Brexit, Covid and Bulgarian migrants in the UK: stay or return?

ABSTRACT Much has been written about Brexit and migration, but little from the perspective of Bulgarians in the UK. Through an online questionnaire followed by in-depth interviews, we study Bulgarians’ ‘stay’ or ‘leave’ intentions in light of Brexit and then Covid-19. Combining cross-tabulation and regression analysis of survey data and narrative analysis of interview scripts, we show how migrants’ plans are dependent on their position in the labour market and document the experiences of those who have returned. We find that stayers are more likely to be highly skilled and longer-resident in the UK. Most returns are only partially related to Brexit and are often finalized following the impact of Covid-19; family considerations are the main trigger to return. Many returnees are dissatisfied with life in Bulgaria. The article demonstrates that EU migrants’ mobility decisions post-Brexit need to be examined in relation to other contingent factors operating at various scales.


Introduction and key questions
Some dates remain iconic in the personal and collective memory: the destruction of the twin towers of the World Trade Centre on 11 September 2001, the London bombings on 7 July 2015, and, most recently, the Russian invasion of Ukraine on 24 February 2022. For the estimated 3.3 million EU migrants and their families living in the UK at the time, 23 June 2016 has an analogous, enduring resonance: the day of the UK's controversial referendum on remaining in or leaving the EU that resulted in a narrow (51.8%) majority for 'Brexit'. The aftermath of the referendum sparked a prolonged period of uncertainty for EU citizens living in the UK, who had previously been the beneficiaries of 'freemovement' rights across the 28 countries of the EU. Subsequently, these rights became part of the protracted and contested withdrawal procedures, reflecting the rhetoric of the 'Leave' campaign, which promoted 'uncontrolled immigration from Europe' as one of its key slogans.
There has been an explosion of 'Brexit Studies' since 2016, including research on EU migrants in the UK who saw their status shifted from 'EU citizens' to a new categorization as 'immigrants.' Recent literature has framed their reactions to such a 'transformative geopolitical episode' (Kilkey and Ryan 2021) as a 'rupture' in their lives , forcing them to abruptly reconsider their future plans. Should they stay and grapple with the bureaucracy, expense and uncertainty of securing 'settled status' and, perhaps, British citizenship? Should they hasten their return to their home country -but what future would await them there? Or should they seek to 'move on' to another European country or elsewhere in the world?
Then, for much of 2020 and 2021, migrants' decision-making regarding their future mobility was strongly affected by the Covid-19 pandemic. On the one hand, the pandemic resulted in unemployment and loss of income for many EU migrants engaged in sectors hit hard by the new restrictions on working, such as agriculture and factory work. On the other hand, bans and strict controls on international travel restricted their ability to return home or lead mobile, transnational lives.
Given the plethora of research on Brexit and its impact on various migrant groups, we strive for novelty and make three claims for originality. First, prior research remains bounded within the timeframe of ongoing uncertainty, which lasted from the referendum until the UK government's announcement of the settlement scheme for EU nationals in January 2019. Our research has taken place since then. Second, we study Bulgarian migrants, a group which has been somewhat overlooked, especially by UK-based researchers who have overwhelmingly concentrated on Polish and Romanian migrants. Third, nearly all research on EU migrants' reactions to Brexit has been qualitative, based on rather small-N interview samples. We employ a dual methodology: a quantitative online survey of Bulgarians residing in the UK (N = 360); and a follow-up interview round with 21 Bulgarian migrants, comprising those still in the UK and those who have returned to their home country.
We address three questions. First, what factors shape Bulgarians' plans to remain in or leave the UK in light of impending Brexit? We say 'impending' because the timing of the online survey, administered in March -April 2019, came during the long transition period between the referendum and the UK's formal exit from the EU in January 2021 and after the rules for achieving settlement in the UK were announced in January 2019. Second, to what extent are migrants' changed plans (i.e., to 'leave' the UK) dependent on their skills and position in the labour market and their perceptions of their future situation? This is again largely answered by data from the online survey, including answers to some 'open' items in the questionnaire schedule. Third, what have been the experiences of those who have returned to Bulgaria, and how can these be compared to those of the 'stayers,' including future plans? This question is mainly illuminated by data from the interviews.
The next section gives a brief overview of Bulgarian migration to the UK. This is followed by a discussion of the Brexit context and some conceptual linkages to migrants' behaviour in the wake of the referendum as an 'unsettling event' (Kilkey and Ryan 2021). A section on methods describes the two main research instruments. Then, the results are organized under three heads. First, descriptive statistics from the online survey, subject to simple cross-tabulation, are presented to indicate who is most likely to leave in the short term, within one year. Second, a logistic regression model is specified for analysing selected variables from the online questionnaire to estimate the maximum likelihood to remain in or leave the UK in the longer term, within five years. Third, answers to questions in the interviews are used to explore more qualitatively the views and experiences of two sets of participants: those who have remained in the UK post-Brexit, and those who have returned to Bulgaria. The conclusion summarizes the results in light of the key concepts deployed and compared to similar studies.

Background to Bulgarian migration to the UK
The fall of the Berlin Wall marked a new era of intra-European travel and migration: Europe became 'a continent moving West' (Black et al. 2010). The scale of East -West mobility was greatly enhanced by the EU's eastern enlargements. With Romania, Bulgaria joined the EU in 2007, as part of the second enlargement, following the much larger eastern expansion of the EU in 2004. Before 2007 Bulgarians entered the UK either legally, through various officially sanctioned schemes (the Highly Skilled Migrant Programme, self-employment visas, the Seasonal Agricultural Workers Scheme), or as visitors/tourists who overstayed. These access routes continued to operate during the seven-year transition period before full free movement, and EU citizenship rights were granted on 1 January 2014.
According to data from the UK's Office for National Statistics, the number of Bulgarians increased from around 5,000 at the 2001 Census to 49,000 in 2011, 76,000 in 2016 and 128,000 in 2019, when our survey research started. 1 The regional breakdown for 2019 shows 61% of Bulgarians concentrated in London and the South-East region. Occupational data reveal them to be a diverse community, ranging from professionals and higher-education students to manual workers. The latter comprises agricultural work, construction trades, tourism and hospitality, domestic cleaning, and care work. Another ONS statistic confirms their clustering in the low-wage sectors of the economy: Bulgarians, on average, are three times more likely than the UK workforce to be employed in 'elementary occupations' − 29.8% vs 10.7%.
Compared to Poles and Romanians, research on Bulgarian migrants in the UK is scarce. The leading authors on Bulgarian migration to the UK are Genova (2016Genova ( , 2017Genova ( , 2020 and Manolova (2017Manolova ( , 2019). Manolova's research is on Bulgarian would-be migrants' imaginings of life in the UK -seen as the destination country where they can aspire to a 'normal life,' as opposed to Bulgaria where, in the words of one interviewee, life 'is just a brutal fight for survival' (Manolova 2019, 68). Manolova interviewed 25 prospective migrants in Bulgaria, mainly 'ordinary people' (only two were university graduates).
Genova's research is on students (N = 21) and graduate professionals (N = 16) and hence refers to a narrow cross-section of Bulgarians in the UK. Bulgarian migrants are seen as occupying a space 'between a rock and a hard place' in that they have to confront stereotyped categorizations in both countries, a situation described as 'double-sided othering' (Genova 2017). From the Bulgarian side, partially reflecting the communist legacy when emigration was banned as a traitorous act, emigrants are seen as disloyal escapees who betray their national cause. From the British side, Bulgarian migrants, alongside others who 'mass-migrated' from Central and Eastern Europe since 2004, are subject to a generalized negative discourse of 'Eastern Europeans' as a new 'Other', perceived as 'poor, badly educated, benefits-driven, potentially dangerous, unskilled migrants' (Genova 2017, 38). As a response to these stigmatizing discourses, Genova (2020) advances the notions of the 'ambassador,' who projects to the host society a 'corrective' image of Bulgarian migrants as educated, hard-working and eager to integrate, and the 'enlightener' who applies a similar corrective lens towards the home country, promoting the benefits that skilled return migrants can bring.
The empirical research behind the above-cited papers by Genova and Manolova was carried out during 2013-2015, before the Brexit referendum. However, in Genova and Zontini (2020), we find 13 Bulgarians (and 17 Italians) interviewed in the postreferendum transition period before Brexit was finally enacted in January 2021. Genova and Zontini explore the 'liminality' of their participants' lives, shaped by the shock of the referendum result, the disruption to their plans, and the emboldened antiimmigrant and nationalist sentiments of the British population.
We also acknowledge that, although Bulgarians (with Romanians) only achieved full freedom of movement rights in 2014, they were subject to a longer history of demonization, especially around the contested issue of 'benefits tourism' and other measures to erode EU citizenship rights (Barbulescu and Favell 2020). This longer-term trend to roll back welfare entitlements serves to 'de-exceptionalize' Brexit and to view it instead as the culmination of inter-related events and policy stances. 2

Conceptual landscape
Our analysis rests on a range of conceptual foundations. A key underlying notion is that the referendum not only changed the legal rules for EU migrants in the UK but also created a different, more antagonistic 'atmosphere' for them. Building on the UK government's self-declared 'hostile environment' towards immigrants, Brexit added new anxieties for migrants from the EU who hitherto had enjoyed relative privileges which gave them more or less equal rights to UK citizens. So, as well as the instantaneous 'rupture' that the referendum result produced, both at a geopolitical level in terms of Britain's relationship with Europe and at the individual level of upset in people's lives, we also observe a marked change in the 'affective atmosphere' of public-sphere discourse and everyday life . Under the post-Brexit conservative governments led by Theresa May, Boris Johnson, and Liz Truss, official rhetoric became more nationalistic, turning away from Europe towards a vague notion of 'global Britain.' At the level of the everyday, there was a noticeable rise in hate speech and other forms of violence towards people from 'abroad,' including those from Europe who, although predominantly 'white', were 'visibilized' by their new status as 'immigrants' and by their names, accents and other cultural features (Guma and Jones 2019; Rzepnikowska 2019). The racialization of 'East European' migrants in the UK was not a new development initiated by Brexit but enhanced a longer-established trend to negatively portray migrants from 'underdeveloped' European countries and discriminate against them in terms of work opportunities and wage levels (Fox et al. 2012), access to welfare benefits (Burrell and Schweyher 2020) and the processes of 'bureaucratic bordering' that permeate and impede the regularization of status and a broader integration in the host country (Manolova 2021). In a nutshell, many EU migrants were made to feel no longer welcome. Therefore, we suggest that those feeling most rejected and discriminated in the 'atmosphere' of Brexit would be most likely to leave.
Second, there are important connections between EU migrants and the UK labour market. Economic principles are key here since most Bulgarian and other EU migrants (especially those from Central and Eastern Europe) relocated to the UK precisely to improve their economic well-being. As many studies have shown, their prime motives were to access better employment opportunities, higher incomes, and better career prospects over the longer term (Botterill 2018;Burrell 2010;Grabowska 2018;King et al. 2018;Moroşanu et al. 2021). We can therefore assume that those who are doing well in the UK labour market, advancing their economic well-being and careers to a level that would not have been possible had they stayed in Bulgaria, would be less likely to want to return to their home country. On the other hand, those who have been more negatively impacted by Brexit, for example, by perceiving that they are discriminated against in the workplace, or left in precarious jobs with low wages, would be more likely to consider return migration.
A third set of conceptual ideas focuses on the time in the migration process. Engbersen and Snel (2013) have argued that post-accession movements within the EU take the form of 'liquid migration', which refers to the circumstances where 'migrants do not settle permanently, but move back and forth from their source country to receiving countries. Many stays are short, while others opt for a medium-term or longer-term stay' (Engbersen and Snel 2013, 33). Liquid migration has been critiqued for giving an exaggerated impression of the fluid and temporary nature of East-West movements within Europe. Bygnes and Erdal (2017), for example, have described how, over time, Polish migrants in Norway become 'grounded' in the host country through improving their work and income situations, settling with their families and having children enrolled in school. Similarly, Grzymala-Kazlowska (2016) documents processes of 'social anchoring' whereby migrants develop strong economic, family and affective ties to the host country, making return migration less likely as time passes. Whilst liquid migration as an operational concept responds to the uncertainties created by unexpected events like the economic crisis, Brexit and Covid -migrants can simply react by moving back or moving on -the longer-term embedding processes described above make a return more challenging and, therefore, less likely to occur by force majeure.

Methods
Two research instruments were used to answer the research questions posed in the introduction and link them to the conceptual ideas outlined above: an online questionnaire survey and a round of follow-up interviews. The online survey, entitled 'Bulgarians working in the UK post-referendum', was conducted from March to April 2019. The questionnaire was in Bulgarian and contained 75 questions organized in three sections: demographic characteristics, labour market participation, and the impact of the EU referendum. Three open-ended questions asked respondents to elaborate on their reasons for migrating from Bulgaria, why they chose to move to the UK and their concerns over Brexit.
Like most online surveys, the resulting sample cannot be claimed to be random or representative. Because the first author was a native Bulgarian speaker, we were able to build trust and reach a diverse range of respondents, hedging against bias through multiple entry points and snowballing on from these. The link to the questionnaire was publicized through Bulgarian community associations, Facebook groups, language schools, ethnic shops and personal contacts. Such a wide spectrum of access routes ensured a respondent sample with a diverse range of socio-demographic characteristics in terms of region of residence in the UK, age, education, duration of stay, and labourmarket status. Nevertheless, we acknowledge that, given the nature of online surveys, certain groups in the Bulgarian population in the UK were over-represented in the respondents' profile, namely women and the better-educated, plus, of course, online media users. The final sample comprised 360 respondents.
Although the questionnaire provided a modicum of qualitative data in the form of (mostly short) answers to the three open-ended questions, more detailed and nuanced answers to the key questions underpinning the research came from the interview phase. Most of the interview participants had already answered the questionnaire. Again, we note caution about the degree of representativeness of the interviewees since they are selfselected from a non-random sample. However, their views are broadly in line with other post-Brexit interview surveys with EU migrants, especially those from the post-2004 accession countries (Duda-Mikulin 2020; Guma and Jones 2019; Rzepnikowska 2019).
The first author conducted interviews with 21 Bulgarian immigrants between 2020 and 2021: 10 who were still in the UK as 'stayers' (for the time being) and 11 who had returned to Bulgaria (again, for the time being). The interviewees were recruited through a process of networking and building trust, with the interviewer sharing their professional affiliation, the purpose of the study and the use of the collected data. Most interviewees were very open and willing to relay their experiences to contribute to the collective knowledge on Bulgarian migration to the UK. Given that these two years were beset by the Covid-19 pandemic, most interviews were conducted remotely. Although screen interface interviewing cannot replace the advantages of in-person interaction, such communication has increasingly become accepted and normalized over the past two years. Moreover, there are significant advantages in terms of researchers' time and the ability to easily contact geographically dispersed populations -including the participants in Bulgaria -without the need for expensive and time-consuming travel.
The interviews did not collect systematic data beyond age, gender, occupation, time in the UK, dates of arrival and (for returnees) departure, legal status and place(s) of residence. Rather, they focused on more discursive accounts of participants' experiences of living and working in the UK during Brexit; and, for returnees, their feelings about their decision to return to Bulgaria and their post-return experiences there. Interviews were carried out within the conventional ethical parameters for research with human subjects, including informed consent to be interviewed and recorded, the right to refuse to answer any of the questions and to terminate the interview at any time. Interviews lasted, on average, 45 minutes, ranging from 20 minutes to 1 hour and 40 minutes. Generally, the shorter interviews were with men. Like the online survey, women respondents outnumber men by two to one. The age range was 28-66. Most interviewees were in their 30s, 40s and 50s, slightly older than the age profile of the questionnaire respondents.

Online survey: respondent profile and bi-variate relationships
From the extensive array of information generated by the questionnaire, we select data most relevant to the issues dealt with in this paper. First, some basic features of the respondent sample. Seven out of 10 respondents were women. The mean age was 38 years; 80% were aged 25-54. The mean length of stay in the UK at the time of the survey was just over five years, ranging from less than one to 37 years. 'Excellent' or 'good' English-language skills were self-reported by 56% of respondents, which is consistent with their overall high level of education (60% had third-level qualifications) and generally high labour-market status (46% were in professional, technical or other qualified jobs at the time of the survey). Three-quarters had other family members in the UK, including 28% with children. The geographical distribution of respondents was close to that portrayed by ONS data: 58% were located in London and the South-East, the remainder distributed in much smaller numbers across other regions, including Scotland and Wales. Regarding the more opinion-oriented questions, one-third said that they had experienced more discrimination in the labour market since Brexit, and the same proportion opined that they would earn more in the UK if they were not Bulgarian. Finally, six in 10 had confirmed their legal status in the UK by acquiring (or applying for) settled status or British (dual) citizenship.
We now link the survey responses to the key questions and operational concepts outlined earlier. The key dependent variable is the dichotomous variable 'stay in the UK vs leave the UK' as a result of Brexit. For the shorter-term (within one year) variation of the 'leave or stay' intention, the sample was rather evenly split: 47% answered 'remain for certain' or 'perhaps remain,' and 53% answered 'leave for certain', 'perhaps leave' or 'uncertain whether to leave.' Following the logic of our earlier arguments, we expect two sets of potentially explanatory relationships to the dependent variable. First, those who expressed intentions to stay would be: (i) longer established in the UK (and hence, likely to be older), (ii) highly skilled and well-paid (they would have most to lose by returning to Bulgaria), and (iii) satisfied with their lives and prospects in the UK (also for their children's future). Second, and by the same token, those most likely to express an intention to return to Bulgaria or move elsewhere would be: (i) recent arrivals in the UK and, hence, younger respondents, (ii) in lower-skilled, low-paid, temporary and precarious jobs, and (iii) feeling discriminated for being Bulgarian or 'East European.' Tables 1 , 2 and 3 confirm these hypothesized relationships for three independent variables: year of arrival (hence, length of time in the UK), the skill level of the profession or employment followed, and perception of discrimination in the labour market. Table 1 shows a clear and statistically significant relationship between the length of time in the UK and intentions to stay or leave. Other things being equal, a longer presence in the UK is likely to result in the development of various kinds of anchoring within British society, including income and career progress, language learning, a sense of feeling increasingly Source: Online survey 2019. Statistical tests: Chi-square = 9.267**, df = 3, p = 0.026. When the duration of residence is measured in individual years, Pearson's r = 0.157***, p = 0.007, 2-tailed. For this and the other tables in the paper, statistical significance levels are * for 0.1, ** for 0.05 and, *** for 0.01.
'at home', and family ties (e.g., British partner, birth of children etc.). For the longestresident category (in the UK from 1989 to 2006), more than twice as many respondents declared their intention to remain compared to those considering leaving. For the larger numbers of migrants in the middle two time periods (2007-2013 and 2014-2016), there is a slight preference for leaving, whilst for the most recent arrivals, there is a stronger tendency to intend to leave. We will explore these bifurcated intentions -and the realities of those who have returned to Bulgaria -in more detail via the interview data later in the paper. For now, we briefly note some frequently posted comments by the questionnaire respondents, which indicate their progressive social embedding or 'grounding' in the UK over time through family relationships and the presence of children (cf. Bygnes and Erdal 2017). Phrases repeated over again, in these or similar words, were: 'I married an Englishman and stayed;' 'Better education for my children;' and 'For the normal development of my child.' As we shall see in more detail in the in-depth interviews, the trope of attaining a 'normal life' in the UK through staying long-term, adapting and planning for the future of one's children is a widespread narrative theme amongst Bulgarians in the UK (Manolova 2019). Table 2 explores the relationship between labour-market status and intention to leave or remain in the UK, reflecting the hypothesis that the higher-skilled would be more likely to stay to enjoy higher incomes and better career prospects, whereas the lowerskilled, in poorly paid and precarious jobs, would have less to lose by returning (Ghosh 1996). We classified work status by scrutinizing what the respondents said about their current or most recent employment. Typical examples of our allocation system would be high-skilled, comprised of academics, doctors or business owners; uppermiddle-skilled, comprised of technical and office jobs; lower-middle-skilled, comprised of tradespeople or residential care-home assistants; and low-skilled, comprised of labourers and cleaners. We find a clear, albeit not statistically highly significant (p =  0.087*), relationship between skill level and intention to remain in the UK post-Brexit, at least over the short term, confirmed by several comments in the online questionnaire relating to security of employment, personal and professional development and the quality of the health system. Table 3 takes on board the third hypothesized relationship, namely that those respondents who feel they would be further discriminated against in the workplace due to antiimmigration sentiments fuelled by Brexit would be more likely to express an intention to leave. Again, this relationship is supported by the survey data, significant at the 0.05 level.

Exploring the stay or leave intention with logistic regression
The second stage of our quantitative analysis involves a logistic regression model to determine which independent variables influence the stay or leave intention over the longer term, five years hence. Before we outline our model, we touch base with two existing large-N surveys of Brexit-related migration intentions.
First, McGhee et al. (2017) analyse data from a targeted online survey (N = 894) of Poles living in the UK. This survey was carried out during the early months of 2016, before the referendum. The survey questions were thus somewhat speculative and asked about possible plans if the referendum vote was for 'Leave,' which it was. Awareness of EU migrants' civic rights was the most significant determinant of plans to remain in the UK after a 'Leave' result. The majority of respondents (72%) were planning for 'civic integration' (51% for what was then called permanent resident status, 21% for UK citizenship). According to the regression analysis, the civic-integration 'remain' plan was more common amongst older migrants, the self-employed, and those using English in the workplace. Insignificant independent variables were gender, length of time in the UK, and educational level.
Second, Jancewicz et al. (2020) deploy logistic regression models on CAPI (computerassisted personal interview) survey data on Poles in the UK (N = 472). Again, the timing of the survey is significant: June -September 2018, two years after the referendum but still in the uncertain 'transition' period before the precise rules for acquiring settled status were announced in January 2019, along with the rather harsh threshold criteria (points based on income and skills) for new immigrants coming to the UK. The survey questions were not aimed directly at the role of Brexit on its own but were designed to tease out whether the Brexit element was part of broader push-pull factors. The regression results indicated that the most important influences shaping decisions to leave the UK or stay were economic factors and migrants' employment and income experiences. Standard demographic variables -age, gender, duration of stay, educational attainment, marital status -could not explain which respondents found Brexit 'discouraging' and were thus more prone to leave. Unsurprisingly, holders of British citizenship or permanent residence status were less likely to be anxious about Brexit.
For our model, we select specific variables from the online survey most logically relevant to the questions at hand. This meant slimming the sample down to 213, since 147 residents did not answer one or more of the survey items required by the model. Table 4 sets out the selected variables and their summary statistics, as well as the expected direction of the relationship with the dichotomous dependent variable of remain or leave. Table 4 shows that the mean age of the reduced sample (like the whole-sample average) is 38 years; six out of 10 respondents have excellent or good English; and nearly half are in professional, technical or other qualified jobs. These three variables are hypothesized to be indicators of longer-term stable integration and, therefore, a desire to remain in the UK (positive sign). On the other, negative side of the hypothesized relationship, onethird of respondents envisioned being discriminated against in the UK labour market, and close to one quarter thought they would earn more in the UK if they were not Bulgarian. Finally, over two-thirds of the sample have legalized their long-term status in the UK, either through dual Bulgarian and British citizenship or via the scheme to acquire settled or pre-settled status -an indication of a commitment to remain in the country.
In modelling the determinants of Bulgarian migrant intentions to remain in the UK long-term, post-Brexit, it is important to consider the binary nature of the dependent variable. We define the decision-making equation as: where Rem � i is a partial latent dependent variable that captures the i th individual's intentions to remain in the UK long-term, x i is a vector of the migration decision-determining variables for individual i and β is a vector of fixed unknown coefficients to be estimated, and u i is the error term, u i N 0; σ 2 ð Þ: Thus, from [1] and [2], the probability that an individual will choose to remain in the UK long-term, i.e., Prob(Rem = 1), is given by: where F is the logistic cumulative distribution of u i .
The log-likelihood function, L, may be expressed as: The logit estimates for the 'leave or remain' decision equation are reported in Table 5. The overall goodness-of-fit is satisfactory by the standards of cross-sectional binary models. Three of the hypothesized effects are reported as statistically significant at the 5% level using two-tailed tests. Age and employment in skilled professions exert a positive impact on the decision to remain in the UK. Contrarily, the expectation of an individual to be discriminated against if he/she were to look for work or establish a business is, as anticipated, a negative determinant of the decision to remain. The computation of the marginal impact effects (Table 5, right-hand column) offers a more detailed interpretation of the findings. Getting older by a year raises the probability of deciding to remain in the UK post-Brexit by 0.6% points, ceteris paribus. Older migrants are more likely to have been in the UK longer, to be married or in a stable romantic partnership (including with a non-Bulgarian), to have children and to own a property -all factors that anchor them to the place where they are currently living and make it more difficult to 'up sticks' and leave. They may be experiencing the social embeddedness or 'grounding' that leads them gradually to regard the UK as their longterm 'home' (Bygnes and Erdal 2017;Schewel 2020).
Second, there is a correlation between having a skilled job and wanting to remain in the UK. Included under the 'skilled' category here are both high-skilled professionsuniversity lecturers and researchers, computer and IT specialists, doctors and other medical staff -and skilled tradespeople such as plumbers and electricians. As Table 5 shows, having a skilled profession increases the intention to stay by 13.1% points. These respondents perceive and experience work and income in the UK as decidedly more advantageous than returning to Bulgaria, where their qualifications would be less well rewarded, if at all, since they may be faced with unemployment (Ghosh 1996). Beyond the income aspect, respondents appreciated the favourable promotion possibilities and the better working conditions and employment protection available in Britain, evident in many written comments in the online questionnaire and the interview narratives covered in the next section.
The third statistically significant relationship in Table 5 is between perceptions of likely discrimination when looking for work, being at work or setting up a business, and future mobility intentions. A person who experiences or perceives this relative deprivation is 14.7% points more likely, ceteris paribus, to intend to leave the UK post-Brexit. Conversely, those who experience career advancement and appreciation at work are more likely to remain. Whilst the three statistically significant relationships evidenced by our analysis in Table 5 confirm expectations, how about the non-significant outcomes, and how does this array of findings compare with other quantitative studies of post-Brexit migration intentions? Our analysis showed that acquiring formal legal status in the UK was not significantly related to the intention to stay long-term, contrary to what we hypothesized. However, this variable, termed 'civic integration' by McGhee et al. (2017), was correlated to plans to remain in their study of Poles' mobility intentions. A similar result was indicated by Jancewicz et al. (2020). However, as some of our interviews with returnees in Bulgaria show, acquiring citizenship or settled status permits migrants to return home for a while, secure in the knowledge that they can move back to the host country whenever that want.

Stayers and returnees: insights from the interviews
In this section, we adduce qualitative evidence from the 21 interviews, divided roughly equally between the 'stayers' in the UK and the 'returnees' in Bulgaria. However, this distinction is not absolute: some of the stayers talked about relocating to Bulgaria at some point, and some returnees discussed possible moves back to the UK.
We might assume that stayers are more inclined towards life in Britain compared to how they remember or imagine life in Bulgaria or that returnees are more critical of post-Brexit Britain compared to their post-return lives in Bulgaria. In fact, it often transpired that stayers were highly critical about some of their experiences in Britain and how things were unfolding in the wake of Brexit, and many returnees were far from happy about their current situation in Bulgaria. In what follows, we pick out common narrative themes that emerged from close scrutiny of the interview transcripts, with some typical interview quotes to illustrate how these themes were articulated. We link our analysis of the interview data to the three questions posed at the outset of this paper -about factors shaping the 'stay' vs 'return' decision of the migrants, about the influence of Brexit (and Covid) on this dilemma, and about the experiences of those who have returned.

Stayers: 'Life is so much easier here [. . .] you can lead a normal life'
Bulgarians have migrated to the UK mainly for economic reasons: to access jobs which are much better paid than in Bulgaria and to have the chance to improve themselves and advance their careers. A few participants arrived as students and stayed on, either for career reasons or because they met someone with whom they became romantically involved. For all participants, including those back in Bulgaria, the economic rationale for coming to the UK is the most dominant narrative thread in their interviews. In the pithy phrase of Jana, aged 41, a medical consultant, 'We are here for the money.' Below are two examples of interview quotes, from different ends of the skills spectrum, about how economic motives shaped both the original migration decision and the subsequent plan to 'stay on' after the referendum result, even if an eventual return to Bulgaria is not ruled out. The cases also illustrate how economic and career motives often intertwine with personal and family circumstances. The first case study is 30-year-old Ivan, who currently works as a parking warden; he has secondary education in Bulgaria; he came to the UK in 2015 to the town where his mother's cousin was already living.
It was always my dream to come to the UK, to see how life is here . . . . I was 24 at the time. When I came here, I said to myself that I wanted to work, save some money, and live well. I could understand English, but speaking was difficult . . . . My first job was a waiter in a Greek restaurant [and] gradually I was promoted and became restaurant manager. It was increased responsibilities, higher pay, but a lot of exploitation. I was working significantly more than the legal limit on hours. I was happy I was paid but it was a lot of hours of work.
[. . .] I have obtained settled status, and now I am preparing to apply for British citizenship . . . I really like life in the UK -life is so much easier here . . . you can lead a normal life and afford many things. [. . .] For the next ten years I plan to stay in the UK. My girlfriend is Albanian . . . we might return to Greece, where she was born, or to Bulgaria or Albania . . . but first I need to save money.
Second, and a good example of a highly skilled migrant committed to staying in the UK post-Brexit, we take the case of Draga, aged 50, who originally arrived in the 1990s to do a PhD and is now a secondary school teacher. By the time of the referendum, she had a master's degree and a doctorate and over ten years of work experience in UK universities. Draga has a daughter and is separated from her English husband. As well as recounting her motivations and career trajectory, she also sums up the stereotypes she observed of Bulgarians on the part of the British population.
The common perception of Bulgarians in the 1990s and 2000s . . . was all about 'prostitutes, corruption, mafia'. Generally, people did not know about the geography of the country . . . , and some people did not even know it was in Europe. After Bulgaria joined the EU in 2007 . . . the perception changed to 'cleaners and builders.' I can say that in my 26 years of living in this country, I have hardly heard anything positive about Bulgaria in the mainstream news . . .

[. . .]
Random people kept asking me, with an unfriendly curiosity, 'Where are you from?' . . . And then, after the referendum, 'When are you going back to your country?', creating a feeling of not being wanted . . . The above quotes illustrate a range of experiences of working in the UK in the context of Brexit and a hostile environment towards Bulgarians and other 'Eastern' Europeans. Key themes that emerge are good pay, regular employment, self-improvement, achieving 'civic integration', acquiring non-Bulgarian partners and various family links and responsibilities. Draga, for example, was torn between care duties towards her ageing parents in Bulgaria and her 'half-English' daughter in the UK.
The following extended quote is from Maria, a 61-year-old divorced mother and now a grandmother. Her account exemplifies two other essential themes pertinent to our broader analysis: the trend of de-skilling widespread amongst Eastern European migrants in the 'West;' and the selective character of post-Brexit and Covid-related return migration. Maria had been a business owner in Bulgaria who came to the UK in 2008 to develop an offshoot in food-industry machinery linked to her main enterprise in Bulgaria.
I came here to expand my business internationally . . . After a few weeks, I realized that nothing was happening. Then a friend suggested that I start cleaning some of the houses she was cleaning in London. Imagine -I am a qualified engineer in food production, I was in my late 40s at the time, and I had to clean houses! Before, I had my own cleaner in Bulgaria, but now I was a cleaner.
After a few months, Maria brought over her two teenage children, who had been living with her parents in Bulgaria. She continued her work as a cleaner to support them while also exploring other options. She re-trained as a maths teacher but was put off by the rough behaviour in secondary schools, so she carried on with the cleaning, which gave her freedom and little stress, whilst allowing her some hours to give private maths tuition. She does plan to return to Bulgaria to retire eventually -'I have a dream to return to my summer house in Bulgaria' -but this depends on what her children and their partners want to do.
After the referendum, I know of many Bulgarians who returned to Bulgaria. Many also returned because of Covid. They lost their jobs; work dropped in value, so their real pay dropped . . . The Bulgarians who returned were mostly in construction and cleaning.
Although the de-skilling phenomenon is common amongst studies of Central and Eastern Europeans in the UK and is borne out by statistical comparisons (Johnston et al. 2015), this is far from the whole story. The de-skilling may only be temporary whilst the migrants find their feet and improve their grasp of English; other studies document the trend towards better employment conditions and pay over time (e.g., Moroşanu et al. 2021). Looking at the participants we have already quoted, recall that Ivan evolved from waiter to restaurant manager. Maria got her teaching diploma whilst continuing with her cleaning job. These examples play into a widespread narrative theme of the UK being a fertile terrain for self-improvement and meritocratic reward for those who work hard and deserve a promotion. This becomes a powerful argument for staying in Britain despite the 'repel' factors of  When asked what they particularly liked about living and working in the UK, all of the participants, including those who had returned to Bulgaria, alluded to the flexible and responsive UK labour market and the fairness and efficiency of official institutions. Here is a typical example: Compared to Bulgaria, things are better organized in the UK. There is less bureaucracy; if you need help, you can get it . . . In Bulgaria, you have to bribe people to help you (Plamen, 41, IT engineer).
Plamen's reference to corruption in Bulgaria links to other aspects of remembered or perceived life in Bulgaria, which act as arguments against return. The next subsection looks at the experiences of the 11 participants who have returned to Bulgaria.

Returnees: 'Return to Bulgaria is not an easy step'
The accounts of life in the UK given by returnees in Bulgaria were not substantially different from those described in the previous subsection on the stayers. What differed was the strength of the pull factors drawing people back 'home'. These were usually not economic factors since Bulgaria is still poor by European standards, with low incomes and limited employment and career opportunities for returning migrants. The most common reasons for returning were the push factor of economic distress in the UK (loss of job or greatly reduced income during the lockdown periods), disillusionment over Brexit and, most frequently cited, family ties and responsibilities. Folded in with some of these return mechanisms were occasional references to nostalgia and personal crises such as divorce/separation. Economic factors shaped return moves in two senses. First, there were those whose economic circumstances in Britain were precarious due to a combination of the work itself (temporary, poorly paid etc.) and the loss of work due to Covid. A good illustration of this is Milena (56), who had spent her time in the UK working in the residential care sector. She arrived in the UK on her own in 2008, leaving behind her husband, from whom she was separated, and their two children. Her daughter subsequently joined her in 2010, whilst her son remained in Bulgaria -a typical illustration of how East European migration to the UK often creates transnationally split families. Since she returned to a city on the Black Sea coast, she had not worked but was called once in the summer of 2021 to work for two months in the hotel in the UK where her son was working. She responded to the acute shortages in the post-Brexit hospitality sector, thereby confirming the in-betweenness and informality of Bulgarian workers who, as Manolova (2017) has shown, had a vulnerable economic existence long before Brexit. I did not originally plan to return to Bulgaria, but I was living in a care home, a place without social life . . . I was on the minimum wage, like all the other foreigners working in the care home. . . . The best thing was that the rent was minimal; that's why I stayed so many years . . . Then I decided to build a flat in Bulgaria . . . and return home. . . . I returned in August 2020. . . . The pandemic sped up a bit the process of return.
The second group of economically motivated return migrants are those for whom good jobs are available in Bulgaria, usually set up in advance of the return and sometimes benefiting from an 'international' salary combined with low local living costs. Assen (31, IT programmer) is a case in point, albeit his story breaks the mould of the highly skilled being less likely to return -but there is a reason for this, as we shall see. Following his first degree in Bulgaria, Assen arrived in 2012 to do a master's in IT programming at a Scottish university. After graduating, he moved to London to develop his IT career for several years before returning to Bulgaria in 2020, having in the meantime acquired British citizenship. His main reason for returning was his family and other aspects of Bulgarian life -'parents, siblings, friends, the food and the weather.' Then he went on to describe the special economic circumstances of his return: The family motive was the most widespread trigger of return: sometimes as the prime or only motive, on other occasions combined in a mix of reasons, as with Assen. Familyrelated return factors range quite widely, involving various combinations of ageing parents, spouses/partners, children and even grandchildren. Often the 'family-return' narrative is combined with nostalgia for the 'homeland.' A good example of this is the following interview extract from Vania. She is a graduate in business administration, worked in marketing in the UK, and now has a business importing and selling baby clothes in Bulgaria: I came to the UK in 2001 with my now husband; we were university students . . . . Later, when I gave birth to my son, the nostalgia was unbearable. Things change when one creates a family. It took us a long time to decide . . . return to Bulgaria is not an easy step . . . . We finally returned in 2010, but eight years later, in 2018, we moved back to the UK . . . [we felt that] the education system is much better than in Bulgaria . . . . We stayed two years in the UK: in the summer of 2020, we decided to return again to Bulgaria . . . . My main reason again was overwhelming nostalgia.
Vania went on to talk about the schooling challenges faced by her son with these backand-forth moves. The reason for the second emigration to London was for him to benefit from a better education system than the one he had started in Bulgaria, but then she became concerned about the overcrowded life in London and the high crime rate. She continued: In Varna, I felt safer for my kid than in London . . . . My son had already started school in Bulgaria when we moved to the UK for the second time. He adapted very quickly, took him only two months. He felt very happy, found new friends . . . . But the change back to Bulgaria was very problematic . . . he went back to the same school class he left two years ago . . . he had not practised Bulgarian spelling and writing for two years. Culturally, too, it was a shock for him. The kids in his class had changed . . . they were adolescents now . . . he was shocked by their language, how they behaved in class, how they treated the teachers, how nobody does anything to stop this behaviour. After a year, he managed to adapt, to distance himself from the vulgar reality at school.
Vania sounded a warning that children born in the UK, who had started school there, would find it difficult to adapt to the tough Bulgarian school environment. She recommended that Bulgarian families either stay abroad for their children's educational continuity and future or return earlier when their children reach school age. 3 Reintegration of children is just one of the problematic aspects of the overall experience of Bulgarians returning from the UK. In our participant sample, accounts of disappointment, disillusionment and frustration were dominant over those of happiness and satisfaction. On the whole, the latter were characteristic when the return was entirely voluntary and planned; the former characterized returns which were 'forced' by family emergencies or shaped by the 'double whammy' of Brexit and Covid. The balance between the plusses and minuses of return referenced several factors. On the positive side were nearness to family and old friends, the general social warmth of Bulgarian society, the relaxed pace of life, better weather and more wholesome food. On the negative side were the low wages, the frustrations of dealing with the bureaucracy, the endemic corruption, and the difficulty of reconnecting with former friends who have either migrated themselves or moved on with their lives in other ways. All these latter features were repeatedly contrasted with the better economic situation, the efficiency of public services and the general transparency and honesty characteristic of their life experience in Britain. Even when the economic and family situation of returnees is positive in Bulgaria, other difficulties of living there are still very challenging, as voiced by Jelyo, aged 28, a specialist in digital media.
What do I not like in Bulgaria? The mentality -we are 45 years behind the rest of the world. People are narrow-minded, even vicious . . . the whole society is confused. I saw bad things in the UK, but here things are 10, even 100 times worse. Maybe because we are a small country, these things are more concentrated; the political situation, the country as a whole. I am proud to be Bulgarian, but I am not proud of Bulgarian society. I don't want to raise my kids in this environment.
As a result, Jelyo is now planning to move to Germany, not back to Britain, because of Brexit. In Germany, he and his Greek partner can take full advantage of their EU citizenship and rights to free movement, no longer applicable in the UK.

Conclusion
We opened this paper with a list of recently historical dates. Thus, on the morning of 24 June 2016, when the referendum result was announced, the three million-plus EU migrants living in the UK felt the EU border passing across them, leaving them statically displaced on the outside. If the day of Brexit seems unconnected to the most recent awful day in European history, when Russia launched its invasion of Ukraine on 24 February 2022, then British Prime Minister Boris Johnson made that connection in a speech to the Conservative Party spring conference on 19 March 2022. He grotesquely compared the Ukrainian resistance to the Russian forces with the UK's decision to leave the EU: 'both reflected a common appetite for liberty,' he said. 4 'An appetite for liberty' was certainly not the emotion felt by Bulgarian (and other EU) migrants living in the UK in 2016. As our analysis has shown, their liberty -to move freely as EU citizens and to be able to live and work without hindrance in the UK -was suddenly withdrawn by the narrow margin of the referendum vote. Bulgarians' reactions to this moment of geopolitical rupture have been the focus of this study, particularly in how the referendum changed their future mobility plans -to stay in Britain versus return to Bulgaria. Less than four years after the Brexit vote, the onset of the Covid-19 pandemic added a further 'unsettling event' (to put it mildly) to the complexity of our participants' lives and decision-making.
Our interview data reveal how Brexit and the Covid-19 pandemic are interrelated factors shaping mobility intentions and decisions. Brexit, as a legalistic act based on the will of the (small majority of) people, reflected a wider history of hostility towards immigrants, exacerbated by the divisive effects of austerity and regional/class inequalities (King 2021), and resulted in a step change in EU migrants' mobility rights and feelings of belonging and acceptance. It was, in a sense, an 'internal shock' brewing for some time. Covid, on the other hand, was an 'external shock' which mostly affected people's employment and health conditions, often with devastating personal consequences.
Three questions framed this paper. First, which factors influence Bulgarian migrants' intentions to stay in or leave the UK post-Brexit? Second, how are these intentions dependent on migrants' skills and position in the labour market? Third, what have been the post-return experiences of those returning to Bulgaria? From the quantitative analysis of our online survey data, we found that key factors affecting the stay vs return decision were: (1) length of time in the UK (positively correlated with intention to stay); (2) skill level (the higher skilled being more likely to stay); and (3) perception of discrimination (those thinking they are discriminated through being Bulgarian are more likely to express an intention to leave the UK). These results proved statistically significant for both shorter-term (one year) and longer-term (five years) leave or stay intentions.
Analysis of the interview data, with both 'stayers' and 'returnees' (recognizing that both are provisional categories), enabled us to nuance these findings with more in-depth and variable insights. Both the stayers and the returnees emphasized the overwhelmingly economic rationale for moving to the UK pre-Brexit, i.e., to seek work, earn a higher income, and build a career. A recurrent trope was that of a 'normal life,' which was achievable in the UK but not in Bulgaria (cf. Manolova 2019). Some interviewees had acquired British (or other non-Bulgarian) partners and had British-born children. These 'anchoring' (Grzymala-Kazlowska 2016) or 'grounding' (Bygnes and Erdal 2017) factors overrode the negative impressions of life in the post-Brexit UK, which had mainly to do with heightened xenophobia (cf. Guma and Jones 2019; Rzepnikowska 2019) and a generally more hostile 'atmosphere' towards migrants, especially those from the post-2004 accession countries.
The balance between push and pull factors for the returnees swings the other way. Some were pushed to return by a combination of Brexit and Covid, although these were usually not the main elements of the return decision. Other factors which were more determinant were the pull of family obligations, nostalgia and quality-oflife considerations, including friendships, cost of living, food and the weather. However, in a final twist, several interviewees remained unhappy with their situation back in Bulgaria and complained about low wages, corruption in politics and the challenges of reintegration.
Our findings both resonate with and complement those of Manolova (2019) on Bulgarian potential migrants' vision of a 'normal life' in the UK and Genova's (2016Genova's ( , 2017 research on Bulgarian students and young professionals living in the UK. On a broader plane, our research reinforces and updates the much larger body of research on European migrants in the UK in times of Brexit, much of which, as noted in the introduction, is on the large Polish community in Britain. As we digest the long-term impacts of Brexit in various domains and emerge from the two-year shadow of the Covid-19 pandemic, the conditions for EU migrants should theoretically become clearer. They have either secured settled status, returned home or moved on elsewhere. For those whose lives were built on 'liquid mobility' (Engbersen and Snel 2013), the future is less certain because of restrictions on residence rights and controls over the border crossing. If Brexit was a 'one-off' event which upset many people's lives (and not only migrants), it has also created ongoing uncertainty and states of limbo for those who wait for clarity on what their post-Brexit, post-Covid futures will be (Hall 2022). As Genova and Zontini (2020) have stressed, many Bulgarians, and other EU nationals, are still leading 'liminal lives' in the 'liminal space' (Owen 2018) of Brexit Britain and suffering an emotional process of 'unbelonging' (cf. Mas Giralt 2020). For these and others, questions arise over how the stay vs leave dilemma intersects with life course and life stage, especially for those Bulgarians whose record of migration and mobility stretches over many years.