1Studies on Ukrainian women migrating to the domestic work sector are still rare. Research into this theme has been carried out in Italy (Scrinzi 2008; Solari 2006; Vianello 2008; Nare 2003) and in Austria (Haidinger 2008). However, the subject of their presence in Poland is virtually a terra incognita. This analysis is based on twenty in-depth interviews conducted between 2005 and 2006 with Ukrainian women working in the domestic sector in Warsaw and its suburbs. It aims at giving insight into the Ukrainian women’s possibilities to develop agency in their relationship with the Polish informal employer.

1. Defining domestic work and agency

2Domestic work is a specific type of employment. It is the commodification of services that were provided in the past by family members, mainly by women, without payment. Domestic work can be defined as services concerned with housework, i.e. cleaning, cooking, washing clothes and caring for dependent persons, such as the elderly, children or persons with disabilities. The term „domestic work” was introduced in the 1970s by the feminist movement. Until then, domestic services were not linked to productive labour (Humm, 1995). The obligations related to paid domestic service have also changed with the changing meaning of household and family, as well as with the introduction of new technologies which have mechanized housework. However, remunerated domestic work should not be associated only with services sustaining the physical existence of the employer’s household, but also with a service allowing the employers to lead a particular lifestyle; in that sense domestic work upholds the employer’s social status. Thus, domestic work is definable “in terms of a role which constructs and situates the worker within a certain set of social and economic relationships” (Anderson 1999:120).

3Remunerated domestic work means engaging in a work relation, which is carried out in a private household. A home or a private household can be described as a specific locale. It is what Anthony Giddens (1979) would refer to as a setting for interaction that has a normative base – there are rules of what to do and what not to do, based on social conventions. Thus, a household is also characterised by specific power relations, to which the migrant has to respond. The presence of a stranger – the domestic worker – in the household, has changed what was perceived as the ‘private’ sphere of household chores into something ‘public’ – someone’s workplace. Trust, that the informal rules of the household will be respected, is required for this relationship to function. Under such an agreement, tension may exist between the fact of employing someone to carry out domestic tasks and the need to preserve the intimate character of the private sphere.

4The migrant domestic workers represent a specific type of domestic worker due to their legal, social and economic status in the country of migration. The main interest of this study concerns the possibility for migrant domestic workers to develop agency in their work relationship with the employer, in light of their different political, economic and social status. According to Anthony Giddens (1979:55), action or agency refers to a continuous flow of conduct by the agent, who is an active subject that has access to a common cultural stock of knowledge which allows him/her to act in a given temporal and spatial context. Being an agent means having an internal structure which is both constraining and facilitating. Structure refers to particular structuring properties, such as rules and resources. Structures exist “paradigmatically as an absent set of differences, temporally ‘present’ only in their instantiation, in the constituting moments of social systems” (Giddens 1979:64). Giddens refers to structure as particular “sets or matrices of rule-resource properties governing transformations”. Having agency means: having wants that prompt action (motivation), rationalizing action and having the capacity to act otherwise.

2. The demand for domestic work in Poland

6Another explanation for the demand for remunerated domestic services is the inability of the State to provide adequate public social services to satisfy this growing need. In Poland, the privatisation of numerous social care services, such as residential homes for the elderly, means that they are no longer affordable for many people. In addition, the Polish beneficiaries of care services have little trust in institutional solutions; they prefer to have someone recommended from a trusted, informal source (Frąckiewicz 2002:18). There is also a cultural lack of acceptance in Poland for placing either an elderly parent in a care-institution or a child before the age of 3 in a nursery. The legacy of the poor quality of service in such institutions during communism is one of the reasons for this negative attitude. This opens a niche for work in home-based care, among others, for migrants.

7Behind the demand for cleaners is the changing lifestyle of Poles, especially those living in large urban centres, such as Warsaw. It should be kept in mind that the construction of the demand for care is different from that for cleaning. With regard to cleaning one can argue that supply creates demand, because if a person has the possibility to pay a relatively small sum to someone in order to carry out domestic chores for them, allowing them to spend that time on something more enjoyable, she/he will do so. The cleaners keep up certain living standards, and a certain social status of their employers.

3. The character and diversity of migrant domestic work

9In many developed countries domestic work is already regarded as a job for migrants (Parrenas 2001; Anderson 1993).  It is unstable and neither well-paid nor socially prestigious, making it unattractive for native workers. The demand for migrant domestic workers, combined with the restrictive migration policy, has triggered the development of informal recruitment practices. While these give migrants access to work, they also make them vulnerable to labour exploitation. Rhacel Parrenas (2001), analysing the situation of Filipino women working in the domestic sector in Los Angeles and Rome, claims that the globalisation of the labour market is the main cause behind the lack of protection of migrant domestic workers by the labour regulations of various countries. According to Bridget Anderson’s (1993) analysis of domestic workers in Great Britain and the Middle East, the labour laws in some of the countries actually generate the legalisation of ‘modern slavery’. In Great Britain, a foreigner’s contract is made with a specific employer, resulting in the lack of possibility to change employers, even in case of abuse, while in Saudi Arabia a domestic worker does not have the legal status of a worker, thus is not protected by the labour laws. According to Anderson (1993), the legal regulations in Great Britain favour the employer, giving him or her extensive power over the domestic worker. In addition, British law criminalises migrants who escape from abusive employers, by classifying them as ‘undocumented’ foreigners who should be deported. Domestic work is also generally part of the informal sector of the labour market.

11Many authors regard gender as one of the main factors influencing the character of migrant domestic work (Murray 1996; Aronson and Neysmith 1997; Anthias 2001; Anderson 2000; Pessar 2003; Pessar and Mahler 2003; Lutz 2005; Solari 2006). Household chores such as cleaning, cooking or care work continue to be regarded as women’s work. Migrant domestic work, in general, replaces or complements the work done without remuneration by the woman in the household. This is one of the reasons why domestic work is not acknowledged as proper labour. According to Bridget Anderson (2000), the resistance to accept care as labour, whether remunerated or not, stems from the belief that it is an emotional need; we care for those we love. The construction of this care myth is based on the conviction that care does not require physical and psychological work, only emotional engagement. This phenomenon seems consistent with the view of domestic work in Poland. Most Poles regard unpaid domestic chores as the essence of the private sphere. One of the results of such a conviction is the attempt to make remunerated domestic work ‘invisible’, in order to ‘protect’ the intimacy of the private sphere. Poles also refuse to accept care as labour, claiming that we do this for ourselves and for people we love (Titkow et. al 2004).

12One of the basic features which diversify the sector is the living conditions, dividing the workers into those living in, who live with the employer or cared-for person, and those living out, who live independently from the employer.  Living-in secures work, accommodation and often food. This is important for women who initially do not know the language of the receiving society or have few migrant ties, many of whom are indebt and have an irregular status (Momsen 1999:6). However, living-in has many drawbacks. Combined with the nature of care-work, it means spending most of the time inside the household. The permanent availability, being constantly ‘at work’, is psychologically highly strenuous. The migrant women have little control over their working hours. In addition, the migrant women have few chances to meet other migrants, exchange information and create their own networks. Living-in is also characterised by a lack of privacy. A migrant is dependent on the employer for her work, accommodation and free time. She is an easy target for labour exploitation, including physical and psychological abuse. On the other hand, migrants who are living out have to respond initially to higher risks than living-in workers. They have to find out about housing and work possibilities, often without knowing the language of the receiving society. They have to cut expenses by accepting poor living conditions. However, they have more independence and better-defined work duties than those who are living in. Because of this, they often earn better than the living-in workers. According to research, the migrant women in this sector often start as living-in and, once having gained experienced, move to living-out (Anderson 2001: 47).

13The risks of migrant domestic work at first glance seemed obvious: loss of skills, no possibility of upward mobility in the labour market and an overall declining social status. At the same time, however, there is financial opportunity and an increased status in relation to the country of origin. This phenomenon was observed by Parrenas (2000: 150, 244) in her study on migrant Filipina domestic workers in Rome and Los Angeles, who referred to it as ‘contradictory class mobility’. This contradiction occurs due to a decline in the migrants’ occupational status and an increase in their earnings. The Filipina women in Parrenas’ study experienced underemployment as an ‘excruciatingly painful experience’. However, when analysing the experience of migrant domestic work of Ukrainian women, the occupational aspect did not come across so strongly. That could be partly because of the pendular character of migration and the overall perception of it, against the evidence, as a temporary strategy.  The argument was maintained more in terms of dignity and social status; it was not so much the type of work, but the treatment of the employer and the power relations within the private household that affected the interviewed migrant women. Having to carry out work that the employer would refuse to do displayed the low status of domestic work. Although the women, through their care and domestic work in the home, were allowing someone else – the employer – to affirm his/her social status, the new aspect – being paid for domestic work – did however allow them to increase their status at home.

14The characteristics of migrant domestic work frequently addressed in literature, such as irregular status, no possibility for social mobility or living-in, can lead to a sensation of temporariness. One way to respond to this is by searching for protection and stability in the relationship with the employer. The development of personal ties between the migrant and the employer has to some degree led to the neutralisation of arbitral decisions and a higher degree of predictability, and has given migrants greater sense of being in control.

4. Ukrainian migrants’ view on personalised relations with the Polish employer

15Several of the migrants underlined during the interviews that their employers treat them like ‘family members’. The notion of the domestic worker being treated ‘like a family member’ was originally developed by Bridget Anderson (2000). According to Anderson, this type of treatment is characterised by expectations from the worker of having personal interest in the family of the employer, but there is little or no expectations that the employer will display interest in the family of the migrant woman. Thus, there are no mutual obligations, no entry into a community with the becoming de facto ‘part of the family’. The domestic worker as a ‘family member’ can also more easily be asked to do additional work ‘out of attachment’. Some of the interviewed migrants gave examples of being coerced emotionally by their employers to do additional work – ‘how can you not do this for me?’ or had to be available even when it was the migrant’s ‘free’ time – ‘it is as if you were leaving your own mother!’.

16However, the experiences of the women interviewed in this study also contrasted with Anderson’s findings. According to the migrants, the employers often did display interest in their lives, giving economic, social and even legal support. In the cases of some of the interviewed migrant women, close ties to the informal employer led to particular forms of support of the migrant’s family, such as the Polish employers sponsoring the education or holidays of a Ukrainian domestic worker's child. It could also lead to improvement of work qualifications. The informal employers of one of the migrants signed her up and paid for a beautician course. Another informal employer, who decided to open up a company in Ukraine, offered the migrant worker a job in that company. These actions go beyond the employer’s interest of providing favours to keep a trusted worker. The development of a close tie to the employer can be defined as migrant social capital, which refers to the ties to people who have social, economic or other capital, which enable attaining the migration goal (Górny and Stola 2001).

17However, the migrants in general were unable to return the employers’ help due to a limited amount of money, low social status and irregular legal status. This meant that the employers held a new form of power over the migrants, which could potentially be abused. It seemed that the only ‘acceptable’ form of gratitude for the employers was additional, unpaid work by the migrant. Irina, a 30-year old migrant from Lviv, worked for over two years as a living-in, taking care of a small boy. She spoke about how migrants that have little migration experience are susceptible to an increasing workload.

A foreigner, who comes for the first time to a country, goes for the first time to work, is not familiar with this, and as a rule agrees to everything “yes, I will do”, “yes, of course”, “yes, I have time”, and later, when at some moment he realises that he wants to do something for himself, to have some private time, somehow and sometimes refuses to come during the weekend to [work for] someone, then there is great offence, great unhappiness [on the side of the employers]. And sometimes there is such exploitation for small payment and it is not worth it. And then it [the additional chore] turns into a duty.

18Irina underlined the power relations between the migrant and the employer, the former being in an unfamiliar environment and lacking migration experience and the latter being in an influential position, expecting the migrant to do increasingly more work. The lack of knowledge of inexperienced migrants about acceptable working conditions gave them a very weak position in their relationship with the employer, placing them at risk of different forms of labour exploitation. The moment the migrant resisted the increasing demands, the employer felt ‘offended’ that his/her authority was questioned. Interestingly, Irina pointed to the subtleties of power relations and justified her employers’ behaviour:

There were no such… no one dictated anything. It was simply, they were as they were, but the problem was they would ask whether I could [do some additional work] and my dumbness was that I always agreed to everything.

19The employers did not ‘dictate anything’ – they attempted via euphemisms to hide their direct affirmation of power, but managed to exert control over Irina by asking her to do certain chores, which she did not refuse. She blamed herself for being submissive and not protesting against the increasing workload. Irina’s submissiveness was related to her dependence on the employers – the accommodation, being a living-in, the need to earn money and her limited social network at that time. Her acceptance of her working conditions was also linked to her initial unawareness of the degree of risk involved. Irina described during the interview the increasing expectations of her employers:

At that time they paid me one thousand (PLN) [per month] [approx. 250 Euros] and I was available from early morning, I ironed, cleaned, washed the windows, the bathrooms… and cooked for the child on their terms. Also the grandma [of the employer’s family] praised me a lot, saying that I am the only one who can cook for the child, and so … The child was taken care of with all his medications, and the house and the child were very well taken care of. Everything was always washed and ironed. With me that is how it is, I don’t feel well in a house when there is disorder. And it was always like that (…), because they had catering, then there were sometimes a lot of dirty dishes and a lot of dirty and clean table-cloths to separate, so this was additional work. And sometimes they were angry that I was not following an educational programme with the child, because I could. But when? When? I also did their shopping, I had this on my shoulders, so when?

20Irina’s lack of control over her workload was an example of how the living conditions affected working conditions. Irina, as a living-in worker, was constantly ‘available’ to the employer’s demands, thus continuously on duty. The initial care work which had been agreed on was combined with other duties, such as house maintenance. The employer accepted the migrant domestic worker’s initial submissiveness as a norm and tried to ‘gain’ the most from her work. By living-in, Irina did not feel initially to be in a position to oppose the employer’s increasing demands. It was the employer, who was the consumer of services, who demanded, set the rules and decided when the work of the migrant started and when it ended. The lack of clearly defined tasks in domestic work gave the employer more freedom to judge the migrant’s work as not completed or not carried out well enough. However, Irina made attempts to empower herself, by claiming during the interview that she did set some of the standards of cleaning and care. Only she ‘knew’ how to prepare food for the child and ‘everything’ was ‘always’ clean when she was in charge. One could assume that during the initial stages of the work relationship, the employer set particular rules, most of which were accepted by the inexperienced migrant. However, with growing experience, the migrant introduced her own work standards, in terms of how she fulfilled the tasks. She also, having solidified her work position, was able to some extent to reject the additional, new chores.

21The demands and expectations of the employer limited severely time off work in the case of several of the interviewed migrants. The interviewed live-in migrant domestic workers were not allowed, in general, to leave and return to the household as they pleased. Again, Irina said:

Officially I had free days… I had those free days, but as I said, those days somehow got entangled with their life, so that in the end… these were not my free days. I remember, at the beginning it was terrible; at the time I didn’t know that it was terrible, only now I understand this. For example, on Saturday, I remember, I asked for permission to go shopping. I asked for their permission. Of course they told me “yes, but when will you be back?”. Only now I realize this… Also when I went shopping they made… because when I went to those shops and so on… I would forget about everything and… a phone call “why aren’t you back yet?”.

22Irina reflected on how in the beginning she did not realize that her employers were invading her privacy. Having to ask for permission to leave the house, being asked for how long one will be away and being checked up on when outside the house were constraining for an adult non-family member. Free time was essential for migrants to build up their social capital, find out about different jobs and working conditions from other migrants. From Irina’s statement it appeared that when living with the employers the boundaries between the migrant woman and the employer’s family became blurred. The migrant became part of their life.

23The migrant’s status within the household – whether living-in or living-out – influenced the effects in working conditions of personalised working relations. When living-in, the negative outcome was of an economic nature and related to an increase in workload without an increase in wages, placing at stake the migrant’s migration goal. The exposure to risks related to personalisation was more characteristic for the interviewed women working in care. Availability is one of the basic features of care-work. A care-worker has to be available to intervene in case of need. However, the continuous presence of the migrant in the household increased the employers' demands, which began diverging from the earlier agreed-upon working conditions and remuneration. The interviewed living-in migrants were isolated from people other than their employer and her/his family. They had little chance to meet other migrants, develop ties, create their own network or exchange information on their working conditions. This isolation limited the migrant’s possibilities to find new employment that would enable leaving an unfavourable working environment. One of the strategies the interviewed women used, in order to improve the working conditions, was to shift from living in to living out.

24However, Ukrainian migrants in this study not only were treated “like family members”, but they also actively personalised their relation with their employers. This was visible in the story of Marta, a 31-year-old divorcee from a small town north of Lviv whom I met through a non-governmental organisation in Poland. At the time of the interview she was working in a Warsaw suburb as a cleaner and care-worker for a person with a mental disability. This person’s mother was, according to Marta, ‘like a mother’ to her:

Mrs. M. is for me simply like a mother (…) If something [happens], I call Mrs. M., right away, immediately, like a mom. She will manage everything; she will tell me how it should be, how it shouldn’t, how to behave, what to say even... I am telling you, I just felt immediately at home, like in a family.

25Marta took an active role and instead of the relation ‘being personalised’ by the employer, she ‘personalised’ her relation with the employer. The fact that she felt in her employer’s house ‘like in a family’ meant that there were good grounds to believe that she had not only the obligations, but also some of the rights of a family member. Marta did not live with the cared-for person, which gave her more privacy, freedom and control over her work. Still there was an asymmetry of power visible in how Marta described asking her employer for advice and the employer telling her how to behave and even ‘what to say’ in particular situations. The employer was the knowledgeable woman, while the migrant woman was the one in need of assistance. This asymmetry was also partly related to the age difference between the women. Marta used the unequal power relations between her and the employer to gain information and protection.

26Personalisation is a double-edged sword. On the one hand, it puts the migrants at risk of being exploited by the employer. On the other hand, it is a potential resource for migrants with an unstable status. Ukrainian migrants, although having limited resources, did not cease to be agents. An agent is an active subject, who has a shared stock of cultural knowledge and whose knowledgeability is bound by unintended consequences of intentional action and unacknowledged conditions of action (Giddens 1979). The migrants responded to the work-related risks by engaging in what Anthony Giddens (1979:39) referred to as a ‘dialectic of control’, where “all power relations express autonomy and dependence ‘in both directions’”. The power was not entirely on the side of the employer, because she/ he needed and was even dependent on the migrant’s services. Migrants used the fact that the employer treated them like a ‘family member’ to construct their own strategies to cope with migration risks or they even ‘personalised’ their relation with the employers. The status of being like a ‘family member’ also allowed escaping the low status of being a domestic worker.

5. Ukrainian migrants’ view on dehumanised relations with the employer

27Migrants, in the relation to the employer, also experience dehumanisation of the work relationship. A dehumanisation of the work relationship by the employer may result in intensifying it, attempting to make it more ‘time efficient’ and cheaper (Aronson and Neysmith 1996). Some of the interviewed migrant domestic workers commented on being treated by their employers like ‘machines’, without human needs. The services they provided, their efficiency and quality, are all that mattered to the employers and was their only justification for their presence in the household.

28During the interview, Chrystyna, the widowed mother of two children and a former book-keeper, referred several times to the memories of her working conditions at her first job in Poland. The migrant who informed her about this job, had said, “Prepare for hell”. Although warned, Chrystyna’s difficult financial situation and limited social network in Ukraine pushed her to take the risk and accept the job offer. Chrystyna became a living-in care-worker, looking after the employer’s eight children. She claimed she had lost 15 kilograms of weight during her three-month stay there. The mother of the children that Chrystyna took care of did not work outside the home. In addition to Chrystyna, a cleaner and a cook also worked in the household. The chores were intense, with the Polish woman having a managerial role. The employer had to affirm her position by successfully communicating power and authority to the migrant domestic worker. However, the Polish woman’s pressure to be efficient at work seemed to have slipped, according to Chrystyna, into exploitation:

You couldn’t just sit down simply and eat, because if you only sat down, she came straightaway. She didn’t say ‘why are you sitting here’, she gave work right away and you had to do this work immediately. That means […] there was no time to eat.

29In Chrystyna’s view, the employer denied her the right to take care of her own needs, such as eating, sleeping, or treating an illness. She was not treated as a person with particular needs, but as a service, a function. The employer considered the migrant’s wants as an excuse for not working. Often, the pressure from the employer to be more efficient was related to the method of payment. When a migrant was paid on an hourly basis, it was in the employer’s interest to set a time limit within which the cleaning was to be done. This was not the case of Chrystyna, who was a live-in domestic worker and was paid the same wage monthly. Chrystyna’s employer was the Polish woman’s husband, who was outside the house most of the day.  This created a pattern of power relationships in the household where the man was not participating directly in any domestic chores or care work, but paying for the services of the domestic workers, while the Polish woman did not work outside the home and was fully dependent on the husband financially. She exercised control within the house by giving orders to the employed domestic workers. The domination of the woman in the household in Poland was termed by Anna Titkow (et. al. 2004) as the ‘matriarchal manager’ and is a possible explanation for the treatment Chrystyna experienced. The ‘matriarchal manager’ meant that Polish women who had limited possibilities to lead in the ‘public’ sphere during communism, as well as after the transformation of the system, took over the ‘private’ sphere. For the migrant woman this meant supervision and complex power-relationships in the household, with the Polish woman not wanting to lose her dominant position with regard to care- and house-work.

30Chrystyna’s ‘manager’ did not allow her to react to the insubordination of the children, who were under Chrystyna’s supervision, even though she was held responsible for the outcome of their insubordination:

(...) when I woke up every morning at five, I had to get up at five, go to every child, wake her up; and the child didn’t wake up, because she ignored me (...). She woke up when it was already daylight and I felt the results ‘Why is the child not awake?’ [in Polish – impersonating the Polish woman]. And what about me.., I guess I am to be blamed, because I woke her up, what does she [the child] care, right? (...) One had to dress this child, by itself she wouldn’t do it, spread the legs, had to put on the underwear, (...) and she was screaming. There were two such horrible children. There were two little children whom I really liked, I liked them a lot. Well, children are not guilty that they are not raised properly. The children I liked, but not their mother, because she was treating me like a machine.

31Chrystyna referred in this quote to the dehumanisation of work relations, claiming to be treated by the Polish woman ‘like a machine’. Chrystyna responded to this functional treatment by criticizing the Polish ‘manager’ for having raised her children badly, thus accusing her of being a bad mother. Chrystyna reached for the power of the ‘powerless’: claiming moral superiority.

32Larysa, a 40-year old woman working for several employers as a cleaner, underlined how she experienced the treatment of Polish employers as lack of trust in her as a worker:

… It was very unpleasant for me… I couldn’t… this was a humiliation for me. When there were four doors, the upper landing, of those four doors and simply I forgot to clean one of them. Simply, it is not that I didn’t want to [clean], but you know how it is, I am human, I am not a machine […]… I thought, well, I will clean it immediately, and something else happened and I forgot. So the husband of this woman came and checked the whole house. This was very strange for me that it was him and not her who went [to check], and later on it was just terribly sad. And she was coming every moment, was checking whether I was just sitting and doing nothing. But I do not have such habits; even when I am alone at home I do not have the habit of sitting. I just simply work at a normal tempo. I just got used to it, although it was difficult at the beginning. And now it is so that… I do not sit normally, I work at a normal tempo (…)

33Larysa treated the first working day at new employers as a warning. It was probable that this checking on her performance would continue during the following cleanings. Larysa felt that she was not trusted as a person; she felt suspected of being dishonest, of not working properly despite being paid for work. The fact that the husband checked the quality of her work was experienced as humiliating for the migrant woman. This meant that she accepted the idea of domestic work as ‘women’s work’ and she experienced this as degradation that a man, who, in her opinion, did not know much about housework, was checking on her work. She therefore reinforced the gender hierarchies in the household. She reflected on the fact that she was only human and can make mistakes during her work. According to her, those employers expected her to behave like a ‘machine’. The services she provided and the efficiency and quality of her work were all that mattered to the employers and constituted the only justification for her presence in the household. The sensation of limited grounds to develop a personal relationship was reinforced by the fact that she was not given anything to eat during her whole day of work. While Larysa experienced the checks on her work as a disgrace, the employers thought they were acting as regular employers and wanted her to return:

And this was it. He said: you did everything alright; well you forgot this element here, well… I am sorry. We are pleased with you, well, until next week. I said: yes. And later on I came home and I was so tired, so hungry and so humiliated that I told myself: I’d rather be sitting here without work; I will not go back there.

34Larysa did not return to work, telling the employers that she was leaving for Ukraine. Leaving insecure work relations was another form of exercising control by the migrant domestic worker.

35Similarly, Lilyana, a 30-year old school-teacher and mother of a 5-year old who worked for the past five years as a cleaner in Warsaw, said:

For example I know a lot, I come [to Poland] all the time and those persons, who are simply awful, I try to leave them. I simply leave them. Now, those people I work for, there has been nothing [for] already such a long time; for two years I did not get into any conflict. Not everyone likes it, when I have my free days, but somehow they manage and it is very nice and… it is good.

36However, the moment a migrant developed an emotional tie to the cared-for person, leaving work was difficult. This was pointed out by Susann Cheever (2003), among others, in the work of child-care workers. As Marijka said:

I was there only three months, and it was very hard for me there, but they dispensed with [the services] themselves. They said that they have some financial problems, that they can’t [have a nanny]. So I was very happy that they gave up, because I found it difficult to tell them, because I get very attached to the children; I like children, I get used to them and it is hard to part from them. But when they don’t need me and that’s all, “thank you” and “goodbye” (laughs).

37The asymmetry of the power relations between the employer and the migrant worker was visible in the strategies to leave work. Face-to-face criticism may have had very negative consequences for the domestic worker. Using alibis and lies, or giving signs of dissatisfaction such as ‘getting ill’, were ways to avoid directly leaving work. As Hondagneu-Sotello (2003:65) wrote:

A domestic employee who tells her employer that she is leaving because she has a better job prospect or because she was dissatisfied with her job may be accused of betrayal and disloyalty. She may be accused of theft, justly or unjustly. She may fear that the former employer will call the immigration authorities in retaliation. Or she may fear that by quitting verbally, she will ignite the flame of unpleasant verbal encounter…

38Leaving work meant a lack of a work routine and was regarded by some migrants as a risk. Exit-strategy means facing new risks. Not wanting to leave jobs, some migrants developed other strategies. Larysa, who rejected the employer’s impersonal treatment, adopted one strategy which reflected what she had already experienced:

I am not afraid of this. I am already (…) you understand, I was in different jobs. I even was in such a job where a woman was standing at the doors, leaning against them and I worked, and she was standing and watching. So I said, I am sorry, but maybe (…) you will not look at my hands. Because now I no longer pay attention, I just work and that’s it. But at that time, at the beginning, my hands were trembling from this when someone was looking like that at me. I could not work. Now I don’t pay any attention. One time I said: I am sorry, but maybe you will not watch me like that, later you come and check. I am not afraid, you can come and check. But she didn’t understand this, so let her stand if her feet didn't hurt…

39Larysa’s ‘professional’ attitude, a form of emotional detachment, involved ignoring the employer and seeing the check-ups on her work as irrational. In Larysa’s relation to her employers she saw herself as the expert who ‘knows everything’ when it comes to how to clean the apartments of her informal employers:

There are many such places where I have worked for 2, 3 years, sometimes longer. They are pleased that I work and they all want me (…) ‘you know everything, I don’t have to tell you anything, just come’ (…) I have some where I have already been 5 years, 5 years, in July it will be five years.

40Other interviewed migrants also tried to strike a balance between a personalised knowledge of their clients and, at the same time, not getting too familiar. They were also acting as labourers without emotions.

6. Sketching conclusions

41The Ukrainian women as domestic workers felt to be liminal figures. According to them, they were treated as someone between a worker and a family member. The liminality resulted mainly from the locale in which domestic work was carried out – the household. Their liminality both increased and decreased the risks present in their relationship to their employer.

42In addition to other factors, their liminality enabled the migrants to enter into a patron-client relationship with the employer, an asymmetrical relationship characterised by a division of domination-subordination (Tarkowski 1994:45). The employer was interested in a particular service provided by the migrant and the migrant wanted, apart from the financial gratification, a form of protection from the employer. The interest of the employer and the migrant to enter into such a relationship resulted from their diverse legal, social and material statuses and from different access to resources. The migrant – who was in a subordinate position and exposed to exploitation in the relationship with the employer, who was in a dominant position – still had the power to influence the employer.

43To balance the risks of exploitation, migrants leaned in the direction of personalisation of work relations, based on mutual trust. The employer was also interested in developing friendly relations with the migrant, to guarantee him/herself loyalty and good quality of service. However, personalisation also led to exploitation as soon as the dependency of the migrant on the employer passed a threshold of acceptability, to the point where the migrant was no longer able to negotiate or exit work.

44The dependency of the migrant on the employer was strongly conditioned by the living arrangements. Migrants who lived in were dependent on the employer for accommodation and work. Their access to information about new jobs and other forms of social support were also limited, due to scarce time off work. Thus, when faced with unfavourable working conditions, migrants who lived in attempted to negotiate their position within the household instead of exiting the work relationship. The weaker the migrants’ position within the household, the less possibility for negotiation and the higher the risks of being forced to accept new workloads without additional payment. Migrants living out, while having fewer possibilities to develop a personal relationship with the employer, were not as dependent on him/her for work and accommodation and thus could leave work that they did not like.

45An important risk related to domestic work was the ability to exit the sector and find work in accordance with qualifications. The migrant women were in general skilled workers, with secondary or even in some cases higher education. Those who treated migrant domestic work as only a temporary remedy did not regard this work as a risk to their social status. However, those who were planning to remain longer in Poland were affected by the low status of domestic work and the resulting treatment of the Polish employer.