers according to previous empirical studies?
3.1. Theoretical model
Economic theories regarding the possible effects of labor migration can be divided into two different main streams, i.e. labor migrants can either be seen as substitutes or complements.
The most basic economic theory considers immigrants and natives as perfect substitutes. Assuming that the labor demand curve is downward sloping, the labor supply curve upwards and capital is fixed, wages will reduce and employment will increase when immigrants enter the labor market. However, natives are less willing to work for a lower wage, causing the unemployment rate for natives to rise (Borjas, 2013).
If labor migrants are complements for natives, these labor migrants will search for other jobs than natives do. By assuming that immigrants are willing to take labor-intensive jobs, the higher skilled natives get the opportunity to specialize themselves in more capital-intensive jobs. This way, immigrants and natives complement each other. The downward sloping demand curve will shift up, due to the rise of the marginal product of natives. The native wage will therefore increase and native employment will also rise due to higher wages (Borjas, 2013).
From Kahanec’s et al (2009) point of view, immigration of low-skilled workers hurts the low skilled and benefits the high skilled. Immigration of high-skilled workers is the other way around; it will hurt the high high-skilled and benefits the low skilled. In Figure 3 (Kahanec et al, 2009) this theory is explained in a graph. By considering an increase of high-skilled immigration, the supply will increase (from to ), whereas the wage will decline from to . As an effect, the demand for low-skilled laborers increases due to the complementarity
(Kahanec et al, 2009). If the market is competitive, the wages for the low-skilled laborers will increase from to . If there is a rigid union wage, which is the case in the Netherlands, unemployment of low-skilled workers will decline from
to . Due to the demand increase, the employment rate for low skilled laborers increases and this also causes an upward shift in the demand curve for high skilled workers. Due to this increased demand, the wages for high-skilled workers will increase, from to .
In case of an immigration influx of low-skilled migrants, the effects are opposite. In a competitive market, the wages for low-skilled workers will decline to . Under a rigid union wage, the unemployment for low-skilled workers increases from - to - . Due to the complementarity, the demand for high-skilled workers increases. As an effect of this, the wages for high-skilled workers will increase to (Kahanec et al, 2009).
Considering these theories the effects of labor migration depends on the substitutability or the complementarity of labor migrants. The effect on wages for natives will be negative if labor migrants can substitute native workers. If labor migrants are complementary, wages will increase.
The effect on job opportunities also depends on these two factors. Considering labor migrants as substitutes for native workers, unemployment will rise as an effect on the lower wages. If labor migrants complement native workers, the unemployment rate will decline.
Figure 3. Immigration in an economy with heterogeneous labor (Kahanec et al, p.4, 2009)
3.2. Effects of labor migration in previous studies
Besides the theoretical model discussed previously, empirical research can be done. Many studies are developed and performed in different countries. Borjas (2003) states that outcomes can differ per country and per study. In this section previous studies are discussed about their results regarding the effects of labor migration on wages and job opportunities. Moreover, the advantages and
disadvantages of the different methods will be discussed, such that the discussed effects can function as a good comparison in the final conclusion of this thesis.
3.2.1. Effects of labor migration on wages
Probably one of the most famous studies about the effects of labor migration is the study of Card (1990). Due to a huge inflow of Cubans into Miami in 1980, called the Mariel boatlift, Card (1990) was able to perform a natural experiment. Considering the effect of this influx on the people who already live in Miami, Card (1990) did not find a relative change in wages for this group. Overall conclusion is that Miami was able to absorb the 80.000 incoming Cubans in their industry and that there were no effects for the low skilled population already in Miami.
Another study done in the US is performed by Borjas et al (1992). In the 1980’s wages fell of less-skilled workers. There were two causalities that could be the reason for this. Migration into the US of mainly less-skilled people or a trade deficit; imports rose from industries with mainly less-skilled workers which means that productivity in the US declined (Borjas et al, 1992). In order to measure the effects and to see what the cause was of the decline in wages, Borjas et al (1992) performed a study using the factor proportions approach. When looking at the education level, Borjas et al (1992) found that 36 percent of immigrant men did not have a high school degree. For native men this was 15.3 percent. Wages of high school dropouts declined by 4.8%, where migrants seemed responsible for about 44% of this widening gap. This outcome substantially differs from the study of Card (1990), where no effect was found.
A more recent study is done by Boeri and Brucker (2005). These authors studied the cost and benefits of East-West migration in the expanded EU using the computable general equilibrium approach. Their conclusion is that when the labor markets are clear, a 1 percent increase in population due to migration is positive for GDP; this will rise with 0.3 percent. However, in case of wage rigidities, which is the case in the Netherlands, this increasing GDP goes hand in hand with a decline in wages for natives. This could be an incentive for governments to close the borders. To make these wage effects smaller, the suggestion is made by Boeri and Brucker (2005) to make the labor market in the EU more flexible.
In contrast with the methods discussed above, some methods have mixed results. Studies that have been done using the area analysis or the skill group approach confirm this. For example, wage effects measured in the US, using the area analysis, are much smaller than in Europe. This is surprisingly, because the wages in Europe are less flexible than in the US. Wages declined at most 1.5 percent for a 1 percentage point increase in migrant share in the US, while in Germany the founded results are -3.3 percent and even -6.4 percent (Okkerse, 2008).
Two studies that used the skill group approach are giving mixed results as well. Borjas (2003) measured the impact of immigrants on the US native workforce used data from the U.S. labor force for the years 1960-1990 and 1998-2001. The results conducted by this study were statistically significant. A ten percent supply shock reduced weekly earnings by about four percent, a sufficient negative result. These results differ per education group, the wages declined by
8.9 percent for the high-school dropouts and only 2.6 percent for high-school graduates.
However, using the same skill group approach, results estimated by Bonin (2005) did not find any significant effect for a decline in German wages. Bonin (2005) found that for every ten percent rise in the immigrant share in the workforce, the natives’ wages reduce less than 1 percent. Surprisingly, this estimated result is in contrast with the results estimated by Borjas (2003).
The UK opened its borders immediately after the EU expansion for eight countries (Estonia, Hungary, Poland, Latvia, Lithuania, Czech Republic, Slovakia and Slovenia). This gave Dustmann et al (2005) a reason to study the effects of incoming labor migrants on the labor market. First thing Dustmann et al (2005) noticed, is the difference in the education of immigrants. In the UK the immigrants are not typically low skilled. The composition of natives’ skill-groups is the same as the composition of the immigrants’ skill-groups. Therefore, the composition of skill groups does not change. This seems to be in contrast to the EU10 labor migrants in the Netherlands, where the majority of the EU10 labor migrants seem to be low-skilled. As a conclusion, Dustmann et al (2005) did not find any effect regarding wages. A second study in the UK (Blanchflower, et al, 2007) had the same conclusion. These authors did not find any effect on wages.
3.2.2. Effects of labor migration on job opportunities
While the wages were not affected in Miami due to the incoming Cubans, Card (1990) did find an increase in the unemployment rate. However, this increase could be signed to the Cubans in Miami. Unemployment rates for Cubans were three-percentage point higher in 1981 than expected. This number can easily be explained by adding up the Mariels. So, this new group did not affect the earlier Cuban immigrants, who were in Miami already. The increase in unemployment can be fully aligned to the new group of Cubans.
Borjas et al (1992) measured, next to the effect on wages, also the effect on job opportunities in his study mentioned in the former paragraph.The reason for this is, that not only wages declined in the 1980’s, also the employment rate decreased. In their analysis Borjas et al (1992) concluded that the amount of labor migrants increased significantly with approximately 35 percent. Despite this increase, the incoming migrants are only for 25 percent responsible for the increase of the unemployment rate.
Regarding the area analysis, not only the wage effects are mixed, results of effects on job opportunities are also mixed. Angrist & Kugler (2003) found a decrease in the employment rate of 1.6 percent in EU countries, while Winter-Ebmer & Zweimuller (2000) did not find any evidence of an effect on job opportunities in Austria.
The estimated results of Borjas (2003) and Bonin (2005) on the effect on job opportunities also differ. Borjas (2003) concluded that the amount of work is reduced by a supply shock; a ten percent supply shock reduced the time of working with 3.7 percentage points. Bonin (2005) also performed a skill group approach in Germany considering the effects on job opportunities. His results showed that there was not a statistically effect on job opportunities.
In the UK, where the characteristics of labor migrants differ from the labor migrants in the Netherlands, a small negative effect on employment in the
intermediate skill group is found by Dustmann et al (2005). However, this effect is offset for the high skill group, so in total no effect is found. Blanchflower et al (2007) reviewed several literature papers about the effects of labor migration on the job opportunities in the UK. After reviewing these papers, their conclusion is that, despite the inflow of immigrants, there is no effect on unemployment rates.
3.2.3. Advantages and disadvantages of previous studies
Now several studies have been reviewed in former paragraphs, some studies need some attention due to their used research method. The results found by Borjas (1992) require some cautiousness while interpreting, just like other results using the same factor proportions approach. The results can be an overestimation due to the assumption that all skill groups, for natives and migrants, have to be perfect substitutes. Another reason for bias using the factor proportions approach is that it might be based on theoretical models too heavily (Okkerse, 2008). It simulates the impact for given elasticity’s of substitution, instead of estimating the impact of immigration on the wage structure. Despite the criticism, this method can be a valuable instrument to gain some insights regarding the effects of migration (Okkerse, 2008).
An advantage of the computable general equilibrium approach is that a distinction can be made regarding effects for different types of households. This is an advantage, as migration can affect just one special type of household. Using the computable general equilibrium approach, it is possible to consider the effects of migration and simulate the changes of economic variables. Another distinction compared to other models is that there is no ceteris paribus condition; other variables can change in this model. Research using this method generates the same conclusion on the outcome of migration. The effects are rather small and are sometimes even positive (Okkerse, 2008).
When considering the area analysis there are two downfalls. First, migrants choose their destination depending on the local wage and/or unemployment level. A possible endogeneity problem is the result. Due to this endogeneity, it is impossible to state that labor migrants are the cause of lower wages or a higher unemployment rate. The causality can run either way. If this problem is neglected, the estimations will be biased. Research held using a natural experiment or the IV estimation can deal with this endogeneity problem. The second downfall is the fact that natives can react on the incoming migrant by moving to another area. If this is the case, no effect in the area will be found. The conclusion that there is no effect on migration is then incorrect. Interestingly, results from studies using this approach differ from each other.
The section above discussed the existing theories in literature and pointed out a few studies regarding the effects of migration on wages and job opportunities. Whilst considering the economic theories, you have to keep in mind that the underlying assumption is that the labor market is in equilibrium before the labor migrants enter the host country. This assumption is however not realistic. This way, it is not possible to get a clear-cut, realistic answer on the effects of labor migration based on a theoretical model.
The different empirical models make clear that it is possible to measure the effects of labor migration on the wages and job opportunities for natives in
the host country in very different manners. As a result, different kinds of studies have shown to give different results. While most of these studies have the same results as the theory suggests, using the area analysis method or the skill group approach results are not always in line with theory. Despite all these studies, there are difficulties in concluding the effects of labor migration. Besides the effects on wages and job opportunities, labor migration effects a lot more aspects. E.g. Boeri & Broecker (2005) concluded that the wage effect is negative, but GDP will rise. While discussing whether a country should allow labor migrants or not, one should keep these other aspects in mind as well. Also important to note is, considering the effects on job opportunities, that it is possible that the unemployed natives are not part of the study. It is possible that no employment effect is found, but it could be the case that it is harder for unemployed natives to find a job (Okkerse, 2008).
3.4 Conclusion
Some studies regarding the effects of immigration were discussed in this chapter. With regard to the effect on wages, the theoretical models have the same conclusions. The effect of immigration on wages of natives depends on the substitutability or complementarity. If migrants are considered as substitutes, the natives’ wages will decline. If migrants complement the natives, the wages will increase. The actual effects of immigration can be measured on many different ways, but reviewing these studies, conclusion is that effects are very small and not always significant. It seems that immigrants not always harm the natives. This indeed depends on the substitutability or the complementarity of immigrants. If labor migrants are mainly low skilled, high-skilled workers did not have any effect due to the labor migrants. Dustmann et al (2005) noticed that the immigrants had the same characteristics and overall there was no wage effect found.
When considering the effect on job opportunities, theory predicts that the employment rate will decline if migrants are substitutes. In case of complementarity, migrants will benefit the natives and therefore, the employment rate will increase. Reviewing empirical studies, outcomes are in line with conclusions made out of theory. However, the effects are much smaller than indicated by theoretical models.
All in all, despite some opposite results, the results seem to be small in the studies discussed. From these results, it can be expected that the effect of labor migration in the Netherlands will be small.
4. Immigration in the Netherlands
Despite some different results between studies, the effects of immigration on wages and employment of natives have been very small or invisible. In this chapter the focus is on the Netherlands. Presenting this chapter will give the opportunity to answer sub-questions 3 and 4:
3. Which effects of immigration on wages and job opportunities occurred on the Dutch labor market since 1950?
4. What is the current situation in the Netherlands regarding labor migration with a particular emphasis on EU10?
4.1. The effects of labor migrants on the Dutch labor market
Immigration has changed over time. From the ‘traditional’ migrant to a ‘modern’ migrant. The first migrants in the Netherlands, after the Golden Century around 1650 and the industrial revolution in 18th century, came from Indonesia. After the decolonization, around 350.000 people arrived in the Netherlands. In the 1960’s the Dutch economy was booming and labor demand exceeded the labor supply. So the Dutch government encouraged people from Southern Europe to work in the Netherlands. Those migrants turned back to their home country, due to the good prospects in their own country. Around 1970 migrants came, most of the time, from Turkey and Morocco. Because the economic prospects in their own country did not get better, these migrants did not return to their home country as much as the European migrants. Around 50.000 labor migrants from Turkey and Morocco got a permit to stay in the Netherlands and due to the family reunification 20.000 extra family members arrived as well (Roodenburg et al, 2003). People from Surinam and the Netherlands Antilles arrived in the Netherlands as well, not only for labor prospects, also due to political reasons. This group was often high educated. After 2000, the greatest flow of migration came from the EU10 countries.
4.1.1 The wage effect of the former migration flows
Hartog & Zorlu (2005) measured the wage elasticity’s in the Netherlands in 1997. The different migrant groups in 1997 can be divided into five groups from where they originally came from, namely:
1. Western countries (EU, Eustern Europe, Indonesia, Japan and other industrialized countries)
2. Turkey 3. Morocco 4. Dutch Antilles 5. Surinam
Dutch native workers were divided into skill levels, and the results from this study are mixed. Immigration from EU seems to have a positive effect on wages for low skilled native workers, and a negative effect on wages for high skilled workers. Nevertheless, these coefficients are not significant. The founded coefficients for non-EU migrants seem to be significant, and have an opposite result. Low skilled natives have a negative wage elasticity, and high skilled natives a positive wage elasticity. A remarkable effect is the fact that Western immigrants have a small negative impact on low skilled native workers, despite their high education level. Theory would suggest that those high skilled migrants are competing with high skilled native workers. By considering the wage elasticity’s per group, different results are estimated. Turkish migrants have a small negative effect on wages of low and medium skilled native workers, whereas Moroccan migrants have a positive effect. Possible explanation for this difference might be that the Turkish migrants have a somewhat better reputation than Moroccan migrants (Hartog & Zorlu, 2005). Due to this
reputation, Turkish migrants can compete with native workers, whereas Moroccan migrants function as complements for natives, which as an effect cause the positive wage elasticity. Surprisingly, for migrants coming from the Dutch Antilles and Surinam Hartog & Zorlu (2005) found a positive wage elasticity. It was expected that these migrants could function as substitutes, because the shared common history and the same language, but the opposite happened. The wage elasticity is positive, so native workers can profit from this group. To see whether migrants have to compete more with native workers or other migrants, Hartog & Zorlu (2005) measured the wage elasticity’s for other migrant groups. All estimated results are higher than the results estimated for native workers. It can be concluded that migrants function more as substitutes for other migrants than for native workers (Hartog & Zorlu, 2005).
The effect of labor migration of these groups on job opportunities for the Dutch native workers is, for these former migration flows, still unknown. Studies regarding these migrants considered the effects of labor migration on the labor migrants. The effect for Dutch native workers is only considered by Hartog & Zorlu (2005) with regard to the wages. Research analyzing the effects of labor migration for job opportunities of Dutch native workers is recommended due to the given gap in available information.
4.2. EU10 labor migration into the Netherlands 4.2.1. Permanent migration
Ooijevaar et al (2013) described the changed group of labor migrants into the Netherlands for the CBS. While in 2007 435.000 EU citizens lived in the Netherlands, in 2012 this was increased to a total amount of 588.000. This expansion can be contributed to the increase of people from EU10 countries; in 2007 98.000 EU10 citizens lived in the Netherlands and in 2012 this was more than doubled to 237.000. Especially Polish inhabitants entered the Dutch labor market in these years, while people from Latvia relatively increased the most. The Gemeentelijke Basisadministratie (GBA) and the Belastingdienst encountered all these numbers. The estimated amount of permanent migrants is stated in figure 4 (Heyma et al, 2008).
4.2.2. Temporary migration
Besides the permanent migrants, a large amount of temporary migrants do stay in the Netherlands. This group often has a temporary job for a season. For the GBA and the Belastingdienst it is difficult to estimate these temporary labor migrants. The estimated amount of permanent migrants is stated in figure 5 (Heyma et al, 2008). As can be seen, the estimated amount of temporary migrants is larger than the permanent migrants.
Figure 5. Amount of temporary immigrants (Heyma et al, 2008, p. 19)
According to Berkhout en ‘t Hof (2012) the amount of temporary migrants have risen to over 200.000 in 2012. However, the authors state that these migrants are a positive factor for the Dutch economy. This is due to the fact that temporary migrants seem to function as complements. They can fulfill vacancies if demand is high and native employers are not available to fulfill this (Berkhout en ‘t Hof, 2012).
Difference between the temporary and the permanent labor migrants is that temporary labor migrants work more hours compared to the permanent labor migrants. Also, it seems that temporary migrants earn less, are younger, are working more often in Limburg and Noord-Brabant, and are working as a temporary employee and in the farm industry.
A possible reason explaining why these EU10 migrants are just temporarily in the Netherlands, just like the first labor migrants, can be the fact that they have the opportunity to come back to the Netherlands again. Due to the free borders for the EU-citizens, labor mobility is free. The Turkish and Moroccan people did not get this opportunity. The decision to return to their home country was therefore much more rigorous (Heyma et al, 2008).
4.2.3 Effects of EU10 labor migration
In figure 6 (Heyma et al, 2008) the wages can be seen from the migrants; it is clear that temporary migrants have jobs with the lowest incomes.
Figure 6. Wage distribution of natives and EU10-migrants (Heyma et al, 2008, p.33)
In 2003 Roodenburg et al wrote a book that has been commissioned by the
Centraal Plan Bureau (CPB). Roodenburg et al’s (2003) conclusion was that in
general an economy is better off with immigrants who work many hours and for a long period of time. The natives that have skills comparable with the skills of the migrants will lose. Their wages will decline, but the effect will be small. This is a conclusion with non-Western immigrants in mind, but according to the assumptions, the EU10 migrants fit into this frame as well. These assumptions are that migrants are young, have followed their education in their home country, are motivated to work and won’t make any use of social security. Above all, before their retirement they will go back to their home country. Regarding to the characteristics of the temporary migrants discussed on page 13, these can be compared with the non-Western immigrant Roodenburg et al (2003) discussed.
To consider the effects on wages and job opportunities, Heyma et al (2008) used the skill group approach of Borjas (2003), but based their groups on different segments. By establishing the groups, they considered a certain region and sector at a certain time. In the end, 600 combinations and all months during 1999-2005, gave 50.400 different groups of the labor market. This way, the total amount of jobs and average wage could be measured and together the total amount of immigrants was included.
Heyma et al (2008) concluded that permanent labor migrants are complements of native workers. Job opportunities of natives seem to increase by a doubling of permanent labor migrants. Nevertheless, for temporary migrants a small negative effect is found for job employment rates for natives. Immigrants and not natives filled new job vacancies. This is probably due to lower wages of immigrants (Heyma et al, 2008). However, this effect on the unemployment rate will increase by only 0,07 percent if the amount of temporary labor migrants double. In other words, for every 1400 jobs, one immigrant fulfills only one vacancy.
Berkhout en ‘t Hof (2012) also studied the effect of temporary labor migration. These authors state that there is not an effect on job opportunities. In the sectors where a lot of temporary labor migrants work, the amount of native employers did not change. Overall, it can be concluded that the actual effect on the unemployment rate is rather small or invisible.
Heyma et al (2008) also studied the effect on wages. To see if there is a difference between groups with low and high wages, they performed the same analysis as for employment rates, but with three different groups: the 40 percent group with the lowest wages, 30 percent middle group and the 40 percent group with the highest wages. All in all, the authors found a small negative effect on wages for people with the lowest wages. This decline seems to be caused due to the influx of permanent migrants. In contrast, temporary migrants cause an increase in the lowest wages. Due to this, temporary labor migrants can be considered as complements for the group with the lowest wages. However, the effect on wages is visible by a doubling of temporary migrants. These effects are so small, that this does not lead to a welfare change. Considering the high wages, no effect was found for people with high wages by Heyma et al (2008).
Berkhout en ‘t Hof (2012) state that, despite the overall positive effect of temporary labor migration, a small effect is seen on wages of low-skilled native workers. However, these results are rather small. On the other side, the temporary labor migration gives a positive effect for high skilled native workers. This comes from the fact that low skilled labor is cheaper for employees. In this way, prices are declining, causing more purchasing power. The low skilled natives do not feel this change in purchasing power, due to their declining wages.
Kahanec et al (2009) discussed the effect of the EU expansion for different countries in the EU15. While the UK, Ireland and Sweden opened their doors immediately after the enlargement, the Netherlands kept some restrictions. For the UK we saw that mainly high-skilled migrants entered the UK and therefore the economy strengthened. Germany also had some years of transition, before they opened their borders. Because of this, Germany can be compared to the Netherlands. The difference in migrants between the UK and Germany is that in Germany the migrants are most of the time male, young and less educated than the previous migrants. These findings suggest that these immigrants have to compete with low-skilled immigrants from outside the EU than with low-skilled natives. While Germany needs high skilled migrants, they attracted the low skilled migrants. This same reasoning can be used for the Netherlands; the high-skilled migrants were not attracted to come to the Netherlands, therefore a lot of low-skilled migrants entered the country (Kahanec et al, 2009). It is therefore questionable whether the Dutch government made the right decision by remaining the borders closed after the EU enlargement. In general, Kahanec et al (2009) states that, after reviewing several studies, that there has not been find on average, a non-negative effect of the incoming EU10 migrants for the host countries. Also, the labor market in the host countries does not show any serious disturbances.
Considering these studies, it has to be mentioned that the data used is somewhat outdated. The EU is expanded in 2004, the Dutch borders opened in 2007 for the EU10 countries, but the studies that have been performed are done before this opening. However, the most recent data about labor migrants is not analyzed yet. Ooijevaar et al (2013) gave an insight of these data, such that is known how large the influx is in the Netherlands regarding the EU10 labor migrants.
5. Overall Conclusion
Last years, the Netherlands are dealing with an influx of EU10 labor migrants. The prospect is that more than half of this group is only temporary in the Netherlands. In the last chapter several studies were discussed regarding the effects of labor migration on wages and job opportunities in the Netherlands, from both former migration flows as well as the current migration flow. These studies estimated the effects of the labor migrants on the labor market of the Netherlands. The overall conclusion is that the effect is clustered around zero. In cases showing an effect, an effect was visible for the low-skilled natives competing with labor migrants. However, this effect was negligible and always very small. Just like the results in chapter 3, the outcomes in practice are smaller than expected in the theoretical models. However, in general these conclusions are comparable with the outcomes of these theoretical models. What we have to keep in mind is that those performed studies, are about former migration flows or, in case of Heyma et al’s (2008) study, about the migration flow before the restriction period ended. Also, a research gap was noticed. The effect of former migration flows on the Dutch job opportunities is unknown. It is not sure that conclusions drawn, can fit into the current situation in the Netherlands. Nevertheless, all conclusions found in this thesis are small and cluster around zero.
In this thesis the question whether EU10 labor migration have an effect on the Dutch natives’ wages and job opportunities is discussed by answering multiple sub-questions in order to answer the main research question. Answers on the sub-questions are found by reviewing the theoretical models and empirical studies about the labor migration effects on wages and job opportunities. Despite the theoretical models and their expected outcomes, all these reviewed studies found a smaller effect on wages and job opportunities than eventually thought based on the theoretical models. The low skilled natives have to compete most of the time with the labor migrants. The high skilled natives do not feel any effect. The UK was an exception, this country benefits from the labor migrants.
By reviewing studies performed in the Netherlands, the same conclusion can be drawn. There is a small negative effect for the low-skilled natives. Their wages will decline and the unemployment rate will increase. Nevertheless, this effect is very small. Although not all studies reviewed in this thesis, can be compared with the Dutch labor market, it is remarkable that all results are cluster around zero or insignificant. Keeping this in mind, it can be expected that the effects of the incoming EU10 labor migrants on the wages and job opportunities on the Dutch labor market is rather small. However, the way the Netherlands have dealt with former migration flows give an indication in what the effect could be from the current migration flow. Mixed results were found for different groups of labor migrants. It can be the case that some groups will affect the Dutch labor market more than others. Future study has to research this by estimating effects per EU10 country.
Not in line with these conclusions, politicians and citizens are still wrongfully assuming that labor migrants are the reason that the opportunities for the natives become worse. Indeed, this country is dealing with a high unemployment rate and people do have a hard time economically. A reason
might be that together with the economic crisis the labor migrants have a stronger effect. This suggestion has to be studied to confirm this thought. Also, this thesis highlights only the effects on wages and job opportunities, but an economy can be much more affected on different levels by labor migrants. It could be benefit from labor migrants, because labor migrants spent their earnings in the host country. Or labor migrants can have a negative impact, due to the social securities where labor migrants benefit from. This is a whole other aspect of effects of the influx of labor migrants in a country. Berkhout en ‘t Hof (2012) already stated that temporary labor migration have a positive effect on the Dutch economy. But to support this hypothesis for all EU10 migrants, temporary and permanent, more research has to be done. Overall conclusion of this thesis is that the effects of the EU10 labor immigration on the wages and job opportunities of Dutch natives are limited. This conclusion could benefit the current discussions regarding labor migrants in the Netherlands.
6. Bibliography
Angrist, J.D. & Kugler, A.D. (2003) Protective or counter-productive? Labour market institutions and the effect of immigration on EU natives. The Economic
Journal, Vol. 113, p. 3-2-331
Berkhout, E. & Hof, B. (2012) De economische bijdrage van tijdelijke arbeidsmigranten; een realistisch beeld. SEO-rapport no 2012-88.
Blanchflower, David G.; Saleheen, Jumana; Shadforth, Chris (2007): The impact of the recent migration from Eastern Europe on the UK economy, IZA Discussion Papers, No. 2615
Boeri, T. & Brucker, H. (2005) Migration, co-ordination failures and EU enlargement. IZA discussion Paper 1600.
Bonin, H. (2005). Wage and employment effects of immigration to Germany: evidence from a skill group approach, IZA Discussion papers, No. 1875.
Borjas, G.J. (2003) The labor demand curve is downward sloping: reexamining the impact of immigration on the labor market. The quarterly journal of
economics (118-4) p. 1335-1374.
Borjas, G.J. (2013) Labor Economics. New York: McGraw –Hill.
Borjas, G., Freeman, R.B. and Katz, L.F. (1992) On the labor market effects of immigration and trade. In G. Borjas and R.B. Freeman (eds), Immigration and the Work Force: Economic Consequences for the United States and Source Areas (pp. 213–244). Chicago, IL: National Bureau of Economic Research, University of Chicago Press.
Card, D (1990) The impact of the Mariel Boatlift on the Miami Labor Market.
Industrial and Labor relations Review, Vol 43, No.2, pp. 245-257
Constant, A.F. (2011). Sizing it up: Labor migration lessons of the EU enlargement to 27. Discussion paper series, Forschungsinstitut zur zukunft der Arbeit, No. 6119.
Dustmann, C., Fabbri, F., and Preston, I. (2005) The impact of Immigration in the British Labour Market. The Economic Journal, Vol. 115, No. 507 pp. 324-341.
Hartog, J. & Zorlu, A. (2004) Economische effecten van Immigratie Ontwikkeling van een databestand en eerste analyses, paper 2004-26. Amsterdam: University of Amsterdam.
Hartog, J. & Zorlu, A. (2005) The effect of Immigration in Three European Countries, Journal of Population Economics, Vol. 18, No. 1 pp. 113-151
Heyma, A. Berkhout, E., Werff, S. van der, Hof, B. (2008) De economische impact van arbeidsmigratie uit de MOE-landen, Bulgarije en Roemenie; een studie naar omvang, aardig en economische effecten van arbeidsmigratie. SEO-rapport no. 2008-70.
Kahanec, M., Zaiceva, A. & Zimmermann, K.F. (2009) Lessons from migration after EU enlargement, IZA discussion papers, no. 4230
Okkerse, L. (2008). How to measure labour market effects of immigration: a review. Journal of Economic Surveys, (22-1) p. 1-30.
Ooijevaar, J., Sluiter, N. & Verschuren, S. (2013) Immigranten en werknemers uit de Europese Unie in Nederland. Centraal Bureau voor de Statistiek.
Roodenburg, H. Euwals, R & ter Rele, H. (2003). Immigration and the Dutch economy. Den Haag: CPB/Sdu.
Winter-Ebmer, R. & Zweimuller, J. (2000). Consequences of trade creation and increased immigration for the Austrian labour market. In Okkerse, L. (2008). How to measure labour market effects of immigration: a review. Journal of
