Abstract
Research on administrative burdens has demonstrated that families experience significant costs in navigating different institutions. Yet studies have often focused more on the nature of the burdens that result from administrative rules than on the types of obstacles that produce these burdens. Less attention has also been paid to how families navigate multiple institutions simultaneously. Drawing on qualitative research with Congolese refugees resettled in the United States, we conceptualize how errors and mishaps in organizations tangled procedures into institutional knots, or complex blockages. We also show how some knots had a ripple effect as problems in one institution reverberated, leading to new, unrelated problems in different institutions. These institutional knots and subsequent reverberations were costly to resolve and a hindrance to upward mobility.
When [the COVID-19] pandemic hit, things were really difficult. We were laid off at work, and we were told to apply for unemployment. I was the only one [in my family] working. . . . Paying for our bills was challenging ‘til we got unemployment after about six to seven months. . . . There wasn’t anyone I could turn to for help.
—Jabari, resettled Congolese refugee
In the United States, institutions provide crucial resources, including food benefits, unemployment assistance, access to health care, and other crucial supports (Sherman, Trisi, and Parrott 2013; Carlson and Keith-Jennings 2018; Sommers et al. 2017; Schanzenbach and Bauer 2016). These services are invaluable for low-income households, and particularly for recently resettled refugee families, many of whom arrive with minimal financial resources (Zhou and Bankston 1998). Yet families often face administrative burdens when trying to access these services. As Pamela Herd and Donald Moynihan (2018, 2020; see also Moynihan et. al. 2022) show, navigating social service institutions entails costs, including steep learning curves, thick stacks of paperwork, and uncertainty and stress. Research demonstrates that burdens are products of formal rules and routine organizational procedures, but institutions also make errors (Perrow 1999; Vaughan 1997). Errors have been documented across a wide range of social service institutions, including the Internal Revenue Services (IRS) (Dalrymple 2003), Temporary Assistance to Needy Families (TANF) (Brodkin, Fuqua, and Waxman 2005; Brodkin and Majmundar 2010), and immigration services (Moynihan, Gerzina, and Herd 2021). Less is known, however, about the role of errors and on-the-ground institutional interactions in the creation of burdens for families.
In addition, studies on administrative burdens, such as those in this double issue, typically focus on challenges in one institution, such as childcare (Bouek 2023), child welfare programs (Barnes, Halpern-Meekin, and Hoiting 2023; Edwards et al. 2023), Medicaid (Rauscher and Burns 2023, this issue), schools (Lareau et al. 2016), housing programs (DeLuca et al. 2023; Pierce and Moulton 2023), legal assistance (Yu 2023, this issue), and disaster relief (Raker and Woods 2023). Yet families are situated at the intersection of multiple institutions, and each one provides a set of administrative burdens for family members to negotiate.1 More attention is needed on the dynamics of the institutional obstacles families encounter across numerous institutions as well as on how these institutional barriers intersect.
In this article, we draw on qualitative research on the experiences of refugee families from the Democratic Republic of the Congo (DRC) who recently resettled to the United States.2 Even though refugee resettlement is rare (fewer than 1 percent of the world’s refugees are resettled), refugees resettled to the United States are an important conceptual case in investigating administrative burdens. In the first few months after they arrive, refugee families encounter numerous institutions. In addition, refugees have access to services that other types of immigrants, particularly the undocumented, do not receive through the federal resettlement program, including access to the social safety net and help from caseworkers and volunteers (Ludwig 2016; Brown 2011; Waters and Pineau 2015). Yet, as we show, even with eligibility, rights, and assistance after they arrive, refugees encounter formidable challenges.
In this article, we advance our conceptual models of the institutional obstacles that lead to administrative burdens. We illustrate two types of institutional obstacles that refugee families faced in their day-to-day navigation and show how these obstacles imposed administrative burdens, which threatened access to resources. First, we demonstrate that in addition to burdensome routine rules and requirements, families faced institutional knots, or complex blockages in institutional procedures. Knots usually began with a modest institutional error or mishap: a computer error, a misspelled a name on a form, a document lost in the mail. Families regularly faced institutional errors, mistakes, and accidents while navigating multiple institutions at once. These mishaps tangled procedures and brought processes to a screeching halt, threatening to delay and even prevent refugee families from accessing social services and benefits to which they were legally entitled, such as Supplemental Nutrition Assistance Program (SNAP) benefits and green cards. Crucially, untying knots involved additional, often complex administrative steps.3 These steps each entailed painstaking work that demanded time, energy, and institutional expertise—additional administrative burdens. Even knots that were eventually straightened out were costly for refugee families and those helping them.
Second, problems in one institution could create institutional reverberations, or ripple effects, that lead to new, unrelated problems in different institutions. As families navigated a nexus of administrative burdens from multiple institutions simultaneously, problems could have a cascading effect, swelling into new unexpected problems in other institutions—for example, a delay receiving a green card created problems for documentation required for employment. Reverberating problems required new administrative steps in different institutions, levying additional burdens on families and threatening to become a barrier to additional resources.
Our study highlights the need for research on administrative burdens to address not only the costs of formal rules and routine procedures but also the obstacles that trigger burdens. We focus on the concept of institutional knots triggered by mishaps and show how they reverberate across multiple interdependent institutions. Moreover, racism and xenophobia within institutions prompted more opportunities for institutional knots to develop and made them more difficult to untangle. Cultural brokers, such as caseworkers and volunteers who helped families navigate institutions, sometimes provided support in overcoming obstacles and reducing the burdens families faced.4 These concepts contribute to the theoretical understanding of the informal organizational mechanisms that create administrative burdens and have important implications for impediments to upward mobility for families. Although our focus is on refugees, our model has potential theoretical implications for a wide range of low-income families.
ADMINISTRATIVE BURDENS, ORGANIZATIONAL ERRORS, AND INTERSECTING INSTITUTIONS
Institutions play an especially important role for refugees and immigrants. As Helen Marrow (2009) shows, immigrants have significant interactions with American institutions, leading at times to a form of “bureaucratic incorporation.” While institutions can provide resources and support, they also have rules and requirements, which create costs for clients as they access benefits. Herd and Moynihan (2018) term these navigational costs “administrative burdens.” Administrative burdens refer to the learning costs, compliance costs, and psychological costs that organizations and their procedures cause for families. Herd and Moynihan argue that these burdens are deliberately constructed by politicians and are consequential for clients, becoming a barrier to services. Administrative burdens can have adverse consequences for immigrants, as the process of legal immigration and maintaining legal status becomes more onerous (Moynihan, Gerzina, and Herd 2021). These burdens may have negative health effects (Herd and Moynihan 2020). Crucially, they are distributive (Herd and Moynihan 2018), racialized in their conceptualizations of groups, and disproportionately disadvantaging to racially marginalized groups (Ray, Herd, and Moynihan 2022). Although studies show differences in administrative burdens across organizations, less is known about how these burdens may compound and interact.
Threaded through studies on social services and public administration are signs that the unfolding of errors and mishaps can be burdensome and threaten access to services. Indeed, examining declines in TANF caseloads, Evelyn Brodkin and Malay Majmundar (2010) conceptualize administrative exclusion as not only due to formal rules and modes of governance but also to the nature of informal practices. For instance, drawing on data from the Public Benefits Hotline Research Project (Brodkin, Fuqua, and Waxman 2005), they highlight the case of Ms. Garcia, whose income support and food stamps were unexpectedly canceled due to missing documents for a work record and earnings statement. Even after she submitted replacement copies and a Public Benefits Hotline advocate made seven phone calls, spoke with a supervisor, and walked them through how to correct the error, her benefits were not reinstated for weeks. Despite no changes in eligibility or formal rules, informal processes played an important role in the organizational mechanisms of administrative exclusion. Similarly, Moynihan, Julie Gerzina, and Herd (2021) show that when an undocumented immigrant left a nonapplicable field blank (rather than writing “not applicable”) while filling out a form for a visa, her application was overruled. Lilly Yu (2023, this issue) finds that immigration attorneys and their clients faced “impossible-to-satisfy requirements,” such as a change in fee waiver requirements that mandated proof of income through tax returns—a requirement that undocumented clients unauthorized to work could not meet. Although these studies do bring to the fore the role of informal practices in administrative exclusion, the nature of these snags in institutional procedures, and especially across multiple institutions, has not been sufficiently conceptualized.
Research on administrative burdens has shown the impact of formal organizational rules and policies, yet service delivery also depends on informal decisions by workers. Social scientists from Michael Lipsky (2010) to Celeste Watkins-Hayes (2009, 2019) demonstrate the role of “street-level bureaucrats,” or frontline institutional agents, in wielding power in the delivery of services. Even with these formal procedures, bureaucrats have significant discretion in their interactions with clients, differentially applying rules and doling out services (Lipsky 2010).5 Frontline workers may adopt a range of strategies and helpfulness—from “moving towards clients” to “moving against clients” (Tummers et al. 2015). Caseworkers favor some clients over others (Levine 2013).6 For instance, the anthropologist Aihwa Ong (2003) argues that health and social workers formed racist views of Cambodian refugees as unworthy. Workers’ assessments about clients’ deservingness, based on client characteristics, may lead to unequal resource dispersion of services (Calarco 2018; Lara-Millán 2014; Lopez 2010; for a review, see Pepinsky, Pierskalla, and Sacks 2017). Some groups of immigrants, such as European and Canadian immigrants, received more assistance than others, such as Mexicans (Fox 2007; Ngai 2004). Further, frontline workers make mistakes in administering services. For example, Internal Revenue Service employees have been found to provide incorrect information on one in four questions from taxpayers (Dalrymple 2003). More attention is needed to the role of informal institutional mechanisms in the creation and distribution of administrative burdens.
Moreover, scholars show that families confront a dizzying array of institutions (Black and Keyes 2020; Paik 2021; Heinrich et al. 2022). Many institutions have crucial similarities, as individuals face rules, schedules, inflexibility, deadlines, and important costs if the institution’s opaque demands are violated.7 However, institutions do not function in uniform fashion. Different institutions vary in their orientation and service workers’ willingness to bend rules or prioritize clients (Tummers et al. 2015). For example, immigrant clients may find different levels of inclusion and service worker orientations in schools than in courts (Marrow 2009; Jones-Correa 2005). Nor do families interact with each institution in isolation. Indeed, Mariana Chudnovsky and Rik Peeters (2021) suggest that administrative exclusion may have a cascading effect across institutions, showing a trickle effect with exclusion in identity registration and documentation leading to exclusion in a social policy institution in Argentina. Yet research on conceptualizing how administrative burdens intersect as families navigate a range of institutions is still limited.
Finally, in addition to the variability in service delivery, organizational sociologists show that errors can occur in organizations, especially in complex systems (Perrow 1999; Vaughan 1997). In his work on nuclear accidents and other organizational failures, the sociologist Charles Perrow (1999, 7) argues that complexity in systems can lead to “normal accidents” when multiple and unexpected errors intersect and interact in some unexpected way: “No one dreamed that when X failed, Y would also be out of order and the two failures would interact so as to both start a fire and silence the fire alarm. . . . The cause of the accident is to be found in the complexity of the system. . . . the failures became serious when they interacted. It is the interaction of multiple failures that explains the accident.”
Although Charles Perrow (1999) theorizes about risks and failures in complex technological systems—not service institutions— his approach is illuminating in the context of social science service delivery as well. In sum, important research on administrative burdens has brought to the fore the navigational costs of rules and routine procedures in accessing institutional resources. Yet more attention is needed on the accidents and institutional errors families experience as they navigate multiple intersecting institutions, especially since they have implications for social mobility.
METHODOLOGY
This article draws on in-depth interviews with forty-four Congolese refugee families resettled to the Unites States and thirty-five aid workers and volunteers who assisted refugees.8 This interview sample is part of a larger project, which included weekly home visits with four of the Congolese refugee families and observations with aid workers (see Sackett and Lareau, forthcoming). At the time of the study, from 2014 to 2022, Congolese refugees were the largest group being resettled to the United States and are an important yet understudied group. The study went through extensive Institutional Review Board review and approvals.
The interviews with the refugee families were conducted in two phases. During the first phase, Sackett conducted in-person interviews with twelve Congolese refugee families resettled to Philadelphia within the previous three years.9 During the second phase, three years later, Sackett and three Kenyan research assistants, undergraduates at our university and native Swahili speakers, conducted interviews with thirty-two additional families. Just as recruitment was scheduled to begin, the university closed all in-person operations in response to the COVID-19 pandemic; face-to-face research was generally prohibited. As a result, Sackett and the research assistants conducted interviews with families by telephone and occasional video call. Because the interviews were virtual, we expanded our geographical reach, recruiting interviewees across the United States. Given the refugees’ trauma of fleeing, distrust of government officials, and unfamiliarity about the research process, we found that recruitment was more effective when someone vouched for the study. In an effort to broaden our sample, we recruited refugees from multiple social networks and sponsors. Refugees were told that the study was about their adjustment to life in the United States; they were given a gift card of $30 for the interview. Interviews varied in length but were often ninety minutes; due to their pressing obligations, some interviews took place over multiple calls. The interviews revealed a wide range of institutional barriers to services, as well as key resources. Although in a few instances, respondents were uncomfortable with us recording, and we honored their wishes, most interviews were audio-recorded, transcribed in Swahili, and translated into English.
Additionally, Sackett conducted interviews with thirty-five aid workers and volunteers, recruiting those who worked with families from the DRC. During the interviews, the aid workers discussed at length the institutional obstacles refugees faced. The majority of aid workers and volunteers were White and highly educated. A few of the caseworkers were immigrants and refugees themselves. These interviewees were recruited through a snowball sample. All were conducted in English, audio-recorded, and transcribed verbatim.
The research team included researchers with varying social locations or positionalities.10 As White women born and raised in the United States, we, the authors, are outsiders to the refugee respondents in terms of race and nationality as well as in our lack of experience in being a refugee. Sackett sought to build rapport through her advanced-conversational Swahili and experience living in a refugee camp (when she conducted research for a separate research project). Our research assistants were all native Swahili-speaking women from Kenya and phenotypically Black. Although they shared a native language with the participants, the RAs and refugee interviewees drew distinctions between their countries of origin (Kenya, a middle-income and stable country, versus the DRC, a low-income nation in conflict) and reasons for migration (to attend university versus to escape war). Surprisingly, we did not find that the accounts of racial discrimination that refugee respondents provided to Sackett differed from those reported to the research assistants. Respondents were more likely, however, to volunteer strong opinions about cultural differences in marriage and parenting practices in the United States, such as constraints against corporal punishment, to the research assistants.
We coded the findings, searched for disconfirming evidence, and followed standard research practices for working with qualitative data (see Lareau 2021). In coding, we sought information on key themes—loss of services, moving, racial discrimination, cultural brokers—as well as for disconfirming evidence. Although we used the Dedoose program early in the study, we subsequently found it more helpful to read the interviews looking for key themes and create extensive Excel tables with the names of the respondents and their experiences on selected themes (for example, food stamps, housing problems, moving, and others). We trained research assistants who helped with the coding, and we had weekly meetings to discuss patterns that emerged. Our open-ended interview guide focused broadly on institutional navigation; patterns on the experience of administrative burdens emerged through our data analysis (for a discussion of the emergent nature of qualitative research, see Lareau 2021). All names are pseudonyms.
FINDINGS: MANY INSTITUTIONS, MANY OBSTACLES
When refugee families needed help with food, shelter, and other necessities, they often walked into large government buildings to fill out complex forms or spent time scrolling the internet to find the right website to make a request. Navigating these institutions entailed costs as refugee families learned how systems worked, exerted time and money to comply, and shouldered psychological burdens in the stress and uncertainty.
Sometimes these routine institutional processes moved smoothly, obstacles were overcome despite costs, and refugee families gained valuable resources. Refugee families received limited but essential assistance from resettlement agencies, nonprofit organizations subcontracted by the federal government to help during the initial months after arrival.11 They were met at the airport by a caseworker, who then took them to an apartment furnished with basic items; and acting as “navigators” (see DeLuca et al. 2023), caseworkers helped refugee families enroll children in school, apply for benefits, and secure medical care (see, among others, Gowayed 2022; Ludwig 2016). Many refugees also received support from the social safety net. For instance, while Zainab and her family had faced hunger in the refugee camp, on their arrival in the United States, they were able to access food assistance from SNAP. As Zainab recounted, these resources made a difference: “After arriving, we were never hungry. They gave us food stamp assistance. We were still new with no jobs, but we were given houses [to stay in]. They told us, ‘You use food stamps when you go to the store’. . . . We used it to buy groceries, and just like that we never went hungry. No kids were complaining of hunger.”
Others benefited from college preparation programs, scholarships for college, and homeownership programs, including financial support for first-time homeownership (see also Sackett and Lareau, forthcoming). Institutions thus offered a wide range of resources beneficial for economic mobility.
Yet refugee families also faced institutional obstacles, which created navigational costs, or administrative burdens, and threatened to block access to valuable resources. We conceptualize two types of obstacles: knots and reverberations. Knots are complex blockages, usually triggered by an error or mishap (see table 1). Moreover, as refugee families navigated multiple institutions, institutional processes intersected within the family. Although these institutions had interwoven demands, they often lacked coordination. Errors in one institution could tangle processes together, forming a knot across institutions and reverberating to create new problems. Overcoming these obstacles created administrative burdens for refugee families and the caseworkers and volunteers helping them. To forefront the cost of administrative burdens, we focus on obstacles that were eventually overcome; however, if left unresolved, these obstacles also led to missed services, entitlements, and opportunities, despite eligibility (see Sackett and Lareau, forthcoming).
TANGLES IN PROCEDURES: INSTITUTIONAL KNOTS
While institutions could provide valuable resources, things sometimes went wrong and procedures tangled into a knot. Officials made mistakes. Documents were lost in the mail. Sometimes, the refugee families (or the volunteers helping them) misstepped: they took the wrong bus and missed an appointment or filled out a form incorrectly. Key infrastructure faltered: mail was not delivered, paperwork was not recorded, or a computer system jammed. Although each institutional error might seem unexpected, unpredictable, and unique, evidence from other research indicates that these system errors are woven into the structure of service delivery (Foner 1994; Lopez 2010; Szymczak and Bosk 2012; Perrow 1999; Vaughan 1997). In complex systems with intersecting institutions and limited slack, even seemingly small mishaps could snag institutional procedures and compound to create complex institutional blockages. Simple problems then became complicated and challenging to unravel. We refer to these complex problems as institutional knots.12 Difficult to resolve, these knots increased administrative burdens for families and those helping them. Some knots were resolved, but they were costly to fix. As we show, some of these problems were tied to the families’ refugee status, but others were more run-of-the-mill organizational mishaps. In addition, racism, xenophobia, and discrimination had the potential to compound these problems.
Complex systems entail opportunity for errors—which could tangle into knots. For instance, due to the decentralized nature of service delivery in the United States, a move across county lines meant that services needed to be transferred to the new county. For refugees, who often did not choose the country or the city where they were resettled, secondary moves were common (Bloem and Loveridge 2017). Several refugees in our sample moved—to a cheaper residence, to be near family or friends, or for other reasons. Moving and transferring services not only brought administrative burdens for the families but also introduced opportunities for error. For example, when Safi moved from one side of the Philadelphia metro area to the other—a move of less than five miles, but across a county line—an institutional knot developed, and her SNAP benefits, unbeknownst to her, were cut off. As Safi’s caseworker Wendy recalled, after the move, “Safi had not received her food stamps. She moved from one county to another, she needed to transfer the benefits. She told me, ‘I went to welfare and I told them about [the move], and they said they were going to transfer my benefits.’ So, I said, ‘It’s fine. As long as you’ve done that, that’s okay.’”
Safi correctly sought to overcome a key administrative burden by visiting the welfare office and filing paperwork to transfer her benefits. However, it turned out that at the same time, the semiannual reporting form to prove that she still qualified for food benefits was also due. The two different parts of the organization made an error by failing to coordinate the different organizational divisions. Wendy recounted what happened: “They sent the semiannual reporting form to Safi’s old address. So, we [Safi and I] missed that form. When I followed up with the welfare office, they told me, ‘We sent the forms to this [older] address.’ I told them, ‘She moved—and she came in here and told you she is moving.’ The officer checked on it, ‘Yeah, she did.’”
The welfare office confirmed to Safi’s caseworker that Safi had not made an error; instead, the office had done so when the recertification division was not apprised of the new address. Further, the back-up system failed when officers were unable to reach her by telephone and subsequently discontinued her food stamps. Wendy shook her head as she recalled Safi’s devastation: “She literally cried at that food stamp place and refused to leave, because she said she has no food to give to the children. They even didn’t understand that this person is a refugee with two kids, a single mom, struggling!”
Wendy helped Safi submit the correct paperwork on Friday morning, but it typically took five to ten business days to be processed—an additional delay of a week or two. At this point, Safi had run out of food. Fortunately, Wendy’s supervisor was able to get the family gift cards for food as a stopgap.
In the complex system, an institutional error tangled procedures into a knot. For Safi, moving (and the requirement to submit a change in address) prompted a knot. Although the specific unfolding of events may have been unique to Safi, eighteen families (41 percent) in our sample reported moving in the United States, and six families faced snags in transferring benefits. Moreover, the knot had consequences: Safi and her children lost their food benefits for weeks. Within this system, the onus was on clients, such as Safi, and their caseworkers and social ties to untie the knot. Solving it was burdensome, involving costs in the time visiting the office, calling administrators, filling out paperwork, and waiting weeks for processing. For Safi, the knot led to precarity—and the risk of food insufficiency for weeks. Because of the error, she used up scarce resources. Even for refugee families, such as Safi’s, who were able to meet the costs to untie knots, they found themselves back where they started.
Institutional agents played a key role in both triggering and untangling institutional knots. Frontline workers could enthusiastically help refugee families overcome errors or drag their feet. Some interactions with institutional agents were hostile, leading participants to raise questions about racial discrimination. For example, two White volunteers, Nathan and Nancy, were aiding a Congolese refugee, Sandrine, secure a driver’s license. The rules were complex. The driver’s license agency required a green card, but refugees are not eligible for a green card until after twelve months in the United States. Befuddled, they reached out to the caseworker who provided the form, and Nathan went to the Department of Motor Vehicles with Sandrine, where they met with a cool reception. As Nathan explained, at first, the clerk simply refused to help: “We waited our turn, then when we got to the desk, Sandrine told the clerk what she was there for. The clerk asked her for her green card. And Sandrine said, ‘I don’t have it.’ And the clerk said, ‘I can’t help you then.’”
The institutional agent made an error in applying the rules. Nathan contested the error, providing the regulations:
I had a copy of the regulations—regulations that the clerk should be aware of. So I handed it to her, and said, “What about this here?” She looked at it a little bit and took off [and] went to see her supervisor. Didn’t talk to us at all. She came back and sat down and started typing. She asked Sandrine for whatever information she needed, and then she said, “That would be $50.” Sandrine gave her a check. There was not a whole conversation about “I am sorry, I didn’t know about all this.” Then when we got up to leave, I had pulled a chair up and I was going to put the chair back. But before I had a chance, she said, “That chair goes back there.”
In interpreting the clerk’s comments, the volunteer interpreted it as “a little racism there.” For the Congolese refugee families in this book, racism and xenophobia in institutions could make matters worse, prompting more potential moments for knots to occur, and making them harder to solve. As others also show, even in the absence of conscious discrimination, administrative burdens are racialized, disproportionally affecting marginalized groups (Ray, Herd, and Moynihan 2022). Institutional agents in immigration offices, schools, and welfare offices could use their discretion as street-level bureaucrats to help, yet refugee respondents reported that they were often skeptical and unhelpful in untying knots. In some instances, as with Nathan and Nancy, volunteers helped refugees overcome obstacles and access services. In other instances, volunteers introduced new errors.13
At times, refugees faced problems with administrative clerks who did not know specialized rules. For example, refugee status led to additional institutional interactions and processes, like applying for a green card, which were specific to the immigrant experience. Often within these institutions, the onus was on refugees—and the caseworkers and volunteers helping them—to know the rules and regulations around refugee status and protect against errors. For instance, in addition to the initial first three months of government support for refugees, federal policy decreed that some refugees, who were not eligible for other funds, including Supplemental Security Income, were eligible to get cash assistance for the first eight months after their arrival. This eligibility was rare, however. Sometimes, when refugees went to apply for assistance, officials mistakenly turned them away. Lisa, an aid worker at the resettlement agency, explained: “Most people who are adults and who are healthy don’t qualify for cash assistance. But if you’re a refugee, you qualify for [the first] eight months. So, several times we’ll hear, ‘You don’t qualify. You’re twenty-two years old, and you’re healthy. You have to get a job.’ And so, we’re like, ‘No! They’re wrong! You’re a refugee.’ They [the client] went in with their proof that they’re a refugee, but whoever was interviewing them didn’t know what the regulations were with regard to refugees.”
When refugee families were turned away from services they were legally entitled to receive, they faced additional administrative burdens to correct the process. They had to return to the office, often with a caseworker or volunteer, who also had limited time and availability. Because applications had to be done in person to verify identity most of the time, they had to pay for an additional roundtrip on public transportation, which further depleted their scarce resources. Thus a small error could spiral into an institutional knot. Social workers emphasized that in this county-based system workers varied across counties. Philadelphia County workers were more experienced. Other county workers made errors and denied entitlements to refugees. As a result, Lisa stressed with a laugh, “There’s just no way new refugees can navigate these different institutions. It’s hard for us [caseworkers], and we do this a lot.” The complexity of the decentralized system, the complication of the special refugee case, and the lack of built-in checks in the system led to errors that often escalated into an institutional knot.
Untying knots was burdensome—entailing learning costs, compliance costs, and psychological costs. For Alphonse and his wife Bahati, one mishap—a missed bus—snagged institutional rules and processes within multiple interconnected institutions: the public school system, the police, and (nearly) Child Protective Services (CPS). Although the missed bus itself was not a knot, it did trigger a complex institutional tangle, and the administrative steps to untangle it imposed additional costs. In the family’s first month in the United States, Alphonse and his wife, Bahati, enrolled in English classes on the other side of the city through their resettlement agency. To get to class, they needed to use public transportation, which was a complicated system. Their caseworker showed them the exact buses and bus stops from their apartment to the class for the scheduled class time. But, on their first day traveling alone, they realized they would need to leave early at 2 p.m. to pick their children up from school. As Alphonse soon realized, the change in time triggered a mishap: “We had been used to our route number. So we boarded the bus, but it kept going round and round. We got lost. The bus routes were changed, and the route numbers changed as well.”
Two hours later, at 4 p.m., Alphonse and Bahati were getting desperate and decided to get off the bus at the next stop, in hope of finding some assistance. After waiting more than an hour, they found a policeman, who called their resettlement agency, got their address, and helped them get home. To their surprise, when they opened the door to their apartment, “the children weren’t there.” By the time they arrived home, their children would have returned home from school hours earlier—to an empty apartment. Their delayed return home conflicted with institutional expectations that parents should not leave their children unattended in the United States. To get their children back, Alphonse and Bahati had to complete new administrative steps:
I called the police. They asked me very many questions, “Do you want an ambulance?” I said, “No.” “Do you want this?” I said, “No.” So finally, I told them that I can’t speak English. They got a Swahili interpreter. I told them what happened—how we got lost, that we had just arrived at home to find the children weren’t there. Then they asked us many questions like, “Which school do your children attend?” “What’s their height?” “What kind of clothes did they have on?” “What color?” So apparently the children came home after school, and when they did not find us, they decided to go back to school. At school, they were taken to the police station.
To get the children back, Alphonse had to comply with police questioning in English. Finally, at around midnight, with a “stern warning,” the police brought the children home. “They had cried their hearts out.” Alphonse and Bahati had a straightforward goal: to get home and pick up their children from school; yet their mishap with the complicated public transportation system conflicted with complying with expectations the school system and legal system, snagging institutional procedures and creating a blockage with compounding complexity and severity. If the children had been left at the police station longer, the police would have formally opened a case with CPS and possibly placed the children in foster care. Indeed, our interviews with refugee families and aid workers revealed six instances of entanglements with CPS. Even though in this instance Alphonse and Bahati narrowly dodged a CPS investigation, the institutional knot was stressful and emotionally exhausting—psychological costs. Alphonse concluded, “That is the biggest challenge I have ever faced here, and I will never forget about it.”14 Thus, untying the knot led to new learning costs, compliance costs, and psychological costs.
ONE PROBLEM LEADS TO ANOTHER: INSTITUTIONAL REVERBERATIONS
At times, institutional knots in one institution reverberated to create new, unrelated problems in other institutions—prompting additional administrative burdens in more institutions. For instance, for Jeanine and her family, a delay with her green card application after a mishap in changing addresses threatened to reverberate, causing a new problem in the workplace. Jeanine, who worked as a nurse in the refugee camp, was resettled to Colorado with her twin middle-school sons and her mother. She got a job in the cafeteria at a retirement home for $13 an hour. A team of church volunteers helped the family set up in a little rental house and while they supplemented the rent things seemed good. At the end of the first year, however, the volunteer team reduced their financial support—including their rent supplement. Money was already tight. Jeanine fretted, “my son wants $100 shoes like the other kids at school.” After the hardships her son endured in the refugee camp, she stressed that she “wants to provide for him, so I need to have savings.” In addition, when she was able, Jeanine sent $20 or even $50 back to her nieces and nephews and “mentally unstable” aunt in the refugee camp. Jeanine decided to look for another place with cheaper rent. A Congolese friend helped her find a place across town. The change in address, though, triggered events that became tangled into an institutional knot, creating a serious delay in receiving green cards. With the help of an aid worker, Jeanine and her family had complied with the green card application procedures, as Jeanine recounted:
We went to [resettlement agency] offices and told them that we wanted to apply for a green card. At the office, we were asked to pay $100 for adults and $50 for children. “All that money?” “Yes.” We felt that was a lot. We were then referred to a guy who helps refugees free of charge. So, we went to him and he asked us if it was our first time [to apply] and we said, “yes.” “How many people are you in the family?” We told him. We booked an appointment, and we went and filled out the forms.
Yet the family’s move during the process complicated the procedures, as Jeanine remembered:
By bad luck, we applied for green cards while in [our old neighborhood], and while waiting for the green cards to be processed, we moved. So, we had to change our addresses at the hospital and everywhere else. So, we started receiving the letters at our new house, including one letter indicating that we would receive our green cards “in August on this and this date.” So, in August we didn’t receive the green cards, yet the letter had indicated that we would receive them by that date. We asked the guy to check that out. He made a follow up and learned that the green cards had been sent. “Where to?” The letter indicated that we would receive them here. We checked the mailbox, and there was nothing. We went to the post office and asked about it, and we were told that the green cards had been sent. We went to the headquarters, and again, “the green cards were sent.” That is how we lost the green cards.
Untying the knot imposed new costs as Jeanine navigated additional administrative steps to track down the lost green cards. She fretted that “if we have lost the green cards, we will not be issued with another.”
Moreover, the problem with the green cards threatened to reverberate and lead to problems in the workplace. In the United States, government restrictions mandate that employers collect documentation on immigration status. During the delay, Jeanine’s manager at her job at the retirement home started asking to see her green card. Her papers and ID on file had expired, and her employer wanted her to have the proper paperwork. Jeanine worried that she would lose her job. In America, she stressed, “money is everything,” and she agonized that she would not be able to get by if her boss decided that he was tired of waiting.
Later, when Jeanine was at the hospital with one of her children for an unrelated appointment, she told her social worker about the problem with the green cards: “The social worker followed up. The response was that they already processed our green cards. So, he told them that they got lost, ‘They received the letter but not the green cards.’ So, they asked that we send them copies of our documents. We made copies and sent them. I thank God because it took about three months after following up, but we received the green cards.”
The social worker was a crucial navigator, and finally, eight months after initially applying for their green cards, the documents arrived in the mail. As Jeanine said, “We were very happy. We thank God that we received them.” Yet the process was harrowing. To understand what happened, the refugee family worked with two caseworkers, each of whom found out one piece of the puzzle. This was a near miss, but the delay nearly cost Jeanine her job, threatening the family’s financial future, and resolving the problem created new administrative burdens.
Reverberating problems also surfaced for another family when a traffic accident and minor communication problem with a police officer ballooned into a court case. Although the car accident itself was in part simply bad luck, the ensuing institutional knot was linked to routine, racialized institutional processes in the criminal justice system.15 In this instance, Kashindi was driving his wife to an appointment at a hospital clinic in Tennessee. Early after he arrived in the United States and started work at an auto supply factory, Kashindi started saving for a car and then bought what he called a “junk car” for a few thousand dollars. He then got a learner’s permit, and a friend taught him how to drive on the weekends—during the week, there was no time between Kashindi’s day shift, his wife’s night shift, and juggling watching their five children. That day, driving down the highway with his wife, Kashindi was in the wrong place at the wrong time. He was hit by another car: “My wife and I were involved in an accident. . . . We were driving as usual, but then there was another driver, who was driving at a high speed. He came and hit our car on the driver’s side. It was such a big accident. The tire on the driver’s side came off, so I lost control of the car. The car slammed to a stop.”
Kashindi was grateful that, despite the damage to the car, they came out with only minor bumps and bruises. In the commotion, someone called the police. Kashindi had a learner’s permit but miscommunication followed when the police officer asked for his ID and Kashindi sought to comply with the request:
The police came and took a statement. They asked me, “Where is your ID?” So, I gave them my American ID—when you come here, you are given an ID card (later you are given a green card after a certain process, and then later you can apply for citizenship). So, I gave them the ID I have always had. “Give us your ID,” and I gave [it to] them. So, they wrote in their statement that I was driving the car and that I only had an identity card—yet they had wanted me to give them a sort of identification permitting me to drive.
The police officer cited Kashindi for driving without a license, and Kashindi’s problem moved to the court system. Navigating the courts required additional burdensome administrative steps. In court, Kashindi explained the misunderstanding:
When I went to court, they asked me, “Why are you driving the car using your ordinary ID card?” In court I told them, “No, I have a learner’s permit. Not just this ID.” And they were like, “Why then haven’t they mentioned here that you have learner’s permit?” And I told them, “They asked me for my ID and I gave them. They did not ask for a learner’s permit. They asked me, ‘Where is your ID?’” And I took my ID and gave it to them. “Do you have a learner’s permit?” “Yes.”
Kashindi was given yet another court date. By then he had taken—and passed—the driver’s license test, and the court instructed to bring his license with him: “[By] my last appointment, I already had a driving license. So, I told them, ‘Here it is.’ And they said, ‘If you have a driving license, then it seems like the fault wasn’t yours. The one who is at fault is the one who came to ask you for your identification documents. This case has been dismissed.’”
Kashindi navigated the DMV, secured a driver’s license, brought the correct documents to court, and successfully appealed the case. What started with a car accident that was not his fault—simply his being in the wrong place at the wrong time—threatened to spiral into legal trouble and financial penalties. The accident triggered a citation. The citation triggered a court case and potential charges. After a visit to the DMV, another institution, and the court, Kashindi was able to stop the ripple from spreading further: the charge was dropped and had no lasting consequence. Still, Kashindi took off work and was anxious. As institutional problems reverberated, refugee families like Kashindi’s faced additional administrative burdens in different institutions with new learning costs, compliance costs, and psychological costs. In sum, small errors, particularly in crucial institutional processes such as green cards, credit scores, driver’s licenses, and social welfare programs, tangled into knots that reverberated, creating cascading and costly new problems in new institutions.
CONCLUSION
Administrative burdens surface in a wide array of contexts across the life course—from childcare programs to Medicaid, as the articles in this double issue demonstrate. Scholars have often focused on one type of obstacle in one type of one institution, each with unique rules and procedures. Although studies on administrative burden underscore the formal organizational rules and requirements which burden clients (Herd and Moynihan 2018; Ray, Herd, and Moynihan 2022; Moynihan, Gerzina, and Herd 2021), room remains in which to deepen and clarify our understanding of the institutional mechanisms that create administrative burdens. In this article, we focus on two types of obstacles: institutional knots and reverberations. By showing how knots and reverberations form and the burdens involved in untangling them, we illuminate conceptual similarities across a range of social service institutions.
Some institutional obstacles were specific to refugees and other immigrants. Yet many Americans may also experience some of the institutional problems that refugee families faced: a misunderstanding triggering CPS, errors transferring food stamps across the county, or a hiccup getting Social Security benefits. Indeed, we found that even after living in the United States for years, refugee families continued to run into institutional knots. The American volunteers helping refugee families also found knots hard to prevent as well as cumbersome and difficult to straighten out. This is not to suggest, however, that all clients have similar pathways. After all, in the context or pervasive racism, Black clients experience unequal treatment in key institutions, including police-citizen interactions, workplace assignments, health-care services, schoolyard interactions, and other social processes. Indeed, the Black refugees in this study reported racist insults at work, in housing searches, in encounters with police, and in other social spheres; in some instances, their worries about racialized police violence even led them to question the wisdom of migrating to the United States (Sackett 2022). These racialized interactions, as others have noted, increased the risk of administrative burdens (Ray, Herd, and Moynihan 2022). We find that these knots and their reverberations led to administrative burdens for families and at times, led to downward mobility.
The conditions for knots were baked into the system—through both the design and implementation of systems. Fragmented, decentralized, and uncoordinated systems created errors. In addition, in a context of cost-cutting, downsizing, and increased surveillance, seemingly small institutional hiccups became knots as overburdened agencies struggled to provide services. Some of these knots were likely unintended, and the product of unforeseen implementation problems. In other instances, as Herd and Moynihan (2018) show, the complex nature of systems are part of a deliberate effort to reduce the number of people receiving services. For example, onerous requirements to demonstrate eligibility for services routinely created knots. Lilly Yu (2023, this issue) finds that the Donald Trump administration introduced a slew of administrative changes, which burdened immigration attorneys, increasing the potential for things to go wrong while reducing the number of pathways to resolve related problems. Policymakers sometimes clearly anticipate that a policy change will create knots that can make eligible clients lose services (described as administrative churning). For example, with the winding down of the public health emergency tied to COVID-19, Medicaid’s continuous enrollment period ended, and the Health and Human Services Office of Health Policy estimated that around 8 percent of clients would subsequently lose Medicaid, even though they were eligible. However, because staffing and control of eligibility assessment were determined at the state level, limited steps were taken to prevent these exclusions.16 Future research might explore how and when different types of knots emerge, particularly differentiating between institutional problems that are a function of unintentional mishaps and those that are designed. In some cases, illuminating the sources of knots may lead to remedies.
Studies on organizational structures suggest that it is possible to design more streamlined and reliable systems. Examining high-risk technologies, such as nuclear power plants, Perrow (1999, 4) shows that in systems designed to have considerable leeway to check and correct for mishaps, errors can be caught before they produce negative consequences (see also Vaughan 1997, 2021). Some organizations are designed to prevent and safeguard against errors that involve grave consequences from hazards, such as when bridges collapse, nuclear power plants explode, and planes collide over airports. In these settings, some organizations introduce systems to “achieve high levels of continuous reliability” (LaPorte and Consolini 1991). These insights might be fruitfully applied to public institutions providing social services.
In social service organizations, fewer rules and requirements, more streamlined processes, and more slack in systems can lead to reduced administrative burdens and increased access to services. For example, when the federal tax credit for families was introduced in 2021, families were required to apply to access the funds (Child Tax Credit 2022). Overall, around 20 percent of families did not receive the funds (Hamilton et al. 2022).17 By contrast, in the fall of 2020, when the IRS mailed out relief checks during the COVID-19 pandemic, distribution was widespread and nearly universal. The paperwork to screen eligibility is costly to administer and provides opportunity for error. Moynihan and his colleagues (2022) find that lowering compliance costs, such as providing information about administrative categories on SNAP forms, increased accurate state group categorization. As we consider policy alternatives for refugees and other families, simple, universal services can reduce administrative burdens.
Drawing on insights from theorizing on high-reliability organizations, we also suggest that built-in slack could help prevent and correct errors in social service organizations. In many instances, refugee families did receive help from third-party actors, such as caseworkers and nonprofits, through resettlement agencies; these helpers provided external sources of slack as they double checked procedures, caught errors, and sought to course correct (see also Sackett and Lareau, forthcoming). This slack helped prevent and untangle knots and halted reverberations from becoming new problems. Yet, because institutional obstacles are rooted in the structure of systems, and because these outside helpers were limited in their ability to change system design, their help, though invaluable, was often not enough. Similarly, Yu (2023, this issue) finds that although immigration attorneys helped clients overcome new administrative hurdles, they were unable to change the rules and requirements that prompted these hurdles. Instead, we suggest that slack should be built into the systems. For example, organizations might implement systems in which staff follow up with reminder messages, deadline extensions were permitted, and other forms of oversight might prevent the cascading impact of organizational snags. Because institutions are fragmented but interconnected, slack in one system can have positive reverberations across other agencies.
In the end, we find institutional knots to be costly and consequential. Even obstacles that were eventually overcome required time and complex information to resolve, and the process was stressful. Other institutional knots were not resolved and led to missed resources and penalties, which could have long-lasting consequences. Most of the families in our study were in precarious positions because they worked in low-wage jobs with limited support, and the knots and reverberations were a barrier to upward mobility. Refugees, and others in the United States, interact with a wide range of institutions, each of which has the potential for knots and reverberations. Social scientists would benefit from more directly conceptualizing knots and reverberations across a variety of different agencies and studying the impact of these burdensome obstacles on life trajectories, as immigrants and others seek to improve conditions for themselves and ultimately for their children.
FOOTNOTES
↵1. For studies that take up the interdependence of institutions, see Black and Keyes 2020; Paik 2021.
↵2. Refugees are people who have fled their home country due to war or other disasters and have been granted legal refugee status by the United Nations Refugee Agency (UNCHR). In the literature on immigrants, refugees are often considered to be a distinct type of immigrant (for a discussion, see FitzGerald and Arar 2018). Unlike other types of immigrants, refugees are chosen for resettlement and assisted in the process by international nongovernmental organizations and the UNCHR, rather than self-selecting for resettlement and relying on social ties. Refugees resettled to the United States are usually unable to control the city they are sent to, the neighborhood and house in which they are settled, and the timing of when they are moved. The federal government funds the resettlement process for refugees, but relies on NGOs to serve as resettlement agencies.
↵3. By definition, institutional knots are problems within one institutional setting. Yet untangling knots may require navigating multiple institutions. For instance, receiving SNAP requires completing a recertification form, which involves documentation from other institutions, such as paystubs from the workplace and copies of government identification from government agencies (see Sackett and Lareau, forthcoming).
↵4. The literature on cultural brokers is extensive and often focuses on the role of children in aiding their immigrant parents (see, among others, Katz 2014; Delgado 2020; for more on the role of outside helpers as cultural brokers, see also Sackett and Lareau, forthcoming).
↵5. Indeed, Melvin Kohn (1971, 473) argues that “what is notable about bureaucratic practice is not how closely authority is exercised but how effectively it is circumscribed” (emphasis added).
↵6. Additionally, as Judith Levine (2013) shows, low-income clients often develop extensive knowledge to interact with these agencies.
↵7. Many of these organizations are bureaucratic and emphasize “objective,” formal procedures, rules, and documentation (Weber 2015). Max Weber (1949, 215) defined bureaucratic objectivity as “a discharge of business according to calculable rules and without regard for persons.” Yet scholars since Weber have debated the extent of this bureaucratic objectivity. John Meyer and Brian Rowan (1977, 343) argue that many organizations are loosely coupled with “powerful institutional rules which function as highly rationalized myths.”
↵8. In this article, we use the term families because family members often live together. People who are not blood relatives but live together may be considered part of the family unit. A number of family members were killed or not permitted to resettle at the time; some men had multiple wives in refugee camps. At times, adults lived with their parents. Thus family structures among research participants varied in form and included a number of single-parent families.
↵9. For a discussion of the immigration context in Philadelphia, see Singer et al. 2008.
↵10. Of course, racial patterns of inequality reverberate through daily life in countless ways, and these persistent inequalities have raised deep, complex questions about the position of researchers doing research as outsiders. Reasonable people disagree about the legitimacy and desirability of different researchers’ positionalities. One common position in the social sciences, however, is that both insiders and outsiders have valuable perspectives to bring to the research process. Insiders—in this case, Black refugee researchers from the DRC—would have intimate knowledge of the refugees’ experiences that could likely more easily build a trusting relationship with research participants and enhance interpretation of their experiences. However, outsiders can bring fresh perspectives too—they can adopt a position of naivete that draws out explicit reflections on experiences and perspectives that otherwise might be assumed, and they can ask questions that might be inappropriate for insiders to raise (a full discussion is beyond the scope of this paper, but see Lareau 2021).
↵11. In recent decades, however, the amount and length of assistance from the federal resettlement program has decreased (Brown and Scribner 2014), and caseworkers in resettlement agencies face increased demands (Fee 2019).
↵12. The idea of institutional knots parallels Charles Perrow’s (1999) concept of system errors, also referred to as normal accidents. Perrow conceptualized system errors in the context of complex technological systems and organizations. He discussed complex organizations, such as nuclear power plants, and how an unanticipated sequence of unimaginable events created disasters. We use his concept in the context of mundane interactions with social service institutions. Many of these errors are routine and thus well known, but the exact sequencing of how the errors unfold is highly variable.
↵13. As we explain elsewhere, volunteers were frequently learning about social service programs for the first time. In one instance, a volunteer was helping a refugee family get their food stamps reinstated, and the employee mistakenly directed the volunteer to leave forms on the application blank, saying that the computer would automatically fill out the birthdays. However, because the family had been terminated, the spaces were not populated. The application was rejected and a longer delay ensued (Sackett and Lareau, forthcoming, chap. 4). In a fragmented system, it is also possible that one person’s actions trying to untangle a knot could reverberate and create a new knot in a different system.
↵14. In addition, because of how he interpreted the warning by the police “to not leave them outside,” after that day, they kept the children indoors after coming back home from school. Although the children were safe indoors, Alphonse lamented, “They do not make friends.” Thus even one-time events, which were resolved without incident according to the agencies, could have serious consequences that cast a shadow on the lives of the families.
↵15. Rob Voigt and his colleagues (2017) show police officers to be more polite and accommodating with White citizens than Black ones in routine traffic stops. In this instance, the failure of the police officer to explore if the refugee had any other forms of identification is consistent with racialized, abrupt, and sometimes lethal treatment of citizens in the criminal justice system which routinely creates institutional knots.
↵16. Although the Department of Health and Human Services can provide to guidance to states, the assessments of clients, staffing, determination of eligibility, and other policy matters are managed at the state level (Office of Health Policy 2022).
↵17. Michael Karpman and Elaine Maag (2022) report that 31 percent of low-income families who did not receive the tax credit reported that they did not know how to claim it.
- © 2023 Russell Sage Foundation. Sackett, Blair, and Annette Lareau. 2023. “Institutional Entanglements: How Institutional Knots and Reverberating Consequences Burden Refugee Families.” RSF: The Russell Sage Foundation Journal of the Social Sciences 9(4): 114–32. DOI: 10.7758/RSF.2023.9.4.05. We thank the refugee families, aid workers, and volunteers who shared their experiences. Maria Odongo, Mirriam Chemutai, and Frida Aloo were exceptionally talented research assistants who helped carry out interviews with refugee families. We also gratefully acknowledge the financial support of the School of Arts and Sciences of the University of Pennsylvania. An earlier version of this article was presented at the 2022 Russell Sage Foundation conference “Administrative Burdens as a Mechanism of Inequality in Policy Implementation.” We are grateful to the feedback of conference participants, especially from the editors. We are also thankful for the input from the doctoral student writing group at the University of Pennsylvania: Minseo Baek, Ashleigh Cartwright, Sherelle Ferguson, Peter Harvey, Katharina Hecht, and Doron Shiffer-Sebba. Maia Cucchiara, Judith Levine, Julia Wrigley, Danni Falk, Shaquilla Harrigan, and Frida Aloo provided invaluable comments. We benefited enormously from the feedback given to us from Cecilia Menjivar, Leslie Paik, Van Tran, Emilio Parrado, Chenoa Flippen, Naomi Schneider, and Elliot Weininger. Andy Jimenez, Lee Ang, Edward Stevens, and Isabelle Alaniz did valuable work as research assistants as we were completing the manuscript. Of course, all errors are the responsibility of the authors and not of the sponsoring agencies. Direct correspondence to: Blair Sackett, at blair_sackett{at}brown.edu, Watson Institute for International and Public Affairs, 280 Brook St., Providence, RI 02906, United States.
Open Access Policy: RSF: The Russell Sage Foundation Journal of the Social Sciences is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.