Abstract This article uses data from face-to-face interviews with recently resettled Burundian and Burmese refugees in Michigan to explore the concept of market citizenship. Market citizenship (Brodie 1997) is defined as the allocation of citizenship rights based on an individual’s economic power and participation in the labour market. While refugees have legal access to certain social rights, through the limitations of market citizenship, they are frequently denied access to those rights. Our data illustrates some ways in which that denial occurs, but also points to ways that refugees use family relations to circumnavigate the barriers to social citizenship that they frequently experience during the immediate resettlement period. Refugee families reassemble household configurations such that they increase the number of work-eligible household members, adjusting what we call the ‘neo-liberal citizenship ratio’. We argue that citizenship is broadly constrained by neo-liberalism, and that refugee families’ creative mobilization of familial and community relations are often the only avenue refugee households have to survive under neo-liberal constraints. Introduction Refugee resettlement is understood as a durable solution to long-term displacement by reunifying refugees with citizenship and belonging to a new state (Arendt 1973). Yet the emergent literature on citizenship questions the inevitable connection between rights and belonging to a state, pointing instead to alternative sources of rights like supranational institutions and ideologies (Escobar 1994; Soysal 1994; Ferguson and Gupta 2002), transnational movements (Fox 2005), non-governmental organizations (NGOs) (Ferguson 2005; Nguyen 2010; Nagel and Staeheli 2015) and global capital (Roitman 2005; Ong 2006). This scholarship emphasizes that an array of social and political rights are increasingly tied to access to resources, not necessarily legal status—a transformation described as ‘market citizenship’ (Brodie 1997). In market citizenship, the state no longer directly provides or guarantees social rights, but rather access to rights depends on economic power and access to the labour market (Somers 2008). Market citizenship has been institutionalized in refugee resettlement by making employment the primary goal and objective of the resettlement process (Nawyn 2011). Refugees are expected to get a job within the first six months of arriving in the United States—and ideally within the first 60 days (Nawyn 2011). The United States resettlement programme makes employment placement a primary goal with the assumption that employment facilitates integration. But the resettlement of entire family units results in requirements for some individuals to care for family members who are either too old or young to work, or ineligible for employment due to reasons such as infirmity. These dual demands create a tension whereby the measure of success (one person obtaining a job, regardless of pay) is disconnected from the daily economic reality of supporting and caring for a family. This disconnect results in refugees not always having access to social rights that are increasingly denied them by market citizenship. However, refugees still sometimes find ways to access social rights through creative mobilization of family networks. In this article, we explore the barriers that refugee families experience in realizing social citizenship, and how they strategize ways to access social citizenship rights otherwise denied them by market citizenship. Our findings illustrate that for newly resettled refugee families, social citizenship access is largely contingent upon what we call the ‘neo-liberal citizenship ratio’—that is, social citizenship rights access relies on families having a certain proportion of able-bodied, employable adults relative to the number of hungry mouths to feed. In doing so, we illustrate the ways in which newly resettled refugee families navigate economic realities in order to reassemble households in pursuit of social and economic belonging. Citizenship and Social Rights T. H. Marshall defined social citizenship as ‘the right to a modicum of economic welfare and … to live the life of a civilized being according to the standards prevailing in the society’, which encompasses the rights that are derived from social institutions like education and health care (Marshall 1950). Marshall’s definition of social citizenship grounds abstract notions of rights in particular institutions and services, illustrating how and where individuals access and obtain rights. In Marshall’s three-part configuration of citizenship, the social aspects of citizenship inform the legal and political citizenship, enabling citizens to not simply engage with the state, but to engage with a modicum of social equality mediated through social services. For example, through free public education, individuals acquire the ability to read, which allows them to read a ballot and fully participate in political rights like voting. With the decline of the welfare state, the expansion of global capitalism and increased privatization, many of the services of social citizenship are no longer provided by the state, instead making rights of social citizenship available for purchase (Somers 2008). With the shift towards neo-liberalism, alternative theories of citizenship have emerged that challenge the idea of the state as the primary analytic for citizenship. Many of these theories have drawn on the concept of globalization, emphasizing that transnational human rights discourse has shaped local manifestations of rights (Soysal 1994), that transnational institutions have increasing influence in sovereign spaces (Castles 2005) or that immigrants situate themselves strategically within and between states to maximize access to rights (Ong 1999, 2006; Fitzgerald 2008; Sadiq 2010). In each of these models, private sources of rights influence the remaining public implementation of services. These debates fundamentally call into question the centrality of the state to the provisioning of citizenship rights to individuals. But they do little to illuminate how individuals actually use and acquire citizenship on a daily basis within a neo-liberal context. The focus on official, institutional perspectives of citizenship (originating from the state or NGOs) has portrayed citizenship as something that is provided, distributed or received, but not as something that is agentively pursued (Brodie 2004). Market citizenship (Brodie 1997), in which rights are accessed through participation in the labour market (most often through purchasing power, such as the ability to afford the cost of private services that had previously been public), provides an opportunity to understand the individual pursuit of citizenship rights. In particular, the emphasis on associational democracy in the United States with its attendant focus on the individualism and locality (Isin and Turner 2002) should provide examples of agentive acquisition of social citizenship. But perhaps paradoxically, even discussions of market citizenship in the United States conceptualize individuals as doing little outside of the market-prescribed actions—like work—to acquire rights. Individuals are assumed to automatically be able to access social citizenship through labour market participation, or alternatively denied them when their economic power is insufficient to purchase the rights they desire. Institutional Restrictions to Social Citizenship Scholarship on citizenship has explored different mechanism for denying the social citizenship rights to those who should have legal access to them. The findings point to the varied ways that institutions serve as barriers to accessing social citizenship, and how the onus of accessing social rights is increasingly put on individuals who must control sufficient resources to access those rights. Some of this scholarship also highlights the ways that individuals circumnavigate institutional barriers by mobilizing alternative resources. Somers’s (2008) study of the governmental response to Hurricane Katrina in New Orleans describes how neo-liberal governmental infrastructure cuts combined with racist discourses about ‘deserving’ versus ‘undeserving’ citizens of government assistance led to the abandoning of many poor Black New Orleans residents to the catastrophe brought on by flood waters. What had been government-supported services were transformed into de-institutionalized private services that residents were required to purchase. In his comparison of the school-to-work transition for Turkish immigrants in Germany and Mexican immigrants in the United States, Faist’s (1995) research demonstrates how exclusionary practices can replicate the results of neo-liberal government. Faist compares Germany’s policy-based system (the state, employers and unions work together to shape the transition into the labour force) with the United States’ market-based system, and finds that both systems produce strikingly similar exclusionary practices. Turkish young adults were denied access to many of the German institutions that facilitated labour market entry because of their lack of citizenship, whereas Mexican immigrants struggled to gain a toe hold in United States unions, and had lower rates of entrepreneurship and thus little opportunity for the co-ethnic employment that is a mainstay for many immigrant groups in the less regulated United States labour market. Whether it was the institutionalized apprentice system in Germany or the more privately regulated union system in the United States, young Turkish and Mexican immigrants generally lacked the cultural and social capital that nationals used to navigate the different institutions that facilitated entry into the labour market. These macro-level studies focus on the institutional barriers to accessing social citizenship, but there is also a micro-level analysis that is necessary to find out how individuals respond to these barriers. Carte’s (2014) qualitative study of Central American migrant women in the Mexican state of Chiapas shifts the focus from macro-legal structures and national and state governmental agencies to micro-level interactions between individuals and local government bureaucracies. In doing this, she uncovers the informal ways that bureaucrats engage in exclusionary practices that are not necessarily codified into law. Carte found that, when Central American migrant women in Chiapas did things like apply for a birth certificate for a child or attempt to access pre-natal care, following laws and procedures was insufficient for accessing the services to which they were legally entitled, as low- and mid-level bureaucrats had power to make arbitrary demands and refuse services migrants should be able by law to access. Carte refers to these exclusionary practices as ‘everyday restrictions’, clearly showing that these restrictions exist separately from formal legal structures. Thus, macro-level policies and laws are not reflective of the actual rights migrants can access in their everyday lives. The migrant women in Carte’s (2014) study made calculated decisions about when to engage with bureaucracies and when to avoid them, and their precarious status in Mexico clearly shaped some of those decisions. Migrants with more secure status have more options to enact agency in seeking social rights. In their study of refugee integration and exclusion, Nawyn et al. (2012) examined the role of linguistic resources in immigrants’ refugees’ access to institutional supports. They found that, in cases in which a refugee household was linguistically isolated (defined as when a household has no one who is fluent in the dominant language), having access to interpretation and translation assistance was necessary in order to access the services to which the refugees were legally afforded through their refugee status. Although laws exist that require reasonable language accommodation for many services (such as at hospitals, government offices and many social service agencies), such accommodation was rarely available in the community they studied. This meant that linguistically isolated households were required to tap into their own social networks in order to avail themselves of social services, which those refugee households that had social ties with settled co-ethnics were able to do. Nawyn and colleagues argue that, rather than thinking of language solely as a measure of integration, it should also be considered a resource that facilitates integration and access to social citizenship rights. These findings point to two sources of potential inclusion or exclusion of social citizenship. The first is the institutional arrangements in the receiving country (or often times more locally in the receiving community) that can either facilitate or hinder migrants’ access to social rights, and the second are the resources that can be mobilized by immigrants, either the cultural, social or financial capital, when their access to social rights is otherwise hindered. As Nawyn et al. (2012) showed, these do not have to be resources possessed by an individual, but rather can be resources accessible through social networks. Family relations, for example, can be an important source for resources mobilized to access social citizenship; it is to this area that we now turn. Family as a Resource for Accessing Citizenship While refugee resettlement is designed to confer social citizenship rights to refugees through temporary assistance immediately after resettlement (Hein 1993), the United States resettlement programme emphasizes not a right to services necessary for incorporation or subsistence, but the necessity of work in order to access even basic services through market citizenship (Nawyn 2011). In fact, the preface to the Refugee Act of 1980, which established the US Federal Refugee Resettlement Program, defines the primary goal of the resettlement process as helping refugees to achieve economic self-sufficiency as quickly as possible. Other services that can facilitate integration such as English-language classes and establishing United States educational credentials are described as subordinate priorities to employment placement (United States Office of Refugee Resettlement 1980). To this end, the Office of Refugee Resettlement (ORR)’s Reception and Placement Program, which is described as the primary means for refugee social integration, consists of only employment placement assistance. Consequently, the following criteria are tracked and reported as resettlement indicators by resettlement organizations (U.S. Department of Health and Human Services 2012: 14–16): Caseload—the total number of employable adults in the employment placement programme. Entered employment—the percentage of employed caseload. Terminations due to earnings—the refugees removed from initial resettlement assistance due to early job placement. Reductions due to earnings—refugee cases whose initial resettlement assistance was cut due to earnings and job placement. Average wage at placement—the average hourly wage, reported by state and county; for instance, in 2010, the national average for average wage was $8.92 per hour. Employment retention—the rate of refugees who maintained employment for 90 days. Entered employment with health benefits—the percentage of refugees who some health benefit available through their employer. While the United States focuses on employment as the primary means of integration, the 1951 Convention and Protocol Relating to the Status of Refugees identifies ‘the family’ as both a right and a critical source of social integration support (Goodwin-Gill 2008: 10–11). In the United States resettlement programme, refugee families are prioritized through family reunification support for extended families separated through the resettlement process and initially resettling families as household units. Consequently, initial case management becomes the functional means of familial social integration—that is, support for health care, assisting families with enrolling children in school, housing, utilities and other assistance. These services are provided at the household level for up to eight months (and potentially as little as four months, depending on the resettlement agency’s type of refugee assistance grant). This short period of assistance is designed to be mediated by wages earned through employment; the initial resettlement assistance often ends when one family member finds employment. At this juncture, the focus on social citizenship services switches from a family-centred model of initial resettlement support to a market citizenship model whereby social citizenship is assumed to be accessible through work by an individual. As indicated in the previous section, the ORR measures resettlement success by the percentage of work-eligible refugees placed in jobs, not the percentage of refugees placed in jobs that provide the means to support a family (or non-employment-eligible household members). So, while the resettlement process initially emphasizes the wellbeing of the family, the goal of the resettlement is process is not family sustenance, but individual job placement. This distinction significantly impacts the experience of wellbeing during the resettlement period. Families rarely disaggregate individual earnings from collective needs. It seems ridiculous, for instance, to think that one family member would withhold water, electricity, food or other resources from other family members because of their lack of employment or employability. This, however, is exactly how market citizenship and consequently employment-as-a-resettlement-indicator function: rights are disaggregated from the family unit and expressed at the individual level, but only for the employable. The result is that producers—able-bodied, non-elderly, non-care-giver, educated adults—become ideal refugees. Resettlement outcomes count only these refugees; refugee children or those who are disabled, elderly, care-givers and those with no or little education are hidden by outcome measures, but their needs persist regardless of official counts. Despite the centrality of family in the experience of daily life, families are largely absent in the literature on market citizenship and neo-liberalism. This oversight creates a tension between the assumptions of market citizenship (rights are available to those who are employed) and the expectations we bring about families and economic survival (that not all members of a household can be economic producers). By examining what refugee families do to negotiate market citizenship and economic family life, we illustrate not only how neo-liberalism constrains opportunities after resettlement, but what individuals do within families in pursuit of social citizenship rights under market citizenship constraints. Refugee resettlement provides an excellent opportunity to examine the family in neo-liberalism because the resettlement process formalizes the tension between family and economic production that exists in the broader society. Resettlement is an explicitly family-based process (i.e. resettlement of a household as the analytic unit of resettlement), yet resettlement outcomes are explicitly individual (job placement). Refugee resettlement is also one of the rare cases when social citizenship expectations are made explicit by the state: as the state attempts to create citizens from refugees, it delineates rights, expectations and desired outcomes. Thus, the refugee resettlement process captures a broader social process—the production of ideal citizens in neo-liberalism—and provides a unique vantage point into the role family plays in this process. Methods In this study, we primarily utilized qualitative data from 36 face-to-face, semi-structured interviews with adult refugees from Burundi who arrived between April and June 2007 and were resettled by the same agencies, so the available organizational and receiving community resources were similar across the group. Michigan is an ideal research site because the greater Lansing and Grand Rapids areas became home to an increasing number of refugees during the time of this research. In 2008, Michigan received the fifth largest number of refugees, behind California, Texas, Florida and New York (Batalova and McHugh 2009). Together, Lansing and Grand Rapids resettled 10 per cent of all of the Burundian refugees resettled in the United States in 2008 (Office of Refugee Resettlement 2009: A2). Michigan has also become a major resettlement site for refugees who do not have anchor families to sponsor them. These patterns of resettlement have led to an increase in ethnic diversity in small cities in Michigan that are more common within larger metropolitan areas that have long been traditional gateways for refugees and other immigrants. And, without family sponsors or established ethnic communities, this diversity presents particular challenges for refugees and refugee service providers, who cannot rely upon anchor families to assist the new arrivals. Additionally, our proximity to these two areas of resettlement allowed us to develop relationships to refugees and service providers that facilitated our access to the communities we studied. Participants had been in the United States for between 120 and 180 days when they were interviewed. We sampled refugees from anonymous lists1 of refugee households provided by resettlement agencies that described the country of origin and the number of adults and children in each household. We used a purposive sampling technique in order to capture spousal relations and intergenerational dynamics within households, as this was an empirical point of interest in our larger study. Therefore, whenever there were two married individuals in a household, we sampled both partners. Many of the Burundian families were households composed of spouses with children. A staff member at the resettlement agency contacted our selected participants and described the project and requested the refugees’ permission to give us their names and contact information. Some resettlement agencies did not have the staff capacity to call refugees; in such cases, we asked our interpreters to identify refugees fitting our eligibility criteria and the interpreters contacted them directly for their permission to release their names and phone numbers to us. For refugees who expressed initial interest in participating, we called them on the phone and through an interpreter described the study in detail. Refugees who agreed to participate gave formal consent before we conducted the interview. Each of our participants received a $40 gift card for their time during the interviews. We conducted interviews from October 2007 through May 2008. All the refugees that we approached for inclusion in the study agreed to participate. The interviews focused on the services refugees have received from their resettlement agency, as well as assistance and support they have received from other organizations and individuals. We also asked refugees about how helpful and effective they found the assistance they received and the challenges they have faced in adapting to life in Michigan. Finally, we asked what changes they would make to the services they received in order to make resettlement more helpful. We conducted the interviews in English and either Kirundi or Kiswahili. We conducted all interviews in participants’ homes. The interviews were tape recorded and lasted approximately one to two hours in length. We used interpreters for all interviews except for one interview with a Burundian man whose English was very good. We transcribed verbatim all the interviews in English, maintaining the particular wording used by our interpreters. The interview translation and transcripts were reviewed by two of the authors who are fluent in Kiswahili for translation precision and clarity. For data analysis, we used Atlas.ti qualitative software to discover patterns and themes in the data. Our analytic strategy was based on the grounded theory approach (Glaser and Strauss 1967) in which we built codes and developed salient themes directly from the data. We first used an open coding process by analysing each transcript line by line using the Altas.ti software. We then engaged in a selective, focused coding process (Emerson et al. 1995) in which we analysed the data to find connections between general codes and sub-codes until we developed key themes. Salient codes captured the different kinds of household strategies expressed by the respondents. We also wrote brief memos and used triangulation to compare the codes we each developed and the memos we wrote, to further develop the meanings of and linkages between the codes. While we were interested in family division of labour and work, the actual variation in strategies used for balancing income and domestic labour emerged unexpectedly as a central theme in our data. For this article, we selected quotes that best represented how individuals expressed the challenges related to division of labour and strategy. After analysing the interview transcripts, we decided that we needed to collect additional data from the Burundian refugees in order to understand their extreme problems with the resettlement process. Therefore, we conducted single-sex focus groups with Burundian men and women (one focus group with each sex) who lived in Lansing. We chose to organize separate focus groups for men and women because we expected that both men and women would be more open and comfortable sharing their stories about spousal relations, household labour and care work in single-sex groups. We conducted the focus groups approximately 18 months after completing the initial interviews. For the focus groups, we chose five women and five men from among our Burundian participants. All the women we sampled were able to attend the focus group, but, because of transportation difficulties on the day of the focus group, only three of the five men attended their group. The topics presented in the focus groups were language challenges, relationships between spouses and with children, work and any continuing difficulties the refugees faced. The focus groups were conducted in Kiswahili and English with the women and Kirundi and English for the men, and interpreters were used for both groups. Likewise, all focus group transcripts were reviewed, translated and transcribed by two of the authors who are both fluent in Kiswahili and have previous experience translating with the Burundian community. The focus groups took place at a refugee community centre in Lansing and lasted approximately three hours each. Data analysis is based on the audio recordings as well as field notes we took during the focus groups. We analysed these field notes in Atlas.ti using a similar coding strategy to the one we used for analysing interviews. Each woman received a $20 gift card and, at their request, men received $20 in cash for their participation. We also served East African food during each focus group. We received Institutional Review Board approval and complied with all protocols. Recognizing the complexities and power relations of working in multiple languages (Temple and Young 2004)—particularly with newly resettled refugees—we worked to ensure the continuity and quality of interpretation, translation and transcription throughout the research course. Two of the authors are fluent in Kiswahili and worked as a team throughout the project to ensure clarity in the instruments at the time of translation and were present in focus groups in order to assist with emergent language needs. These authors oversaw the translation and transcription of focus groups. One of the author-translators was a native speaker and the other was a non-native speaker with specialized linguistic training in research and interviewing with refugee populations. In addition to focus groups and interviews with refugees, we participated in meetings of service provider collaboratives in Grand Rapids and Lansing throughout the project period. These observations included discussions among service providers about what types of assistance they were offering their refugee clients, for how long that assistance was offered and identifying additional client needs that were not being met. We draw upon this data to provide a richer context for the data from refugees. Results Even after only being in the United States for a matter of weeks, the confusion of living and navigating a new place, bureaucracy and culture was already wearing on our Burundian participants. They described issues with health care, housing, understanding a new culture, learning a new language, parenting, finding work, understanding the role of resettlement agencies and maintaining relationships with family members still in refugee camps. Resettled as an ethnic community en masse, but scattered across the United States in clustered communities, Burundian refugees arrived in Lansing, Michigan, with little English and few connections to United States society outside of their resettlement organizations. Upon arrival in the United States, Burundian refugees believed that rights were available through the state; after all, the United States government interviewed them in refugee camps, ran background checks, selected individuals for resettlement, provided pre-departure orientation on life in America, arranged transportation for them to the United States and, ostensibly, created the opportunity for them to become part of American society. But, despite all the lessons in refugee camps about participatory democracy and assurances of social citizenship services like housing, health care, water, food, education and safety, the state and the government were conspicuously absent from daily life. The Burundian refugees we interviewed were perplexed by where social rights were supposed to come from and who, exactly, was supposed to ensure rights access. In the following three sections, we discuss how Burundian refugees began to understand the assemblage of social rights after resettlement in the United States. In the first section, House and Home, we present the moment when our refugee participants described realizing that the enforcement, distribution and guaranteeing of rights in the United States did not emerge from the state, but the market. In the second section, Unlikely Sources of Rights, we describe how refugee families became important means for accessing rights via labour markets. Finally, in Limitations of the Family, we examine what happened to refugees without extended family support mechanisms. House and Home Upon arrival in the United States, refugees are not greeted by government officials, but by caseworkers employed by the refugee-focused non-profit organizations (called voluntary agencies, or VOLAGs) that subcontract with the government to provide resettlement services. Long before the Burundian refugees arrived in the United States on flights from refugee camps in Tanzania, their new homes had already been selected by resettlement organization caseworkers and paid for using part of the refugees’ resettlement allowances. Yet resettlement assistance and case management by the resettlement organization are only provided for the first six months after arrival in the country; after this initial period, refugees are on their own to navigate their new country. This initial time period after arrival is largely designed around job placement. Resettlement agencies’ efficacy is rated according to refugee job placement (as the measure of successful integration). Consequently, resettlement activities in this initial period after arrival focuses largely on helping refugees find jobs. For the Burundian refugees we interviewed, this initial resettlement period may have placed them in jobs, but the refugees still lacked the bureaucratic knowledge necessary for social citizenship rights. For instance, one family struggled to identify rights largely because they lacked the language skills necessary to demand a basic standard of housing: It’s difficult for me to know what to do. The landlord came to my house—I wanted to move out—and he said he wouldn’t give me my deposit back. I don’t know English, I can’t ask him why. I can’t argue with him, I can’t tell him my ideas. When I moved in, I didn’t know the laws, the rules, to take pictures of the things that were bad in the house. I just thought I had to accept this place to live. I didn’t get my money. It was a lot of money. When he asked for assistance from the resettlement agency to resolve this problem, the refugee was told that his assistance period was up; it was time to be self-reliant. The combination of the lack of the bureaucratic knowledge and lack of English skills left many Burundian refugees feeling isolated and overwhelmed with the economic and cultural expectations placed upon them after their resettlement assistance expired (Nawyn et al. 2012). This feeling of isolation reached a tipping point after one Burundian family experienced a broken sewage pipe in their home. Already past their initial resettlement period, they tried asking their former case manager at their resettlement organization for help, but were told they no longer had a right to assistance. They tried to ask the landlord for help. He came to their house one day, surveyed the basement flooded with sewage but never fixed the problem. Each month, he would return for the rent check and harass the family’s elementary school-aged (English-speaking) children for the rent check, threatening to evict and imprison the family. Without sufficient English skills or knowledge of American bureaucracies and housing rights, the adults in the household failed to confront the landlord about fixing the pipe and they feared his empty threats. They tried—and failed—to fix the pipe themselves. Again, they went to their resettlement agency to ask for help and advice, but were again turned away because they were past their initial eight months of support. They went to city hall and the local county court house to ask for help, but were turned away. Ultimately, they had no way of knowing whether the landlord had the power to imprison them, whether he was required to fix the sewage problem and who else might help them hold the landlord accountable. They failed to understand if, how and when housing was a right and who (or what entity) was responsible for ensuring their access to a safe, clean home. Without the linguistic and bureaucratic know-how to advocate for help or identify the responsible government entity, they contacted the last government authority they knew of: they dialled 911 and silently waited on the phone until a police officer showed up at their doorstep. An officer responded to the call and was shown the sewage-filled basement (that had been flooded with excrement for months at this point). Although neither a medical emergency nor an immediate security threat, the police officer assisted the family with finding a temporary place to live and forcing the landlord to fix the problem. Although the individual police officer used his position of power to assist the family, through a translator, it was made clear to the family that they were not to call 911 again for this kind of assistance. While the 911 call worked to solve the problem of the flooded basement, it also reinforced to the Burundian community that the government was not interested in, what they called, human rights (haki ya binadamu). This story was narrated in our focus groups with both Burundian men and Burundian women to make a larger point about rights: although the United States government effectively brought the refugees to the United States, it remained mysterious and inaccessible to the refugees, and the refugees perceived it to be uninterested in helping them meet their basic needs. Although the resettlement organization initially seemed to be their access-point for social rights, its support ended only months after arrival. The result was that the refugees lacked the language and bureaucratic skills to identify their specific rights, make rights claims and hold individuals accountable. The only way that the government seemed accessible was through emergency services. Even then, it fixed small problems but did little to address the larger social conditions that the refugees found themselves in. As one Burundian woman described, When I came to the United States, I thought, ‘now I can live in a good, clean house [nyumba safi]. I will have electricity and water. I will have a toilet. I will be safe.’ Now I see these things cost money and I don’t have enough for these [things]. Just like in refugee camps, sometimes I live without water and electricity. Money emerged as the only undeniable and enforceable source of rights. True to theories of market citizenship, the Burundian refugees found rights in the United States, but only at a cost. Even the most basic of services like water and electricity were not guaranteed due to a lack of financial resources. The cost of living in the United States was often a passionate topic during our focus groups with Burundian refugees. The crux of this problem was the relationship between knowing English and getting a job that paid a living wage. Expected to learn English in order to get a job to pay for housing, food, utilities, health care and other services, refugees were given a limited time—often less than four months—to learn English and were often encouraged to quit class in order to have a more flexible schedule for work (see Nawyn et al. 2012). The English classes they attended focused only on the most basic of English skills necessary for work; yet, without English skills, their opportunities for finding work that paid a sufficient amount to sustain a family were few and far between. In order to work within this cycle, families experienced periods without water, electricity and food when they could not afford to pay these bills or ran out of SNAP benefits (food stamps). Basic subsistence was a daily challenge and, in focus groups, Burundian refugees reported that they realized they were outside of the government’s purview for support and had few options for accessing the services of social citizenship through the market. With few options for support in their new community, the Burundian refugees began to look within their Burundian community for help. Unlikely Sources of Rights Very few Burundians were formally educated before coming to the United States and very few had formal skills or employment histories that provided them with the kinds of human capital that lead to success after resettlement (Zhou and Bankston 1999). With very few members with sufficient language skills and educational training, Burundian refugees drew upon the resources they did have to provide alternative entry into market citizenship: family. Similarly to Zhou and Bankston’s findings that intra-ethnic communities become a source of support for resettled Vietnamese refugees, Burundian refugees in Michigan became a unique form of support. Unlike the Vietnamese refugees that Zhou and Bankston describe, the Burundian refugee community in Michigan was not robust or large enough to create social capital alternatives necessary for social integration and advancement in American society. What the Burundian extended families provided was a very basic level of social support necessary for survival at the household level. For Burundian refugees in Michigan, reassembling families at the household level provided a way through which they could access basic social rights through the market. The United States refugee resettlement system generally tries to resettle families in American-style nuclear families (mother, father and children living in one dwelling)—a definition of family that is further emphasized through the distribution of social services like food stamps, health care assistance and temporary refugee assistance at the household level. However, this configuration is often expensive, especially for families with small children that require care. For the Burundian refugees we interviewed, the government’s ideal nuclear family resettlement model with two wage earners often failed to generate sufficient financial resources to cover rent, heat, water, food and transportation costs every month. After their access to refugee cash assistance ended, many Burundian refugees addressed this challenge by reconfiguring families in order to access two different forms of capital necessary for neo-liberal citizenship through two strategies. The first strategy redistributed non-wage-earning community members in order to meet household labour demands, including language and translation. The second strategy redistributed wage earners in order to maximize the amount of income generated within a household. In the first strategy, household labour was met by redistributing non-wage-earning family members to address domestic needs. This strategy was primarily used to equip households with care workers or with access to translation. Pre-teens and teenagers were most often moved within in extended families in order to maximize access to those with English skills. Because children learn new languages more quickly than adults and because children attended formal school with English as a Second Language (ESL) programmes after resettlement, refugee children often spoke English better and more quickly than older family members. Redistributing children within extended families ensured those most people had access to someone within the household that knew English. Children also played an important role in household labour. Pre-teens and teenagers often watched younger children as adults worked, freeing care-givers to seek wage-employment. This allowed families increased access to market citizenship by increasing access to wages. In addition to teenagers, elderly family members who were unable to seek work outside of the home also were moved between households in order to help with housework and care work. Often, in assessing household needs and human resources, families engaged in elaborate exchanges that involved reassembling multiple households: My mother in-law stays with us and helps with the cooking. This way I can send my daughter or my son to live with my sister and help with her two young kids so that my sister can go to work and her husband can work, too. This way we can keep our house with bedrooms for our children and my sister can keep her house and we can afford life here. … Still there is pressure here. If I stop working we cannot pay [the bills]. If we lived like Americans [nuclear family] we could not pay our bills, my mother-in-law could not pay her bills, and my sister in law could not pay her bills. Second, Burundian families redistributed wage labourers within extended families. Some teenagers moved to the homes of their aunts and uncles or grandparents and got afterschool jobs to help pay the bills. Single adults often moved in with their siblings or parents—sometimes relocating from another state or city—in order to help contribute to the household income. In some cases, elderly family members entered the workforce to assist households with income. This strategy sought to increase the amount of financial resources within a household in order to help families directly access rights through market citizenship. With increased income, families were able to afford social citizenship rights like water, electricity, housing and food by increasing the number of workers engaging in wage labour. In this way, Burundian extended families strategically rearranged households and reconfigured paid and unpaid labour in order to meet the requirements of market citizenship. By maximizing access to wage labour through redistributing forms of unpaid labour within and between households and by maximizing human capital access through relocating English speakers and care-givers, Burundian families responded to the pressures of market citizenship by rearranging family units. But, by doing so, they also manipulated the terms of market citizenship. Market citizenship, as the rights distribution mechanism of neo-liberalism, presupposes that rights are accessible by individuals, eliding the inaccessibility of rights for those unable to work. Here, Burundian refugees attempted to redistribute the benefits of market citizenship by reworking the distribution of unpaid labour to those unable to engage in wage labour outside of the home. This strategy sought to maximize the number of wage earners outside of the home while also addressing the linguistic needs of a household that are crucial to engaging with bureaucracies and ensuring rights. In essence, they used household reassembly to increase the neo-liberal citizenship ratio in needy households. This strategy led to a more even distribution of resources and helped provide a basic level of survival—there was enough food for all members and families worried less about utilities like water and electricity being cut. But not all Burundian refugees were resettled in extended family units. In the next section, we examine Burundian families without extended family assistance. The Limitations of a Family Not all families had the extended familial resources to reassemble for rights. Some nuclear families were resettled in the United States while their extended families remained in Tanzanian refugee camps. Some families lost the majority of their adult members in conflict or in refugee camps or were resettled with a disproportionate number of orphaned children. Other families had large numbers of disabled members who were unable to work due to their disabilities. For families without extended family ties or multiple wage earners, market citizenship became the ultimate burden: with deficient resources from wage labour and exacerbated need, these families failed to access even the most basic of rights. For instance, in our focus group with Burundian women, a mother of a child with disabilities lamented that she was trapped by the system that linked rights with labour (market citizenship). Even though she received a small amount of social security assistance on behalf of the child and basic cash and food stamp assistance for her other child, the assistance was not enough to pay the rent in a safe area of town or to purchase healthy, fresh food. As a single mother, she had no partner to rely on for additional income and she could not work outside of the home because of the extensive daily needs of the child with disabilities. The rest of her family remained in refugee camps in Tanzania, where they frequently requested remittances. She felt extreme shame for failing to provide for her children after resettlement, jealousy of those in her community with large extended families that formed a sort of safety net, disappointment for not being able to provide for loved ones abroad and the desolation that comes from confinement combined with culture shock. She relied on other Burundian families for food when her food stamps ran out and for the clothing and other material needs of her children. ‘What is my future,’ she asked in the focus group, ‘[but to] wait for my child to die so that my other child can have a life?’ While this woman technically had her basic rights taken care of by the state—assistance with food, housing, health care and education—the assistance was so meagre that she often went for periods without food, electricity or water. Her disabled child’s extensive and specialized medical needs often went beyond the basic assistance that the state provided. She feared that her other child was receiving a subpar education because they lived in a low-income area. Worse, she predicted that her family was likely to remain in this liminal state until her child with disabilities passed away; as long as she, the only adult in the family, was confined at home with her child and unable to work, her family had no hope of social mobility. While the broader community provided her with what support they could (most often clothing, food or the occasional financial assistance), her extensive needs kept others within the community from being able to assist her beyond a level of basic subsistence. She remained trapped within a liminal state of rights—surviving but not thriving—defined and limited by her child’s condition. Without the ability to work outside of the home, she struggled to learn English and establish relationships in her new country. This loneliness and poverty was not what she expected when she found out she had been selected for resettlement in the United States. Without extended family to rely on for language assistance, to subsidize her household income or to help with care work, she had limited opportunities to improve her living conditions. The woman described above was not the only refugee in this position; households without extended families in the United States and households with a disproportionate number of young children, individuals with special needs and elderly members struggled more than others. In these families, the number of individuals with need exceeded their number of individuals able to work. Market citizenship posits that rights come from an individual’s ability to work and some Burundian families were able to access a basic level of rights by changing the ratio of employed adults to individuals requiring assistance. But, for those nuclear families unable to draw on broader support networks by changing this ratio of work to need, market citizenship simply guaranteed inescapable poverty. These numbers fluctuate based on the human and social capital of the employable bodies and the types of need of the hungry mouths. It is the dynamism of this ratio—what is ‘enough’ one day can be suddenly insufficient the next—that led the Burundian refugees to describe life in the United States as war—not an insignificant statement coming from individuals who fled actual war: We are paying [bills] for things from even before we even came [to the United States2]. Compared to where we lived in Africa, the difference is that here it is good, you can go to sleep. There is no [actual] war. If you just sit, you are okay, but if you start to think about your bills, it is like war. Money, food, heat, it’s all gone. It looks easier here. In the refugee camp there was no peace, here there is peace… but all I can think about here is where am I going to get the money to pay for the lights, for the food? It was better in Africa. In Africa we had each other3 …. But where I live … there’s problems where I live. But these bills, this is war. While some families were able to respond to the neo-liberal citizenship ratio by reorganizing family arrangements, it did little to address the larger structural obstacles that they faced to long-term success in the United States. They still struggled to learn English while working full time; families still lacked a guaranteed level of subsistence, and still felt under siege and unstable. That which is enough today can easily be too little tomorrow. Conclusion When the state failed to provide basic services and when individuals lacked the resources to purchase rights, some Burundian refugees reassembled their extended families to meet market demands. This happened in a variety of ways: teenagers or unmarried adult children were sometimes were ‘loaned’ to other families in order to equip them with translators, cooks or babysitters who helped free up parents—most often mothers—to seek outside employment and increase the number of incomes supporting a family. Others joined families to add an additional income to a household. These swaps were arranged within extended family units in order to ensure that families had sufficient incomes and that household needs were met. But not all individuals had access to extended family networks after resettlement and others faced the problem of having more need than the capacity to earn. And, when this happened, there was no source to call on for help. These families endured long periods with insufficient food and water and without electricity. Even those with enough to meet their immediate needs recognized that their lives were inherently insecure; they were still ‘at war’. Aihwa Ong (2006) contends that neo-liberalism is not simply an economic arrangement, but a mobile technology composed of an assemblage of politics, economics and culture that ultimately form a shift in governmentality. The technological ‘work’ (in Ong’s terms) that neo-liberalism accomplishes in refugee resettlement is that it redefines refugee resettlement from a process that seeks to return refugees to the community of states through the delivery of social citizenship to a process that provides individuals with a territory in which to pursue employment. While there is an extensive literature emphasizing the individuality embedded in neo-liberalism, our research illustrates one of the ways in which refugees respond to this shift: by working within extended families to manipulate the terms and conditions of a family in order to distort the ratio of earners to mouths—what we have called the neo-liberal citizenship ratio. For the Burundian refugees we interviewed, market citizenship shifted the source, expectations and enforcement mechanism for rights from the promised source, the United States government, to the market, accessible and enforceable only via individual labour. As a technology, neo-liberalism placed the onus of belonging on individuals instead of as part of the social citizenship rights guaranteed by the state. When the cost of citizenship rights were too extensive to provide for all family members, Burundian refugees strategically reassembled households in order to ensure broader access to rights. This shift from the state to individuals to the family was made necessary by ORR’s adoption of market citizenship in the resettlement process by using individual employment placement as the indictor of resettlement success, while ignoring the reality that a single income is insufficient for supporting an entire family—especially since the majority of refugees were employed at minimum wage or only slightly above minimum wage, according to ORR data (Brick et al. 2010; ORR 2015).4 When combined with refugees’ minimal English skills and overall lack of social connections, market citizenship renders refugees economically, socially and linguistically isolated. The result is a form of social isolation and structural violence that renders refugees unable to identify their rights or demand rights enforcement outside of what protections a minimum wage-based market citizenship could provide. This social isolation is not just an immediate problem for refugees; it has potential to inhibit long-term incorporation of adult refugees and even their United States-born children. Bloemraad, Sarabia and Fillingim (2016)’s research on Mexican immigrants and their United States-born children indicate that few citizen children with non-citizen parents (whether legal residents or unauthorized) defined ‘good’ citizenship in terms of political participation, including voting. Additionally, attachment to the state in the form of patriotism was mentioned by very few children of immigrants but those that did include patriotism in their definition of ‘good’ citizenship all had parents who permanent status in the United States (almost all had citizenship). The authors postulate that the non-citizenship of parents (and perhaps the liminal legal status of many people in their communities) causes a weakening in their children’s connections of citizenship to the state. By the state playing the role of adversary rather than a source of protection and rights (which has been a common experience for people targeted by the enforcement-first emphasis of United States immigration policy; Aranda et al. 2014), resettlement enacts a form of legal violence (Menjívar and Abrego 2012) that threatens to exclude refugees and potentially their United States-born children, and may lead to the kinds of second-generation uprisings that are more common in some European countries, and which up to now the United States has largely avoided. It is a significant flaw in the resettlement system to treat family life as occurring apart from economic life. So, while the resettlement process does its best to keep families together and to reunify families after resettlement, it fails to ensure that economic resources produced in resettlement are sufficient to sustain a family. Our findings suggest that using the measure of individual employment to capture refugee integration elides the fact that a single wage—and often a minimum wage, at that—is not enough to sustain an entire family and obscures the production and reproduction of poverty in the resettlement process. Footnotes 1. The anonymous lists provided by the resettlement agencies included a household number, a column indicating the relationship of household members (e.g. wife, husband), an individual case number (instead of their actual names) and country of origin. 2. Refugees’ travel to the United States is administered through an International Organization for Migration (IOM) loan. Refugees begin repaying loan after arriving in the United States. Heads of households are responsible for the loans for all of their family members and thus may enter the United States with thousands—if not tens of the thousands—of dollars of debt. 3. This phrase in Swahili uses the word raha, which literally means joy. But here it was used to differentiate between being together with family—not joy in the sense of happiness, but joy in the sense of being together come what may. The phrase is contrasted with the loneliness of the United States. 4. For details on how the ORR calculates employment placement as a measure of integration, please see CFR 400.154 (https://www.gpo.gov/fdsys/granule/CFR-2012-title45-vol2/CFR-2012-title45-vol2-sec400–154/content-detail.html) as well as the conclusion (p. 105) of the FY 2013 Office of Refugee Resettlement Report to Congress (https://www.acf.hhs.gov/sites/default/files/orr/arc_2013_508.pdf), which reflects the alignment of integration as employment placement. References ARANDA E., MENJÍVAR C., DONATO K. ( 2014) ‘ The Spillover Consequences of an Enforcement-first US Immigration Regime’. American Behavorial Scientist 48( 13): 1687– 1695. Google Scholar CrossRef Search ADS Arendt H. ( 1973) The Origins of Totalitarianism . New York: Houghton Mifflin Harcourt. BATALOVA J., MCHUGH M. ( 2009) Top Languages Spoken by English Language Learners Nationally and by State, 2009 . Washington, DC: The Migration Policy Institute, pp. 2– 9. BLOEMRAAD I., SARABIA H., FILLINGIM A. E. ( 2016) ‘ “Staying Out of Trouble” and Doing What Is “Right” Citizenship Acts, Citizenship Ideals, and the Effects of Legal Status on Second-Generation Youth’. American Behavioral Scientist , p. 0002764216664941. BRICK K., CUSHING-SAVVI A., ELSHAFIE S., KRILL A., MCGLYNN SCANLON M., STONE M. ( 2010) Refugee Resettlement in the United States: An Examination of Challenges and Proposed Solutions . New York: Columbia University School of International and Public Affairs. BRODIE J. ( 1997) ‘ Meso-discourses, State Forms and the Gendering of Liberal-Democratic Citizenship’. Citizenship Studies 1( 2): 223– 242. Google Scholar CrossRef Search ADS BRODIE J. ( 2004) ‘ Introduction: Globalization and Citizenship beyond the National State’. Citizenship Studies 8( 4): 323– 332. Google Scholar CrossRef Search ADS CARTE L. ( 2014) ‘ Everyday Restriction: Central American Women and the State in the Mexico-Guatemala Border City of Tapachula’. International Migration Review 48( 1): 113– 143. Google Scholar CrossRef Search ADS CASTLES S. ( 2005) ‘ Hierarchical Citizenship in a World of Unequal Nation-states’. Political Science and Politics 38( 4): 689– 692. Google Scholar CrossRef Search ADS EMERSON R. M., FRETZ R. I., SHAW L. L. ( 1995) Writing Ethnographic Fieldnotes . Chicago: University of Chicago Press. Google Scholar CrossRef Search ADS ESCOBAR A. ( 1994) Encountering Development . Princeton, New Jersey: Princeton University Press. FAIST T. ( 1995) Social Citizenship for Whom? Young Turks in Germany and Mexican Americans in the United States . Aldershot: Avebury. FERGUSON J. ( 2005) ‘ Seeing Like an Oil Company: Space, Security, and Global Capital in Neoliberal Africa’. American Anthropologist 107( 3): 377– 382. Google Scholar CrossRef Search ADS FERGUSON J., GUPTA A. ( 2002) ‘ Spatializing states: toward an ethnography of neoliberal governmentality’. American Ethnologist 29: 981– 1002. Google Scholar CrossRef Search ADS FITZGERALD D. ( 2008) A Nation of Emigrants: How Mexico Manages Its Migration . Los Angeles: University of California Press. FOX J. A. ( 2005) Unpacking ‘Transnational Citizenship’ . UC Santa Cruz: Center for Global, International and Regional Studies. GLASER B., STRAUSS A. ( 1967) The Discovery Grounded Theory: Strategies for Qualitative Inquiry . London, England: Wiedenfeld and Nicholson. GOODWIN-GILL G. ( 2008) ‘Introductory Note to the Convention Relating to the Status of Refugees: Protocol Relating to the Status of Refugees’. United Nations, http://legal.un.org/avl/pdf/ha/prsr/prsr_e.pdf (accessed January 2016). HEIN J. ( 1993) ‘ Refugees, Immigrants, and the State’. Annual Review of Sociology 19( 1): 43– 59. Google Scholar CrossRef Search ADS ISIN E. F., TURNER B. S. ( 2002) Handbook of Citizenship Studies . New York: SAGE. MARSHALL T. H. ( 1950) Citizenship and social class . Cambridge: Cambridge University Press. MENJÍVAR C., ABREGO L. ( 2012) ‘ Legal Violence: Immigration Law and the Lives of Central American Immigrants’. American Journal of Sociology 117( 5): 1380– 1421. Google Scholar CrossRef Search ADS NAGEL C., STAEHELI L. ( 2015) ‘ International Donors, NGOs, and the Geopolitics of Youth Citizenship in Contemporary Lebanon’. Geopolitics 20( 2): 223– 247. Google Scholar CrossRef Search ADS NAWYN S. J. ( 2011) ‘ “I Have so Many Successful Stories”: Framing Social Citizenship for Refugees’. Citizenship Studies 15( 6–7): 679– 693. Google Scholar CrossRef Search ADS NAWYN S. J., GJOKAJ L., AGBÉNYIGA D. L., GRACE B. ( 2012) ‘ Linguistic Isolation, Social Capital, and Immigrant Belonging’. Journal of Contemporary Ethnography 41( 3): 255– 281. Google Scholar CrossRef Search ADS NGUYEN V. ( 2010) The Republic of Therapy: Triage and Sovereignty in West Africa’s Time of AIDS . Durham, North Carolina: Duke University Press. Google Scholar CrossRef Search ADS OFFICE OF REFUGEE RESETTLEMENT ( 2009) Office of Refugee Resettlement Annual Report to Congress 2008 - Office of Refugee Resettlement - Administration for Children and Families. Department of Health and Human Services, Washington, DC. OFFICE OF REFUGEE RESETTLEMENT (ORR) ( 2015) ‘Key Indicators for Refugee Placement FY 2014’. United States Department of Health and Human Services, Washington, DC, http://www.acf.hhs.gov/sites/default/files/orr/fy_2014_placement_briefing.pdf (accessed January 2016). ONG A. ( 1999) Flexible citizenship: The cultural logics of transnationality . Durham: Duke University Press. ONG A. ( 2006) Neoliberalism as Exception: Mutations in Citizenship and Sovereignty . Durham, North Carolina: Duke University Press. Google Scholar CrossRef Search ADS ROITMAN J. L. ( 2005) Fiscal Disobedience: An Anthropology of Economic Regulation in Central Africa . Princeton, New Jersey: Princeton University Press. SADIQ K. ( 2010) Paper Citizens: How Illegal Immigrants Acquire Citizenship in Developing Countries . New York: Oxford University Press. SOMERS M. R. ( 2008) Genealogies of Citizenship: Markets, Statelessness, and the Right to Have Rights . New York: Cambridge University Press. SOYSAL Y. N. ( 1994) Limits of Citizenship: Migrants and Postnational Membership in Europe . Chicago: University of Chicago Press. TEMPLE B., YOUNG A. ( 2004) ‘ Qualitative Research and Translation Dilemmas’. Qualitative Research 4( 2): 161– 178. Google Scholar CrossRef Search ADS UNITED STATES OFFICE OF REFUGEE RESETTLEMENT ( 1980) The Refugee Act with Commentary on US Law. Vol. H.R. 3061; H.R. 3061, http://www.acf.hhs.gov/programs/orr/resource/the-refugee-act (accessed January 2016). U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES ( 2012) FY 2011 Report to Congress - Office of Refugee Resettlement. Administration for Children and Families - Office of Refugee Resettlement, Washington DC. ZHOU M., BANKSTON C. ( 1999) Growing Up American: How Vietnamese Children Adapt to Life in the United States . New York; Plymouth: Russell Sage Foundation. © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com
Journal of Refugee Studies – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
It’s easy to organize your research with our built-in tools.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud