from inside and outside the EU in the context of the German labor shortages.

(1)

Master thesis

University of Groningen

Evaluating the differences in labor market performance between migrants

from inside and outside the EU in the context of the German labor shortages.

Author Supervisors

Tsjerk Jehannes Bottema Dr A.A.Erumban

s2052857 Dr H.J.Drogendijk

International Economics & Business

International Business & Management

(2)

Abstract

This study uses data from the German Socio Economic Panel (GSOEP) to investigate the differences in labor market performance between migrants from inside and outside the EU. The results of this study show migrants from outside the EU earn on average less than migrants from inside the EU. This gap is smaller for migrants that have been in Germany for a longer period of time and migrants with good German writing skills. Having a high education on the other hand increases the gap significantly, which is likely the result of job-skill mismatches.

(3)

List of Figures

2.1 Migrants(foreign-born people) as a percentage of the total population. . . . 6 2.2 Inflow of migrants per country in 2009 and 2013. . . . 7

5.1 Percent change of the natural working age population between 2015-2025. . . . 26

8.1 Employment & participation rates of migrants from inside and outside the EU, and German natives . . . . 47 8.2 Education levels migrants from inside and outside the EU, and

German natives. . . . 48

(7)

List of Tables

1.1 Raw real wage gaps between migrants from inside and outside the

EU . . . . 2

3.1 Overview of the theory on the labour market performance migrants. 11 3.2 The effect of worker characteristics on the wage rate and unemployment chance of workers . . . . 13

4.1 Fiscal contributions and benefits of natives and migrants, 2007- 2009 average. . . . 18

4.2 Overview of the labor market effects of migrants. . . . 20

4.3 Overview of the fiscal effects of migrants. . . . . 20

4.4 Overview of the productivity effects of migrants. . . . 20

4.5 Overview of socio-cultural effects. . . . 22

5.1 Estimated changes in the labor force 2010-2020 in comparison with 2000-2010 (in percentage). . . . 27

5.2 Net immigration stocks of college graduates 2000-2010. . . . 32

7.1 Expectations on the relative effects of characteristics of a migrant from outside the EU in comparison to the effect of characteristics of a migrant from inside the EU . . . . 43

8.1 Gross real wage differences 1984 - 2013 (e) . . . 46

8.2 Gross real wage differences 2013 (e) . . . 47

(8)

8.3 Results of Mincer wage regressions on migrants from outside the EU and German natives . . . . 51 8.4 Results of Mincer wage regressions on migrants from inside the

EU and German natives . . . . 52 8.5 Unemployment and participation regressions on migrants form

inside the EU . . . . 55 8.6 Unemployment and participation regressions on migrants form

outside the EU . . . . 56

(9)

Chapter 1

Introduction

In the last couple of years Europe is dealing with a large stream of incoming migrants (Eurostat, 2015; OECD, 2015). Even though motivations to migrate may differ and can have its origins in political, economic and family circumstances researchers agree that immigrants can have both an economic and socio-cultural impact on the host-country (Borjas, Vreeman & Katz, 2015; Kerr, 2011). The potential positive and negative effects of these impacts are a hot topic for discussion for politicians and citizens alike, as they seem divided on whether to welcome or repulse immigrants. On an international level this has led to a very broad spectrum on the willingness to accept migrants, running from a welcoming and open Germany to an almost closed Hungary (Eurostat, 2015). The intensity of the influx of immigrants and the polarized ideas on how to deal with them has instigated the usage of the term “migrant crisis” by both politicians and the media.

Historically, one of the most prominent discussion points in the migrant debate is the increased labor market competition instigated by immigrants entering the labor force (Bailey, 1987; Borjas, 1987; Scheve & Slaughter, 2001). Nationals typically fear that they will be substituted by a migrant worker or that their wage will decline due to the relatively low wages of migrants (Card & Altonji, 1991; Friedberg & Hunt, 1995). However, the occurrence of significant labor shortages across several European countries has put this discussion in a very different perspective, as attracting skilled migrants seems to be one of the most feasible solutions for this problem. Therefore, instead of being a possible substitute for nationals, migrants seem to become a necessary complement to the native workforce in order to retain economic stability (Kahanec, 2014).

The country facing the most severe labor shortages in the EU is Germany (Ette, Hess & Sauer, 2015; The Conference Board, 2014). Because of an aging population and a low fertility rate its working age population is declining rapidly. Pro- jections show that this could lead to a natural decline of the workforce of almost 9 percent by 2025 (The Conference Board, 2014). In order to counter the negative effects such a decline might have on the economy the German government

(10)

has implemented legislation which is specifically aimed on attracting skilled migrants (Die Bundesregierung, 2016). As a result, the German acceptance rate of migrants, even without counting refugees, has increased significantly over the past few years (Eurostat, 2015).

In the context of these developments the IMF (2015) has executed an extensive study on the labor market performance of immigrants in Germany. The study investigates the effect of specific (migrant) worker characteristics¹on labor market performance, which is represented by the chance someone is unemployed and wage level. The outcomes show that migrants have a higher chance to be unemployed and earn a lower wage than natives, where the size of the difference critically depends on specific migrant characteristics.

However, from a country of origin perspective, the study sees migrants as a homogeneous group, while researchers indicate that the German labor shortages can mainly be addressed by attracting migrants from outside the EU (Bauer &

Kunze, 2004). The raw real wage gaps between migrants from inside and outside the EU, as presented in table 1.1, show that the labor market performance of migrants from outside the EU, in terms of wage, deteriorated relative to migrants from inside the EU. This leads to the belief that there might be discrepancies in the valuation of migrants and their characteristics from both respective groups.

Table 1.1: Raw real wage gaps between migrants from inside and outside the EU

EU Outside EU

1984-1993 e14,53 e14,59 1994-2003 e15,94 e15,70 2004-2013 e16,75 e15,14

Source: GSOEP, 2016

Therefore, this study wants to build upon the study by the IMF by investigating the differences in wage rate and unemployment chances between migrants from inside and outside the EU. The differences are assessed by developing empirical models of the determinants of unemployment and wage and looking at their relative impact on the performance of both migrant groups. By doing so, this study wants to answer the following research question: “Are there differences in the labor market performance of migrants from inside and outside the EU in Germany? And if so, where do they originate?” The data used in the empirical models to answer this question is provided by the German Socio-Economic Panel (GSOEP) and the English language proficiency index.

Investigating the differences in labor market performance of migrants from inside and outside the EU is a relevant extension of the existing literature because it gives insight in the valuation of specific worker characteristics in the German

1Characteristics are for example a migrant’s working experience and level of education received.

(11)

economy. Moreover, from a more practical perspective, the outcomes of this study could present focus points for the German legislation on labor market integration of migrants. As a good labor market performance of migrants that are already present in Germany could be a vital argument in attracting other migrants, improving this performance might turn out to be critical when com- peting for talent with other countries. However, besides ensuring optimal labor market performance, legislation regarding migrants is based upon multiple impacts of immigration. Therefore, these impacts are well accounted for in the discussion on labor market performance and the construction of the legislative focus points.

Section 2 starts with explaining the concepts of migration and the migrant population. Hereafter, the section continues with an overview of migrant numbers in the EU in order to give an indication of the significance of migration. Sec- tion 3 follows with a discussion of the labor market performance of migrants in general while section 4 outlines the effects of migration on the host-country.

Then, section 5 introduces the labor shortages occurring in both the EU and Germany, the policies that have been implemented to counter these shortages and the success of these policies. In section 6 the hypotheses that will be tested in the empirical part are formulated, which are based on the differences between migrants from inside and outside the EU than can be found in the literature.

Section 7 goes over the methodology that is used to investigate the hypotheses and sections 8 presents the results of the regression. The conclusion of this study can be found in section 9 and the subsequent discussion in section 10.

Finally, the references are stated in section 11 and the appendices are in section 12.

(12)

Chapter 2

Overview of migration

This section introduces the concept of migration, explains which people can be identified as a migrant and outlines the different motivations to migrate. This serves as a context for understanding the different concepts that are discussed in the literature review. The second part of this section gives an overview of the number of migrants that are currently living in the European countries and shows recent data on the influx of migrants in order to get familiar with the magnitude of migration in Europe.

2.1 Defining migration and the migrant popula- tion

The United Nations (1990) defines migration as the crossing of a boundary of a political or administrative unit for a certain minimum period of time¹ . In line with this definition the OECD (2015) defines a migrant as a person who moves to another country with the intention to stay for a minimum period of time (i.e. it excludes tourists and business visitors). This definition includes the movement of refugees and displaced persons as well as economic migrants; which makes that the term “migrant” is subject to a lot of different interpretations.

In the public debate it is therefore sometimes used to explicitly discuss asylum seekers and refugees.

Hawkins (2016) proposes three more specific migrant definitions that can each be used and which give a more precise description of the intended migrant population. Firstly, a migrant can be someone whose country of birth is different to their country of residence. Secondly, a migrant can be someone whose nationality is different to their country of residence, and thirdly can be someone who changes their country of usual residence for a period of at least a year, so that

1There is no specification on the length of a certain minimum amount of time

(13)

the country of destination effectively becomes the country of usual residence. In this study a migrant is identified as someone whose country of birth is different to their country of residence. The reason for this choice is that the other two definitions would exclude all people that have migrated in the past but received the nationality of the host-country. This would indicate that the migrant population would mainly exist of people that have been in the-host country for a relatively small amount of time, as it takes a certain period to receive a new nationality (OECD, 2013). In that case the effects of migrants over their en- tire period of stay in the host-country might diminish, which could significantly influence study results as migrants tend to do better in the host-economy over the years (OECD, 2013). Additionally, a further distinction is made between migrants from any other EU country than Germany and migrants from outside the EU. These two groups will be indicated as migrants from inside the EU and migrants from outside the EU, respectively.

2.2 Motivations to migrate

Notwithstanding, as the United Nations (1990) point out, this broad definition reflects the current difficulty in distinguishing between migrants who leave their countries because of political persecution, conflicts, economic problems, envi- ronmental degradation or a combination of these reasons and those who do so in search of conditions of survival or well-being that does not exist in their place of origin. Broadly, migrants can be split in four distinct groups (OECD, 2015).

The first groups is composed by so called long-term migrants within a free mo- bility zone, which could for example be people moving from one EU country to another. The second group is formed by family migrants, which are people that migrate due to family circumstances like relatives living abroad. A third group, which is the humanitarian group, consists of all people that receive some form of protection in their country of residence, which could be a refugee status but also temporary protection or something similar. Refugees, more specifically, are persons who owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, are outside of the country of their nationality and are unable, or owing to such fear, are unwilling to avail themselves of the protection of that country (The UN Refugee Agency, 1951). The last group of migrants are the labor or economic migrants, in which case the decision to migrate is taken freely by the individual concerned for reasons related to ‘personal convenience’ and without intervention of an external compelling factor (United Nations, 1990).

The choice for picking a certain destination is however not solely driven by income disparities between the home- and destination country, as it can also be partly motivated by geographic proximity and historical links such as common language and colonial ties (Katseli, Lucas and Xenogiani, 2006).

Focusing on economic migrants could be of particular interest for this study as they are most likely to be attracted by policies implemented to offset labor shortages. In that way the success of these policies could be best assessed.

However, people’s initial motivations to migrate have not always been registered,

(14)

which makes dividing migrants in groups based on their motivation to migrate very difficult. Therefore this study accounts for all migrants and does it not try to single out economic migrants, refugees or others.

2.3 An overview of migration in Europe

This sub-section provides an overview the of size of the migrant population in the EU on 1 January 2014. The reason to show data up until this date is that later data is unavailable or incomplete. Figure 2.1 shows the percentage of foreign-born people in the total population per country. A distinction is made between the total percentage of foreign-born people and the percentage of people that is a citizen of a non EU country.

Figure 2.1: Migrants(foreign-born people) as a percentage of the total population.

Source: Eurostat, Statistics on migration and the migrant population, 2016.

It can be seen that most Western – European countries accommodate a relatively similar percentage of non-foreign born people in relation to their population size.

Luxembourg is the only real exception as almost 45 percent of its population is foreign born. On a significant distance Belgium, Sweden Austria and Ireland also have a relatively large foreign born population. The picture is very different however for most of the Eastern-European countries except Estonia, Latvia and Slovenia. In the other Eastern-European countries less than 5 percent of the population is foreign-born. A reason for this large difference could be the higher attractiveness of the Western-countries due to their prosperity. Another reason could be that the Eastern-European countries were part of the Soviet- Union, which was not very open to immigrants (Buckley, 1995; Mansoor &

Quillin, 2006). Finally, both the governments and citizens of these countries are currently not very welcoming to immigrants as well (Sides & Citrin, 2007).

They have a mainly negative perception of immigrants and their impacts. It is also noteworthy that non-EU citizens account for the largest percentage of the

(15)

foreign-born population in almost every country except Luxembourg, Cyprus, Ireland, Hungary and Slovakia, Estonia and Latvia. For Luxembourg this can be the result of the relatively large amount of workers from other EU countries that are present (Agu & Fekete-Fargas, 2015), while Hungary, Estonia and Latvia actively keep their migrant numbers from outside the EU low because of their negative perception of the cultural characteristics and backgrounds of foreigners as outlined by ministers from these countries (Avramov, 2009). The reason the other countries all have a high percentage of foreigners that are born outside the EU could be that people from outside the EU are relatively often refugees or dispersed. These groups of migrants are more likely to stay on a long term basis than economic migrants from other EU countries that merely come over to work (Dustmann and Weiss, 2007).

Figure 2.2 shows the yearly inflow of migrants per country in the period 2009- 2013. Not all EU countries are present as there was no sufficient amount of data available for these countries. Furthermore the United States have been added to make a comparison with the European countries. This is relevant because the US is well known for the high economic contribution its immigrants have (Wadhwa, Saxenian & Siciliano, 2012). This will be further discussed later on in this thesis.

Figure 2.2: Inflow of migrants per country in 2009 and 2013.

Source: OECD, 2015.

It can be noted that Germany by far had the largest influx of migrants of all European countries during this period. While this might be expected as Germany is also the largest EU country in terms of population the volume of the difference is disproportional with the differences in population numbers. In 2013 Germany has taken in almost three times as much migrants as the United Kingdom, which is the EU country that ranks second on migrant intake. The inflow increased considerably over the observed years as the number of incoming migrants per year almost doubled when comparing 2013 with 2009. This could

(16)

be due to Germany’s positive political stance towards refugees and its related high acceptance rate of migrants (Die Bundesregierung, 2016).

Other countries that receive a lot of migrants in absolute numbers are Italy, Spain and the UK. Even though the numbers of Italy and Spain decreased quite significantly since 2009. Relative to their size Belgium, The Netherlands and Austria also receive their fair share. These observations are mostly in accordance with the expectations that were raised by figure 1. France and Luxembourg however have a relatively small number of incoming migrants in comparison with its foreign-born population. The Eastern-European countries in general have the smallest inflow of migrants, which is also what was to be expected after investigating figure 1. The United States initially took in more migrants than Germany but over the years the numbers of The United States decreased while those of Germany swelled significantly, making the latter the biggest in taker in 2013. The significance of the German intake is highlighted by the notion that the population of the US is almost 4 times as big as the population of Germany.

(17)

Chapter 3

The labor market

performance of migrants – a review of literature

This study wants to investigate the differences in labor market performance between migrants from inside and outside the EU. Therefore, in order to develop a truthful empirical model of the determinants of unemployment and wage, the existing literature on the labor market performance of migrants is reviewed.

The first sub-section discusses the performance of migrants in general while the second sub-section specifically focuses on Germany, as it is the country of interest in this study. As mentioned in the introduction, this study follows the interpretation of labor market performance by the IMF, where it represents wage level and unemployment chance.

3.1 Labor market performance of migrants

Researchers have investigated the employment rates and wage levels of migrants in a number of countries. However, an evaluation of the quantity of literature shows that the interest in the labor market performance of migrants seems to be much less than the effect the migrants have on the labor market performance of natives. This could possibly be the result of the fear for job losses and wage declines that is expressed in the media and amongst nationals; making studies on that matter assured of attention.

Nevertheless, the general consensus of researchers seems to be that there are large discrepancies in labor market performance between migrant groups, which depend on their country of origin, education and language skills (Kanas & Van Tubergen, 2009; Galor & Stark, 1990). Bratsberg, Raaum and Knut (2014) for

(18)

example show that immigrants from high-income countries were able to perform on similar levels as Norwegian natives in terms of labor market and social insur- ance outcomes. Immigrant from low-income countries on the other hand showed much lower employment rates than natives and had a much higher chance to participate in disability program over their working span. In a later study (2016) they add that the chance of job loss is twice as big for migrants from low-income countries than for natives and migrants from high-income countries.

Mattoo, Neagu and Ozden (2006) find that there are large differences among highly educated migrants from different countries. Migrants from Latin Amer- ican and Eastern European countries are more likely to become occupied in a low skilled job than immigrants from Asia and industrialized countries. They state that the main reason for this phenomenon can be found in the attributes of the country of origin that have an effect on the quality of human capital, like educational expenditure* and the use of English as language of instruction.

Not surprisingly, the Asian and industrialized countries score high on the PISA performance index of students*, which measures students capabilities on math- ematics, science and reading. This is in contrast to the Latin American and Eastern European countries, as they can be found in the lower segment of the spectrum. This indicates that the quality of education in the home-country is a strong predictor of a migrants labor market performance in the host-country.

Clark and Drinkwater (2008) on the other hand state that the relatively low employment rates and wage levels of migrants they find do often not result from a low education as migrants in the UK are often relatively well educated in comparison to natives. They argue that the low employment rates and wage levels are mainly the result of a lack of host-country-specific skills that are needed to perform on the level that is required. Because these skills can only be gained over time and through significant effort migrants have a serious disadvantage on the host-country labor market. Another explanation for the relatively poor labor market performance of educated migrants could be bad job-skill match- ing. Chiswick & Miller (2008) show that migrants are often overeducated for the jobs they fulfill. Being employed on the level a person is educated for could therefore significantly raise the performance.

Similar to Mattoo et al.(2006), Dustmann and Fabbri (2003) find that English language proficiency is a factor that increases employment chances and wage levels in the UK. This is not only the case in English speaking countries as Yao and Van Ours (2015) show the importance of knowledge of the Dutch language in the Netherlands while Kaida (2013) outlines similar results for French in Canada. In a sense speaking and understanding the host-country language can be identified as a host-country-specific skill that Clark et al., (2008) described.

Perhaps not surprising in the context of the discussion on wage discrimination of women in a lot of countries (Biddle & Hamermesh, 2013; Lips, 2013), female immigrants have worse labor market prospects than men (Sarvimaki, 2011; Yao et al., 2015). The unemployment rate amongst them is higher and they receive a lower wage.

The final factor that seems very important for the labor market performance of migrants is a particular country’s legislation on their acceptance, as the con-

(19)

ditions of entry vary widely across countries (Buchel & Frick, 2003; Chiswick, 1997; Hawthorne, 2013; Sarvimaki, 2011). Having policies that are strict for certain groups of migrants but aim to attract other groups of migrants results in a more preselected migrant population in which people likely possess one or more of the above mentioned characteristics. This on its turn leads to a better overall labor market performance and a faster assimilation to natives in terms of wage level. For this reason, migrants in countries like the US, Canada and Australia do very well, as these countries have very distinctive acceptance pro- tocols (Chiswick, 1997; Hawthorne, 2013). As a result there are large differences between countries in terms of the time it takes for income differences between natives and immigrants to disappear. In Finland this takes for example an approximate 20 years while it is 10 to 15 years in the US (Sarvimaki, 2011). The differences in legislation will be further discussed in section 4.

In summary, the education and language proficiencies of migrants are a large predictor of their chances of employment and wage rate. Therefore, migrants from high-income countries generally fair better on the host-country labor market. However, learning host-country-specific skills and having a job-skill match seem important as well. Finally, the labor market performance of migrants is heavily influenced by a country’s conditions of entry, as these predetermine the composition of the migrant population. An overview of the theory on the labor market performance of migrants can be found in table 3.1.

Table 3.1: Overview of the theory on the labour market performance migrants.

Finding Authors

The labor market performance of migrants is heavily influenced by their country of origin, education and language skills.

Kanas & van Tubergen, 2009;

Galor & Stark, 1990 Immigrants from high income countries perform better on

the host-country labour market than migrants from low income countries.

Bratsberg, Raaum & Knut, 2014;

Mattoo, Neagu & Ozden, 2006 Level of education is an important predictor of a migrant’s

labor market performance in the host country; the better one’s education, the better the labour market prospects.

Mattoo et al., 2006 A poor labour market performance in the host-country is

not the result of a low education but a lack of host-country specific skills.

Clark & Drinkwater, 2008

Knowledge of the language spoken in the host-country leads the higher employment chances and wage levels.

Mattoo et al., 2006; Dustmann

& Fabbri, 2003; Yao & van Ours, 2015; Kaida, 2013

Female migrants have worse labour market prospects than

men Sarvimaki, 2011; Yao et al., 2015

The labor market performance of migrants is heavily influenced by legislation in the host-country.

Buchel & Frick, 2003; Chiswick, 1997; Hawthorne, 2013; Sarvi- make, 2011

3.2 The labor market performance of migrants in Germany

As Germany is the country of interest in this study the next section specifically discusses the labor market performance of migrants in Germany. While there are multiple studies on this subject the focus of this section will be on a recent study

(20)

by Beyer, who, on behalf of the IMF, investigated the labor market performance of migrants in Germany in the period 1984-2013, as this study wants to expand on the outcomes of that study. The motivation to deepen the outcomes of the IMF study by investigating the differences in labor market performance between migrants from inside- and outside the EU originates in the European labor shortages, which will be explained in section 5. In order to expand on the outcomes of the IMF study the same definition of labor market performance as well as a similar methodology is used.

The IMF paper starts with Mincer type equations¹, which investigate how wage rate is influenced by certain (migrant) worker characteristics. Its most noteworthy finding seems to be that employed migrants earn approximately 20 percent less after their arrival than employed natives with similar characteristics. In the first years after their immigration the migrant’s wage catches-up with one percentage point per year. However, this paste slows down over time and the wages never fully converge. The importance of the duration of stay is stressed by a variable which shows that migrants who came to Germany after 2007 received 17 percent less wage than migrant who migrated earlier.

One of the migrant characteristics that has a significant influence on the wage gap is a migrant’s ability to write German, as this can decrease the gap by almost 7 percent in case of good German writing skills. The importance of education in general becomes apparent as well, as having a high education*

instead of a medium education* can increase the wage gap by 30 percent, while a medium education increases the wage gap by 10 percent compared to a low education.* However, surprisingly, a migrant with a medium or high education receives less wage than a low skilled migrant. This seems to be the result of a job-skill mismatch as migrants with a high education doing a high skill job do receive an approximate 20 percent more wage.

Working experience is measured by the number of years someone is employed and it increases wage with almost two percent per year. This speed of this increase does however decrease over the years. Being trained for a job seems to be more important as well, as it increases wage with almost 9 percent. Next to that females receive 12 percent less wage than males, which is relatively low in relation to the general European wage gap between men and women of more than 20 percent (Eurostat, 2014). Furthermore, as one might expect, having a job with a high level of autonomy decreases the wage gap between migrants and natives with more than 10 percent. Finally, being born in an advanced country is a large advantage, as it increases wage by more than 10 percent as well.

The paper continues with investigating the unemployment and participation probabilities of migrants and the characteristics that influence these probabilities. The regressions investigate mostly the effects of the same characteristics as in the Mincer equation.

Every year of experience a person has lowers his or her chance of being unem-

1A framework that measures the labor market rewards related to productive attributes of workers (Mincer, 1974).

(21)

ployment with 0.1 percent and increases his or her participation rate with 1.4 percent. Being a migrant has a strong negative effect as migrants are 6.8 percent more likely to be unemployed and 15 percent less likely to participate. This is different for migrants from advanced countries as they are 9 percent more likely to be employed and more than 5 more likely to participate. The number of years a person is in Germany does have a significant but marginal effect of 2 percent points.

In general, having a good education has a positive effect on the chance to be employed, as a medium education decreases the chance of unemployment with almost 2 percent and increases participation with almost 4 percent while a high education decreases the chance of unemployment with 7 percent and increases participation with 9 percent. This is different for migrants as, in line with the outcomes of the wage regressions, medium educated migrants have a 1 percent higher chance to be unemployed while high educated migrants have a 5 percent higher chance.

Surprisingly, being female decreases the unemployment rate with 1 percent.

However this can perhaps be explained by the female participation rate, which is almost 10 percent lower. Finally, German writing skills decrease the unemployment rate around 2 percent while it does not have a significant effect on the participation rate.

In summary, most of the migrant characteristics have the effect that was to be expected after evaluating the outcomes of section 3.1. Training, German language skills and (working) experience have a positive effect on wage and employment chances while being a migrant or a female has a negative effect.

However, the effects of education are counter intuitive, which might be the result of job-skill mismatches. An overview of the effects can be found in table 3.2 Table 3.2: The effect of worker characteristics on the wage rate and unemployment chance of workers

Wage rate Unemployed chance

Experience + -

Medium education + -

Migrant with medium education - +

High education + -

Migrant with high education - +

High-skill job match +

Immigrant - +

Female - +

Trained for job +

Autonomy +

Years in Germany + -

Good German writing + -

Born in advanced country + -

Migrated after 2007 -

Migration can have multiple effects on the host-country. These effects are important to determine when evaluating the labor market performance of migrants as they put the performance into context by showing possible positive and negative spill-over effects of migration. As the effects can have a significant impact on the legislation related to migrants they are discussed in the next section.

(22)

Chapter 4

Host-country effects of migration - a review of literature

It seems logical that the influx of people from all around the world can have an impact on the host-country, as these people participate in both the economy and the society. Therefore, researchers have done extensive research on the host-country effects of migration. Because these host-country effects seem in- tertwined with the labor market performance of migrants and have a significant influence on the legislation on migrants they are discussed in this section. The effects are divided in economic or socio-cultural. Besides the effects on the host- country there are also multiple home-country effect of migration. These effects are however beyond the scope of this paper and are therefore not discussed. A short overview table of home-country effects can be found in appendix 1.

4.1 Economic effects

From an economist’s point of view migration seems a great instrument for the achievement of income and productivity gains as Clemens, Montenegro and Pritchett (2009) indicate that, in the ideal situation, moving workers from developing to developed countries could increase the value they add per hour up to 5 times more. Moreover, The World Bank (2006) points out that an increase in international migration of a few percent of the workforce of the developed countries could lead to hundreds of billions in gains. However, beyond the economist’s ideal picture lays the actual situation. Therefore, researchers have extensively investigated the effects migrants have on the host-country economy, which can broadly be split up in labor market, fiscal and productivity effects.

Each of these effects will be discussed in the following sub-sections.

(23)

4.1.1 Labor market effects

In case large groups of migrants enter a country this is likely to have an effect on both the supply and demand side of the labor market as one or more members per migrant household are anticipated to join the workforce and the economy has to provide for an increased number of people, respectively. As a result, researchers have given the labor market effects of immigrants a significant amount of attention.

Meta-studies on the subject are in relative consensus that immigration, in general, has no large negative effect on the host-country labor market position and wages of its natives (Borjas, 2003; Carrasco, Jimeno & Ortega, 2008; OECD, 2003; Longhi, Nijkamp & Poot, 2006). After analyzing 9 different studies Longhi et al. (2006) find that a 1% point increase in the share of immigrants in the host country’s labor market in general decreases the wages of native born citizens by 0.029

Nevertheless, even though the overall effects of immigrants on the employment rate and wage level of natives are rather low the effect might be more intense across different skill levels and occupations. Borjas (1999) states that the impact of immigrant workers does critically depend on the skills and attributes that the immigrants possess in relationship to the skills and attributes the current workers possess and the skills and attributes needed in the economy. It has to be noted that these needed skills are directly influenced by the changes in supply as firms can change the composition of output and production methods following an immigration induced labor supply shock (Longhi et al., 2006).

In case an immigrant possesses a similar skill set as a native worker they can be identified as substitutes (Borjas, 1999; Borjas, 1987; Grossman, 1982). This means that the native can be easily replaced by the immigrant in his or her occupation. Similar skill sets are the result of comparable educational patterns and job experiences (Borjas, 2003). Consequently follows that the more complex an occupation is the more difficult it is to find a suitable substitute. Friedberg and Hunt (1996) find that immigrants will lower the price of factors for which they are perfect substitutes, have an ambiguous effect on the price of factors with which they are imperfect substitutes and raise the price of actors with which they are complements. So in case migrants provide an abundance of unskilled labor firms will substitute away from capital and skilled labor to the cheaper unskilled labor. However, because the output of unskilled labor is now higher employers are stimulated to utilize more of all inputs, raising the prices and employment opportunities. They do however also mention that the same would be true in case there would be a large inflow of skilled labor. In that situation employers will initially substitute away from unskilled labor and capital but will later need more from these inputs to complement the skilled labor as well.

Some researchers are opposed to the idea that natives and immigrants can be seen as perfect substitutes due to small discrepancies in wages between natives and immigrants with the same education and experience (Cortes, 2008; Gross- man, 1982; Ottaviano & Peri, 2011). However, these results are solely based

(24)

on a discrepancy in wage level, which can be the result of other reasons than the less effective fulfillment of tasks (Orrenius & Zavodny, 2007). Moreover, the general consensus seems to be that especially low-skilled natives can be substituted with low-skilled immigrants (Borjas, 2008; Altonji & Card, 2005). Jobs that mainly require manual and physical labor skills do not exclusively have to be executed by natives (Peri & Sparber, 2008). This is different for high-skilled jobs, which often require both quite extensive knowledge of the common language and country specific skills. (Peri & Sparber, 2008; Clark and Drinkwater, 2008).

Research shows that immigration indeed has a small negative effect on the employment rate and wages of low-skilled natives (Card, 2007). On the other hand immigration complements high-skilled natives as they have cheaper and additional human capital at their disposal (Borjas, 1999). Even though the negative effect for low-skilled natives can partly be explained by their uncomplicated substitutability the effect does still critically depend on the skill level the incoming migrants possess. Longhi et al. (2006) address this in their meta-study on the effects of immigration on the host-country labor market circumstances.

They show that immigrants historically had a much lower level of education than the average native population of the EU countries and the US. As a result the immigrants could only do low-skilled work and so they would substitute low-skilled natives, which explains the negative effects low-skilled natives experiences. Moreover, Docquier, Ozden & Peri (2010) outline that the effects for low skilled-natives become more positive when the share of high-skilled immigrants amongst the total amount of immigrants increases. This because high-skilled workers often create additional employment opportunities for low-skilled workers.

The conclusion of this sub-section is that the overall employment rate and wage level does not seem to be very negatively affected by incoming immigrants.

However, the effects do differ across low- and high skilled natives as the former group generally experiences more negative effects. Moreover, the precise effects depend on the skills and attributes that are needed in the host-country and the skills and attributes that the migrants possess.

In the next section the fiscal effects of migrants will be discussed. In terms of direct financial benefits and losses this seems to be the most straightforward predictor of the economic impact of migrants.

4.1.2 fiscal effects

Similar to the labor market effects, the fiscal effects of immigrants have received a significant amount of attention from researchers as well. The fiscal effects show whether immigrants have a positive or negative effect on a country’s GDP. Be- sides the general interest in the immigrants’ influence on GDP the large amount of research that has been conducted can be explained by small or large discrepancies in the methodologies that are used (Boeri, 2010; Cully, 2012; Ekberg, 2006). Therefore, the fiscal effects cannot be described by one undisputable fig-

(25)

ure per country, as its measurement depends on key assumptions that are made in the methodology.

Boeri (2010) shows that immigrants, in general, do not receive more social- benefits than nationals. He does however outline that this does depend on the immigrants motivation to migrate. In case a large part of the immigrant population came for humanitarian reasons instead of economic reasons it is highly likely they will take up more social benefits. Ekberg (2006) makes a similar point as he states that the composition of the migrant population is to be considered when evaluating social-benefits. Besides the immigrant’s motivations he also mentions the family size as a factor that could influence the fiscal effects, a larger family tends to receive more benefits.

A study of the OECD (2013) finds that, under normal circumstances*, possible discrepancies between the net-fiscal positions of nationals and immigrants do not originate in the latter group receiving more benefits. They do however tend to pay less taxes and social security contributions. This seems to be related to the relative high unemployment rate amongst immigrants, which in 2013 was 1.5 times higher for immigrants as for natives of OECD countries, making employment the single biggest determinant of immigrants’ net fiscal position (OECD, 2013). As a result, immigrants are less well able to contribute to the economy while they do have to rely on social benefits in order to come around. Therefore, the unemployment rate of the immigrants population as well as the reasons for unemployment seem to be important to consider when making statements about the net fiscal position of immigrants. Related to employment is the skill level of immigrants. High-skilled immigrants generally make a higher fiscal contribution than low-skilled immigrants (Rowthorn, 2011).

They pay more taxes and social security contributions and they generally receive less benefits. Therefore, the percentage of the migrant population that is high- skilled can also be relevant when evaluating the fiscal effect of the migrant population.

The general outcome of studies on the fiscal effects of immigration is that it is rather small (OECD, 2013). The impact, whether it is positive or negative, not often exceeds 0.5% of GDP per year. The impacts per country can be seen in table 4.1, which shows the contributions that are paid, the benefits that are received and the net contribution by both natives and migrants. As mentioned, migrants indeed tend to contribute less than natives with Luxembourg as the only exception. However, in most countries the immigrants also receive significantly less benefits than natives, which limits the difference in net contribution.

In Germany and Poland migrants nevertheless do have a significantly lower net contribution than natives. This is especially the case in Germany as migrants contribute more thane11.000 less.

Rowthorn (2008) confirms the view of the OECD and states that the fiscal effects of immigrants are not a very strong argument for or against large-scale immigration, the desirability of the phenomenon should perhaps be decided on other grounds. Therefore, the next sub-section investigates the effects of immigrants on labor productivity.

(26)

Table 4.1: Fiscal contributions and benefits of natives and migrants, 2007-2009 average.

Net Contribution(e) Contribution(e) Benefits(e)

Migrant Native Migrant Native Migrant Native

Luxembourg 9.178 -1.228 23.732 20.043 11.285 21.270

Italy 9.148 3.980 12.310 15.346 3.162 11.366

United States 8.274 8.534 13.145 15.527 4.871 6.993

Greece 7.728 5.008 9.476 13.246 1.748 8.238

Spain 7.496 3.106 10.057 10.518 2.561 7.412

Belgium 5.560 9.159 13.707 18.856 8.147 9.697

Portugal 4.479 950 8.320 8.024 3.841 7.074

United Kingdom 3.029 2.604 10.803 11.503 7.774 8.899

Slovenia 3.006 4.450 10.491 13.316 7.485 8.866

Netherlands 2.544 9.940 12.415 21.175 9.871 11.236

Denmark 2.368 7.362 11.041 17.574 8.673 10.211

Austria 2.353 3.375 12.334 16.705 9.980 13.330

Hungary 1.864 1.081 6.643 6.531 4.779 5.450

Finland 1.314 5.706 8.942 15.188 7.628 9.482

Sweden 896 6.815 11.005 17.041 10.109 10.226

Estonia -2 4.514 3.990 7.528 3.992 3.014

Czech Republic -184 3.474 4.914 8.465 5.100 4.990

Ireland -1.274 -2.487 7.309 9.527 8.583 12.014

France -1.451 2.407 9.961 13.359 11.412 10.952

Slovak Republic -2.171 2.148 2.439 6.151 4.610 4.003

Germany -5.633 5.875 8.094 15.373 13.727 9.498

Poland -5.691 291 2.319 5.470 8.009 5.178

Source: OECD, 2013

4.1.3 Productivity effects

The effects of immigrants on productivity seems to have received less attention than other economic impacts. Nevertheless, researchers that did investigate the subject came to noteworthy conclusions, which will be discussed in this section.

Kangasniemi, Mas, Robinson & Serrano (2012) outline that migrants can influence productivity levels through multiple channels. The most straightforward channel is the individual skill but migrants can also spread technological developments and contribute to innovation. Thereby, immigrants can have an impact on the relative price of inputs, which can influence the choice of production tech- nology (Kangasniemi et al., 2012). An increase in low skilled labor supply can for example shift the output mix towards production of goods that are more unskilled labor intensive (Dustman et al., 2008; Gonzalez and Ortega, 2008).

Quispe-Agnoli and Zavodny (2002) find that states in the US with high numbers of migrants experience lower productivity increases than states with a relatively low number of migrants. They find these outcomes in both low-and high skilled sectors. Similar results are found by Gonzalez and Ortega (2011), who conducted research in Spain. Kangasniemi et al. (2012) test the effects of migrants on productivity in both the UK and Spain. They conclude that both countries experienced a positive impact on output growth in aggregate. However, the labor productivity growth in both countries is negative, although very small in the UK, because the effect on capital deepening¹ is negative and larger in absolute

1The capital per worker is increasing in the economy

(27)

value than the quality effect. There are however large discrepancies between industries. The growth accounting methodology that they use does nevertheless also show that migrants have a positive effect on the total factor productivity² on the long term. They point at complementarity between migrant workers and other inputs as potential cause of this effect.

Huber, Landesman, Robinson and Stehrer (2010) find that there is little ev- idence to suggest that migrants raise overall productivity levels. They mention over-qualification as the possible reason. However, they also show that highly-skilled migrants have a robust positive influence on labor productivity in industries that can be classified as “skill intensive”. Thereby, they state that migrants from non-EU countries do seem to have a positive effect. They point at the selection process for migrants from outside the EU as the cause of this effect. Moreover, migrants from other EU countries are more often used in low- skilled jobs. Nahm and Tani (2014) find similar results as their outcomes show that skilled migrants contribute to a country’s productive efficiency in virtually all sectors except finance.

All in all researchers find different results. High-skilled immigrants do however seem to have a more positive impact than low skilled immigrants. Furthermore, the industry of employment seems to have an influence on the labor productivity contribution.

4.1.4 Evaluating the economic impact

Nationals seem to have a somewhat contrasting view when it comes to the economic performance of migrants. On the one hand they want migrants to contribute to the economy by working and paying taxes while on the other hand they don’t want migrants to compete for scarce jobs or substitute them in their current occupation. The two seem to be mutually exclusive as the more successful immigrants are on the labor market, the higher their net economic and fiscal contribution to the host-economy will be (Algan et al., 2009). As a result, it seems nationals might always have diversified opinions on the economic effects of migrants.

Nevertheless, the literature also shows that the chance of substitution critically depends on the skills and knowledge that are needed in the host-country and the skills and knowledge the incoming migrants possess. Therefore, it is important to evaluate both the developments on a country’s labor market and the government legislation on the acceptance of migrants. This is why Germany has become this study’s country of interest as it is faces significant labor shortages and has plans for migrants to counter them. The German labor shortages, their origins and the role that migrants can play in lowering them will be discussed in the next sections. An overview of all economic effects can be found in tables 4.2, 4.3 and 4.4.

2All growth that is not accounted for by labor and capital

from inside and outside the EU in the context of the German labor shortages.

Master thesis

Evaluating the differences in labor market performance between migrants

from inside and outside the EU in the context of the German labor shortages.

Abstract

Contents

List of Figures

List of Tables

Chapter 1

Introduction

Chapter 2

Overview of migration

2.1 Defining migration and the migrant popula- tion

2.2 Motivations to migrate

2.3 An overview of migration in Europe

Chapter 3

The labor market

performance of migrants – a review of literature

3.1 Labor market performance of migrants

3.2 The labor market performance of migrants in Germany

Chapter 4

Host-country effects of migration - a review of literature

4.1 Economic effects