‘I Would Never Have Come If We’d Know It Might Be Like This’: On the (Un)Intended Consequences of Welfare Governance of EU Roma Migrants in Britain

This paper presents the findings from a small-scale pilot study which explores the experiences of accessing welfare benefits by the migrant Roma European Union (EU) citizens in the UK. It compares administrative barriers and individuals’ knowledge of welfare entitlement both prior and after the implementation of changes to the welfare regime in 2014, when a tranche of ‘policy hardening’ legal enactments came into force. For the migrants who participated in this study, precarious, low paid post-migration work has brought several hazards, including a non-eligibility for certain social protections and an inability to demonstrate documentation which enable access to ‘passported’ welfare benefits. The combination of problems in accessing welfare benefits and the resulting state interventions, including expulsion from the UK in some cases, suggest that EU Roma citizens experience disproportionate negative impacts of welfare hardening, adding to the much vaunted ‘hostile environment’ to EU migrants in the wake of the Brexit vote. As such, we find the practice of ‘bordering’ migrant EU Roma citizens to the UK is taking place through covert state enforcement action against families and households, discouraging effective and genuine use of their free movement rights guaranteed under European Union law.


Introduction
This paper examines the experiences of Roma European Union (EU) citizens resident in the UK and asks whether explicit anti-migrant discourse (Cap, 2017), widespread public scepticism towards the benefits of migration 1 , and policy 'hardening' (British Social Attitudes 31, 2014;Hopkins, 2017) towards EU migrants, particularly in the wake of the 2016 Brexit vote (Khaleeli, 2016) have coalesced so as to disproportionately impact 'bordered' 2 and socially excluded Roma EU citizens, who mainly arrived to the UK from the so-called Eastern European countries.
Our research focus on migrant EU Roma citizens is deliberate.Since the EU enlargements post-2004, many of the 'old' EU Member States became increasingly concerned with the 'welfare tourism' (Giulietti and Wahba, 2012), trying to find ways to restrict free movement for economically non-sufficient EU citizens (Blauberger and Schmidt, 2015).Such policies often counteracted the EU law guarantees on free movement not only for those who have a job offer or provide commercial services as the self-employed, but also for those who seek work, including their family members. 3 Arguably, next to the intersectional interplay between ethnicity, nationality and gender (Vrăbiescu and Kalir, 2018), Roma EU citizens who arrived in the UK after 2004 bring to the host countries new and additional policy challenges based on class and socio-economic status, which may be different to those faced by the 'older migrant' Roma communities, 4 as the research on Gitanos/Roma in Spain illustrates (Magazzini and Piemontese, 2016).In this context, while focusing on this particular population sample, we acknowledge the problematic use of the homogenising term 'Eastern European Roma migrants' in public discourse: it is a social/political construction that fails to acknowledge not only individual migration stories and experiences but also assumes a possible simplified policy 'quick fix', without seeking to understand meaningfully the various communities and the socio-economic challenges they face (Magazzini and Piemontese, 2016;Vrăbiescu and Kalir, 2018).
Therefore, this discussion piece asks whether in Britain targeted 'covert enforcement' action is undertaken against EU citizen Roma families and households as a way of discouraging 'undesirable' migration (Bulat, 2017)  implementation (Travis, 2016) that the UK is not as welcoming and desirable a destination as may have been portrayed in online discourse or family migration narratives (Boehmova, 2016;Travis, 2016;Parutis, 2014;Grill, 2011;Dekker et.al., 2016).In this way, our research enhances the understanding of the consequences of welfare governance of EU Roma migrants in Britain, contributing to the wider literature on migrants' agency in dealing with administrative removals and deportations from the UK (Sardelić, 2017;Schweitzer, 2017) and on the growing body of work that explores EU minority rights and the social justice dimension of the European Union citizenship (Kochenov and Agarin, 2017;O'Brien, 2017).
While our research sample is relatively small and cannot be considered fully representative of all EU Roma citizens in the UK, our findings are similar to those of more recent large-scale studies (Martin et al., 2017).Thus, despite the positive benefits of European migration (Portes, 2018), when compared to the often deeply exclusionary circumstances experienced by Roma in countries of origin (FRA, 2016), our findings suggest that migrant Roma, especially since the toughening of the UK welfare regime in 2013-2014, are subjected to particular scrutiny: they are viewed through the 'welfare gaze' and perceived of as the 'undeserving poor' (Romano, 2017) who must be treated in a disciplinary manner to ensure compliance with the preferred (or imagined) norms of British society (Nagy, 2016).Whilst non-Roma 'poor' or 'undesirable' EU migrants (typically those who are street homeless) are also subject to the state treatment and have been increasingly targeted by interventions aimed at requiring or enforcing their removal from the UK (Cooper, 2017), 5 we propose that there is a growing body of evidence to suggest disproportionate levels of 'bordering' (Van Houtum and Van Naerssen, 2002) of Roma families, impacting on their migration intentions or influencing decisions to return to countries of origin.Building upon the findings from our pilot report (Dagilyte and Greenfields, forthcoming 2018) and on the data on de facto deportations (Home Office, 2017a), or expulsions 6 of EU citizens (Mantu et al., 2017), we suggest that EU Roma citizens and their families may experience particularly high barriers to demonstrate their entitlement to welfare benefits, in result impacting on their residence rights and migration intentions, as well as diminishing their trust in state agencies. 5For a wider European context, note the organised large-scale expulsions in 2010 in France that are widely documented (Ciulinaru, 2018;Rieder, 2012;Bennett, 2011) and the recent call for a new Roma expulsion policy in Italy (Salam, 2018). 6In the context of this paper, this term refers to the gamut of means by which EU citizens can legally be required to leave the UK with a particular focus on administrative removals and voluntary returns.Deportations/expulsions occur where an individual is removed from the UK on the decision of the Secretary of State for the Home Office if their presence is not 'conducive to the public good' (as provided under sections 3(5)(a) and 5(1) of the Immigration Act 1971) or following a court decision resulting in a prison sentence.Administrative removals are the enforced removals of EU citizens and their family members who no longer have legal rights of residence (the right to reside) or on the grounds of 'mis-use' of rights (see: Evans in: Mantu et al., 2017;Home Office, 2017b).Voluntary departures are the departures of individuals against whom enforced removal proceedings have been initiated but who have opted to leave the country prior to such enforced removal taking place (Migration Observatory, 2017).Following deportation or removal by the means outlined above (including voluntary return), there is reentry ban imposed which may be for up to five years, depending on the expense to the state or circumstances in which they left the UK (see further: Home Office, 2017c;Mantu et al., 2017).
In the remainder of this paper we present the findings from our pilot study and consider whether multi-factorial social exclusion, both pre and post migration, places Roma migrants in a situation of unique disadvantage, leaving them particularly vulnerable to negative welfare governance and at risk of expulsion.To do so, we will firstly outline the national political and economic context in which the EU Roma citizens' migration has occurred.This is followed by the overview on how our empirical research study was designed and conducted, proceeding to highlight the key findings and offering conclusions on the institutional 'bordering' EU Roma citizens via the hostile welfare state policies and such migrants' eventual voluntary or enforced returns from the UK.

The UK Political and Economic Context of Welfare 'Bordering'
In the British context, there continues to exist a considerable debate about 'pull factors' influencing migration from the EU (Migration Observatory, 2016).Hence the complex and continuously restructured social welfare system seems to have been deliberately (re)designed to diminish the 'attractiveness' of low paid employment in the UK (Fontanella-Khan and Warrell, 2013;Alberti, 2017).One of the reasons is to curb employment related migration from the EU, where, subject to regulation, some EU citizens can legally access in-work welfare benefits and 'top-up' their income where children or disabled persons are part of a household.
Analysis of Hansard debates in the UK House of Commons7 , both prior to and following the UK Brexit vote (especially during the period 2013-2017), reveals considerable political preoccupation with the fiscal 'cost' to the state of EU migration, whilst reporting of discourse around the Welfare Reform and Work Act 2016 evidences a preoccupation with EU citizens exercising their legal rights to move to Britain and to claim welfare benefits.
Typical examples of such utilitarian political discourse include statements rooted in a presumption that migration -especially from poorer EU states -is directly associated with the ease with which migrants can claim financial support, and who in turn (by implication) diminish the public wealth available to British citizens: Mark Harper (MP) 29/11/2017 'the migrant workers in Britain who do not earn significant salaries but have access to benefits such as our welfare system are not making a net contribution to public finances.I am not suggesting that they are not working; they are, but they are earning a lower salary and are therefore entitled to things like in-work tax credits and […] universal credit.[…] In effect, British citizens and those already working here are subsidising some of those migrant workers.' The examples reveal a toxic focus on the 'burden' to the State of EU migrant workers, ignoring research indicating that EU migrants generally pay more in tax than they claim in benefits (although variables exist depending on the country of origin, see further: FullFact, 2017).There are also concerns that the EU has been associated with loss of control of borders and diminished sovereignty at the expense of the wellbeing and economic security of British citizens.These economic implications of migration were at the heart of the success of the 'Leave-EU' campaign in 2016, in a process dubbed 'project fear' by Moore and Ramsey (2017).Similarly, the Migration Observatory (2016) found that prior and after the Brexit vote on 23 June 2016, print media from across the political spectrum (tabloid to broadsheet) had an increasingly explicit focus on immigration.Commonly, 'tabloid', right-leaning newspapers framed the argument for voting to leave the EU around concerns over 'sovereignty' and resentment of the European 'project' which led to 'uncontrolled' migration.
Although debate over the level and speed of EU migration to the UK first came under public and media scrutiny in relation to unexpectedly high levels of East European migrants exercising their right to enter the UK for work8 after the 2004 enlargement of the EU (Watt and Wintour, 2015), it was not until the wake of A2 enlargement (Romania and Bulgaria) that Roma EU citizens became more noticeably framed as low-skilled undesirable workers, likely to seek to settle in the UK in large numbers, given the perceived ease of access to welfare benefits (Richardson, 2014;Yuval-Davis et al., 2017).
A particular strand of discourse both within Westminster and outside among the media and general public concerned the fact that EU migrants who fulfilled certain criteria were able to access both UK 'universal benefits', such as payment of an allowance for children (FullFact, 2016), even if the child was not resident in Britain, and also other 'passported' benefits which were relatively uncommon elsewhere in the EU and included the right to apply for financial support to meet the costs of housing (Alberti, 2017;Dagilyte and Greenfields, 2015).
Accordingly, it became a truism widely reported in the media, and in political discourse, that the 'easy' availability of welfare benefits had had a disproportionate impact on migration rates from A2 and A8 countries (Portes, 2015;Riley-Smith, 2015;Dawar, 2015;Guentner et al., 2016).Alongside these enhanced levels of explicit anti-EU migrant discourse, with even moderately mainstream newspapers and online media outlets increasingly using terminology and tropes which would have been  POLITICS, 4(3): 81-105.virtually unthinkable a few years previously other than in the most extreme of the right wing press, the political response was an obvious policy-hardening which was in apparent response to public concerns over 'welfare tourism' and alleged downward pressures on British citizens' wages (British Social Attitudes Survey, 2014;Hopkins, 2017).
Whilst rarely explicitly framed as anti-Roma reportage, xenophobic representations of EU migrants are coupled with persistent 'tabloid' media hostility towards EU Roma citizens who are portrayed as the undesirable migrant par excellence (Yuval-Davis et al., 2017;Kóczé, 2017;Clark, 2015): tabloid newsprint often associate anti-migrant stories with images of 'Roma villages' or visibly identifiable Roma individuals (on which see further : Tremlett et al., 2017;Richardson, 2014).At the same time, EU Roma welfare benefit claimants in the UK were found to be only occasionally consulting civil society organisations for help, even though compared to non-Roma peers they experienced disproportionate monitoring of their welfare claims and ever-increasing documentary hurdles when they sought to access benefit entitlements (Dagilyte and Greenfields, 2015;Martin et al., 2017).Moreover, they appeared to face an additional layer of 'Romaphobic' othering (McGarry, 2017), in addition to the significant linguistic or bureaucratic barriers placed in the way of all EU migrants who were seeking to access in-work welfare benefits (Alberti, 2017;Shutes, 2016).

Research Design and Methods
This small-scale pilot project, co-funded by the Socio-Legal Studies Association and Bucks New University, emerged following the consultation of the Czech Republic NGO, Odlisnost, where we developed a welfare benefits advice briefing for their Roma clients.It became clear that there was scope for developing a pilot project to explore themes which were emerging as important to civil society agencies dealing with enquiries from Roma clients who had migrated to the UK.
The project fieldwork was undertaken in 2014-2015 and was thus in essence a scoping study to evaluate the extent of the impacts of welfare regulation change on recent migrant EU Roma populations to the UK with the view to exploring whether in work benefit changes were impacting migration intention for primary migrants and their family members.A subsidiary theme looked at whether the poverty associated with low-paid employment and difficulties accessing welfare benefits were increasing the likelihood of Roma households considering return migration, or indeed being expelled to their country of origin.
The research involved close collaboration with two leading civil society Roma rights organisations running comparative field-work sites.Thanks to ongoing collaborations and existing contacts, it was possible to undertake interviews with NGO support staff and EU Roma migrants in a small Northern city with a large Roma population (Derby) and also in London.The project used multiple methodologies and incorporated the following activities: 1. Three focus groups facilitated with known NGOs in the two geographical areas (exploring Roma migrants' experiences of seeking work; intentions on moving and the process of claiming welfare benefits); INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.
2. A group interview with NGO support workers, focused on their knowledge of employment patterns and welfare claim experiences of Roma client groups (London); and short discussions with support staff in Derby.3. In addition, we prepared a short online survey covering the same key questions asked of support staff, which was distributed via relevant enetworks.This final phase produced an additional five responses from law centres, advice drop-in agencies and local authority staff.
We collected data from Roma participants via three focus groups on their knowledge of welfare benefit entitlement pre and post migration, employment access strategies and intent on entering the UK (e.g. to work or claim benefits), actual experience of claiming welfare benefits, complexities pertaining to relevant documentation, success rate in claims, appeals processes, and linguistic barriers pre and post the 2014 welfare benefit changes.Our research focused on two benefits in particular: the income-based Jobseeker's Allowance and the Housing Benefit (for reasoning on the selection, see Dagilyte and Greenfields, 2015).
In total, 28 Roma respondents were interviewed.These consisted of 12 Polish and 14 Czech or Slovak and two Romanian respondents.The majority had been resident in the UK from between one month and three years.In several cases, respondents had been resident in the UK for less than three months: a key time frame post-April 2014, when the minimum period of residence was extended to evidence migrants' 'Right to Reside' and 'Habitual Residence' that are required to access welfare benefits.
Three Polish Roma interviewees had been resident in the UK in excess of five years and accordingly had established permanent residence status: a fact which was to prove critically important in terms of attempted administrative removals recounted by one respondent; and indeed, in evidencing inconsistent decision making by officials pertaining to refusal of welfare benefits.
The gender and age sample of Roma respondents was split approximately 60:40 male to female and equal numbers of 20-34-year-olds (within which group there were more than 80 per cent who were new arrivals resident for less than three months in the UK) and 35-65 years of age.More than three quarters of respondents had dependent children.
Thematic analysis was undertaken for transcripts of focus groups and of the online survey responses received.Empirical findings were supplemented by secondary analysis of more recently published materials (for example, the large-scale Salford study on Roma and Welfare Benefit access, see: Martin et al, 2017).The section below summarises key findings from the final report, to be published in 2018.

Key Findings: The Institutionalised Process of 'Bordering'
The overall findings indicate a worrying picture of EU Roma migrants' rights after the 2014 welfare reforms in the run up to the Brexit vote.Despite moving to the UK to find work, when falling in need of welfare support this group of people faced strict rules for compliance with the UK legal tests that are applied by authorities for EU citizens to be eligible for welfare benefits, often without considering the migrants' and INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.their family members' personal circumstances.We found such decisions were often taken despite the welfare agencies staff's lack of knowledge about the ever-changing conditions for welfare benefits entitlements, without a possibility to interview the applicants in their own native language, accompanied by a lack of transparency as to exactly what documents are required to file a benefits claim.As our interviews with the advice workers show, such institutionalised barriers of 'bordering' often cannot be challenged judicially, even though they cause delays and result in many EU Roma citizens abandoning welfare benefits applications.
In particular, findings from our soon to be published pilot study (Dagilyte and Greenfields, forthcoming 2018) and a discrete yet associated inter-university project due to be reported on later in 2018, have identified a set of circumstances which appear to disproportionately 'border' low-skilled, low-paid Roma migrants who have practised 'whole-family migration' (Ryan and Sales, 2013;Moskal and Tyrrell, 2016).Such patterns of family group migration, which are particularly common among Roma households moving from the Czech/Slovak Republics and Romania (Grill, 2012;Matras and Leggio, 2017), have tended to be favoured by 'pioneer' migrants among Roma communities.This is precisely because the 'pioneers' are viewed as opening up opportunities which enable a family group to access various financial and practical support mechanisms when settling into work and a new life as long-term migrants into the UK (even though, over time, post-migration social and economic mobility may stall, negatively impacting second generation migrants, as persuasively argued by Beluschi-Fabeni et al., 2018).

Purposes of Migration to the UK
The primary finding from this project was contrary to the populist narrative of welfare 'tourists' as outlined in the Introduction to this paper above.Our findings, however, are fully in line with those identified by Martin and his colleauges (2017) in which participants in a series of focus groups held with Roma migrants in five locations highlighted that, far from being attracted by the UK welfare system, the opportunity to work and create a better future for their families are the primary motivating factors and that accessing benefits proved extremely difficult as a result of poor literacy and increasingly restrictive criteria excluding them from the UK welfare system.Thus, we found that, like other EU migrants, Roma people relocated to the UK predominantly to work and achieve a better standard of living for themselves and their families (IPPR, 2013;Okólski and Salt, 2014), albeit often with the additional driver of escaping intolerable levels of racism and discrimination in countries of origin which limits employment opportunities (FRA, 2016;ERRC, 2007).
In our study, we found that 'first wave' migrants, i.e. individuals who were pioneers amongst their family or community group and who had not migrated to join established relatives or participate in existing networks with jobs awaiting them, typically had no clear idea of the type of work which they would find on migrating to the UK.As such they migrated with a core priority of seeking an opportunity to work and earn money to assist relatives at home, in a manner which was largely unavailable to them in the post-communist system.Such work was typically low paid, dirty and involved harsh working conditions: Regardless of their country of origin, Roma respondents indicated that a combination of hardship in their member state and the perceived opportunities in the UK (often relayed via family or friend networks or online communication) had been the main migration driver leading them to move to Britain.Those respondents who were the longest established (for example, Polish Roma migrants who had been resident in two cases for over 15 years) reported how they had initially obtained work doing 'anything' on arrival in the UK: accepting temporary unskilled jobs such as street cleaning, undertaking field labour, working in chicken slaughtering factories, road sweeping, collecting scrap metal, or distributing leaflets.This was until they were able to obtain more stable employment through developing networks with local employers.For example, one respondent's wife was working as a teacher.Such employment patterns were also found among more recent migrants from the Czech and Slovak Republics and is further borne out by other research in the UK with diverse Roma populations (e.g.Beluschi-Fabeni et al., 2018;Grill, 2012;Martin et al., 2017).
In time, the more established migrants (for example, early migrants from Poland or the Czech Republic) had been able to advise and support their own relatives and members of other Roma networks migrating to the UK, advising them on how best to access agency work (generally through compatriots) until English language skills developed adequately to permit of more mainstream, albeit generally fairly low skilled, work.
Very few respondents reported engaging in self-employment, such as was carried out by highly visible Roma 'Big Issue' sellers, who were identified as most likely to be Romanian or Slovak migrants, present in the UK for long enough to access these 'sale' networks but who had not yet managed to establish themselves in 'mainstream' employment.Pendulum migration was also as common among such 'sellers', while our participants were either well established in Britain9 or indicated that such was their intention: 'there [are] chances here for children to have education and for work.At home this is not possible for Roma.My wife will come, [and] my oldest daughter -in a little while six month -eight -I will send for them.' (focus group participant - Slovak Roma, resident in the UK for less than one year).
For migrants who were recent newcomers to the UK, e.g.newly arrived Polish, Czech and Slovak migrants present in the country for less than three months, employment was typically achieved through the goodwill of their family and friends who formed part of a pattern of chain migration.More than two-thirds of newly arrived migrants in one focus group were working within a month of arrival, typically through INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.introductions made by relatives or their house-mates who connected the new arrival with someone (who could be Roma themselves), or who employed groups of workers on something akin to the 'gang master' system.i.e. working as cleaners (women); factory work (male), etc.Overall, 70 per cent of migrants indicated that their key intention on reaching the UK was to seek work.The exceptions were all married women with dependent children, or in one case an adult daughter with caring responsibilities who assisted her wider family with looking after a disabled household member.
Respondents repeatedly noted that housing costs were kept low by the process of co-residence in over-crowded accommodation, thus reducing costs for each individual.This mitigates the impact of low wages and the three-month new migrant 'penalty period' (post-2014), during which time new migrants are unable access Housing Benefit or other forms of welfare support in the UK.Accordingly, the location and format of these 'new migrant enclaves' are typically high-density, lowquality, private rented housing in excluded neighbourhoods populated by diverse migrant communities (Brown et al., 2013).In some cases, housing and employment were both arranged by a middle-man who may themselves be a Roma who had become more established within a local neighbourhood, a finding which Nagy (2018) has also highlighted.
It was particularly noticeable, for both recent and longer-established migrants, that patterns of low-paid, often 'grey-market'/cash in hand work were the norm on first relocating; with such jobs typically obtained within a few days or weeks of moving to Britain.This form of 'word of mouth' employment access was overwhelmingly obtained through being hired by a person from the same country of origin.Therefore, we found absolutely no evidence that Roma migrants were drawn to the UK as a result of perceived ease of access to welfare benefits: instead, they emphasised the value of work that provided a tangible sense of empowerment, especially for men.

Barriers to Accessing Benefits
Accessing precarious, low paid work of the type outlined above, while an effective short-term financial strategy, and one which potentially enables funds to be remitted home to support relatives prior to their migration, is unsustainable if welfare benefits need to be claimed as additional support.Such sporadic and not always 'official' employment history makes it difficult to demonstrate administrative requirements which enable access to 'passported' welfare benefits, i.e. assistance in meeting housing costs or low-pay 'top-up' benefits available to EU migrants after a period of working in the UK, subject to fulfilling certain criteria.
From our focus group interviews and the online survey it became clear that for respondents who had attempted to access welfare benefits (most noticeably after the regulatory changes which occurred in April 2014) a significant number of barriers existed, which appeared to disproportionately impact EU workers, and we posit, EU Roma migrants in particular.
Major themes which emerged from analysis of data gathered from Roma respondents who had applied for benefits, supported by more nuanced commentary on findings from advice/support workers, presented an almost unanimous picture of a INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.confused and inadequately administered welfare benefits system in which administrative staff employed by the Department of Work and Pensions themselves appeared to lack knowledge over the precise legal status enjoyed by migrant claimants: 'Very bad experience, because he applied for housing benefits because of his illness; he is not fit.When he went to […] City Council he was told to bring old pay slips, e.g.[last] two months, and when he brought it to the Council, they told him he had to bring something else.When he got all the documents, they wanted something else, so he has a bad experience.He has to pay for rent, otherwise he will be homeless, it is very complicated and he is not fit to go to work and he does not have enough money to pay the rent.He expected a more compatible communication with the council regarding benefit' (Translator explaining the experience of Roma migrant from Slovakia -resident in the UK for five years, formerly employed).
A key area of confusion (and one which is strongly implicated in cases of administrative removals) concerns the dual legal test which migrants need to meet in order to be eligible for welfare benefits: the Habitual Residence Test (HRT) and Right to Reside Test (RTRT).Both must be fulfilled before an individual can claim noncontributory, means-tested benefits which exist to retain a minimal income for unemployed people without other finances (see: Dagilyte and Greenfields, 2015 for a full explanation).
The obvious problem, however, lies in the fact that an individual who is undertaking very low paid work (below the legal minimum wage), or for 'cash in hand', will be unable to establish their right to fulfil the requirements for the HRT and RTRT.Even for individuals legally employed on minimum wages, the hurdles to be overcome are substantial.Roma migrants who were often not literate in the language of their country of origin, or who had extremely limited knowledge of how to obtain advice, were found to be particularly vulnerable to refusal of benefits.Evidence was supplied on several occasions of translation services (where they were offered) being provided in only a partially comprehended language or via a translator from a 'country of origin' who was hostile to Roma and either did not fully translate responses, or implied to the investigating officer that the claimant was not seeking to work but wished to claim benefits.
The issue of lack of translation services when seeking to apply for services arose repeatedly, with Roma respondents often using informal networks of support to obtain information or assist them with claim-making.One respondent in a focus group who applied for housing benefit, having been resident in the UK for over three years, told us that he used his current language skills (Russian and Polish) to communicate with migrants from other EU Member States (Lithuania, Poland), whose English language comprehension and expression were more advanced, and asked them for help when dealing with welfare benefits authorities: 'When he went to the Jobcentre, [they] gave him the declaration in English.So when he first came to the country [as a single man], he couldn't read or speak [English], so it was hard for him, as he didn't understand anything.He made a INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.
claim, but he had to provide some documents, he didn't speak English at that time, so it was difficult for him, so he just left it.And now, this is the first time claiming since he became ill […] because he is with a partner, and he has an illness, and she [the partner] speaks English' (Translator in focus group explaining the situation of a Roma migrant participant).Indeed, we found evidence in several cases of Roma migrants giving up welfare claims which they were entitled to make as they believed that they would not be able to satisfy eligibility requirements given complex regulations and documents requested by bureaucrats.It was particularly problematic that there appears to be no single set of guidance on what form of documentation will satisfy officials, leading to at times arbitrary decision-making on whether the HRT or RTRT have been fulfilled.

From 'Bordering' to Expulsion
In turn, lack of demonstrable documented official employment history increases the risk of administrative removal if an individual or family were deemed by the state benefits agency not to be exercising their EU rights of freedom of movement for the purposes of seeking employment; instead they can be perceived as an unwarranted burden on the State, as popularly portrayed in the 'welfare tourist' media discourse.In such circumstances, several respondents in both London and Derby either recounted that they themselves (or relatives and friends) had been subject to interviews by the migration authorities enquiring as to their intent and circumstances and pressure was applied to encourage 'voluntarily return' to countries of origin following the refusal of welfare benefits.In one case of a Roma family who had the right to reside, a homemaker mother with an employed spouse were unlawfully threatened with removal, as she was not personally seeking work or enacting behaviours associated with 'genuinely seeking work/genuinely likelihood of work'.In another case, a Roma man who had acquired British citizenship was repeatedly called in and was subject to disciplinary welfare interviews until it transpired that he had British citizenship and the approach of benefits agency staff abruptly changed towards him.
To illustrate the processes aimed at facilitating a 'willingness' to return to countries of origin, an investigative newspaper article of October 2017 (published shortly before the issue of increasingly stringent Government guidance on the process for undertaking administrative removals of EU citizens)10 reported that as part of the Government's pledge to create a 'hostile environment' for individuals regarded as undesirable, a homeless.'[A] Romanian national [held] in an immigration detention centre [was advised] that his request for emergency accommodation has been rejected and he should consider another country.[The letter he received] states: "You could avoid becoming destitute by returning to Romania or another EU member state where you could enjoy access to all your ECHR [European Convention on Human Rights] without interference"' (Townsend, 2017b).
While it is not stated whether that the individual in that case was Roma, the Roma Support Group working in collaboration with a leading UK homelessness charity, St. Mungo's, found that Romanian nationals were the fastest growing group of rough sleepers in London, accounting for 1,388 Romanian rough sleepers (18.7 per cent of all rough sleepers in Greater London) with a 'sharp rise in the number and of percentage of Romanian rough sleepers thought to be of Roma ethnicity' (RSG, 2016: 3).Moreover, their report explicitly referred to the difficulties in accessing welfare benefits experienced by Romanian migrants and increasing distrust of officials as a result of an emphasis on 'reconnection' or voluntary return to Romania.By the summer of 2017, evidence was increasingly emerging of a distinct policy (perhaps unwittingly assisted by civil society agencies supporting vulnerable migrants) pertaining to deportations of homeless or otherwise socially excluded EU citizens, in particularly those from A2 countries (Townsend, 2017a).
This dramatic increase in administrative removals or 'quasi-voluntary departures' (Mantu et al., 2017), reported initially by Roma respondents to our study, who were often living in poor quality over-crowded housing prior to being refused welfare benefits, may perhaps be posited as indicating that Roma were the 'trial group' for enactment of this policy.This policy shift quietly determined rough sleeping or homelessness as an 'abuse' of EU citizens' right of freedom of movement (Cooper, 2017), enabling enforcement action, including detention prior to removal, to be taken.Subsequently, and in the light of further evidence of increasing numbers of formal deportations of homeless EU citizens which illustrated that the group most likely to experience 'enforced returns' from the UK between June 2016-2017 were Romanian nationals (Home Office, 2017a), a judicial review was brought by a coalition of charities, which led to the policy of a blanket detention and return of homeless EU migrants being declared unlawful by the British courts in a 2017 case.11It is clear, however, based on evidence from our study and anecdotal report to civil society agencies that the declared figures do not include all those who are pressurised to leave the UK, as official statistics do not include 'quasi-voluntary' returns that are driven by fears of child protection interventions or the risk of destitution following unsuccessful applications for welfare benefits.
Moreover, despite the court ruling, in December 2017 new Home Office guidance on administrative removals (2017b) has, while more cautious in tone, still reiterated a sense of suspicion towards EU migrants, indicating that evasion of taxes, entering into a marriage of convenience with an individual entitled to reside in the UK, or persistent low-level offending may all be taken into account when administrative removal decisions are made.
As can be seen, for Roma migrants the risk of becoming entangled in a Kafkaesque bureaucratic system leading to denial of entitlement of financial support, spiralling risk of state interventions and ultimately removal or in the absence of any alternative a strategic decision to return to their country of origin, is relatively high.This risk of respondents coming into unexpected contact with migration authorities staff following referrals made to the immigration authorities by welfare benefits agencies (particularly when considered in relation to the issue of 'evasion of taxes' being grounds for administrative removals) was exacerbated by use of 'word of mouth' and grey-market (cash in hand) employment strategies used by new migrants (see also: Nagy, 2018) which impacted their ability to demonstrate that they had been employed prior to seeking welfare benefit support.Even lawfully employed Roma migrants frequently appeared to have failed to collate suitable evidence to support welfare benefit claims as, contrary to popular discourse, prior or upon arrival they were overwhelmingly unfamiliar with the British welfare regime and legal expectations upon them.Even after the initial three months 'migrant penalty period', following which migrants could access welfare benefits and 'passported' housing benefits, somewhat surprisingly to the research team, recent (and not so recent) arrivals still often lacked understanding of welfare eligibility criteria.NGO agencies and staff, too, reported frequent encounters of Roma who were near destitute, oblivious about their entitlements: 'quite often we first come across a client when someone, another Roma, perhaps someone they are working with -will bring them along to us or phone and say 'can you help this person, they've got three kids under seven, and they aren't getting xx and I get it so I know […] for example, people we've had in weren't claiming child benefit although they were working and entitled, or didn't realise that they could get help if there was someone disabled in the family… that sort of thing' (NGO staff member).

Discussion and Conclusions
As can be seen, there is a complexity to, and multi-layered levels of, exclusion experienced by EU Roma citizens in the UK.These include their prior experiences of engagement with the home state, marginality on arrival in the UK, precarity in both the employment and housing markets, exacerbated by the dramatic shrinkage in migrant workers' rights impacting both Roma and non-Roma who need to seek to access welfare benefits (Martin et al., 2017;Alberti, 2017).
These experiences of poverty and precarious employment are coupled with the impacts of low levels of literacy and limited access to agencies working with and supporting Roma clients (Morris, 2016), language barriers, low levels of knowledge of welfare entitlements in the UK, an increased risk of exploitation by employers, perhaps associated, too, with anti-Roma racism (McGarry, 2017), and a willingness to accept housing and working conditions which may be worse than those available to non-Roma migrants (Brown et al., 2013).Hence, the potential mix for a toxic and declining situation exists which greatly diminishes the security of EU Roma migrants in the UK.
As is illustrated above, the discussion on administrative removals emerged in several interviews.This is worrying, as the policy consists of practices of tacit bordering and policing by state agencies, which operate to 'encourage' return migration of EU Roma citizens to their home countries, in contradiction to EU law INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.
which will continue to apply throughout the transitional period before Brexit.The UK does not seem to be alone in this approach: Tervonena and Enache (2017) found similar 'everyday bordering' of Roma migrants in Finland, operationalised through the processes of 'mobilizing municipal workers and local police as everyday gatekeepers ' (2017: 1114).It would appear that in several European countries regarded as beacons of equality and possessing relatively good reputations for civility towards migrant populations and no record of targeted anti-Roma racism, there is a slow drip-feed of pressure which is disproportionately impacting Roma as a result of their precarity and position in the labour market.Indeed, this focus of the welfare gaze on EU Roma citizens can be seen as directly linked to van Baar's proposition (2011) that there is 'new norm' of 'problematising' Roma as migrant criminals.Nagy (2016) theorised that disciplinary practices are enacted by state agencies to ensure compliance with preferred (or imagined) norms of society and thus that Roma migrants to the UK engage with these pressures by a series of strategies -subversion, compliance or 'ethnic denial' (blending in and perceived a Czech or Romanian national rather than Roma); a finding which has marked similarities to Acton's (1974) thesis of Romani adaptive strategies.
Despite the adaptive attempts adopted by some EU Roma migrants in the UK, and a clear determination by the vast majority of our respondents to seek work and settle in the UK on a long-term basis, our findings suggest that the 'taint' of being a migrant, working in a low-paid sector and in some cases being openly identified as Roma -for example, by translators who share a country of origin -created a particular level of risk for members of this community over and above that experienced by other EU migrants with similar skillsets or working in low-paid jobs.
Professionals who had contact with migrant Roma households often appear to hold deeply embedded opinions based on 'common sense knowledge' which has created an image of the communities fuelled by stereotypical media representations of anti-social Roma behaviours which, to utilise the concept coined by the late Professor Stuart Hall, embodies 'inferential racism' (Hall, 2000).Highlighting this point, our online survey of NGO advice workers showed that not only the UK authorities dealing with welfare benefit claims were not familiar with issues of Roma culture but they also lacked empathy with EU Roma applicants, resulting in the breakdown of trust and cooperation with the UK migrant Roma community.Given that Roma welfare claims are being scrutinised more than other EU citizens' claims, especially if submitted by Romanian nationals, one our online survey respondent went as far as to describe the UK welfare benefits administration system as 'institutionally racist', while the other highlighted the continuing 'stigma of being a Roma'.
When via state bodies such ethnicised association of particular (usually negative) behaviours and attributes are attached to a particular community (in this case the Roma) and acted upon as if those tropes and concepts bore some relationship to lived reality, such vicious circle inclines state actors to treat individuals as though they are automatically a suspect population, furthering the process of 'bordering'.For example, we identified cases of deep concern where, as a result of poverty arising from rejection of Roma migrants' welfare benefits claims, or due to delays in receiving in-work benefits, families 'voluntarily' elected to return to their countries of origin (to once again experience racism, exclusion and joblessness), so as to avoid becoming totally destitute: 'He made a claim but he had to provide some documents, he didn't speak English at that time so it was difficult for him so he just left it […].They are saying that some people will go back to Slovakia, some will stay here, he says there is going to be more homeless people, more people living in poverty and children being poor, it is impossible [...] It is so scary to think about' (Translator -Slovak participants, average period in the UK 6 months).Some EU Roma citizens resident in the UK are comfortably self-sufficient and blend with diverse well-established populations (particularly perhaps in super-diverse London).However, for others, who may be particularly visible by dint of high density of residency (for example, in Govanhill in Glasgow, or Page Hall in Sheffield), 12 there is a danger that the way in which they are positioned in public discourse -as lowskilled, anti-social 'welfare benefit tourists' with a predilection for criminal behaviourmay destabilise their situation.In the light of the post-Brexit anti-migrant toxicity and dramatically increased rates of anti-migrant hate crimes (Dearden, 2017a), it would appear clear that the situation of Roma in Britain, while not perhaps as marginalised as in many countries of origin, is still highly precarious and potentially worsening.
Overall, our research presents compelling evidence to suggest that -contrary to basic EU principles of freedom of movement of people and the EU's commitment to diversity and human rights -policing and bordering is occurring through enforcement action and active discouragement of settlement of Roma and other marginalised migrant households.Such treatment continues to be directed at those who seek legitimately to use their (current) ability to exercise their EU rights and simultaneously make use of welfare regimes within the UK.While we do not suggest this is explicitly formulated into a policy of 'active discouragement' of Roma in a way found in some other member states, such activities are in part an artefact of Romaphobia, that plays out in a wider post-Brexit xenophobia, as resistance to East European migrants who are seen as problematic or failing to comply with British norms of household structures and behaviour.Such highly politicised discourses do not allow for taking into account of individual stories and enable further 'bordering' of EU Roma citizens.
Accordingly, and in the light of fiscal retrenchment and increasingly stringent welfare benefits requirements even for UK-born citizens, the tools exist which enable bureaucrats to operationalise pressures (or enact direct administrative removals) upon migrant Roma who come to the attention of the State, so that they 'voluntarily leave' the country.
In relation to welfare benefits claims, we consistently found that there was a lack of an accessible, precise list of documents required to support evidence required to meet the Habitual Residence Test; imposition of an impersonal, one-size-fits-all application of the Right to Reside Test; long, artificially-created delays in assessing welfare benefits applications; and rejection of welfare benefits applications without INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.administrators providing clear reasons for such decisions, which ultimately may lead to commencement of removal of claimants from the UK.Moreover, there was a high burden of proof on the jobseekers to show 'compelling evidence' of meeting the genuine prospect of work test.
The complex relationship between poverty and the welfare governance of EU Roma migrants is thus deeply entwined.The risk of slippage into destitution for migrant Roma households (including those in 'grey market' or low-income employment such as is frequently obtained following first migration) is profound in the light of employment exclusion and low pay, racism, marginalised accommodation opportunities and migrants' lack of awareness of employment/welfare benefits rights.
Accordingly, migrant Roma household potentially find themselves in 'receiving' member states in a precarious situation requiring them to traverse and balance risk factors which, while qualitatively different from those in their countries of origin, often expose them to worse environmental health and appalling working and living conditions than native-born citizens or even other migrant populations.
In conclusion, we posit that an unarticulated and perhaps barely recognised governance of Roma can occur through using welfare benefits agencies to discourage residence in Britain for all but the 'ideal', self-sufficient, self-supporting, wellintegrated, English speaking migrant (preferably well-qualified and perhaps married to a British citizen).This approach reignites the debate on the undeserving poor (Romano, 2017), deeply embedded into British welfare policy for decades, but which in the current climate exudes a particularly xenophobic tone.The impact of a toxic media environment (Balch and Balabanova, 2016) stigmatising A2 migrants as 'welfare tourists', coupled with experiences of anti-Roma discrimination (McGarry, 2017;Tremlett et al., 2017) has accordingly created a sense of insecurity for members of these migrant communities (, 2016).Thus, we see evidence of a situation coming from the strictest interpretation of EU free movement rights that is very much outside the spirit of European Union law and integration objectives.In the face of imminent Brexit, we fear that circumstances may become even harsher for EU Roma citizens on the margins of British society.
through the mechanisms of welfare benefits regulation and other state interventions, seeking to persuade current and would-be migrants eager to settle in Britain prior to Brexit INTERSECTIONS.EAST EUROPEAN JOURNAL OF SOCIETY AND POLITICS, 4(3): 81-105.

'
they all [are] working 12 hour shifts in one of the four or five major factories that are around, [e.g.] in plastics [industry] […] flour factory, rice factory, chicken factory' (translator for a focus group/NGO advice worker referring to Czech and Slovak Roma recent migrants).