The Influence of Immigration Policymaking on Income Inequality: A Comparative Study

(1)

The Influence of Immigration

Policymaking on Income

Inequality: A Comparative Study

Master Thesis

Author: Lorenzo Di Francescantonio

Student number: s1964844

Supervisor: Dr. Alexandre Afonso

Second reader: Dr. Natascha van der Zwan

MSc Public Administration

Economics and Governance Track

Faculty of Governance and Global Affairs

Leiden University

11.06.2018

(2)

1

Abstract

This study aims to investigate how income inequality is affected by immigration policymaking. Governments may employ specific immigration policies to address the macroeconomic situation of their country. The question is: with rising income inequality and migration flows worldwide, due to globalization, can immigration policy be used to narrow the income inequality gap? And in what conditions? This investigation aims to shed light on this issue by attempting to answer the following research question: “how does the degree of openness of immigration policies in European Member States affect income inequality in Europe?”. Previous work mainly addresses the relationship between immigration (as a phenomenon) and income inequality or poverty. Previous work, however, has not yet broadly addressed how specific immigration policy choices may affect income inequality. This study approaches the research question above stated by employing a mixed method (quantitative and qualitative) comparative study. The relationship will therefore be analysed based on qualitative and quantitative data gathered in the three chosen samples: the Liberal welfare state sample (Ireland, UK), the Conservative welfare state sample (Italy, Spain, Portugal, Germany, Austria) and the Social-Democratic welfare state sample (Norway, Sweden, Denmark, Finland). The analysis of such data has shown that the relationship between immigration policy openness and income inequality tends to be negative in the Conservative and Liberal samples and positive in the Social-Democratic sample. The implications of these results suggest the following conclusions: higher social and welfare spending nations (especially towards non-contributory welfare plans aimed at poverty alleviation and education) are able to maintain a constant, if not decreasing, income inequality gap despite their immigration policies being increasingly open. The opposite is observed in countries whereby the focus of social welfare spending is laid on contributory programs.

(3)

2

Introduction

This study aims to investigate how immigration policy openness affects income inequality in Europe. This topic has been chosen for a number of reasons which render it worthy of investigation. To understand the interest behind the investigation of this topic, it is necessary to understand the societal importance of the topics involved. To begin with, income and wealth inequality are witnessing an unprecedented surge, ranking increasing higher as the most aggravating social and economic issues of our time (Berman, Aste, 2016, p. 1029). The European Commission, the United Nations, the Organization for Economic Cooperation and Development, along with many other major international organizations, have been focusing their efforts partly towards the deterrence of these contemporary issues.

A number of economists and political scientists argue that states could use immigration policy as a tool to address the problem of income inequality. On broad lines, they have theorized several possible solutions, which could potentially be successful both on the international scale as well as within countries. One of the main theories in favour of using immigration policy to address income inequality is that rich states should decrease their immigration restrictions and open their labour market to foreign workers, allowing them to remit a portion of their higher income back home their families (Oberman, 2015, p. 239). Remittances are a significant source of income for developing and underdeveloped countries. This effect could potentially increase at an exponential scale would immigration restrictions be lifted. Some economists and political scientists, however, disagree with this point of view. They might argue that migration policy may not only be unhelpful in tackling income inequality: it may backfire. This is due to the fact that, in theory, migration has the tendency to benefit already richer families which have the ability and opportunity to have a member working abroad. In this situation, the rich would get richer, and the poor would get poorer, further increasing the income inequality gap (Oberman, 2015, p. 239).

Arguments in favour of restricting immigration policy openness to tackle income inequality exist, too. In the past decades, society is witnessing the rise of the phenomenon of brain-drain, which entails the migration of highly educated individuals away from their home country to seek better opportunities. In this case, for instance, rich states could be encouraged to apply more restrictive immigration policies, to help underdeveloped or developing countries hold on to their most promising young talents and high potential individuals (Oberman, 2015, p. 239). On broad lines, there are two main ways in which immigration policy could be used to address income inequality: increasing restrictions in cases when immigration is causing inequality and decreasing restrictions in cases when immigration is fostering equality. In this light, there are normative issues which feed into this picture: should countries become hostile towards their prospective immigrants to safeguard the economic equality of their nationals? Or is there a possibility to reconciliate immigration and, thereby, the

(5)

4 expansion of the labour market whilst improving the equality of wages? Which kind of policy would be more effective in countering inequality: open or restrictive?

This study attempts to empirically address these contemporary queries by seeking to answer the following research question: “how does the degree of openness of immigration policies in European Member States affect income inequality in Europe?”. This question is worth investigating not only due to the fact that it covers issues of contemporary relevance. In addition to that, the particular design that this study employs to tackle the research question is unique thus far. This study, therefore, aims to target an open gap in the migration and inequality literature (which will be discussed briefly in the empirical debate section). In fact, this study employs a mix of qualitative and quantitative analysis to establish the relationship between “openness to immigration” (independent variable) and “income inequality” (dependent variable). The results in the chosen samples will be compared and contrasted, rendering this a mixed method, comparative study. Namely, the samples chosen for this investigation are the three main European welfare state clusters: the Liberal welfare states, the Conservative welfare states and the Social-Democratic welfare states (see Esping-Andersen, 2013). Methodology, sampling techniques and shortcomings thereof will be discussed in depth in the appropriate section.

This study employs a theoretical framework which leads to two main hypotheses on the relationship between the dependent and the independent variable: the first hypothesis expects increasingly open immigration policies to lead to increasing income inequality, based on the theory that immigration leads to a situation whereby the rich get richer and the poor get poorer. The second hypothesis expects increasingly open immigration policies to lead to decreasing income inequality, based on the theory that wage enhancing conditions may result as an effect of immigration, such as increasing education, increased political participation and increasingly progressive taxation regimes. This study ultimately finds that a relationship between immigration policy openness and income inequality exists: it may be both positive (supporting the second hypothesis) or negative (in support of the first hypothesis) alike, depending mostly on the kind of welfare state in which this relationship occurs. The evidence suggests that higher social-spending nations (especially towards non-contributory welfare plans aimed at poverty alleviation and education) are able to maintain a constant, if not decreasing, income inequality gap despite their immigration policies being increasingly open. The opposite is observed in countries whereby the focus of social welfare spending is laid on contributory programs

Finally, this investigation is structured in the following way: the first section, “Theoretical Framework and Methodology”, will define the main concepts, empirical theories and hypotheses related to the relationship between immigration policies and income inequality. It will additionally provide a brief empirical discussion, and discuss the methodology adopted throughout the study in depth. The second section, “Case Study Background and Contextual Information”, will provide a background

(6)

5 overview of the chosen samples, in order to establish the contextual knowledge needed to undertake the analysis. This section will furthermore lay out the quantitative and qualitative data per sample. The third section, “Linking Theory and Reality – Mixed Comparative Analysis of Case Studies and Population”, will encompass the empirical analysis, by combining the qualitative and quantitative data discussed in the previous section with the theoretical framework. The analysis will aim to provide the results necessary to answer the research question, by establishing the acceptance of rejection of hypotheses and theoretical expectations laid out in the theoretical framework. This will be done both at the sample and population level. Finally, the conclusion will recapitulate the main empirical findings and provide a final answer to the research question.

(7)

6

Theoretical Framework and Methodology

The following section will curate the theoretical framework which will be used to assess, correlate and discuss the data on immigration policy openness and income inequality in the chosen samples. Theoretical expectations and hypotheses on the relationship between the two mentioned variables will be laid out, and subsequently tested using the proposed methodology. However, before discussing the framework, it is necessary to provide the definitions of the key variables of this investigation. This will achieve the establishment of common analytical grounds and avoid ambiguity when discussing the variables. In addition, this shall ease the process of operationalizing the variables.

Definitions and Concepts

To understand the independent variable, being the openness of immigration policies in the chosen sample, one first has to understand what is intended by “immigration policy”. The latter concept, for much time, has gone undefined and lacked rigor. For this reason, Bjerre et al (2015) have filled the gap by providing a state of the art definition. Immigration policy is defined as “government’s statements of what it intends to do or not do (including laws, regulations, decisions or orders) in regards to the selection, admission, settlement and deportation of foreign citizens residing in the country” (Bjerre et al, 2015, p. 559). This definition, according to Bjerre et al, is the product of four key considerations in the immigration policy sphere. The first regards which groups are conceptualized as being the targets of a government’s migration policy. The second considers whether a measure of immigration policy should take into account policy output, implementation and outcome. The third attempts to distinguish immigration policy from the similar field of integration and citizenship policies. Finally, the fourth dwells into whether laws which regulate immigration can be grouped and classified (Bjerre et al, 2015, p. 559). Each of these four consideration provides room to manoeuvre one’s decision into establishing exactly how to analyse immigration policies and what to compare amongst them. This is of key relevance to this study, seeing as it is a comparative study of the relationship between immigration policy openness and income inequality.

As per the first consideration, the targets of a government’s migration policy are usually classified in four groups: labour migrants, asylum seekers (including recognized refugees), immigrants for reasons of family re-unification. In addition, one may consider a fifth group: illegal immigrants (Bjerre et al, 2015, p. 559-560). They may not fall under any of the previously mentioned categories, but the fact that they have crossed the borders of a country makes them subject to immigration policy provisions. However, considering that this study is focused on economic inequality as a dependent variable, illegal immigrants will not be considered in this study’s definition of immigration policies. This is because, with the focus being on economic inequality, considering illegal immigrants without a

(8)

7 work permit may skew the results towards openness to immigration leading to more economic inequality.

As per the second consideration, as to whether a measure of immigration policy should take into account policy output, implementation and outcome, this study follows the ideal that it is important to be able to isolate the effects of outputs as opposed to outcomes (Bjerre et al, 2015, p. 561). The output of a migration policy considers the provisions linked to the policy per-se, whilst the outcome is the rate at which the policy affects real life migration indicators. This study aims to assess the effects of policy output, rather than outcome, therefore shifting the focus on laws.

As per the third consideration, a clear classification in political terms can be made between immigration control and immigrant integration. These two areas follow very different political frameworks. Immigration policies concern entry to the country of destination, whilst integration and citizenship policies govern the process of integration and gaining citizenship (Bjerre et al, 2015, p. 561-562). The question arises as to whether integration and citizenship policies should be fully regarded as part of a country’s immigration policy. Scholars, such as Meyers (2000), Brochmann (1999) and Andreas (2003) debate that immigration policies should be understood as the “rules and procedures governing the selection, admission and deportation of foreign citizens” (Bjerre et al, 2015, p. 562). This study follows such framework, and will predominantly focus on provisions regarding the openness to admission of foreign citizens. Integration policies, on the other hand, as they include political, social and cultural aspects, will not be prioritized when considering immigration policies.

The final consideration focuses on the laws which regulate immigration, and how they can be classified. Bjerre et al (2015) explain that the immigration policy of a country is based on a vast number of different laws (p. 563). The main division can be made between regulations, which are binding legal provisions creating or constraining rights, and control mechanisms, which monitor the compliance with the laws and regulate enforcement. These two aspects outline how the laws function (Bjerre et al, 2015, p. 563). This is considered in this study, as both aspects are needed to assess the openness of an immigration policy: lenient regulation and enforcement will generally entail a higher degree of openness, and vice-versa for immigration policies built upon stricter regulation and subject to severe enforcement. This aspect will therefore be critical to assess the openness aspect of immigration policies, which is the independent variable of this study.

Having established the base for a clear understanding of this study’s use of the term immigration policy, and how their openness shall be assessed, I now turn to define the dependent variable of this study, income inequality. A European Parliament’s briefing on economic inequality defined income inequality as “the extent to which income is distributed in an uneven manner among a population” (European Parliament, 2016, p. 2). Income, in turn, is defined as “individual or household disposable income in a particular year, […] any revenue stream coming from wages, interest on savings, dividends

(9)

8 but also public cash transfers like pensions, after taxes and social security contributions” (European Parliament, 2015, p. 2). The OECD (2011) similarly defines inequality as “an indicator of how material resources are distributed across society”. This study adopts such a straight forward definition of income equality, as formulated by the two aforementioned sources.

Theoretical Framework and Hypotheses

The Rich get Richer, the Poor get Poorer: Immigration and Increasing Income

Inequality

The first theoretical causal relationship which is identified between the dependent and the independent variable is the so called “Matthew Effect”. In other words, the Matthew Effect is a connotation of the phrase “the richer get richer, the poorer get poorer”. Sanderson (2013) provides a solid theoretical framework which supports this thesis. In his empirical study, assessing how immigration affects international income inequality, he finds that immigration has a Matthew Effect in the international economy: it has the tendency to benefit high income countries only (p. 683). He therefore rejects the idea that immigration should be politicized as a tool to solve international inequalities. Neo-classical economic theory holds that the availability of a cheap and flexible supply of labor is key for capitalist economies and their growth. More specifically, Sanderson (2013) bases his theoretical arguments on Lewis’s (1954) dual-sector model of economic development.

According to the latter, the development of capitalist economies relies on the existence of two economic sectors: a capitalist sector and a non-capitalist, subsistence sector focused mainly on low-skilled labor (Sanderson, 2013, p. 684). The capitalist sector sustains itself through a cycle of growth, based on low wages and an unlimited supply of labor. This leads to surplus value, which is re-invested to increase capital accumulation. This stimulates the capitalist sector, with an increase in jobs and a higher labor supply coming from the subsistence sector (Sanderson, 2013, p. 684). This, however, leads to an increase in wages in the long run, as the capitalist sector makes use of all the labor supply from the subsistence sector (the Lewis turning point) (Lewis, 1954). This is when immigration comes into the play: in order to restore labor supply competition and, therefore, lower wages, the capitalist sector searches for new sources of labor supply abroad, creating an international labor supply system (Sanderson, 2013, p. 685).

Consequently, the availability of an increasingly larger supply of foreign labor fostered the growth of capitalist economies, stimulating large scale investment and innovation in production methods (Sanderson, 2013, p. 685). In the Global North, as observed Sanderson (2013), this led to an increasingly lenient, laissez-faire neo-classical approach towards immigration, which translated itself in lenient economic policy, for the end-goal of economic growth and capital accumulation. However,

(10)

9 whilst the effects of immigration on the economy are seemingly positive, they are highly uneven. Within most developed countries, in fact, downward shift on wages in the subsistence sector, as well as increasing incomes of the top income tiers, has increased the wedge of inequality in incomes amongst workers in developed economies. On one hand, leading-edge countries benefit higher levels of innovation and investment, as their economic activity is driven by industries in leading sector which can afford lower-wages due to a higher supply of labor from abroad. In addition, the industrial sector in leading-edge countries is equipped with state-of-the-art technological infrastructures and a higher educated labor force, generating higher levels of surplus value (Sanderson, 2013, p. 685). On the other hand, peripheral countries which are currently lagging behind the leading economies are not subject to the equal benefits from immigration. Innovation and investment effects are weaker: economic production, in these countries, remains predominantly based on the primary sector and secondary sectors with little innovation. This leads to lower wages due to decreasing labor supply competition, and an overall lower economic surplus (Sanderson, 2013, p. 685).

In this theoretical fashion, the income inequality gap within and between nations in the international economic system is widening. Rich countries are getting richer, whilst poor countries are stuck or experiencing decreasing economic surplus. This leads to the first hypothesis of this study:

H1: a higher degree of immigration policy openness – and thereby – the potential increase of immigration towards open countries, leads to an incrementally widening income inequality gap within high-income European countries in the international economy

Industrialization and Benign Economic Forces: How Globalization and

Immigration May Lead to Decreasing Income Inequality

As explained in the theoretical analysis discussed so far, events put in motion by the Industrial Revolution, such as the structural change of the economy and urbanization, drove inequality upwards in the western world, both amongst and within countries (Milanovic, 2016, p. 53). The movement into an increasingly diverse manufacturing sector, intertwined with the rise of globalization, has created a situation whereby the ‘rich are getting richer and the poor are getting poorer’. The 1980s were subject to a severe, all-round technological revolution, defined by outstanding changes in globalization, information technology, and the exponential development of the tertiary sector of the economy. Similarly to previous technological revolutions, as already mentioned, the 1980s brought by heavily increasing income disparities (Milanovic, 2016, p. 53-54). This was predominantly due to the fact that new technologies mainly benefitted the newly created high-skilled, IT-oriented manufacturing upper class, further marking the divide between skilled and un-skilled labour, the latter being characterised by an increasing immigrant workforce. Economies of rich countries entered competition with newly rising world economies which, due to lower costs, could quickly adapt to technological changes, such

(11)

10 as China and India (Milanovic, 2016, p. 54). Thus, the structure of the international economy moved towards services, with demand becoming mainly service oriented. The underlying issue is that the newly predominant tertiary sector of services is highly polarized in terms of wages: on one hand, it offers jobs at the lowest pay-tier, characterized by a hefty migrant workforce, and on the other hand, in services sectors such as finance jobs are very highly remunerated. This widened the income inequality gap (Milanovic, 2016, p. 54).

However, is structural economic change, industrialization and globalization to entirely to blame for increasing economic inequality? The answer is no. This leads us to a second, alternative theoretical framework which provides an analytical perspective opposite to the previous one. To begin with, Branko Milanovic (2016) in his book Global Inequality: A New Approach in the Age of Globalization, explains that the abovementioned factors offer much room to decrease inequality just as much as they may potentially increase it. In fact, income inequality may decrease in capitalist societies as “the supply of more-educated labour and the demand for redistribution increased […] and return on capital (which is closely associated with higher inequality) went down” (Milanovic, 2016, p. 53). Milanovic (2016) thereby speaks of “benign” mechanisms which reduce inequality. These effects entail rising education, greater political participation and rights, and a well-informed population, demanding social protection from the welfare state. (Milanovic, 2016, p. 55). In the next paragraph, the study will explore each force in more depth.

The first benign economic force which may push inequality down in a rich, two-sector industrialized society is higher progressive taxation policy. It is important to note that this force arises out of the “median voter hypothesis”, thus in assumption that in an unequal society, voters are likely to vote for policies favouring higher progressive taxation (Milanovic, 2016, p. 113). However, the power of this specific benign force is not to be overestimated. In modern industrialized societies, in fact, there has been a general trend to lower taxes due to globalization, as part of a race to the bottom to attract foreign investment. The second benign economic force, which will be briefly mentioned, is the increasingly dissipating value of rent capital accrued by the upper class in the early stages of the technological revolution (Milanovic, 2016, p. 113). In fact, as the revolution turned into the mainstream and became established, the individuals and companies slowly began to catch up with early innovators, reducing or even eliminating the capital gap between these two classes. (Milanovic, 2016, p. 114). The third benign economic force which may put downward pressure on inequality is the harmonization of income levels worldwide. Considering the recent rise in wages in the BRIC countries, the middle classes of these respective nations shall climb the wage ladder and shorten the income gap both within-country as well as on an international comparative level (Milanovic, 2016, p. 114). It is important to note that these forces may result primarily from economic and demographic outcomes, and lack in societies with, as Milanovic (2016) explains, a ‘stagnant mean income’ (p. 55). In fact, it is only in countries with

(12)

11 growing mean incomes, and thus, advancing economic growth, that the benign effects of the most recent technological revolution may fully unfold.

As a society, we normally think that technological progress automatically complements high-skilled labour and replaces low high-skilled labour, thus increasing the wage gap. This is what Milanovic (2016) calls “pro-rich” policies. However, it is not entirely possible to rule out the argument that there are some types of technological progress which may foster the productivity of low-skilled labour, too (“pro-poor” policies) (p. 55). This progress is part of what is known as “low-skill-biased technological change”, and can theoretically be considered a fifth benign economic force (Milanovic, 2016, p. 115). However, there is substantially lower evidence of such policies in comparison to pro-rich policies, and, in fact, this remains a rather speculative force. Speaking of policies, it is undeniable that benign forces are caused by, in addition to previously discussed sources, caused by political forces and their behaviour. For instance, after the Second World War, Europe experienced a rise in “benign” economic policies that were mainly characterized by a newly found congruence between left-wing political parties (which gained prominence after the war) and corporate interest (Milanovic, 2016, p. 55). The latter, comprising a rich, property owning class, motivated by fear towards new potential socialist far-left movements which may lead to expropriation of their capital resources, agreed to policies proposed by mid-left-wing parties that created a new broad, increasingly equal middle class. These policies broadly emphasized free education, free or affordable healthcare and non-restrictive immigration, amongst other provisions (Milanovic, 2016, p. 55). There is therefore room for immigration policies, as will be addressed in the next paragraph, to lead and complement to shrinking income inequality, provided (according to Milanovic’s theory) that the country in questions keeps a positive income and economic growth.

Immigration comes into the play due to the fact that migration flows often result in an increase in low-skilled labour base in the subsistence sector, as previously explained by Sanderson through the Lewis’s two-sector model. Following this theoretical logic just presented, it is arguable that increasing migration flows through advancing open immigration policies can both foster the receiving country’s economic growth (therefore, allowing for the “benign” factors to take place) and at the same time reduce income inequality, as both locals and immigrants would be theoretically better off than their previous situation due to overall economic growth.

Consequentially, the rise of the services sector, and in turn, the sub sequential increase in average wage levels in most developed and developing countries, has been a central motive behind the migrations prospects of many, if not all, economic immigrants. Workers from abroad, motivated by wanting to work in a wealthier country with higher pay checks, are able to access such high paying jobs. Earnings are then remitted back home, to the foreign workers’ families. Remittances are a crucial source of income for sending countries – therefore, more open immigration policies would allow for more

(13)

12 remittances, and thereby, an increase in average lifestyle in the poorer – sending countries (Oberman, 2015, p. 239). Higher wage conditions have therefore been a primary motivation for economic immigration to take place, both in benefit of the migrant, due to higher wages, and in benefit of the receiving country, due to economic growth and the possibility to increase median wage as discussed. Under the theoretical framework herewith presented, this leads the study to a second hypothesis:

H2: a higher degree of immigration policy openness – and thereby – the potential increase in immigration flows towards European countries with increasing economic growth, lead to a narrowing income inequality gap within such European countries in the international economy.

The section will now proceed to operationalize the variables involved in this study, to settle how they will be measured in light of the proposed theoretical hypotheses.

Empirical Debate and Further Theoretical Causal Relationships Between the

Variables

The relationship between immigration and income inequality is not exclusively a one-way relationship. It has been studied in several contexts and using a multitude of methods and tools. It is important to note that, whilst this study focuses on the causal relationship between the openness of immigration policies as an independent variable and the level of income inequality as a dependent variable, the facets of the relationship between immigration and income inequality have been broadly studied and processed. This section aims to provide broad empirical observations on the alternative relationship between the variables, by drawing in the empirical results and findings stemming from the empirical debate on this topic.

To understand alternative points of view, we may take the fact that countries which are suffering from increasing income inequality have the tendency to close their borders to alleviate any potential pressure on their labour market and on wages as a starting point. This is due to the theoretical intricacies which see capitalism and capitalist economies link with immigration. In fact, as discussed by Afonso and Devitt (2016), on average, more liberal political economies seem to receive a higher number of migrants than coordinated, closed economies (p. 6). Migration has, in fact, the tendency to strengthen entrepreneurship and economic innovation. This is partly because, due to their weaker social and political resources, migrants have the tendency to settle for worse employment conditions, with less favourable and more flexible terms of employment and wage structure, as opposed to local workers (Afonso, Devitt, 2016, p. 4). Another question which Afonso and Devitt (2016) explore is whether immigration enhances or hinders liberalization. On one hand, they argue that trade unions have a harder time controlling the labour market if a large supply of cheap labour enters the national economy, which

(14)

13 paves the way for liberalization of migration policies. On the other hand, importing cheaper labour may be a better alternative, however, than exporting jobs to developing countries with lower wages (p. 7).

Another way in which income inequality affects the openness of immigration policies is through the “welfare magnet thesis” as reported by Afonso and Devitt (2016) based on a thesis by Borja (1999). The thesis explains that increasingly generous welfare benefits, and therefore, the prospect of higher social redistribution, leading to more equal wages, tends to significantly attract immigration. However, successive studies by other scholars have shown that the pull effect exacerbated by higher welfare benefits may only be marginal, as, some argue, it mainly attracts asylum seekers rather than labour migrants (Afonso, Devitt, 2016, p. 13). For this reason, there has been a general concern in the European Union that the free circulation of labour may cause a race to the bottom amongst European countries in terms of wage levels (Afonso, 2012, p. 706). The same effect could potentially affect the welfare state and redistribution, too. Whilst migrants tend to rely less on contributory welfare benefits, such as pensions and other programs financed through taxes, they rely much more on non-contributory welfare programs, such as social assistance, public schooling and public benefits (Burgoon, 2014). On a political level, the perception that immigrants may therefore constitute a burden to the welfare state is one of the main drivers for wage competition, lower redistribution policies and ultimately the closing of immigration policies. In this light, social fragmentation, less social interaction, lowering of solidarity and trust are all potential outcomes which open societies must potentially face, diminishing social and political support for redistribution and openness (Burgoon, 2014).

However, arguments in favour of redistribution as an outcome of increasing immigration exist, too. According to Burgoon (2014), immigration has the potential to increase the elasticity of labour supply and demand (due to labour market competition), leading to uncertain employment prospects and potentially lower wages for locals. In turn, this may cause the natives to demand more redistribution to compensate for the risks which increasingly open immigration policies impose on them (Burgoon, 2014).

Borjas (2003) has shown however that education is a crucial variable when analysing the relationship between such variables. In fact, according to Borjas, locals’ wages without a high school diploma would be heavily compromised by immigration (Berman, Aste, 2016, p. 1029). Card (2001, 2009) expands on this, explaining that immigrants may truly only threaten locals in the labour market in the event that their degrees of education are equivalent, however not otherwise (Berman, Aste, 2016, p. 1029). Hao (2003), however, argues that immigration is not strongly correlated with an increase in inequality in the US, particularly in terms of wealth inequality. He argues that low skilled immigrants are able to maintain the share in wealth of the lower educated section of the workforce, contributing only marginally to the rise in wealth inequality (Berman, Aste, 2016, p. 1030).

(15)

14 Overall, multiple studies such as Borjas (2003), Card (2001) and Rienzo, and Vargas-Silva (2012) prove that there are a number of variables which feed into the correlation between immigration and income inequality. In fact, job skills and training, education and wealth play crucial roles in establishing whether immigration is detrimental or beneficial towards the host country’s equality (Berman, Aste, 2016, p. 1030). The characteristics of immigrant waves, based on their skill and education level, have different effects on inequality. It is important to take this into account. For this reason, scholars such as Oberman (2015) argue that using immigration policy to combat poverty (and thereby, income inequality) may not be the best idea for policymakers. Despite the fact that his argumentation favours the use of immigration policy to reduce poverty, as it can be effective, he explains it is best if used marginally, or as an option of last resort (Oberman, 2015, p. 249).

Operationalization of the Variables

Independent Variable: The Degree of Openness of Immigration Policies in

Europe

As the variables have been defined, and the theoretical framework laid out, it is now crucial to establish how the variables can be measured and, consequently, where the data to achieve such measurements can be sourced. Being that this study employs a mixed method, including quantitative analysis, quantitative data for both variables is necessary. As for the independent variable, the openness of immigration policies, the IMPIC database will be employed. The latter is a landmark study in its purpose which has successfully managed to quantify the degree of openness of immigration policies by the provision of a comprehensive dataset, conceived by Helbling, Bjerre, Romer and Zobel (2017, p. 79 – 80). IMPIC stands for “Immigration Policies in Comparison”. The scholars responsible for its creation, led by Marc Helbling, have emphasized the need for the creation of a way to measure the openness of immigration policies, due to the fact that immigration as a subject is slowly reaching the mainstream of political science due to its growing importance (Helbling et. al, 2017, p. 80). Until this study had been released, most scholarly studies dedicated to the creation of measurements and/or indices of immigration policies entailed one-off, lightly-comparative single case studies, mostly conducted by economists. Helbling et al. (2017) discuss that most of these studies lacked three basic index-building challenges that were not tackled properly: conceptualisation, measurement, and aggregation (p. 80).

It had therefore been difficult, until the IMPIC database, to know certainly how an immigration policy index or measurement could be reliably used. The IMPIC database remedied to such limitations by improving conceptualization, empirical scope, number of cases (thirty-three OECD countries), time span (1980-2010) and policy dimension (Helbling et. al, 2017, p. 80 – 81). To begin with, the conceptualization of immigration policy used in the IMPIC database resembles the one provided by Bjerre et. al (2015), which has already been explained above in the conceptualization section. To

(16)

15 summarize, this was defined by (1) differentiating between types of immigrants targeted by immigration policies, (2) differentiating between policy outputs, implementation and outcomes, (3) differentiating between immigration, integration and citizenship policies and (4) differentiating between policy dimensions (Bjerre et al., 2015, Helbing et al., 2017). Having developed from this conceptualization of the term immigration policy, Helbing et al. (2017) curated the following aspects of the study in order to improve the reliability of their database: firstly, the selection of items to be measured. Those were selected based on the importance attributed to them by policy experts, their relevance and existence in most OECD countries, and their comparability. Secondly, the choice of sources: data collection was based on legally binding immigration regulation, mainly primary law and secondary law. These documents, which were mostly not in English and available only in print, were analysed extensively by country exerts and legal scholars. Thanks to their thorough analyses and expertise, they provided valuable and reliable insight and added to the development of the database (Helbling et al., 2017, p. 86 – 87).

Consequently, the level of measurement of the items of the IMPIC database varies between 0 (open) and 1 (restrictive). This scale indicates the extent to which a regulation “limits or liberalises the right and freedoms of immigrants” (Helbling et al., 2017, p. 88). Of course, this data set is far from bulletproof, due to research limitations. In addition, as this field of study has not been heavily academically fostered, a benchmark to compare this database to does not exist. This makes the IMPIC database seminal work in the field of immigration policy measurement, and can be considered as “an important step forward in (…) laying the basis for important future work” (Helbling et al., 2017, p. 16). To end with, the IMPIC data considered will range from 1980 until 2010, which is currently the full scope of the collection of the IMPIC data.

Dependent Variable: Pre-Fiscal Income Inequality Levels in Europe

As for the dependent variable, income inequality, there are a variety of strategies that can be employed to operationalize said variable. To begin with, this study will quantify this variable using its most common indicator, the Gini coefficient. The Gini coefficient is deriving from the Lorenz curve framework, which aims to show the percentage of total income earned by a cumulative percentage of the population. In a perfectly equal world, the bottom 25% of the population would theoretically earn 25% of the total world income, the bottom 50% would earn 50% of total world income and so on (De Maio, 2007, p. 850). However, this is never truly the case, with the highest earning

(17)

16 percentage of the population receiving exponentially more income than the lowest earning percentage of the population. This disproportionality between the state of equality and the state of inequality is measured and interpreted as the Gini coefficient, and would equal the area A divided by area A + B, as shown in Figure 1 (De Maio, 2007, p. 850). The Gini coefficient ranges from 0 to 1, or can be depicted as a percentage. A coefficient equal to 0 represents a perfectly equal society, such as the utopian scenario described in the example above. A coefficient of 1 represents a perfectly unequal society, where essentially all income is earned by one individual (De Maio, 2007, p. 850).

The main shortcoming of the Gini Coefficient, stemming from the very limitations of the Lorenz framework itself, is that it is unable to differentiate between the various kinds of economic inequalities. In addition, the Gini coefficient tends to be increasingly sensitive to inequalities in the middle range of the income spectrum (De Maio, 2007, p. 850). This may have a confounding effect on results – which makes the Gini coefficient an insufficient measure to be used alone. This study plans to couple the use of the Gini Coefficient with the use of Decile ratios. Decile ratios are a rather straight forward and effective way to operationalize income inequality. The ratios are calculated, by default, by dividing the income earned by the top 10% of the population by the income earned by the bottom 10% of the population, hence ‘decile’ (De Maio, 2007, p. 850 – 851). However, the ratios are not necessarily fixed at 10%, and can be established by the researcher (e.g. dividing the earning top 20% by the bottom 30%, and so on) (World Bank, 2005, p. 101). The Decile ratio does not come short of drawbacks either. By analysing the income spread between the top and bottom percentages of a population, it tends to ignore information about the distribution of income at the middle (World Bank, 2005, p. 101). However, as previously stated, the Gini coefficient tends to focus on this cluster: therefore, the cooperative use of the two indicators should somewhat provide a comprehensive image of a country’s income inequality picture.

Therefore, having established the indicators chosen for the operationalization of income inequality, this study requires reliable sources of data in order to quantify and/or calculate each indicator. The database which is going to be used for this study is the Comparative Political Data Set (hereinafter “CPDS”) (Armingeon, 2017). The CPDS is a comprehensive Dataset whereby researchers group most economic, social and political indicators per country from other databases and national surveys. For Gini coefficient, the CPDS comprises the data provided by the Luxembourg Income Study (hereinafter “LIS”). The LIS dataset collects its data and documentation by “referring to harmonised microdata for one country […] and one year” (FAQs, n.d.). Again, it allows for comparison across several points in time as the LIS database has organized data over several decades. This makes it ideal for longitudinal analysis (FAQs, n.d.). Besides from the use of these two specialized databases, additional data on individual countries will be gathered and used as necessary from the World Bank Database. It is furthermore worthy of mention that the Gini data considered will range between 1980

(18)

17 and 2010, in order to match the data for the independent variable from the IMPIC database (which is only available from 1980 to 2010).

Methodology

This section began by establishing a clear definition of the key variables subjected to this study. Then, the theorization of the relationship between such variables, as well as their operationalization, was laid out with resulting hypotheses. To conclude this section, the methodology designed to execute the analysis with the chosen variables and theoretical framework will be presented. The starting point is the research question. This study aims to answer the following research question: “what the effect of the degree of openness of immigration policies on income inequality in Europe?”.

To begin with, this is a mixed method comparative study. Mixed method entails the employment of both qualitative and quantitative methods to analyse the chosen case studies and population. As such, the first step entailed the finding of reliable quantitative and qualitative data on immigration openness and income inequality. Qualitative data has been sourced from academic journals, newspapers and official government websites and publications. Quantitative data has been gathered from statistical databases. To begin with, data on immigration policy openness derives from the IMPIC database. This databased contains indices on the degree of openness of the immigration policies in the majority of world countries. Data indicators on income equality will be gathered mainly from the CPDS, which in turn sources its income inequality data from the LIS database. In addition, figures from the World Bank database will too be discussed. Once gathered and presented per-country, the data, both quantitative and qualitative, will be analysed in conjunction amongst the chosen countries, in order to provide a comparative analysis of how economically equal/unequal countries with varying degrees of immigration policy openness are. This is therefore a mixed method, comparative analysis, which employs a Most Similar Systems Design I, due to the deductive focus on one major causal relationship with only two major hypotheses (Toshkov, 2016).

The rationale for employing a mixed method is that it allows to achieve a higher level of analytical depth, which would not be possible using quantitative methods only. This choice benefits the completion and understanding of both variables. For instance, immigration policy openness is a variable which cannot simply be interpreted using merely numeric indicators. There are a multitude of intricacies and factors which affect this variable, and the value thereof. It is therefore important to employ qualitative research in order to fill in the gaps which quantitative research alone is unable to reach. This will allow this study to provide a much larger and complete picture of whole policy development process and rationale behind the opening and/or closing of immigration policies in the chosen samples. By mixing the qualitative with quantitative data gathered from reputable databases, this method consequently allows for a much more accurate interpretation of the quantitative values on policy

(19)

18 openness. This is important, as the latter are difficult to be interpreted on their own. The same applies to the other main variable, income inequality. The quantitative measure chosen for this variable, which is pre-fiscal (pre-taxes and transfers) Gini coefficient, is recorded only every few years in the CPDS. Therefore, there is a considerable amount of data missing. The addition of qualitative methods will allow, once again, to fill in the gaps and provide a more complete picture of the situation in the chosen samples.

There are additional motives which have contributed to the choice of employing a mixed method. In fact, the quantitative analysis alone is not sufficient to statistically analyse the relationship between the two main variables in each sample. This is because there are not enough datapoints per cluster to rely solely on statistical results. For this reason, the statistical analysis per cluster is less useful on its own; however, it can be used successfully in conjunction with qualitative analysis. The latter will provide plenty of useful information which will help process and interpret the situation in each cluster. Consequently, quantitative methods will be used to complement the qualitative discourse in each cluster. This provides increasing reliability in comparison to opting merely for correlating and regressing the variables. This is additionally due to the fact that the main statistical method employed, which will be linear regression, has its flaws in this design, namely that it is likely to overestimate the relationship between the variables due to autocorrelation. Qualitative analysis will allow therefore to mitigate the limitations of the quantitative approach chosen.

Finally, there is a rationale for utilizing quantitative methods, too. By nature, income inequality is represented by the Gini coefficient. This numerical indicator is crucial for the measurement of this variable. Therefore, it is necessary to use qualitative methods in order to be able to make full use of this indicator. As for immigration policy openness, it would be very difficult to quantify this variable using only qualitative data. Luckily, the IMPIC Project, which employed legal and policy specialists to quantify the openness of the immigration policies of several world countries, fulfils this need. Therefore, with the ability to obtain quantitative data on each variable, it seems opportunistic to take the chance to use this data to complement and/or contrast the qualitative discourse in a way which adds more depth, insight and reliability to the overall analysis.

Sampling Method: The Three Worlds of Welfare State Capitalism

This study aims to assess the casual relationship between immigration policy openness and income inequality in Europe. However, the comparative analysis will not merely encompass all European countries altogether. A major flaw would plague such a design: Europe is a constellation of countries which are similar under certain aspects, whilst being very different per-region on a geo-political level. For instance, it would not be valid to compare economic inequality in relation to immigration policy openness between Italy and Sweden. These are two countries which have completely differing geo-political targets and needs, rendering such a comparison void from a reliability

(20)

19 perspective. Therefore, the process of case selection in this study follows the theory provided by Esping-Andersen’s in his book The Three World of Welfare Capitalism. Esping-Esping-Andersen’s work is seminal in its contribution to the comparative analysis of welfare states, especially in Europe and other capitalist democracies. This is because The Three Worlds of Welfare Capitalism establishes three ways to classify welfare states based on their differences and similarities.

The first typology is the liberal welfare state. Its main goal is to achieve a state of universal benefits which are based on public provisions (such as services and insurance), which are allocated using means-tested approaches and targeted at low-income recipients. This type of welfare state encourages market solutions to social problems, with an anti-state political tradition (Esping-Andersen, 2013). The state, therefore, is usually weak in influence. The main advantages, in comparison to the other two types, include less sensitivity to demographic changes in the population, lower taxes, and a tendency for jobs to grow. The main disadvantages of this system include high inequality, and a generally negative stigma attached to those receiving welfare benefits (Esping-Andersen, 2013).

The second typology is the Conservative welfare state. Its main interest is to hold high national values through the maintenance of order. It aims to accomplish this interest through the establishment of solid pension, unemployment and disability insurances. The state, therefore, is strong and present, and in part contributes to the welfare of citizens, whilst encouraging family assistance (Esping-Andersen, 2013). This system enjoys a number of advantages, such as a high level of state support by the public, a proportional increase in benefits based on contributions, and a relatively non-burdensome tax system. The disadvantages of this system include high sensitivity to demographic changes, costlier labour supply, and lower support towards individuals in unstable working conditions (Esping-Andersen, 2013).

The third typology is the Social-Democratic welfare state. It promotes an equal society, whereby welfare provisions are allocated on individual need rather than on individual contribution. In addition, its main goal is to render its supporters as independent as possible, by the provision of extensive welfare programs. Very service oriented, this system works hard towards establishing solid services, many of which are focused on insuring the weak, the disabled, the elderly and the unemployed. The state is therefore strongly present, outplaying its main insurance providing competitor – the private sector itself, in the provision of pensions, full-employment policies, disability and unemployment insurance, and so on (Esping-Andersen, 2013). Income and class differences are therefore less prominent. This system has, therefore, a number of advantages, such as universality, population support, high benefits and well-developed services, and reduced social division. However, even a system as such comes with drawbacks. Due to the state’s heavy involvement, this system is very difficult to administer, leading to extensive and expensive bureaucracy. In addition, in order to be able to afford the welfare

(21)

20 state, social-democratic welfare states have to rely on the highest tax burden amongst the three typologies (mostly progressive tax systems are employed) (Esping-Andersen, 2013).

Consequently, European countries employed as cases for this study are classified based on the typology of best fit, and will therefore be divided in three major clusters. This will allow for the comparative analysis to occur within each cluster, in a way which is void of major social, political and strategic differences between the cases being compared. Esping-Andersen’s framework therefore effectively allows for the provision of a solid sampling method. The samples, which follow right below, have been established with the aid of existing literature on this topic, such as Esping-Andersen (1990), Isakjee (2017) and Gough (2008):

• Liberal welfare state sample: Ireland, United Kingdom

• Conservative welfare state sample: Austria, France, Germany, Italy, Portugal, Spain • Social-Democratic welfare state sample: Denmark, Finland, Norway, Sweden

It is very important to note, however, that the Conservative welfare state cluster presents welfare states which, despite their similarities, may not entirely be classified “conservative” as a whole. This relates primarily to the southern European cases, such as Italy, Spain and Portugal. Whilst Austria, Germany and France present much more developed welfare states of classic Conservative Bismarckian background (Germany and Austria in particular), Italy, Spain and Portugal are arguably at a much earlier stage of welfare state development in comparison (Kvist, 2012, p. 166). According to Kvist (2012), Esping-Andersen has undermined these differences, paying less attention to them than was supposed to. In fact, due to their very high degree of family support and less developed welfare state programs thereof, the Southern European countries could, to an extent, be considered as a typology of their own. This is subject of much discussion amongst scholars, in particular Ferrera (1996), whom has produced a study focused on the search of the Southern European model of welfare state. He explains that there are traits which strongly link the Southern European welfare states: firstly, the fact that their welfare programs have only more recently been developed. Secondly, the influence of Catholicism in forming these programmes. Thirdly, the fact that family still plays a crucial role as a social support structure (Ferrera, 1996, p. 18). According to Ferrera (1996), the mentioned traits are crucial towards policy shaping, opening the possibility for these countries to potentially be considered their own cluster in an empirical analysis. Therefore, their communality and comparative potential with the rest of the Conservative cluster, particularly the likes of Germany and Austria, is questionable.

In this investigation, for statistical reliability purposes, the analysis will be carried out keeping the Southern European welfare states in the Conservative cluster, in order to secure a higher amount of data points (instead of working with multiple clusters of merely 2 or 3 countries each). However, this study acknowledges that this is a shortcoming of this design, and that it is not entirely desirable to compare countries as such.

(22)

21

Recapitulation of Working Hypotheses and Theoretical Predictions

The hypotheses stemming from the theoretical framework are hereby summarized:

H0 = There is no relationship, to a statistically significant degree, between the degree of openness of

immigration policies and income inequality in the chosen population and/or samples.

H1 = Open immigration policies – and thereby – the potential increase of immigration towards open

countries, lead to an incrementally widening income inequality gap within high-income European countries in the international economy.

H2 = Open immigration policies – and thereby – the potential increase in immigration flows towards

European countries with increasing economic growth, lead to a narrowing income inequality gap within such European countries in the international economy.

In addition to the hypotheses, sample specific predictions will be tested, to reinforce and/or diminish the strength of one of the hypotheses mentioned above. The sample specific predictions are based on the expectations laid out by Esping-Andersen. Namely, the following will be tested:

a) The degree of public spending and government involvement is significantly higher in the Social-Democratic (Nordic) welfare state cluster. Higher spending on welfare and social safety nets such as unemployment benefits, sickness benefits and higher pensions lead to increasing redistribution of wealth and therefore lower income inequality. It is therefore expected that the degree of openness to immigration has a merely fair statistically significant relationship with income inequality in comparison to the other two clusters. In addition, this relationship is expected to be weak/moderately negative (meaning less openness leading to decreasing income inequality).

b) The degree of public spending and government involvement is lowest in the Liberal welfare state cluster. Welfare provisions in this cluster are, on average, the lowest in terms of percentage share of GDP. Lower market regulation and government involvement is likely to make way for other factors, such as openness to immigration, to affect the relationship between the degree of openness to immigration and income inequality with greater statistical significance (to a good extent) in comparison to the other two clusters. In addition, this relationship is expected to be strongly negative.

c) The Conservative welfare state cluster is then expected to perform in between the two other clusters in terms of the strength of the significance of the relationship between the dependent and the independent variable. This is due to the high levels of government spending in this cluster, mostly directed towards pensions and unemployment/disability insurance, which may not benefit low-skilled immigrants to a great extent. Therefore, as per the theoretical expectations laid out in the theoretical framework, the statistical significance of the relationship

(23)

22 between the dependent and the independent variable is expected to be fairly significant for this cluster. The relationship itself is expected to be weak/moderately negative.

(24)

23

Case Study Background and Contextual Information

This section of this study will focus on providing a general background on the current state of inequality and openness to immigration in the chosen country samples. As previously explained, the chosen countries have been divided in three samples: the Liberal, Conservative and Social-Democratic welfare states, based on the framework established by Esping-Andersen (2013). To begin with, the study will assess the situation and background in the Liberal sample, followed by the Conservative and the Social-Democratic. For each country and sample, the income inequality situation will be discussed, along with factors which may have determined the situation. In addition, a brief explanation of the country’s policymaking process towards immigration will be covered. The period covered will roughly range between 1980 and 2015 (this may vary), which is the data collection period for the empirical analysis. This section is necessary to understand the background situation per country cluster, to provide context for the analysis which will follow.

Economic Inequality and Immigration Policy Development in the

Liberal Welfare State Cluster

Starting with the broader data provided by World Bank estimates, Ireland presents a Gini coefficient (in 2014) of 31.9, with a ten-year high in 2005 of 33.8 and a ten-year low in 2008 of 30.9 (GINI index). According to the TASC, a Dublin-based independent research think-thank focused on challenging economic inequality, the top 11% in Ireland earn over 45% of all income. This breaks down to 10% of income earned by the top 1% of the population and the remaining 35% earned by the top 10% (O’Connor, Staunton, 2014). Ireland has therefore been experiencing a heavily declining labour share of income: trends show that the top 10% have been heavily overtaking the bottom 90% in labour share of income since 1975, despite the country’s economic growth. In total, Ireland interestingly presents the highest gross income inequality in Europe, equal to a Gini of 46.0, which however is adjusted to 31.9 in net terms (O’Connor, n.d.).

Let’s now explore the macroeconomic situation, and possible causes of Ireland’s current level of inequality. To begin with, the cost of living in Ireland is about 20% higher than the EU average, whilst social public spending is relatively low. In fact, public spending on health and education is around €12,191 per household; social protection incomes per person equals €9,776 per year for the unemployed, €10,608 for unemployed carers and €11,976 for pensioners (O’Connor, n.d.). This is on the low-end of the European spectrum, with merely 4.4% of GDP spent on total security contributions, less than half than European average (11.1%) in 2012. This renders Ireland’s percentage-to-GDP of public spending the second lowest in Europe (O’Connor, n.d.). In turn, childcare fees and the overall cost of living are respectively 27.4% and 20% higher than European averages. The rather contradicting

(25)

24 figures are however better understood in light of Ireland’s taxation system. In fact, Ireland presents one of the lowest tax rates in Europe, and is of progressive nature (O’Connor, n.d.).

World Bank estimates show the United Kingdom’s level of income inequality to currently stand at 34.1, with a ten year high at 36 and a ten year low at 32.3 points (GINI index). Practically, the average income of the richest 10% bracket in the UK is nearly tenfold in comparison to the bottom 10% tier. This is slightly above OECD average (which averages at 9.5x) (OECD, 2015, p. 1). Overall, the UK has been one of the most unequal countries in the OECD, especially in the last decade. The income of the top 1% earners has seen a gradual increase. Similarly, to the Irish case, the proportionality of the tax system decreases as we rise in income brackets: marginal income tax rate dropped from 60% in the 1980’s to around about 45% today, further marking the divide between the top and bottom tiers of income earners (OECD, 2015, p. 1). Even though taxation and welfare benefits have marginally been able to curb the level of income inequality in the UK (by roughly 25%), the latter is still lagging behind the likes of France, Germany or the Scandinavian countries in this aspect. Overall, despite the tax system has not been as effective as the welfare state in reducing household income inequality in the UK, as the latter has been mainly driven by changing in the benefits system (OECD, 2015, p. 1). Policies as such included an increase in the income tax basic allowance and rise in child tax credit, which benefitted families in unemployment and with low earnings (OECD, 2015, p. 1). Currently, the Coalition government has stated its commitment in reducing its current balance deficit. Policies to achieve this goal will predominantly entail cuts in welfare state spending rather than increases in taxes (OECD, 2015, p. 2). Some programs, however, such as health and education, will remain untouched.

Moving on, Ireland’s immigration policy’s openness history will be explored in this paragraph. To begin with, Ireland’s economic boom in the 1990s caused a previously unseen stream of immigration towards the country, due to unprecedented levels of prosperity. This wave of immigration, which lasted up until the early 2000s, included most immigrant types ranking from skilled workers to asylum seekers (Ruhs, Quinn, 2009). However, the sudden surge of immigrants towards the country caused a need for the rapid development new immigration policies. According to Ruhs and Quinn (2009), this came in three stages: first, a list of safe and unsafe countries was compiled, and applications began being prioritized. In addition, between 2003 and 2005, citizenship laws were changed to eliminate Irish-born child’s automatic right to citizenship if their parents were not Irish (Ruhs, Quinn, 2009). Finally, Ireland shifted away from its liberal work permit system, prioritizing merely low-skilled immigrants migrating from within the EU-10. EU migration flows therefore increased over the last decade. Consequently, due to the growing number of un-skilled immigrants threatening the employment of locals, and the pressure this has been exerting on the already low social welfare system, Ireland has exerted increasing strict immigration provisions on non-EU immigrants, favouring only highly skilled immigrants (Ruhs, Quinn, 2009). Ireland’s immigration policy openness has thus been on the decline.

(26)

25 Ireland, Descriptive Statistics

N Minimum Maximum Mean Std. Deviation

Immigration Policy Openness

31 .554 .704 .66636 .045226

Gini index of pre-fisc income (before taxes and transfers) among household members

8 35.30 51.10 43.8875 4.44054

Valid N (listwise) 8

In terms of immigration openness, the UK, to a worst degree in comparison to Ireland, has been increasingly adopting a stricter stance towards immigration, even from inside the EU. The UK completely opened its doors to immigrants from the Eastern Enlargement countries (Bulgaria, Romania and Croatia) in 2004. The government then decided to backtrack on this decision in 2007 and 2013, when respectively increasingly strict transitional restrictions were applied to Eastern European immigrants applying for work permits. In addition, in 2010, it introduced new restrictions for migrants attempting to access welfare benefits, to avoid a “pull factor” for migrants to move to the UK (Marczak et al., 2015). Since 2010, the government has in fact been committed to reducing migration, with several policy changes aimed at rendering immigration more difficult for those wanting to work or study in the UK. The first major policy was an increase in eligibility criteria to enter the UK, becoming much more selective for non-EU nationals (Marczak et al., 2015). International students and professionals without a major job sponsorship suffered from this provision. In addition, the new provisions reduced working hours for international students and raised language requirements. Former international students would not be allowed to remain in the UK without a solid job offer, and universities have been required to increase international eligibility standards (Marczak et al., 2015). Along with stricter naturalisation policies, the government of the UK has altogether been fostering its stance of strictness towards immigrants to a great extent.

United Kingdom, Descriptive Statistics

N Minimum Maximum Mean Std. Deviation

Immigration Policy Openness

31 .339 .498 .38944 .045559

Gini index of pre-fisc income (before taxes and transfers) among household member

9 40.80 46.40 44.5333 1.83576

The Influence of Immigration Policymaking on Income Inequality: A Comparative Study