• No results found

Equal rights but only for some? : guest worker programs, exploitation and the rights trade-off

N/A
N/A
Protected

Academic year: 2021

Share "Equal rights but only for some? : guest worker programs, exploitation and the rights trade-off"

Copied!
50
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

University of Amsterdam

M. Sc. Political Sciences

Track: International Relations

Master Thesis

Ely Faris

10641270

Dr. Darshan Vigneswaran

Words: 17760

Equal rights but only for some? Guest worker

programs, exploitation and the rights trade-off

(2)

Table of content

Introduction

2

Literature review

4

Theoretical framework

16

Case selection and methodology

22

Analysis

26

Conclusion

42

(3)

Introduction

Political justice: a democratic political community must be open, and equally open, to all those men and women who live within its territory, work in the local economy, and are subject to local law - Walzer, 1983, p. 59

This paper will be focusing on guest worker programs and how we can better protect migrant workers. Because of globalization and the growing demand for cheap labor, guest worker programs are a popular strategy to fulfill low-skilled jobs left vacant by the population of the host country (OHCHR, 2015, p. 17). Guest worker programs can be beneficial to the receiving and the sending countries, as well as for the migrants themselves. They can provide economic development, higher wages and better living conditions for migrants and their families. Additionally, guest worker programs can provide cheap labor and a large number of workers for the receiving countries (OHCHR, 2015, p. 18). However, to keep foreign labor affordable and to control migration flows, states use different kinds of mechanisms and trade-offs. These mechanisms and trade-offs can have adverse effects, especially for the migrant workers themselves. Decreasing the amount of rights migrant workers receive in order to keep labor cheap, for example, which can cause cases of abuse and unequal treatment of foreign workers (Anderson, 2012, p. 644; McKenzie & Gibson, 2010, p. 2) Especially for lower-skilled migrant workers there is a high chance of exploitation. This is mostly seen in the Gulf countries. The

kafala system in these countries has showed many instances of modern-day slavery

circumstances (Anderson, 2012, p. 638).

While a comprehensive set of rights for migrant workers has been set out in the ‘International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families’ (UNHCR, 1990), almost no migrant receiving countries have signed this convention. No country can be forced to sign the convention and there is no global government that can regulate labor migration or labor migrants’ rights (OHCHR, 2015, p. 19). Conversely, the ‘Universal Declaration of Human Rights’ states that everybody is entitled to all the rights and freedoms of the declaration (OHCHR, 2015, p. 22), but states are reluctant to sign these conventions. Therefore, we see many cases of migrant workers being allocated a very minimal amount of rights, which has been found to be a source of abuse and exploitation.

This research is relevant because guest worker programs are an important part of today’s global economy. The decline of populations and workforces has become an increasingly widespread problem and global migration as well as guest worker programs will likely play a big part in ensuring that economies continue to thrive and remain competitive. It is likely that states would want to increase their openness towards lower-skilled migrant workers. Because of the rise of these programs, it is key that we focus more on the rights of migrant workers and for

(4)

countries to address cheap labor migration and migrant rights (OHCHR, 2015, p. 22). The fact that lower-skilled migrant workers are more at risk of dealing with abuse, poor working conditions, unfair treatment and a decrease in rights (OHCHR, 2015, p. 149), makes this research even more crucial.

When considering the available literature, a discrepancy between the possibility of giving migrant workers a large amount of rights combined with the willingness of states and employers to employ them was encountered. Therefore, the research question is:

Is it possible to uphold a high openness and offer a large amount of rights towards lower-skilled migrant workers in a liberal democracy in order to minimize exploitation?

A case study will be used in an attempt to see if it is possible for states and employers to be willing to employ migrant workers while at the same time granting them their basic human, social and labor rights. This can lead to a win-win-win situation: it would be beneficial to the state and to the migrant workers and hopefully leads to less exploitation. In the first part of this paper literature regarding guest worker programs will be discussed, as well as the rights trade-off theory and exploitation. This paper will also explore the right which is most commonly restricted: the right to change employers. This will be followed by a theoretical framework where the relevant concepts and variables will be discussed, and how they are connected. Then, the case selection and methodology of this research will be explained. Here, the motivation for the chosen research design and methods will be explained, as well as how the case and data were selected. The final part of this research will consist of an analysis guided by reports and literature. In this part of the paper, the research question will be answered which will induce a conclusion regarding the research and findings.

(5)

Literature review

A focus on guest worker programs provides the opportunity to analyze the correlation between openness, offered rights and exploitation. Additionally, the literature can help scrutinize whether a large amount of rights for migrant workers is possible combined with a high openness and whether this will lead to less exploitation. In the literature review Martin Ruhs’ theory will be explained as well as other authors and their stances on openness and rights. The main right that is restricted in guest worker programs will be explored: the right to change employers. The implications of this restriction will be discussed and possible ways to work around it to better protect migrant workers.

Consterdine and Samuk (2018) point out that there are three dominant perspectives regarding the rights of migrant workers:

1. There is a trade-off between openness and rights

2. Some exploitation is acceptable as long as migrant workers are still benefiting from the program

3. Guest worker programs can be improved without lowering openness or numbers by gradually extending the amount of rights that migrants are offered

Especially the first and third perspective will be discussed in the literature review. The first perspective is the most popular with Martin Ruhs’ (2013) trade-off theory at the fore-front, in which he states that there is a trade-off between openness and rights. The second perspective is supported by some authors who say that although guest worker programs have a lot of negative consequences, it does offer a lot of migrants with job opportunities (Weyl, 2018). The third perspective is hardly discussed but the most important to better protect migrant workers. It focuses on a balance between openness and rights, as well as whether we can keep guest worker programs functioning while also increasing rights. Lenard and Straehle (2010) discuss this third perspective.

Ruhs’ trade-off theory

Ruhs (2013) talks about the different mechanisms and trade-offs countries implement regarding foreign labor in his book ‘The Price of Rights’. He compares lower-skilled, mid-skilled and higher-skilled guest worker programs. By doing so, he comes to the conclusion that it is mainly a trade-off in rights that exists regarding labor migration: the more open a program is towards migrant workers, the lesser rights these workers get offered. Ruhs concludes that this is mostly

(6)

because of ‘the price of rights’: migrant workers can cost the receiving country a lot of money if they can make use of certain rights, like welfare benefits, housing, and enjoy the same wages as the local population. Countries want to balance out these costs, so they have two options:

1. High openness: employing large number of workers but keeping them ‘cheap’ by offering them less rights, mostly rights that cost the receiving country a lot of money

2. Low openness: offer migrant workers a large amount of rights but balancing out the costs by accepting a low number of migrant workers

“Employers face a downward sloping demand curve for labor, meaning, higher labor costs will be associated with fewer workers employed. Certain employment rights for workers, such as the right to a minimum wage, work related benefits, and health and safety standards, increase labor costs for employers, thereby generating a trade off between numbers and rights” (Ruhs, 2013, p. 50). Employers are more likely to employ migrant workers when they are able to restrict their rights (Ruhs, 2013, p. 45). The reason why many countries have not ratified the International Labour Organization’s (ILO) convention regarding the rights of labor migrants is because granting all these rights can cost the receiving country, as well as the employer, a lot of money (Ruhs, 2013, p. 19). Giving migrant workers the right to public funds, for example. This would also create an unnecessary pull factor (Ruhs, 2013, p. 20).

One of the main benefits of guest worker programs is that migrant workers can offer the receiving countries cheap labor. When the main reason countries and employers are open towards migrant workers is because of cheap labor, then offering them a high amount of rights can hinder that immensely. Therefore, a high amount of rights will have a negative impact on openness (Ruhs, 2013, p. 27).

For the conceptualization of the term openness Ruhs states: “Refers to policy openness, defined as the number, scope and strength of restrictions that nation-states put in place to regulate the inflow of migrant workers, something that is not necessarily the same as the actual scale of migration to a particular country” (Ruhs, 2013, p. 49). The more restrictions there are in place to control migration flows, the less open a country is towards migrant workers. To calculate this, Ruhs looks at different indicators, like certain policies, to decide how open a country is towards migrant workers. In Ruhs’ definition he says that openness is not the actual scale of migration to a particular country. He hardly looks or makes use of numbers and mostly analyses policies.

(7)

Ruhs studied different indicators in each migrant labor program in order to calculate the openness number. He does this to conclude whether a high openness negatively influences the amount of offered rights. The indicators he used were the amount of demand-side and supply-side restrictions that the country had implemented to control migration (Ruhs, 2013, p. 217-219). By looking at these restrictions he decides how open a country is. The higher the number, the more open a country is, which would lead to less rights being offered to migrant workers in order to decrease costs.

Ruhs concludes that the trade-off between openness and rights is most visible within programs targeted towards lower-skilled migrant workers. The reason why the trade-off between openness and rights differs for different skilled workers is because there are less higher-skilled workers employers can choose from. To attract high-skilled workers, countries and employers are more likely to offer them a large amount of rights: “In contrast, there is an almost unlimited supply of potential migrants in low income countries willing to accept low-skilled jobs in higher income countries with wages, employment conditions and rights that are significantly lower than those mandated by local laws and international norms” (Ruhs, 2013, p. 43). Because lower-skilled workers earn less, they are expected to pay less taxes and are more likely to make use of welfare benefits. This is why in most cases offering certain rights to migrant workers can have very high costs, and therefore these rights will be restricted (Ruhs, 2013, p. 44).

The right to free choice of employment is restricted in almost all analyzed programs in Ruhs’ research. In almost all cases migrant workers, especially lower-skilled, are only able to work in one specific sector and for one specific employer. Most of these migrants get employed to deal with labor shortages. Giving them the opportunity to work for other employers or in other sectors would minimize the benefits of the guest worker program. It can also cause unwanted competition between local workers and migrant workers (ibid).

Ruhs, in the specific case of the kafala system in the Gulf countries, says this program is highly beneficial for the receiving countries because migrants are mostly considered temporary workers and get offered very little rights. This program employs a very large amount of lower-skilled migrant workers in the private sector, whilst most citizens work in the public sector. The Gulf countries have a very high openness towards lower-skilled migrant workers, and are willing to employ a large amount of them, because workers get offered very little rights and are kept very cheap. Workers come to the receiving country through a direct sponsorship system that binds them to their employer, who has financial and legal responsibility over them. It is not uncommon that workers get their passport taken away by their employer in order to exercise control and power over them (Ruhs, 2013, p. 98). There is also less need for integration policies or projects that will cost the state money: “The kafala system has provided a cheap migrant workforce that has facilitated the low-cost provision of goods and services in the private sector

(8)

and the development and maintenance of a generous welfare state whose benefits and services are largely limited to citizens” (2013, p. 268). They are able to employ a large amount of migrant workers because dictatorships “find it easier to deny migrants access to the welfare state than democracies. Therefore, they find it easier to admit more migrant workers than in democracies” (Ruhs, 2013, p. 48).

Other authors on the trade-off theory

Walsh (2014) also make a case that immigration control, like guest worker programs, are a good way for states to keep control over their territories, populations and development trajectories. He also agrees that guest worker programs can do a lot for economic development when a rights trade-off is in place: “In mobilizing workers who will accept wages and working conditions below domestic standards, and attracting foreign investment, importing skills, and establishing international connections and exchanges, migration is managed to boost economic development, output and profitability” (2014, p. 587). Hiring foreign workers can boost economic growth within the receiving country. These new workers will have a taxable income and consumption that benefits the state. The state will generate even more profit when migrant workers are not allowed access to most public goods, which makes these workers more profitable for the state than permanent citizens (Walsh, 2014, p. 590).

Cook (2010) says international human rights law is still very weak and it is mostly up to the state which rights migrants receive: “Advanced industrial democracies that violate rights are also less vulnerable to international pressure or condemnation than are poorer nations. No rich country has ratified the ILO conventions, whereas most developed and migrant-receiving countries have” (2010, p. 149). Basok & Carasco (2010) mention that human rights treaties only grant rights to migrants in very few situations. It is mostly up to the migrants themselves to claim their rights, seeing as policy makers and employers are likely to pay little attention to their rights (2010, p. 344). They state why it is less likely for states to pay attention to the rights of migrants: “It is one thing for states to be pressured to better protect their own citizens and quite another for states to take on the task of protecting individuals not considered national members” (Basok & Carasco, 2010, p. 345). But they also argue that the UN ‘Universal Declaration of Human Rights’ has made rights universal. This means that citizens as well as non-citizens should enjoy the same amount of rights (Basok & Carasco, 2010, p. 346).

Arneson (2018) discusses how the fairest way to deal with migrant workers seems to be to give them full rights, but this causes most countries to close their borders, or to admit them on very harsh terms, which is less desirable to most liberal democracies. He then states that a trade-off might be the fairest way of dealing with migrant rights, and that maybe standards should be lowered. This seems to be the best-case scenario if this leads to lower-skilled migrant

(9)

workers getting employed and countries to be willing to employ them. It depends on whether you value numbers more, or rights (2018, p. 173). He then mentions that giving rights priority over numbers is maybe not the best way to go. Full rights make lower-skilled workers less desirable to employers, and this will prevent them from getting hired in the first place. Full rights means a very little chance of lower-skilled migrants getting employed (Arneson, 2018, p. 180). Weyl (2018) mentions how guest worker programs, especially in the Gulf countries, can only work with a limited set of rights. He states that even a little improvement in rights can cause migration flows to stop. This would negatively affect the future of migrants (Weyl, 2018, p. 23). Regarding openness, one of the biggest issues with an open migration system is the fear that it will deteriorate the welfare state. Foreign workers are not just pulled by new employment opportunities and higher wages, they are also looking for benefits that they are unable to receive in their own country: state provided benefits, certain employment protections and rights workers receive in most welfare states (Freeman, 1986, p. 55). This is one of the main reasons why states want to put restrictive policies regarding labor migration in place. At the same time, states are able to attract a large amount of foreign labor because of their welfare system and the benefits it creates, and it is that same welfare system and its high costs that asks for more foreign workers: “[the welfare state] helps set in train the sequence of events that creates the demand of the receiving countries for migrant labor in the first place” (ibid). The cost of the welfare state can be partly covered with foreign labor. However, unrestricted labor migration combined with full access to the welfare state can cause labor migration to cost the welfare state more money than it receives, therefore making the costs of the welfare state even higher (Johansson, 2014, p. 101). Pritchett (2006) points out that claims about the deterioration of the welfare state because of labor mobility are incorrect, unless you assume that everybody that enters the country makes use of benefits. Laws still determine the distribution of benefits and are structured differently in every country (2006, p. 97).

Restricting rights

Ruhs addresses how almost every guest worker program restricts some rights of migrant workers. The most common examples are a limited time of residency and employment in the country they get to work in. There is also a very little chance for lower-skilled migrants workers to ever become a permanent resident or to earn permanent residency rights. Other ways in which migrants receive restrictions are: only being allowed to get employed in a specific sector, not being able to work for another employer and having to leave the country as soon as their job ends. “Under many proposals for new guest worker programs, migrants are also likely to have very restricted access to unemployment and welfare benefits and no rights to family

(10)

reunion” (Ruhs, 2010, p. 251). Especially for lower-skilled migrant workers there is a high chance of exploitation. This is mostly seen in the Gulf countries (Anderson, 2015, p. 638). The fact that many migrants are tied to their employer, like in the kafala system, that works as a sponsorship system, means employers have a large amount of power. Migrants can lose their worker status and can be forced to leave the country once they get fired or their work offer stops. These situations can cause migrants to receive abuse, and because they are worried about losing their job or getting forced to leave the country, they will most likely endure these abuses (Anderson, 2015, p. 644). Anderson (2012) mentions how ‘managed migration’, like guest worker programs, has a downside of “enforcement, detention, and deportation. This flip side risks undermining win-win-win approaches. It can be troubling to a political culture that values physical integrity and individual rights” (Anderson, 2012, p. 1242). Walsh also talks about how “in hierarchically sorting workers and establishing unfree relations of production, these schemes provide comparatively powerless and competitive workers and contain the potential for exploitation” (Walsh, 2014, p. 595).

Martin (2007) and Piper (2015) point out something very important when talking about states restricting the rights migrant workers receive. Martin points out that decreasing the amount of rights offered to migrant workers is a case of controlling and managing migration flows: “[...] policy makers often turn to the one instrument they control directly, the rights of migrants, in an effort to manage international labor migration. As a result, the rights of migrants may be abridged by governments in order to keep migration “under control […]" (Martin, 2007, p. 15). Piper discusses how lower-skilled migrant workers do not come out as winners in guest worker programs. She mentions that “labour has suffered in the forms of downgrading of labour standards and a loss of traditional union rights [...] and the weakened position of the ILO [...] As a result, the dominant policy approach focuses on primarily controlling migration through state cooperation and extracting economic benefits of foreign workers whilst paying mere lip service to the human rights of migrants” (Piper, 2015, p. 791). They both point out something important: a state does not only restrict the rights of foreign workers because of the price of rights, but also to gain control and power over migrant workers.

The right to change employers

One of the main rights that is restricted is the right to change employers. It is hard to challenge this restriction because restricting the right to change employers is beneficial to guest worker programs and keeps them functioning. Simultaneously, this restriction causes the worst abuse and exploitation. A direct employer sponsorship, where a guest worker comes to a country to work for a specific employer, brings a lot of costs and risks. It is therefore not reasonable or

(11)

realistic to ask employers to agree with giving guest workers the freedom to search for other jobs or employers (Weinstein, 2002, p. 7; Ruhs, 2013, p. 175; Brickenstein, 2015, p. 112; OECD, 2011). However, this causes obvious dangers of exploitation to the migrant worker who is working in an unknown place for an unknown employer. Normally, in exploitative situations, workers have the ability to search for a better job or employer. But these rights are taken away in guest worker programs (Weinstein, 2002, p. 7). This restriction has huge benefits to employers and gives them incentives to hire a guest worker instead of a local worker. It is easy to assume that an employer rather hires an employer that costs less, but “if, however, the terms of employment differ considerably between the native and migrant, it is entirely possible that the employer may develop a preference between equivalent candidates” (Weinstein, 2002, p. 8). When migrants are bound to their employer by law, employers do not have to worry about employer loyalty or offering resources like money to gain loyalty. These worries are irrelevant concerns with guest workers, because migrants can only work for one employer (ibid). This causes employers to have a preference for guest workers, and therefore guest workers can out-compete local workers. Weinstein sees this restriction as a natural consequence of guest worker programs. Instead of looking for ways to eliminate this restriction, he looks for ways to work around it (ibid). One of his main propositions to reform guest worker programs is therefore: Reform 1: migrant workers must be permitted the freedom to roam within the admitting sector of the host labor market by shifting employer outlays away from direct visa sponsorship and towards a system of tradable work permits (Weinstein, 2002, p. 17). With ‘a system of tradable work permits’ he means that employers need to purchase a work permit for a particular guest worker for a specific amount of time. Whenever the guest worker or the employer decides to end the work relationship, the remaining time on the work permit can be sold to another employer who wishes to hire a guest worker (ibid). This prevents workers from being heavily dependent on one employer for work and the ability to stay in the country (2013, p. 175; Ruhs, 2006, p. 26). Ruhs also points out the problems with restricting the right to change employers. He states to protect migrant rights, there should be a system where work permits and the labor of guest workers can be used by different employers. Then you do not give total freedom to guest workers, but you give them the chance to change employers whenever necessary. He then states: “A more realistic policy objective would be to facilitate the portability of temporary work permits within a defined job category and after a certain period of time” (Ruhs, 2013, p. 175). Ruhs’ vision could align with Weinstein’s vision on how we could better protect migrant workers without giving them the total freedom to work wherever or for whomever they want.

(12)

Martin points out that migrant workers could protect their own rights better if they would have the freedom to move around, from one employer to the next. He gives the example of the European Union, where European citizens are free to move from one country to another, to search for jobs and compete with local workers. Some migrant advocates even go as far as saying that unauthorized migrants are better protected than working migrants with a legal status, because they still enjoy their freedom of movement, and are not tied to an employer (2007, p. 30). The OECD, the Organisation for Economic Co-Operation and Development, states that one of the main critiques towards guest worker programs is the fact that in most cases workers are tied to their employer. Because in some cases the employer also has to provide the workers with accommodation, the workers are too dependent on their employer (OECD, 2011, p. 14).

The problem with taking away the right to change employers is that it can also undermine the right to join a union. Brickenstein (2015) states that it is vital to analyze the right to change employers and to join unions because taking away these rights makes migrant workers even more vulnerable to exploitation and abuse. Unions are the ones that can help migrant workers in difficult situations and are an important actor in protecting migrant worker rights (2015, p. 112).

Giving migrant workers more rights

Lenard and Straehle (2010) pose an interesting view regarding migrant workers and the rights they should get. They use a quote by Daniel Attas: “The market and the state are not truly distinct. The economic terms on which guestworkers are employed are a consequence of the exertion of political power, the creation and enforcement of rules” (2000, p. 86). They point out that restricting rights is mostly about exerting power and controlling migrants through the power of the state (Lenard & Straehle, 2010, p. 215). This was also mentioned before by Martin (2007) and Piper (2015). Political justice, as described in the quote by Walzer at the beginning of this paper, seems to be reached whenever we give migrant workers the chance to join unions. But key decisions regarding guest worker programs, like the rights workers should receive, are hardly handled by unions. These decisions and rules are made within the political sphere, and guest workers are not part of this political sphere. They further state that the lives of guest workers are “affected by political decisions made in a country in which they do not have a say” (Lenard & Straehle, 2010, p. 215). Migrants should be able to eventually have citizenship rights so they can be fully included into society. Without these rights, they will remain vulnerable and rights violations will keep on occurring (Lenard & Straehle, 2010, p 215; Goldring, 2014, p. 221; Basok, 2010, p. 50). Extending the rights of migrant workers will have two consequences. First, it will minimize abuse and exploitative conditions because the migrant workers will have a

(13)

chance to get citizenship, and states do not want to fail in protecting their own future citizens. And secondly, it will give workers the ability to fight for their own rights because they do not have to fear immediate deportation (Lenard & Straehle, 2010, p. 215; Mayer, 2005, p. 316). It seems like extending the rights of migrant workers is hardly discussed because “rather than facing up to the exploitation of migrants, employers and governments congratulate themselves on helping those who would otherwise be earning much less at home” (Datta, Mcilwaine, Wills & Evans, 2007, p. 44).

Regarding Ruhs’ trade-off theory, one of its shortcomings is its distorted conclusions regarding a trade-off between openness and rights because of its indicators. Ruhs’ theory entails that indicators on the openness side have a negative or positive effect on the indicators on the rights side. You can envision it as a seesaw on a children’s playground. Figure A1 shows the effects of a high openness on rights, and figure A2 shows the effects of a low openness on rights:

!

Shortcomings regarding this trade-off theory involve indicators that are viewed as a policy which can also be seen as a right. And some rights can also be seen as policies. This causes distorted conclusions regarding his trade-off theory and how openness and rights influence each other. The critique mostly involves wage parity/minimum wage, sectoral/occupational restrictions and the right to change employers, as will be discussed next.

Ruhs states that wage restrictions are an effective way to control migration flows and to keep foreign labor affordable. He states that the most open wage restriction towards migrant workers is to hire them at the same minimum wage as local workers. The most restrictive policy is to

(14)

offer migrant workers wages that are collectively agreed on in wage agreements (Ruhs, 2013, p. 62). It is unclear whether wage parity/minimum wage should be viewed as a right that migrant workers receive or if we should rather view it as a policy restriction. In the ‘Protection of All Migrant Workers and Members of their Families’ (UNHCR, 1990), it is stated that a foreign worker should have certain rights, and ‘The right to equal treatment with regard to remuneration, other conditions and terms of employment, and social security’ is one of them. Therefore, it should be considered as one of these rights. This view differs from Ruhs’ considering Ruhs views wage parity/minimum wage as a policy restriction, and therefore it fits within his description of openness. Ruhs only discusses wages when he discusses demand-side restrictions (Ruhs, 2013, p. 61).

The second critique involves sectoral/occupational restrictions and the right to change employers. The first thing that needs to be pointed out is that Ruhs does not look at the right to change employers as a policy. He only discusses this when he looks at rights. Just like the wage parity/minimum wage indicator, you can see this indicator as a policy and as a right. When it is viewed as a policy, it is part of Ruhs’ openness calculation. But it can also be seen as a right, therefore being a part of Ruhs’ rights calculation. Therefore, it should be considered both. Decreasing the amount of rights offered to migrant workers, especially the right to change employers, can be viewed as a way to control and manage migration flows (Martin, 2007, p. 15). It fits within Ruhs’ conceptualization of openness. The fact that Ruhs does not view the right to change employers as a policy influences the conclusions he makes when he applies the trade-off theory to a certain country. The same critique applies to sectoral/occupational restrictions, but the other way around: Ruhs only views this as a policy restriction, and not as a right. But you could view not being able to work in whatever sector or occupation you want as a right that is taken away.

A third critique regards Ruhs’ definition of openness which also distorts his conclusions regarding some countries and how open they are. Ruhs states that he does not look at the actual number of migration, and he hardly looks at numbers to decide how open a country is towards migrant workers. But he does look at work permits to decide whether a trade-off is in place or not, like in the case of Sweden: “This, together with the Swedish language, is likely to help explain why the number of work permits issued to non-EU nationals remains relatively low” (Ruhs, 2013, p. 102) and he therefore concludes that Sweden’s openness is not as high as it seems on paper. Ruhs concludes that this causes Sweden to be able to offer a large amount of rights (ibid). This illustrates that he does look at numbers regarding work permits to conclude whether a country is open or not, even though this is not stated in his definition of openness.

(15)

Therefore, his conclusions regarding work permits, and how we should view the amount of work permits, are very vague. It is unclear what number Ruhs views as a high amount of work permits, and what he states as a low number of granted work permits. The assumption is that he compares the issued work permits in countries to the numbers in the Gulf countries, because they have some of the highest numbers of issued work permits. Based on these numbers, he then decides whether a country’s openness is high or low. The problem with this conclusion is that these systems differ immensely, and different policies and rules are implemented (Cummins and Rodriguez, 2010, p. 283). The Gulf countries employ a very large amount of lower-skilled migrant workers because they are considered very cheap, they have almost no rights, and employers exercise a large amount of control and power over them. This explains why they have a very high amount of issued work permits. It is unfair to compare these numbers to the amount of issued work permits in a liberal democracy. In a liberal democracy it would be almost impossible to keep migrants workers very cheap, decrease their rights immensely, and exercise too much control and power over them. Therefore, the numbers in liberal democracies are a lot lower, hence Ruhs’ trade-off theory. It would be more helpful if Ruhs explicitly stated what he views as a high or low number of issued work permits for a liberal democracy, instead of comparing it to the immensely high numbers in the Gulf countries.

The last critique regarding how Ruhs looks at rights is that granting or restricting some rights influences other rights. You can give a migrant worker the right to join a union on paper, but when you are tied to one employer, this also influences your ability to join a union. It is not hard to imagine that workers would rather not join a union whenever their employer has a lot of power over them, their work visa, and whether they can stay in the country or not.

From this literature review it becomes clear that most authors agree that states limit the amount of offered rights to migrant workers to keep them cheap and to control migration flows. Offering more rights would lead to states and employers being less willing to employ these workers. The general view is that a low openness and a low amount of offered rights is non-ideal for lower-skilled migrant workers and can cause cases of abuse. A low openness is disadvantageous for lower-skilled migrant workers because this limits their work opportunities. The literature mostly talks about how the trade-off is visible in the Gulf countries and how a low amount of offered rights is not beneficial to migrant workers, and leads to exploitation. It is assumed and expected that countries like the Gulf countries, that mostly have authoritarian systems in place, make use of a trade-off in order to keep workers cheap and to employ a large amount of them. However, the literature and theories do not discuss how a liberal democracy can combine a high amount of rights with a high openness, which would be the most ideal situation for lower-skilled migrant

(16)

workers. Rights and protecting the freedom of individuals is a core principle of a liberal democracy. Protecting the rights of migrant workers should be a core principle, but how to achieve this is hardly discussed as well as how this can be combined with a high openness. No literature was found contesting Ruhs’ statement that the price of rights is the main explanation for the amount of rights states offer to migrant workers.

Lenard and Straehle point out that we can keep guest worker programs functioning while better protecting migrant workers by increasing rights. This will cause states to work harder to protect their future citizens, and it gives migrant workers the ability to fight for their own rights. The puzzle that arises is whether it is possible to uphold a high openness and offer a high amount of rights in a liberal democracy towards lower-skilled migrant workers to protect them better. Lenard and Straehle were the only authors discussing offering a larger amount of rights to migrant workers combined with a high openness, and how this is beneficial. This paper wants to build further on their theory. Therefore, the research question is:

Is it possible to uphold a high openness and offer a large amount of rights towards lower-skilled migrant workers in a liberal democracy in order to minimize exploitation?

A research regarding a high openness and a high amount of rights could also reveal whether the price of rights is the main explanation for the amount of offered rights.

(17)

Theoretical framework

In this theoretical framework important variables and concepts will be discussed that are relevant for this research. What they entail, what their significance is for the research and how they are connected will be discussed. The three important variables for this research are: migrant rights, openness and exploitation.

Different skilled migrant workers

As discussed in the literature review, the trade-off is most prevalent when discussing lower-skilled migrant workers. Therefore, they will be the main subject of this research. Abuse and exploitation is most noticeable with these workers because they are already vulnerable and in most cases do not have better work options. It is important to mostly focus on the rights of these workers and how the rights trade-off affects them in order to better protect them.

Migrant rights

Migrant rights are: the social/economic/political rights migrants receive in the country of employment. The rights for regular migrant workers that are laid out in the ‘Protection of All Migrant Workers and Members of their Families’ (UNHCR, 1990) convention are:

● The right to life

● The right to be free from forced labor

● The right to equality with nationals before courts and tribunals ● The right not to have identity documents confiscated

● The right to equal treatment with regard to remuneration, other conditions and terms of employment, and social security

● The right to join and take part in meetings and activities of trade unions ● The right to form associations and trade unions

● The right to equal treatment with nationals in relation to access to education institutions, vocational training, housing (including social housing), and social and health services ● The right to seek alternative employment in case of the termination of a work contract

prior to the expiration of the work permit

● The right to freely choose remunerated activity after 5 years of residence in the host country

● The right to equality of treatment with citizens in respect to protection against dismissal, employment benefits, and access to public work schemes intended to combat unemployment

(18)

● The right to redress in case of violation of the terms of the employment contract (Ruhs, 2013, p. 15).

It is helpful to divide rights into different categories: political or civil, economic or labor and social (Lenard & Straehle, 2010, p. 212). Political rights give people the ability to participate in a country’s political system and government. In some guest worker programs guest workers get these rights after a certain amount of time. In some programs, they do not have a chance to ever get these rights (ibid).

Civil rights are frequently thought of as rights that people are granted simply for being human. These rights can be seen as independent from citizenship. Sometimes civil rights are also seen as political rights because without civil rights it is almost impossible to participate in a community. Civil rights include free expression and the right to free association (ibid).

Social rights are the most discussed when discussing which rights we should grant to guest workers, because they are hugely influenced by citizenship (ibid). They are seen as a human right or as a right you earn through labor (Berg & Spehar, 2013, p. 157). Social rights include welfare benefits, health care and education (ibid).

Economic or labor rights are rights that protect people in the market and workplace. This includes fair wages and the right to join trade unions. Lenard and Straehle also point out that it is hard to define rights in certain categories because they can be included in multiple. They use the right to join trade unions as an example to explain that this right can be seen as an economic as well as a political right (2010, p. 212).

Openness

For the conceptualization of the term openness Ruhs states: “Refers to policy openness, defined as the number, scope and strength of restrictions that nation-states put in place to regulate the inflow of migrant workers, something that is not necessarily the same as the actual scale of migration to a particular country” (Ruhs, 2013, p. 49). The more restrictions there are in place to control migration flows, the less open a country is towards migrant workers. To calculate this, Ruhs looks at different indicators, like certain policies, to decide how open a country is towards migrant workers. In Ruhs’ definition he says that openness is not the actual scale of migration to a particular country. He hardly looks or makes use of numbers and mostly analyses policies. For this research, the amount of work permits will be analyzed which differs from Ruhs’ conceptualization of openness. By looking at these numbers we can tell whether policies that are meant to increase a country’s openness also actually increases the amount of workers that get employed. This differs from Ruhs’ conceptualization of openness because by

(19)

looking at policies alone you are unable to tell if they are actually working, and if the policy is actually providing more work to migrant workers.

Ruhs studied different indicators in each country and their migrant labor program in order to calculate their openness number. The indicators he used were the amount of demand-side and supply-side restrictions that the country had implemented to control migration (Ruhs, 2013, p. 217-219). By looking at these restrictions he decides how open a country is. The higher the number, the more open a country is. The actual openness numbers Ruhs calculated for all the guest worker programs will be discussed in the case selection part of this paper. In the literature review it is stated that Ruhs’ openness conclusions are lacking because of his indicators. Some indicators can be viewed as a policy as well as a right. Therefore, wage parity/minimum wage, sectoral/occupational restrictions and the right to change employers will be viewed as both. The Indicators to measure openness are:

Demand-side restrictions

Job offer: Some guest worker programs require the migrant worker to have a firm job offer

before he is allowed to move and work in the receiving country. In these cases, it is the employer rather than the worker that initiates the work permit. This is a way to control migration flows and to make sure that only employed workers get into the country (Ruhs, 2013, p. 16).

Labor market test: These tests are mainly meant to protect and prioritize local workers. For most

guest worker programs this means that the employers first need to advertise the job to local workers before trying to employ migrant workers (ibid).

Sectoral/occupational restrictions: Migrant workers can only be hired for a certain sector or

occupation (Ruhs, 2013, p. 61).

Economic work permit fees: Ruhs states that all programs that require firm job offers have

employers pay work permit fees when hiring migrant workers. These fees are used to regulate employers’ incentives to hire foreign workers (ibid).

Wage restrictions: Ruhs states that wage restrictions are a very efficient way to control migration

flows and to keep foreign labor affordable. He states that the most open wage restriction towards migrant workers is to hire them at the same minimum wage as local workers. The most restrictive policy is to offer migrant workers wages that are collectively agreed on in wage agreements (Ruhs, 2013, p. 62).

Trade union involvement: Trade unions’ main concern is to protect local workers as much as

possible. Therefore, in some cases, trade unions have control over the employment of migrant workers to make sure foreign labor does not negatively affect local workers and their wages (ibid).

(20)

Supply-side restrictions

Nationality and age restrictions: Some countries develop programs that require the migrant

worker to have certain characteristics to get employed. Bilateral labor agreements, for example, give preference to workers from particular countries (Ruhs, 2013, p. 63).

Gender and marital status restrictions: A small group of countries will put gender and marital

status restrictions in place to prevent women from performing dangerous work. Under point-based systems marital status can matter because a spouse with a certain skill set will be provided with extra points (ibid).

Skills requirements: A skill set can refer to a certain type of education, qualification or work

experience that is needed for a work permit or to get hired for a job. Guest worker programs that are aimed towards higher-skilled workers, for example (Ruhs, 2013, p. 64).

Language skill requirements: Some countries require migrant workers to speak a certain

language in order to get hired for the job (ibid).

Self-sufficiency: A lot of guest worker programs want the migrant worker to be self-sufficient.

This means not relying on public funds to be able to take care of themselves and their families. In some cases this means that migrant workers need to prove that they have a certain amount of savings in their bank account (ibid).

Supply-driven and demand-driven systems

It is important to understand the difference between supply-driven and demand-driven systems, the two systems most commonly used to select migrant workers.

In supply-driven systems the state is the one that selects migrants already knowing how many workers they need, which skills, and how long they are needed. With these requirements in mind the state will select workers and how many can come to work. Most countries use a point-system to determine if a worker has the right requirements. The worker will be selected based on, for example, education, work experience and age. If a worker ticks off most of these points, they get offered a work permit to work in the receiving country, and in most cases a permanent residence permit. Quotas are mostly determined by the state and used to determine how many foreign workers can get employed. Countries like Australia, Canada and New Zealand use the points system to select the most equipped workers (Emilsson, 2016, p. 44).

In demand-driven systems the employers are the ones choosing migrant workers and decide for themselves how many and which skill sets are needed. In these systems a labor market test is used to set up certain criteria. There have to be unfilled jobs and a certain salary has to be provided. After approval, a temporary work permit is usually granted which ties the foreign worker to his employer in the receiving country. After a certain amount of time, the

(21)

temporary work permit can be changed into a permanent residence permit. Demand-driven models can vary according to who can grant work permits: the employers, a group of employers, state representatives or the trade unions (Facchini & Lodigiani, 2014, p. 45). Employers exercise a lot of power whenever it is up to them to decide who gets a work permit. In most cases, workers are depending on their boss for a work permit and visa. This is why “demand-driven labour migration models tend to be open to abuse and employer manipulation” (Emilsson, 2016, p. 55).

Universalism

When discussing which rights migrant workers should receive the term ‘universalism’ is helpful to understand. Universalism has to do with who gets what rights. It is believed that some ideas like rights should apply to everyone. Some Nordic welfare states are known to practice solidarity because all members of society are part of the welfare system. These states are described as universalist because all members of society are contributing to its welfare state, as well as receiving from it (Johansson, 2014, p. 100). Concepts like universalism help explain who has access to a state’s welfare system.

Exploitation

This research is important because abuses and violations are taking place regarding lower-skilled migrant workers. It is therefore important to conceptualize exploitation. Lenard and Straehle discuss four different views on the exploitation of migrant workers and what exactly causes these exploitations:

1. The vulnerability of workers: migrants get exploited because of their vulnerable position as a guest worker and the fact they do not have other work options. They accept conditions that are not good for them. This view sees guest worker programs as exploitative because workers are trying to get better work opportunities in another country and therefore lack bargaining power and accept unjust work environments. 2. Poor work conditions: this view highlights the inability of a state to better protect guest

workers and the violations of a work contract they go through.

3. Exploitative contracts: this view states that the contracts migrant workers sign are the core source of exploitation.

4. Exploitative bargain: this view points out that in most cases migrant workers do not get the chance to ever become a permanent citizen. Under these circumstances, migrant

(22)

workers are always seen as second-class citizens and treated that way, with chances of deportation whenever their work contract ends (2010, p 285).

In the analysis these four views will be discussed to understand what causes exploitation in the chosen case. These four points also make it clear whether a high openness and a high amount of offered rights offer the ideal situation for migrant workers, and whether it minimizes

(23)

Case selection and methodology

This research will consist of a case study regarding a liberal democracy that is considered open towards migrant workers as well as offering them a high amount of rights. A case study can reveal if and how it is possible to balance a high openness and rights and whether this leads to less exploitation.

For this case study a liberal democracy is needed that offers a high amount of rights while still maintaining a high openness. Ruhs discusses short case studies in his book ‘the Price of Rights’. Sweden stands out immediately because it seems like they are the exception to Ruhs’ trade-off theory. While looking at Ruhs’ calculated openness numbers, Sweden, especially compared to other liberal democracies, scored very high. The first thing that stood out is that Sweden is almost the only liberal democracy offering a general program including lower-skilled migrant workers that is not a seasonal workers program. The second thing that stood out is that Sweden’s openness number is almost as high as countries that have the kafala system in place. The first table shows Sweden’s openness number compared to other liberal democracies with programs in place directed towards lower-skilled migrant workers from Ruhs’ research.

(Ruhs, 2013, p. 203-205)

Country Program Openness number

Canada Seasonal agricultural

worker program 0.500

Canada Low-skilled Pilot Project 0.639

France Seasonal workers 0.667

Greece Seasonal work permit 0.667

Italy Seasonal 0.569

Netherlands General labor scheme 0.639

New Zealand Seasonal workers 0.639

Norway Seasonal worker 0.667

Spain General regime 0.625

Sweden General work permit 0.722

United States H2A 0.708

(24)

Canada, the Netherlands, Spain, Sweden and the United States are the only countries with a program in place that is targeted towards lower-skilled workers which is not a seasonal workers program. Because seasonal worker programs are mostly holiday jobs that give workers a very short time in the host country and limited rights, they are not the right programs to analyze for this case study.

Eventually, the case of Sweden appeared to be the perfect case for this research because of a couple of reasons. First, Sweden is, besides Spain, the only liberal democracy with a general work program in place targeting workers of all skill levels with the same requirements for everyone. This is unlike the program in the Netherlands, which has different rules for lower-skilled and higher-skilled workers requesting a work permit (Government of the Netherlands, 2019), and the United States, that has different programs in place for different skilled workers, offering different rights (DPE, 2017). All foreign workers get the same job opportunities and rights in Sweden. Sweden is preferred over Spain because it is stated in Ruhs’ book that a change in labor migration in 2008 dramatically increased Sweden’s openness towards migrant workers (Ruhs, 2013, p. 114). This policy reform went against some of the dominant theories about immigration policy that predict a Nordic welfare state to use a selective policy instead (Emilsson, 2014, p. 3). This policy is perfect for this research to find out how this has affected offered rights towards lower-skilled migrant workers. Sweden is also a perfect case for this research because it has provided lower-skilled migrant workers with a high amount of rights for many years. The Scandinavian welfare states seem to have found a way of providing a high openness and a high allocation of rights to migrant workers: “The Scandinavian welfare states have tended to define the principle of universalism in an inclusive manner, granting legal migrants access to a package of social benefits and services” (Johansson, 2014, p. 101). There seems to be a general agreement that Sweden has a very high openness towards migrant workers after these policy changes. The OECD calls their system “the most open in the OECD” (OECD, 2011, p. 11).

When analyzing Ruhs’ calculated openness numbers, Sweden being considered as open as the Gulf countries is a surprising outcome. Sweden’s openness number, that is comparable to the numbers in the Gulf countries, also makes Sweden more suited for this case study than Spain. It is already discussed how in the kafala system hardly any rights are granted towards migrant workers and according to Ruhs’ research, this keeps migrant workers cheap, and therefore these countries have a very high openness number.

(25)

(Ruhs, 2013, p. 203-205)

Although Gulf countries offer a very low amount of rights to lower-skilled workers, and Sweden offers a high amount of rights, they both have, according to Ruhs, comparable openness numbers. This already shows that Sweden is an outlier case in Ruhs’ research. Ruhs himself also points out that Sweden seems to be an outlier case: “It combines a liberal market economy with an extensive state-run welfare state. Most comparative analysis of social policy consider Sweden the archetype ‘social democratic welfare state’ that aims at universal coverage along with rights and benefit equality” (Ruhs, 2013, p. 100). It shows that Sweden is the perfect case to find out how it is possible to combine a high openness with offering a large amount of rights towards lower-skilled migrant workers, and whether this is the ideal scenario to minimize exploitation.

By analyzing qualitative as well as quantitative data, like government reports, NGO reports, political debates and literature, it will be made clear why Sweden chose certain policies regarding openness and rights, how they were able to implement these policies, and how they are able to balance openness and rights. A case study will help understand and explore the mechanisms that influence openness and the amount of rights migrant workers receive.

Using qualitative as well as quantitative data means analyzing texts, reports and statistics to make a case. For the qualitative data, in-depth reports, like NGO and government reports, will be used in order to understand which policies and mechanisms were used to achieve a high openness while still offering a high amount of rights. NGO and government reports will be analyzed that are available online to analyze the selected case of Sweden. Detailed and trustworthy reports from NGOs that write policy reports like the Organisation for Economic Co-Operation and Development (OECD) and the European Migration Network (EMN) can be found online. Both of these organizations have written reports on the 2008 labor reform in Sweden. The DEMIG database will also be used. It tracks which migration policy changes were enacted by 45 different countries, including Sweden. This database will be helpful regarding this case study to research the different implemented policies regarding migration and

Country Openness number

Sweden 0.722

Kuwait 0.708

Saudi Arabia 0.722

Oman 0.722

(26)

how this has affected Sweden’s openness (DEMIG, 2015). Literature that discusses Sweden, its migration policies and offered rights will also be included in order to understand the case better.

Quantitative data showing the number of migrant workers will be analyzed. This data shows the amount of migrant workers that have been hired throughout the years and the amount of accepted work permits. This information is available on the Swedish Migration Agency’s website. These numbers can clarify whether policy changes that affect openness and rights have an actual effect on the number of migrant workers that get employed.

The occurrence of data in the Swedish language was anticipated. To minimize this problem, attention was mostly focused on finding data about Sweden through international organizations, English databases and English literature. The 2008 labor reform in Sweden has been researched by many organizations and authors, providing enough English data to analyze. The number of issued work permits was only available on the website of the Swedish Migration Agency but these tables were also translated to English.

Seawright and Gerring define a typical case study, which is used for this research, as a case where a cross-case relationship can be analyzed. The case is often described as a representative case for the specific research question. For this research, it is a case where a country offers a high amount of rights and is still considered open towards lower-skilled migrant workers. A case study gives the researcher the possibility to delve into a case to understand the relevant causal mechanisms that explain the cross-case relationships more. For this research, it is the relationship between openness and rights, and how this relationship affects exploitation. The goal of a case study is to find out whether the causal mechanisms in the chosen case are different than the ones according to the theory, or whether they are the same. A case study can also provide evidence that shows that the causal mechanisms do not explain the cross-case relationship, and therefore there are other mechanisms at work, leading to a different causal relationship contesting the existing theory (Seawright & Gerring, 2008).

A case study is preferable to other ways of research to answer the research question because it provides the ability to analyze causal mechanisms and what specific policies are used to balance out openness and rights, and which specific cases of exploitation have occurred. By analyzing a specific case it can also be specified which contextual factors influenced the variables and outcomes.

With this case study and the selected data conclusions will be made regarding openness, offered rights and exploitation, and how they influenced each other in the case of Sweden. With the help of the selected variables, indicators and theories from the theoretical framework the data will be analyzed to come to a conclusion.

(27)

Analysis

This next part contains the analysis of the reports and literature that were found on Sweden. First, labor rights and Sweden’s openness towards migrant workers throughout the years will be discussed. The labor reform that took place in 2008 in Sweden is an important part of this research and analysis. Political debates will be analyzed to find out why Sweden choose to implement certain policies regarding migrant labor, as well as debates about openness and rights. Implemented policies and which changes Sweden had to go through in order to change their labor migration will then be discussed, as well as how they were able to balance a high openness and rights. Finally, it will be analyzed whether changes in openness and rights have influenced the occurrence of exploitation towards lower-skilled migrant workers.

Overview of labor and rights policies in Sweden

Sweden had a surge of labor immigration in the 1960s that went along with offering migrants easier access to citizenship and granting foreign workers rights close to that of Swedish citizens (Mešić, 2017, p. 23; Brochmann, Hagelund, Borevi, Jonsson & Petersen, 2012, p. 35). Sweden was the first European country that made it possible for migrants to vote and run for office in elections in 1976. These policies were put in place because Sweden wanted to underscore and emphasize their multiculturalist society (Mešić, 2017, p. 23; Focus Migration, 2009; DEMIG, 2015).

The Swedish welfare state has for a long time been viewed as exceptional in the way they grant universal benefits to all workers. The view is that citizens should enjoy the same rights equally under all circumstances and not be dependent on the market. But this makes their welfare state also extremely expensive. From the 1950s to the 1970s their welfare state evolved to keep this system up. It is described as a three-pillar system. Their wage policy, the first pillar, is seen as the world’s “most centralised collective bargaining system, between strong-blue collar unions and employers associations” (Mešić, 2017, p. 24). To keep unemployment as low as possible, they invested in re-qualification programs to ensure the labor market participation was kept high. The second pillar is providing citizens with high welfare benefits including: “child and elderly care, healthcare, education at all levels, state subsidized pensions and unemployment insurance” (ibid). The third pillar is upholding a strong social democratic political system that gains support by providing the benefits in the second pillar to all citizens (ibid). In this system, because of the centralized collective bargaining system from the first pillar, trade unions are very powerful actors in the Swedish labor market.

Sweden always enjoyed a high union member density but this “declined from a peak of 85% in 1995 to 71% by 2012. During the same 17-year period, union density among blue-collar

(28)

workers had fallen by more than one-fifth, from 88% to 68%” (Woolfson, Fudge & Thörnqvist, 2014, p. 696). A fall in union membership is causing more non-ideal work environments regarding low-skilled and low-wage jobs that are becoming increasingly reliant on migrant workers (ibid).

Most EU countries have minimum wages that are negotiated with trade unions, and in most countries wage negotiations happen within the different companies. However, in Sweden there is no minimum wage. Negotiations over wages and working conditions are handled by local trade unions that negotiate these terms for everyone within a certain sector. This causes wages in sectors to be the same for big companies as well as smaller companies. Swedish trade unions can influence wages in companies without union members and legally force companies to industry agreements when they are part of an employer’s association. Even when a company is not a member of a trade union, the trade union “has the right to take industrial action to force the employer to either apply for membership to the relevant employer’s association or to sign the most relevant existing agreement. Most importantly, if there is not an existing local/firm-level collective agreement, the union and the individual employer can settle an affiliated agreement that for the most part is a blueprint of the industry-wide agreement” (Woolfson et al., 2014, p. 698). The trade union can boycott the company whenever the company refuses to do this. The trade union can also call for a strike of its members or block the company if it does not have any trade union members. Trade unions mostly try to bind workers to them by offering workers better conditions than when they are not part of a union and try to prevent non-union members of accepting wages that are lower than those of union members. But an increase in union dues has resulted in a decline in union membership (Woolfson et al., 2014, p. 699).

The 2008 labor reform in Sweden

In Sweden’s previous labor system, it was very hard for employers to hire migrant workers from outside the EU. Every work permit had to go through labor market tests and the work permit had to be analyzed and accepted by the Swedish Migration Board (OECD, 2011, p. 57). At the end of the 60s, workers could only obtain a work permit if they could proof that they had housing and employment in Sweden. Sweden’s strict approach to granting work permits that started in the 1970s up until 2008 contributed to a very minimum granting of work permits to migrant workers (OECD, 2011, p. 57; MPI, 2006; DEMIG, 2015). The 2008 reform was introduced after debates of future labor shortages because of the retirement of the baby boom generation. With this new reform the entire labor system became demand-driven, and it is up to the employers whether they want to hire workers from abroad. The only rules they have to abide to is that they have to advertise the job under local workers first, before looking for migrant workers, as well as

Referenties

GERELATEERDE DOCUMENTEN

De locatie en het uiterlijk van deze functies werden echter niet voorgeschreven door de plan- ners van de stad Wenen, die de grootte van het project alleen op een inhoud

BAAC  Vlaa nder en  Rap p ort  239     Figuur 14: CAI‐kaart van het onderzoeksgebied met de archeologische vindplaatsen in de omgeving 36  

The documentary specifically portrays how European states have increasingly come to make use of surveillance technologies to control people’s mobility at –and beyond- the external

determinants (or driver variables) contributing to the altered state of the fish communities at this site were identified as habitat state alterations – including

The study reveals that the OEM needs to focus on employee satisfaction, improvement regarding the quality of their service exchange component repairs, the elevation of the

Superfoods zijn natuurlijke producten, dus op basis van deze onderzoeken wordt er verwacht dat supermarkten gebruik maken van het natural goodness frame, waarin

gewys is, kan hierdie 'n be· hierdie laaste twee sektore se perkende faktor word, maar ontwikkelingspotensiaal in ag aangesien die interafhanklik· geneem word,

Daarnaast is het van belang te weten wat voor installatie het gebouw gebruikt voor het verwarmen, dit omdat daarmee direct bekend is wat de standaard opties zijn die