EU migration,out-of-work benefits and reciprocity: Are member states justified in restricting access to welfare rights?

Introduction

Immigration is a permanent fixture in current debates about the nature and the future of the EU. Freedom of movement and provisions for non-discrimination make skills-selective immigration difficult to implement within the EU. However, UK-EU negotiations before and after the UK referendum brought forward arguments to undermine the pertinence of both principles for the EU and to add qualifications to their implementation. One measure is significant restrictions on the access to welfare rights for EU migrants to discourage low-skilled EU migration and so-called benefit tourism (HM Government, 2014 ) . Are member states justified in restricting access to welfare rights for EU migrants as a matter of justice? What normative reasons are used to justify these restrictions?

This article examines whether the restrictions are appropriate from a normative point of view. The EU is unique and challenging for normative approaches to migration in at least four ways. First of all, there is disagreement about whether the EU constitutes an institutional structure that sufficiently approximates the nation state (Nagel, 2005: 133–143; Van Parijs, 2003: 7–20). The EU clearly resembles the institutional structures of a state in some policy areas but not in others, so resolving disagreements is hard and drawing relevant thresholds of sufficient resemblance can be challenging and arbitrary. Second, and more significantly for our discussion, the EU has no internal border checks. Hence, it physically permits freedom of movement for the citizens of its member states. But also in this case this right is qualified in ways that freedom of movement within its member states is not (De Witte, 2015). Those seeking work in other member states and those who are not economically active or self-sufficient are increasingly subject to restrictions on its exercise (European Union, 2004b). Third, and relatedly, an analogy between EU workers and immigrant guest workers is a difficult one to draw, as EU law gives all EU citizens who are working the same rights as nationals (European Union, 2004a). Nonetheless, those seeking work and those who lack the means to support themselves are entitled to a significantly smaller bundle of rights than nationals and their bundle of rights increasingly resembles that of immigrants from third countries (Attas, 2000). Fourth, EU immigrants are citizens of relatively developed countries so discussions of brain drain in the EU are less pressing (Brock and Blake, 2014; Oberman, 2013). Again, however, differences in economic and social development among member states create incentives for movement across borders that could have similar brain drain effects, especially during economic crises. These four dimensions call for a tailored answer, drawn from the current literature on the ethics of migration but attuned to the special circumstances of the European Union.

This article critically examines one of the most promising arguments in the context of the four dimensions. This is the claim that reciprocity-based considerations justify restrictions on access to the welfare state of host countries for both EU students and jobseekers, as well as more generally for EU citizens who have not made enough relevant contributions to the host member state (Sangiovanni 2013, 2017, forthcoming; as well as Carens, 2013: 281; Miller, 2016). Until then, the argument goes, EU immigrants must not be treated as full members of the host society’s system of social cooperation and their case must not trigger considerations of reciprocity. This article shows that a concern for reciprocity does not deliver a valid argument for restrictions on EU immigrants’ access to welfare rights.

The structure of the article is as follows. The second part outlines the restrictions to welfare rights for EU immigrants and the reciprocity-based arguments in favour of them. The third part identifies four problems with these arguments and shows the inherent dilemma of restrictions: advocates either have to abandon a restricted conception of reciprocity at the price of granting immediate access to welfare rights for EU immigrants or exclude EU immigrants at the price of also excluding nationals from the ambit of domestic social justice. The fourth part examines a number of possible objections, while the fifth section concludes.

The argument for restrictions

Migrants’ access to welfare is a subset of a wider literature on the ethics of migration in political theory (see e.g. Obermann and Miller in Fine and Ypi, 2016). This debate is overshadowed, if not superseded, by the reality of the EU, its laws and its institutions. EU member states have already agreed to lift restrictions on freedom of movement reciprocally, not least because the economic stability of their common economic area, where freedom of movement of labour, capital, goods and services reigns supreme, demands it (De Grauwe, 2012). This constraint clearly renders the EU a special case. Physical barriers to migration – border controls – have been removed in the EU. Therefore, restrictions or qualifications to freedom of movement must be exercised by other means, mainly economic, but also discursive, and their invocation by member states raises the issue of their justification. ¹ Another set of questions seems more urgent in the case of the EU. Are they justified in implementing policies that de facto limit access as a ‘softer’ means to implement a selective immigration policy? Are member states justified in restricting access to welfare, through measures such as waiting periods, to prioritise and channel resources to their least advantaged nationals?

Justifications as to whether and when immigrants should be granted welfare rights fall roughly into two categories. On the one hand, there are those who argue that what predominantly matters is that migrants stand in a particular type of relationship to the host state. Migrants who reside outside their own state are subject to, and affected by, coercive laws and public policies that they cannot influence due to their restricted access to social and political rights (e.g. Lenard and Straehle, 2012; Walzer, 1983). The counter-argument provided by another group of theorists is that this is not a sufficient condition for triggering duties characteristic of social justice. The relationship in question needs to be pervasive and enduring (e.g. Blake, 2002; Shachar, 2011): it is not merely the type but also the degree of the relationship that matters for the triggering of duties characteristic of welfare rights. ² Is the EU immigrant’s relationship with the host state of a type and/or of a degree that justifies current restrictions on access to welfare rights?

The literature on access to welfare rights is scarce, particularly for non-economically active migrants. This is primarily because the loci classici in this literature depart from the assumption that host states have the right to grant admission to non-citizens on the basis of their expected economic contribution. Therefore, discussions of immediate access to welfare rights in this literature are usually limited to migrants who have met such selection criteria. ³ Even in this case, however, both David Miller (2016: 125, 120–121, 124–125) and Joseph Carens (2013: 281, 118–122, 280–283) agree that the host state can postpone access to welfare rights to ‘underline the reciprocal nature of the scheme’ and to prevent EU immigrants from taking ‘immediate advantage of the receiving state’s more generous welfare programs’, respectively. But if probationary periods are permissible for those whose right to reside is based on their potential economic contribution then it seems to follow that at least the same, if not more strict, probationary periods are justified for those who seek work. ⁴

The common denominator behind these calls for restrictions is the view that immigrants must contribute for a significant period of time to the society that hosts them before they are entitled to welfare rights. This argument echoes concerns among citizens in the UK, the US and some European countries that immigrants must ‘pay in’ before they ‘get out’. ⁵ It makes debates about economic migration more politically salient and urgent. However, if this consensus is unqualified, then both the political theory and public policy of EU migration need to be revised. Thus, the question is: should mobile EU citizens have the same access to welfare rights as nationals? If the answer is ‘no’, then restrictions on the access to welfare rights are justified; if the answer is ‘yes’, then restrictions are unwarranted. ⁶

The current literature on immigration and access to welfare in the EU in political theory is limited. A popular position is that newly arriving EU migrants in work should accrue the same benefits as nationals but that there should be a waiting period before newly arriving EU immigrants can access out-of-work benefits (Bellamy and Lacey, 2018; Sangiovanni, 2017). Following the aforementioned trend in the literature on the ethics of migration, this recent work draws on the relationship between duties of reciprocity and the right to access the goods that welfare rights are meant to safeguard. This link could motivate an argument in support of restrictions on welfare rights. The argument from reciprocity used to support this thesis is: ⁷

Reciprocity in the mutual provision of the basic collective goods condition the content, scope and justification of welfare distributive equality in general and welfare rights in particular. ⁸

If the above argument is sound, then the conclusion that follows is:

Argument for Restrictions (AfR): EU immigrants who have not contributed enough over time in one of the ways necessary to sustain the state must not be beneficiaries of the duties of reciprocity characteristic of welfare rights. Therefore, access to welfare rights for EU immigrants must not be immediate but rather subject to restrictions, such as waiting periods, that correspond to the required level of contribution.

A few points of clarification are needed at this point. First, this article does not challenge the foundational premises of reciprocity-based approaches to social justice, but rather questions the conclusions drawn from it with respect to transnational justice in the EU, and more specifically restrictions on access to welfare for EU migrants. Therefore, steps (1), (2), (3), (4) and (5) of the argument above are not necessarily in dispute here. Those who are not committed to reciprocity-based conceptions of justice or to contribution-based accounts of social justice more broadly are invited to read what follows as an immanent critique and not as a qualified defence of the necessity of a reciprocity-based approach to social justice. Those who believe that states must reserve the right to exclude economic immigrants from their territory on grounds of national self-determination, as well as those who reject welfare rights as incompatible with justice, ¹⁰ will find little in this article to change their minds.

Second, this article shares with proponents of the reciprocity-based approach a commitment to practice-dependent political theory. This has a key advantage in the context of the EU: it relies heavily on an account of justice that treats the state as the primary terrain of principles of social justice. Therefore, steps (6) and (7) of the argument for restrictions (AfR) are not challenged. Hence, if successful, the criticisms of the third and fourth sections of this article deliver an argument for granting full and immediate access to welfare rights for EU immigrants without appealing to more demanding cosmopolitan understandings of social justice and the EU. If the argument in this article is successful, then statists who believe that the causes of wealth and poverty of member states are internal to a system of social cooperation and not influenced by external factors must also endorse it. In that respect, the analysis is also compatible with adopting explanatory nationalism as the explanans behind current inequalities among the EU’s member states. If valid, it converts (some) statists to proponents of transnational social justice.

Third, I assume for most of this article that international justice is observed in current and historical relations among the EU’s member states. This is to show that any injustices at this transnational level, between nationals and EU immigrants residing within the same member state, are not necessarily the aftermath of injustices between member states at the international level. ¹¹

Four problems for AfR and a dilemma

The reciprocity view in the second section tells us that no transnational obligations exist in the case of welfare rights because the EU is not a globalist project, but rather an international union that aims to strengthen its member states’ regulative capacities via international coordination to make it easier to improve the overall position of their least advantaged citizens. The reciprocity view also tells us that to become and count as a member of a member state’s scheme of social cooperation, a person must make the requisite type and degree of contribution over a certain period of time. Therefore, after that period of time one must be considered a member of a scheme of social cooperation. All nationals have contributed enough over time to be granted access to welfare rights as a matter of reciprocity, whereas EU immigrants have to spend a certain period of time making the requisite type and degree of contribution.

In support of the above conclusion, we need an argument that will show that the type and degree of EU immigrants’ contributions is insufficient to grant them access to welfare rights, but also sufficient to grant access to nationals. Hence, the argument for restrictions (AfR) must be supported by an account of reciprocity that determines: a) what counts as contribution; b) what is the threshold contribution; and c) when an individual becomes a participant in the host society’s scheme of cooperation. The focus of this section is on how different interpretations of the requisite type and degree of contribution to a scheme of social cooperation undermine rather than support either AfR or cooperation-based approaches to social justice as reciprocity.

In this section I show that:

IF justice as reciprocity is taken to imply determining the requisite TYPE and DEGREE of contribution in the way that reciprocity theorists suggest, and if it is needed to reach the conclusion that EU immigrants’ access to welfare rights must be restricted, THEN nationals must also be subject to similar restrictions.

This raises a difficulty for the reciprocity approach. What about those who are too young (e.g. infants and children) to contribute in any meaningful sense, as well as in a deliberate manner, to a system of social cooperation? A successful reply to this objection requires extending further the notion of contribution at play here: such individuals may lack the capacity to contribute at t₁ but they retain the capability to participate later at t₂ when they reach maturity. It follows that a member state can grant rights to welfare to both economically and non-economically active nationals, because all of them have the capability to contribute in different ways as adults (Abizadeh, 2016; Sangiovanni, 2017; Song, 2016).

But does broadening the type of requisite contribution justify access to welfare rights for non-economically active nationals only? If the capability to contribute in different ways is a sufficient condition for granting welfare rights to non-economically active nationals, then newly arriving EU immigrants also qualify, as they clearly also have the capability to do so. If, on the other hand, capability is a necessary but not sufficient condition, because a sufficient degree of contribution over time to a society’s system of social cooperation is also necessary, then it follows that non-economically active nationals must also contribute enough over time to be entitled to social benefits (Song, 2016: 235–237). Hence, they must also be denied access to welfare rights until they have contributed sufficiently. This second condition is consequential for restricting access to welfare rights: EU immigrants, and nationals, can be excluded from welfare state entitlements until they have contributed enough to the domestic system of social cooperation. Hence, waiting periods before accessing such rights are justified.

It is arguable that nationals typically meet one of the three criteria of contribution over the first few years of their adulthood, and that therefore they must be granted access to welfare rights. But again, the same objection is relevant here. EU immigrants, either as individuals or as a group, may reach the contribution threshold more quickly if their average degree of contribution is higher than that of nationals. Therefore, also in this case, if waiting periods are relevant to both nationals and immigrants as a matter of justice, then those periods must be tethered to the degree of contribution to the relevant types. In fact, as we will see below, if waiting periods for nationals must be short, then waiting periods for EU immigrants might need to be even shorter or omitted altogether when they have a higher expected capability to contribute. ¹³

This leads to a conclusion that grants welfare rights to EU immigrants and nationals on the basis of the degree of their contribution via one of the requisite types of contribution. However, it faces problems as a contribution-based account of social justice. First, and regardless of the currency of contribution, it seems plausible to assume that contribution to a scheme of social cooperation varies from one person to another. This is primarily because the capability to contribute varies from one individual to another. So shouldn’t everyone receive according to their individual and varying contribution to a system of social cooperation, irrespective of whether they are nationals or EU immigrants?

Reciprocity views are different from the wider family of contribution approaches on this point. They differentiate between a view of society as a scheme of individual contribution and reward, and one where individual contribution occurs within a system of social cooperation that gives individual contribution its gross value (Fine and Sangiovanni, 2015; Rawls, 2001: 52–54; Rawls, 2005: 265–269; Sangiovanni, 2015). The question remains: why should the state distribute the fruits of social cooperation in an egalitarian fashion? Why not allocate resources according to every individual’s marginal contribution, restricting access to welfare rights and barring those incapable of contributing via their taxes and their political and economic activities from accessing them at all?

The move from the claim that someone is a joint author of the state to the claim that someone is entitled to an equal share of the product of her cooperation (see the move from the 4th to the 5th step of the argument in the previous section), therefore, seems unqualified. It also renders the argument for restrictions vulnerable to a criticism aired earlier: if an immigrant’s individual expected contribution is higher than that of a national’s, shouldn’t the immigrant be granted earlier access to welfare rights than the national on grounds of reciprocity? The response given here by reciprocity theorists is an empirical one: 90% of our income is generated by positive externalities of living in a well-functioning state with a wide and reliable provision of collective goods, rather than by our individual talents and efforts (Sangiovanni, 2015; Simon, 2000). Therefore, as a matter of justice, we are entitled to only 10% of our income as this is our pure or net marginal contribution. The rest of the income falls under the scope of egalitarian justice and can be fully taxed (Sangiovanni, 2015: 355).

One can use this empirical response to move from step 4 to step 5. The average individual contribution is set at 10% of the expected national income, whereas the rate of social contribution is set at 90%. ¹⁴ This evidence motivates the argument for restrictions on access for EU immigrants because they, unlike nationals, have not contributed to the social system of cooperation, which brings about the externalities that account for 90% of the expected income of nationals. Therefore, EU immigrants’ waiting periods before accessing welfare rights must be determined, and their access to welfare rights granted, on the basis of their individual expected contribution to the host society’s scheme of social cooperation. That is, at least until they fulfil the criteria of membership to the scheme of social cooperation and count as co-authors of the relevant externalities (see also e.g. Miller, 2016: 124–125; Yong, 2016: 827). ¹⁵ EU immigrants cannot be beneficiaries of duties of reciprocity in the form of welfare rights until that point is reached. But note that even this argument does not deliver the conclusion sought by AfR. What it justifies is a common waiting period before accessing welfare rights for all nationals who reach the age of maturity, and newly arriving EU immigrants, due to individual invariance in contribution. However, focusing on the social nature of contribution in any scheme of social cooperation still leaves many questions unanswered with regard to EU immigrants’ access to welfare rights.

There are three issues. First, if a person’s gross expected contribution is not equal to her net contribution, then focusing on an EU immigrant’s expected gross contribution during the waiting period overlooks the fact that her contribution is the aftermath of her membership to both her home and host society’s scheme of social cooperation. In simple terms, 90% of that gross value is also a social and not an individual product. Two things follow. First, EU immigrants surely also owe other members of their home member states, including those who may now reside in the same host member state, duties of reciprocity. The host member state could easily facilitate the discharge of those duties during those waiting periods without an international transfer scheme. For example, part of the taxes paid by EU immigrants in work in the host member state could be used to pay the social assistance benefits of EU immigrants out-of-work or on low income, providing immediate access to welfare rights (until EU immigrants qualify as full members of that scheme of social cooperation). Therefore, nationals have a duty not to obstruct the discharge of such duties of reciprocity among EU immigrants and, where necessary and reasonable, to facilitate them. ¹⁶ One need not qualify as a full member of the host society’s system of social cooperation to be owed such access as an instantiation of duties of reciprocity.

Second, EU immigrants’ expected contribution to the host state is by a factor of 90% the product of positive externalities found in the host state and not the product of their individual talent or effort as soon as they start to reside in the host state. They immediately become subject to the same social forces as other members of that system of social cooperation when they state their intention to reside in the host member state. ¹⁷ Treating EU immigrants’ expected rate of contribution as an individual rather than a social product ignores the fact that all forms of contribution take place within a system of social cooperation. If net-individual contribution is low for anyone subject to the rules of a system of social cooperation then this should be the case for both nationals and EU immigrants within that structure. Within those social structures, the degree of net individual contribution is ex hypothesi determined by policies that shape the opportunity to find and keep a job. Not all those capable of contributing via work are able to contribute as much as they would have liked to, or wished to do, due to structural involuntary unemployment as well as irregular and precarious work patterns. Being economically active or inactive in those circumstances is not a matter of individual choice for both nationals and EU immigrants. Hence, the average degree of required net individual contribution should be the same for both groups in those circumstances. Nonetheless, the degree of required net contribution to welfare schemes should not be the same for every individual within both groups. For those with ample opportunities it will need to be higher, and for those with a limited set of job prospects lower, than average; while for those willing but unable to contribute due to structural unemployment the net required contribution could be zero. ¹⁸ Therefore, a non-economically active EU immigrant need not wait to become employed to become a member of the host society’s system of social cooperation. As soon as one is subject to such a scheme of regulations that determine one’s level of economic activity but willing to contribute to and participate in that scheme, one must be considered a member of that scheme and be granted access. ¹⁹

Third, determining the effect of positive externalities on our income due to the reliable provision of collective goods by one’s compatriots doesn’t tell us much about the precise threshold for each individual type of contribution required. However, it crucially and significantly lowers that threshold when applied to an individual contributor regardless of whether one defines contribution strictly in terms of economic activity or more broadly. A net 10% of individual contribution to a scheme of social cooperation implies an ever lower, if any, threshold of contribution over a specific period of time. Even if elapsed time is a good proxy for measuring the degree of contribution, it doesn’t follow that the contribution of EU immigrants must be measured and compared to that of nationals between an entry time t 1 and a stipulated threshold that is approximately met at t 2 . Measuring contribution over a longer period of time makes more sense. This is because one could contribute less than average, or not at all, between t 1 and t 2 , but more than average between t 2 and t 3 or at any other point over time t n . This is crucial for how we compare the contribution of nationals to that of immigrants because EU immigrants tend to contribute significantly more over time. ²⁰ Therefore, a concern for contribution does not warrant the imposition of waiting periods. ²¹ On the contrary, such concerns are compatible with immediate and full access. ²²

Given the problems that these arguments from degree (of contribution) face, one might be tempted to resort to arguments from type (of contribution). For example, one could argue that the three types of contribution are incommensurable in nature, as well as individually necessary and jointly sufficient conditions (with the same or different minimum waiting periods) for granting anyone access to welfare rights. However, this pluralisation of contribution, coupled with socialisation, also implies a complex and long waiting period for all participants, nationals and EU immigrants alike. As a result, some nationals must be excluded from the welfare state, or serve a waiting period, most likely after they reach their 18th birthday. But such a policy would deny social justice to some of the most vulnerable nationals in a critical period. It would also result in such a two-tiered workforce with two different sets of rights (undermining each other’s reservation wage) due to the constant flow and presence of a subset of EU immigrants with no access to welfare. ²³ Both are counterintuitive implications for social justice and a high price for justifying the exclusion of EU immigrants.

A response to these objections could be to broaden the requisite type even further. But once we have started broadening the concept of contribution, why should we limit ourselves to work and taxes, and voting and legal compliance? Surely, these do not exhaust the repertoire of participation to a system of social cooperation that aims to bring about basic public goods or welfare goods in general. ²⁴ Isn’t it then more apposite to also treat other types of contribution, such as education, training, care-giving and volunteering as sufficient conditions for triggering reciprocity on the part of the state? ²⁵ Certainly, these are all ways in which an individual can participate in a system of social cooperation, but contribution-based reciprocity approaches somehow ignore them, depriving in this way not only EU immigrants but also nationals from having their contributions recognised. Such broadening of the type also appears inconsistent with AfR. The broader the set of individually or jointly sufficient conditions of contribution, the less likely that EU immigrants will not satisfy any of them.

The above criticisms could be avoided if one adopts a narrower conception of contribution, but at a price. One could opt for a narrow type of contribution that equivocates contribution with economic input. Despite being reductionist and too restrictive with regard to contribution, this has two clear advantages. First, it makes it easier to measure the contribution of immigrants, so it avoids problems of indeterminacy. Second, the link between those who fund via their activities the welfare state and those who enjoy its provisions becomes clearer. Saying that someone has a right to housing benefits (i.e. to a welfare system good) because they voted in the last election or didn’t break the law (i.e. contributed to other types of collective goods) may sound like a huge theoretical jump to many (Arneson, 2014).

This is a plausible defence, but it has two serious consequences. First, under stringent contribution requirements a significant number of worse-off nationals would be outside the scope of domestic social justice. Second, at the group level, EU immigrants are as likely as nationals to contribute to the host society and economy (Dustmann and Frattini, 2014; Dustmann et al., 2010, 2012). Since according to this narrow conception of contribution it is not the type of contribution (as economic input in this case is the only type of contribution) but the degree of economic contribution that matters, it follows that EU immigrants will be as likely to be granted access to the welfare rights as nationals – if not more, given their higher expected average contribution over time due to their higher average skills.

This raises a dilemma for advocates of AfR. They have to accept either the broadening of the type of the contribution at the price of loosening restrictions, or a narrow type of contribution at the risk of excluding a greater number of nationals from welfare rights than EU immigrants. If the above is correct, then AfR can’t convincingly link contribution to restrictions on access to welfare for EU immigrants. Narrowing what falls under contribution, and hence under a conception based on reciprocity, will have adverse implications. It will deny welfare to citizens who are ‘economically inactive’ and most in need. At the same time, under plausible empirical assumptions, it is likely to make it easier for the average EU immigrant to access welfare than the average national. Broadening what falls under contribution appears to be more promising, but it is unclear whether it can justify restrictions on access for EU immigrants by the force of its own argument. We can conclude that further conceptual clarity comes at a heavy price for AfR.

Objections

In this section, I consider a number of objections. First, that the proposal is overtly conservative because it is considered within the current institutional setting and upon contribution-based conceptions of justice. Instead of challenging restrictions on access to benefits as direct violations of global and egalitarian cosmopolitanism, this approach is not far from the status quo; instead of challenging the structural injustices that global capitalism imposes on EU workers by compartmentalising them into a two-tiered citizenship and assigning them to groups with different sets of rights within the same labour market, it proposes rather timid reforms to EU law and institutions.

However, the reciprocity-based view sees the EU as an institution that aims to strengthen the capacities of its member states, including their capacities to realise socioeconomic rights, and not as a union intended to realise international equality of opportunity among its member states. The latter might be a noble ideal, but it is clearly beyond the scope of a reciprocity-based and practice-dependent theory of justice for the EU. Furthermore, welfare is not merely a means to compensate people for adversities produced by a capitalist economy, but is also a way to empower them by improving their bargaining power and to make it easier for them to contest structural injustices. ²⁶ Access to welfare is a structural issue because it strengthens the power of labour relative to capital. In the absence of welfare and facing the prospect of unemployment, EU immigrants are more likely to accept precarious and low-paid jobs. A two-tiered workforce serves the structural interests of capital as a collective agent but not the interests of workers as a collective agent. The aim of this article is to transcend existing practices while staying committed to the broad principles already embedded in the EU’s institutions. This is not because I endorse one particular way of theorising over others, but because I want to start from the same methodological premises as proponents of AfR but deny that AfR follows from those premises. There is one more reason, however, that defenders of more ideal forms of theorising have to engage with: if a moderate version of justice at t 1 is practically necessary to achieve more justice of the egalitarian type or other varieties at t 2 , then those who subscribe to those views of justice must strategically, and in that sense instrumentally, endorse the reciprocity-based view sketched in this article. This is because if the above is true, then the proposal sketched here becomes instrumentally necessary for achieving what those conceptions of justice demand. ²⁷

Second, one could ask whether an EU-wide brain drain tax, or similar, could be an alternative to freedom of movement and more generous access to the welfare systems of host countries. Such a tax is clearly insufficient for resolving the problems arising from restrictions on access to welfare rights in the context of the EU. One potential problem is that high-skilled labour, and talent, is not easily replaceable and hence directly compensable (see Brock and Blake, 2014; Oberman, 2013). Investment in education and training does not necessarily result in equally good outcomes if the most talented and the most ambitious emigrate in significant numbers. Access to welfare rights tackles this issue. It serves as an enabling condition, in the form of welfare payments, that makes the transitional costs of immigration lower. ²⁸ In this way, immediate access to welfare balances the outflow of high-skilled labour with a greater outflow of low-skilled workers and reduces the pressure on the welfare system of sending member states. Further, it makes it more difficult for high-earning and high-skilled individuals to escape their duties of reciprocity by exercising freedom of movement across EU member states, as easy access to social assistance also allows low-earning, and low skilled, workers to become members of the same systems of social cooperation. Transnational access to social assistance benefits, therefore, also prevents the EU’s internal borders from becoming socioeconomic filters; it hinders their use as a means to cherry pick and corral people.

A third objection is as follows: the reciprocity-based approach relies on a simplified picture of EU migration. It is geared to cases of EU nationals who are making a new life in the host country and who intend to reside there for a long time, if not permanently (see e.g. Ottonelli and Torresi, 2012). But it is less convincing when it comes to cases of temporary EU migration. Does an EU national who only intends to work seasonally in a host member state or who looks for work periodically in that state also qualify as a member of that system of social cooperation? There are two responses.

First, what matters for duties of reciprocity is that they take place within a system of social cooperation. But they are duties owed not only within the group of nationals but also within the group of EU migrants. What matters for justice as reciprocity then is not whether person A leaves country C₁ to go to country C₂ only to go back to country C₁ soon after, but whether that person is interacting with others that reside in C₂ in ways that trigger duties of reciprocity. Temporary residence does not change the fact that duties of reciprocity are triggered by such interactions. In the case of non-economically active EU migrants, as argued in the third section, duties of reciprocity could be triggered by the fact that those who are willing but unable to contribute via work are owed by those who are able to contribute via their employment. This could be the case regardless of whether the person is a short-term or long-term jobseeker. What matters is that this person genuinely looks for work for the period of time she resides in country C₂. For that period of time, therefore, she should be entitled to social assistance.

Second, it is important to emphasise that a person qualifies as an associate to that system of social cooperation only for the period of time she resides in country C₂. Therefore, she is owed, and owes, duties of reciprocity for that particular time period. In the case of EU migrants, this entails access to social assistance and social insurance. EU immigrants should be able receive benefits but also to pay taxes and social insurance contributions for the period they have worked and lived in country C₂. They should be entitled to the same benefits as nationals who have contributed for the same periods, and have those benefits directly paid by the host state. Finally, it has to be noted that some of these benefits, most notably those following from contributions towards pension schemes, need not be tied to residency and should be portable. Mobile persons are often participants in different systems of social cooperation over their lifetime. The portability of such insurance benefits ensures that the host society continues to reciprocate for the immigrants’ contributions over time.

Finally, one could object: what makes EU immigrants special vs. other groups of immigrants? Shouldn’t the restrictions also be lifted for non-EU immigrants on grounds of reciprocity? Would that be an unreasonable financial burden for host member states? EU immigrants are a special case for two reasons: the EU’s member states have reciprocally agreed to allow freedom of movement of human capital and hence to extend the range of choices that their citizens can exercise. There are no such reciprocal agreements with non-EEA member states. It would be unreasonable to expect EU member states to give US jobseekers access to welfare if the US refused to do the same with EU citizens. Second, the profile of EU immigration would probably be different to that of non-EU immigration under freedom of movement with immediate access to welfare, due to large inequalities in GDP per capita and welfare state provisions in EU member states vs. some countries of the developing world. The financial costs of extending freedom of movement globally while granting immediate access to welfare could increase exponentially, making the financial burden for host member states unreasonable. ²⁹

Conclusion

In the current political constellation in the EU, freedom of movement is increasingly conditional and the conditionality targets one group: poor, low-skilled and multiply disadvantaged EU immigrants. It is a form of discrimination not so much on the basis of nationality, which is used instrumentally, as on the basis of an individual’s potential economic contribution. The rationale is clear: those individuals who cannot prove that they can imminently contribute to the economy of their host state, and who cannot support their job search via their own means, should find it more difficult to move to and remain in another EU country. I argued that this is inconsistent with what justice as reciprocity requires in the case of the EU. I showed that to justify such restrictions, one needs to dramatically overstate both the type and degree of contribution that both nationals and EU immigrants are required to make to the host member state’s system of cooperation before they access welfare. Further, such overstatement comes at a high price: it leaves the most vulnerable members of both groups without access to welfare rights. Finally, I defended those claims from a number of objections showing how they are relevant in both ideal and non-ideal circumstances, as well as how they could be intrinsically linked to other less practice-orientated approaches to social justice in the EU.