Introduction
Many encounters with welfare bureaucracy tax citizens’ physical and mental energies (Soss, Reference Soss2002). In the current study, we make sense of these encounters by drawing on the emerging administrative burden literature (Herd, Reference Herd2015; Moynihan et al., Reference Moynihan, Herd and Harvey2015), which has demonstrated how these encounters often involve learning, compliance and psychological costs. Our inquiry contributes to this scholarship by unfolding three elements responsible for the emergence of administrative burden: waiting, communication breakdowns, and administrative errors. Drawing upon other bodies of knowledge, mostly public administration, sociological and anthropologic studies on citizen-state encounters, we discuss the subtle and underexplored mechanisms underlying these elements, as well as their multifaceted and context-dependent nature. In particular, based on thematic analysis of semi-structured interviews, we examine the interactions of disability and income support beneficiaries with the Israeli National Insurance Institute (NII). We ask, what elements in encounters with welfare bureaucrats trigger beneficiaries to experience learning, compliance and psychological burdens?
This examination proceeds from a multifaceted and dynamic interpretation of welfare encounters that views them as taking place in various and changing forms and sites, involving interpersonal, structural, physical and technological aspects. However, our inquiry should be interpreted as context-specific, in that it examines the specific welfare encounters in the ‘decision-making factories’ known as the social security system (Bovens and Zouridis, Reference Bovens and Zouridis2002).
We begin by presenting the concept of administrative burden and its key theoretical developments, following which we outline some useful directions in studying three potential triggers of administrative burden: waiting, communication breakdowns and administrative errors. After introducing our methodology, we describe our findings, broken down into these three triggers. We conclude by discussing these elements as experienced by prospective beneficiaries.
Literature review
Administrative burden
Encounters between citizens and social security bureaucrats are complex events often involving anxiety, confusion, and frustration. Recently, public administration scholarship has drawn on the concept of administrative burden to better understand this phenomenon (Herd, Reference Herd2015; Moynihan et al., Reference Moynihan, Herd and Harvey2015, Reference Moynihan, Herd and Ribgy2016). Administrative burden refers to ‘an individual’s experience of policy implementation as onerous’ (Burden et al., Reference Burden, Canon, Mayer and Moynihan2012: 742). The underlying assumption is that formal and informal practices shape our encounters with state programs and their costs. Many such encounters are burdensome, leading also to various material, adversary consequences, as in the form of non-take-up of welfare rights (Finn and Goodship, Reference Finn and Goodship2014; Heinrich, Reference Heinrich2016), or in outsourcing some of the claiming process to other take-up agents (Holler and Benish, Reference Holler and Benish2022).
Recent years have seen efforts to generate a more direct, systematic and integrative conceptualisation of administrative burden (Herd and Moynihan, Reference Herd and Moynihan2019). A key direction in these efforts draws upon Moynihan et al. (Reference Moynihan, Herd and Harvey2015), who identified three main costs: learning costs, which refer to citizens’ difficulties in learning about a program and about their eligibility; compliance costs, related to application and recertification processes; and the psychological costs of both applying and receiving the benefit (see also: Herd and Moynihan, Reference Herd and Moynihan2019; Masood and Nisar, Reference Masood and Nisar2021). Note that although conceptually distinct, these costs are highly interconnected in practice (Moynihan et al., Reference Moynihan, Herd and Harvey2015; Baekgaard and Tankink, Reference Baekgaard and Tankink2022). Finally, the conceptualisation of adminstrative burden emphaises the personal experiences of state actions. In this sense, burden is contingent upon how citizens interpret the world, as well as on other contextual factors, rather than being solely dependent on objective measures of state actions (Moynihan et al., Reference Moynihan, Herd and Harvey2015; Baekgaard and Tankink, Reference Baekgaard and Tankink2022).
The triggers of administrative burden
The administrative burden literature has hitherto focused on identifying the costs or consequences of citizens’ encounters with state bureaucrats, with less emphasis on mundane, everyday practices that trigger administrative burdens (Chudnovsky and Peeters, Reference Chudnovsky and Peeters2020; Baekgaard and Tankink, Reference Baekgaard and Tankink2022). This article contributes to this emerging scholarship by deepening our understanding of three elements that have the potential of triggering burden in the benefit claiming process: waiting, communication breakdowns, and administrative errors. All three have emerged in our case study and resonate with familiar insights in public administration, sociological and anthropological studies on citizen-state encounters. Our analysis further identifies the subtle and underexplored mechanisms that underlie them, concluding that they should be viewed as red flags.
This analysis draws upon claimants’ perspectives and lived experiences. Since administrative burden is a subjective, contextual experience (Nisar, Reference Nisar2017; Barnes and Henly, Reference Barnes and Henly2018; Peeters et al., Reference Peeters, Trujillo Jimenez, O’Connor, Ogarrio Rojas, Gonzalez Galindo and Morales Tenorio2018; Herd and Moynihan, Reference Herd and Moynihan2019; Baekgaard and Tankink, Reference Baekgaard and Tankink2022), which is highly ambiguous and affected by informal, hidden practices that are difficult to identify (Peeters and Widlak, Reference Peeters and Widlak2018), qualitatively examining claimants’ everyday experiences of its triggers is invaluable.
Waiting
Public encounters with state bureaucracy often involve waiting. Administrative burden studies have recognised this critical aspect. However, these analyses have tended to address waiting as a general component of compliance or learning costs (for exceptions, see Auyero, Reference Auyero2012; Herd and Moynihan, Reference Herd and Moynihan2019; Holt and Vinopal, Reference Holt and Vinopal2021). This leaves us with several unresolved questions, including how waiting is experienced in practice, and under what conditions it turns from a mere nuisance into a significantly burdensome experience.
Some useful directions for answering these question lie in previous public administration works (e.g. Goodsell, Reference Goodsell1984; Soss, Reference Soss2002; Lipsky, Reference Lipsky2010), which referred to time and waiting in modern bureaucracy. These, however, were mostly interested in specific encounters at a single point in time, often through onsite observations, rather than as part of the overall claiming process. Further, as many of these studies were conducted in the 1970s-90s, their valuable insights reflected more traditional forms of bureaucratic provision. Therefore, the role of waiting in current benefit claiming encounters, which have witnessed the rise of e-government as well as other organisational transformations, requires further exploration (Lundberg and Syltevik, Reference Lundberg and Syltevik2016; Larsson, Reference Larsson2021).
Drawing on these earlier works as well as more recent bodies of scholarship, mainly sociology and anthoplogy, which take time seriously, some valuable insights can be drawn. One is that waiting can take different forms. It can be chronic, spanning a long period, even generations (Carswell et al., Reference Carswell, Chambers and De Neve2019), as in the case of people who are chronically unemployed or potential immigrants (Conlon, Reference Conlon2011). Two more short-term kinds of waiting are ‘on-the-day’ and ‘to-and-fro’ waiting. The former refers to time spent in queues or in daily waiting to submit paperwork, whereas the latter – also known as the ‘go today, come tomorrow’ type of waiting – involves multiple visits to offices to complete the claiming process. All three intersect, so what may have started as ‘on-the-day’ waiting could under certain conditions become ‘to-and-fro’ and eventually even ‘chronic’ waiting (Carswell et al., Reference Carswell, Chambers and De Neve2019).
Another key insight points to the sociopolitical construction of time (Cohen, Reference Cohen2018). ‘Time is money’, for example, can only be meaningful with the rise of industrial capitalism and the free market logic (Adam, Reference Adam2003). Similarly, ‘waiting for the state’ (Carswell et al., Reference Carswell, Chambers and De Neve2019), as a temporal practice, is experienced differently in different social contexts (Goodsell, Reference Goodsell1984; Seefeldt, Reference Seefeldt2017). Thus, while waiting for two hours in hot, unconditioned public waiting rooms can be highly burdensome, waiting at home for a couple of days can be experienced as rather burdenless (Soss, Reference Soss2002). Crucially, however, as our findings illusrate, the experience of waiting could potentialy be shaped by factors other than location (Goodsell, Reference Goodsell1984), including uncertainty (Ryan and Valverde, Reference Ryan and Valverde2006). Finally, waiting is embeded in unequal power realtions, and is often also being used strategically as an efficient technique of domination and regulation (Schwartz, Reference Schwartz1974; Soss, Reference Soss2002; Lipsky, Reference Lipsky2010; Auyero, Reference Auyero2012; Holt and Vinopal, Reference Holt and Vinopal2021; for the Israeli context, see Helman, Reference Helman2021). In that sense, it is often the powerful segments of society that can escape the costs of waiting.
Communication breakdowns
Another potential trigger for administrative burden is communication breakdowns: bureaucracies’ unwillingness and inability to convey and receive information to and from clients. Communication breakdowns can occur in both face-to-face and online contacts, and represent a crucial factor in adding to or easing administrative burden. Spotlighting communication enables us to highlight three crucial, largely underexplored aspects of administrative burden: the need for personalised communication, the bilateral nature of communication, and the role of ICT.
Personalised communication. While both administrative burden scholarship (e.g. Herd and Moynihan, Reference Herd and Moynihan2019) and take-up studies (e.g. Finn and Goodship, Reference Finn and Goodship2014) highlight the costs of learning about state programs, they tend to focus on difficulties in learning about such programs in general, including complex eligibility rules. Similarly, when bureaucracy is considered to be failing to communicate with its clients, this is usually understood as a failure to communicate such general information. While surely struggling to understand eligibility rules, for example, could be burdensome, that is only part of the story. In particular, we suggest that difficulties in receiving concrete information about what clients need to do in their specific circumstances is another crucial, overlooked aspect of citizen-state communication. Struggling to receive such personalised information could be a highly burdensome experience.
The bilateral nature of communication is another underexplored aspect of communication. Poor citizen-state communication is not limited to the state’s failure to inform citizens, but also includes citizens struggling to convey (personalised) information to state. This occurs, for example, when benefit claimants find it difficult to make an inquiry, reach frontline personnel or report administrative errors (Widlak and Peeters, Reference Widlak and Peeters2020). This kind of communication breakdown can trigger not only learning and compliance costs, but also psychological ones, causing claimants to feel frustrated and disempowered.
ICT. Recently, social policy and administrative burden studies have started to conceptualise the complicated role of technology in shaping claimant-state encounters (Hetling et al., Reference Hetling, Watson and Horgan2014; Madsen et al., Reference Madsen, Lindgren and Melin2022), showing that the use of ICT yields mixed results. On the one hand, it can reduce administrative burden by minimising face-to-face encounters, and can speed up clients’ inquiries on eligibility and application status (Hetling et al., Reference Hetling, Watson and Horgan2014). The use of ICT, particularly digital self-service technology, also means that citizens can time their interactions with the agency (Tregeagle and Darcy, Reference Tregeagle and Darcy2008). On the other hand, this ‘electronic turn’ (Garrett, Reference Garrett2005) can produce additional administrative burden (Herd and Moynihan, Reference Herd and Moynihan2019), due to dysfunctional, highly detailed or inaccessible technology as well as citizens’ limited capacity to use it (Breit and Salomon, Reference Breit and Salomon2015; Hansen et al., Reference Hansen, Lundberg and Syltevik2018; Schou and Pors, Reference Schou and Pors2019). Moreover, in some cases the ICT can result in reduced availability of other more traditional channels of communication, including face-to-face meetings or call centres (Schou and Pors, Reference Schou and Pors2019).
Administrative errors
While administrative errors have been addressed in social policy and administrative burden literatures, they too have usually been treated as a by-product of other organisational mechanisms, and not as one with its own logic. Widlak and Peeters’ (Reference Widlak and Peeters2020) recent study is a good starting point for understanding this trigger. Three of their insights are of particular interest. Firstly, errors are a common cause for administrative burden with significant material and psychological consequences. Secondly, when claimants seek to correct errors, the system produces a further administrative burden, making the correction mechanism, or lack of it, a key element in the process. Thirdly, errors are not disconnected from organisational structures, and are especially rooted in organisational ICT.
In the present study, we point out the centrality of administrative errors in producing burden. A key type of administrative error identified in this study is what Krumer-Nevo and Barak (Reference Krumer-Nevo and Barak2006: 781) call the ‘missing document syndrome’. Importantly, while paperwork could easily be considered a trigger of administrative burden in its own right, by framing it as administrative errors we emphasise that what is important is not the paperwork per se but rather the errors around it. Moreover, the concept of administrative errors includes other types of errors besides those related to paperwork.
To conclude, this review indicates that the elements of waiting, communication breakdowns and administrative errors are all often involved in the claimant-bureaucracy encounter and can be seen as potential triggers of administrative burden of the three kinds: compliance, learning, and psychological costs.
Method
The present study examines the encounters of Israeli citizens with the NII, a semi-autonomous entity responsible for providing most of the social security schemes in Israel. We focus on two key schemes: general disability allowance (GDA) and income support (IS).
The selected cases
GDA is an out-of-work scheme paid to claimants whose capacity to work is heavily restricted due to a medical condition. Obtaining a GDA requires claimants to prove the impact of their medical condition on their ability to work. While theoretically based on standardised guidelines, the NII’s decision often involves relatively high discretion. IS is a public assistance program intended for citizens whose income is below the legally defined minimum. IS claimants must prove their financial need by providing the NII with various kinds of personal information and proof documents. They also need to register as job seekers at the Employment Service. Throughout this process, vast discretionary power is granted to frontline workers. Both GDA and IS were selected as both are emblematic of bureaucratic procedures with a high risk for administrative burden since they embody complex legal requirements, selectiveness, and discretionary power. In addition, both are targeted at vulnerable populations for whom losing access carries high consequences.
Participants
Twenty Israelis were interviewed for this study: all eligible were successful in claiming GDA and IS. Participants were first recruited using non-representative convenience and snowball sampling, with the aim of serving our theoretical purpose (Emmel, Reference Emmel2013): mapping key triggers of administrative burden. Participants were recruited nationally and were contacted prior to the interview in order to verify their qualification for the study and provide them with further information on the study and their role in it. To limit potential biases, claimants were recruited through several channels: Facebook groups on the NII and means-tested benefits; mailing lists of social policy and security professionals; leaflets in various NII branches nationwide; and personal contact with advocacy organisations. Snowball sampling was limited to one referral per participant. Data collection came to a halt upon reaching thematic saturation, and after the participation of diverse groups of participants, including from the Muslim minority, had been secured. The researchers met after every ten interviews in order to decide on saturation.
Of the twenty participants, ten were recruited based on having claimed GDA and IS, respectively. Their reasons for claiming were diverse. Some GDA recipients applied based on an innate, permanent disability (e.g. visual impairment); others applied based on acute conditions and illnesses discovered during the life course (e.g. fibromyalgia). The IS claimants applied based on inability to work due to life conditions such as career changes, health issues, family related difficulties or inability to find a suitable skills-oriented job. Fourteen participants were female (70 per cent); sixteen were Jewish (80 per cent), and four were Muslim. Their ages ranged between twenty-five and sixty-three (M = 44.7). Ten were unmarried (50 per cent); seven married (35 per cent); and three divorced or separated (15 per cent). Finally, all participants were successful claimants, who managed at some point to take up their GDA or IS rights.
Procedure
Semi-structured paired interviews were conducted by at least one of the researchers and a research assistant. The Arabic interviews were conducted by an Arabic-speaking research assistant after careful preparation by the authors. Participation in the study was conditioned on informed consent and it received ethical approval from the university ethics committee.
During the interviews, the participants were encouraged to reflect on their experience of claiming their benefits, while follow-up questions were used to focus the discussion around specific issues. These included their chronological experience and important interactions with the bureaucracy and actions of the NII that impeded or facilitated take-up. The main questions were: How did you experience the state-claimant encounter during the entire benefit claiming process? What helped you and what made the encounter difficult? How did you cope? Each interview was recorded, transcribed, and anonymised.
Data analysis
The interview transcripts were analysed inductively and thematically (Braun and Clarke, Reference Braun and Clarke2006) using MAXQDA. This involved several formal and interdependent stages. The first stage was a careful reading of all transcripts as a single unit in order to familiarise the authors with the experiences of participants. Next, each transcript was broken down into small segments of texts, units of meaning. After these units were coded in an open, bottom-up process, similarly coded units were clustered into initial themes emerging from the text. As reading progressed, themes were changed, dropped and added to better fit the reported experiences. This stage resulted in finalising the themes presented in this study. The process as a whole was accompanied by ongoing reflective discussion between the authors with the aim of fine-tuning the thematic map.
Interviews and analyses were conducted in both Hebrew and Arabic, while quotes were translated into English at the manuscript write-up stage. To ensure accuracy, the authors double-checked the translation. Lastly, to enhance trustworthiness (Nowell et al., Reference Nowell, Norris, White and Moules2017), audit trail and peer debriefing were applied and rich quotations from participants’ accounts were provided.
Findings
Many participants experienced encounters with the NII as difficult if not painful, and reported a variety of administrative costs, in some cases delaying their take-up for several years, and burdening them throughout the process and even long afterwards. Analysis of participants’ perceptions unfolds the role of three elements – waiting, communication breakdowns, and administrative errors – in constituting administrative burden and its three main costs: compliance, learning, and psychological. Often, these elements were highly interrelated, resulting in an interactive impact on the claimants’ burden experience.
‘They have all the time in the world’: waiting for the state
Waiting was experienced in different ways and triggered administrative costs of the three known types: learning, compliance and psychological. One key example is the burden involved in completing the application process and in particular the time involved in filling out forms, including the formal applications and the attachment of proof documents. The participants described this process as highly time consuming, leading to intense negative feelings of overload and frustration and even constituting a significant barrier in the take-up process. Moreover, these costs were not always merited by the circumstances, which only exacerbated their burdensomeness. This is illustrated by Heli, a forty-seven-year-old woman who claimed GDA:
And finally, I said I’d fill out the forms because I had to. It took me a month to organize the materials, because I had to get every medical document from March 2016 to the present. And I’m a meticulous accountant. […] I collected so many documents! A folder this big. It took a month – and that’s just the physiological aspect.
Our participants experienced various types and lengths of waiting. One common description addressed short-term waiting, mostly in queues at NII or Employment Service branches. Short-term waiting could potentially end with no real results, forcing the claimant to start the application process all over again:
Yes, it’s about arriving at the National Insurance and waiting in line […] Sometimes it’s an unbearable queue, up to two hours’ wait, even two and a half…. And sometimes even after you wait that long, and it’s your turn, she either tells you, for example, that some pages are missing, material is missing, ‘you have to submit everything together’ […]. So even on that day […] your problem isn’t solved (Rubin, thirty-two, f, IS).
Another common description addressed long-term waiting. For example, many claimants received no answer for months. For some, these long waiting times resulted in financial harm, as they did not receive the benefit on time: ‘They have all the time in the world. And when one is under pressure, because he knows everything is on hold because of that stupid note […]. In my case, that note meant 13,000 NIS that could really come in handy’ (Ron, sixty-two, m, IS).
A key source for waiting was the need to obtain proof documents from third parties, such as banks, former employers, or health authorities:
We tried to get the [records of the] last three months from the bank. […] You need all kinds of bank forms and all that. It was complicated for us because our bank moved the branch elsewhere and that other branch didn’t want to give us, and they sent us back and forth. It was quite exasperating (Noga, twenty-eight, f, IS).
This account and others about the difficulty in obtaining third-party documents implies that the scope and quality of ICT usage are key to the waiting experience. In Noga’s case, for example, she would not have had to burden herself with obtaining these documents had the NII provided a system for sharing data held in the private sector, or had she been provided a practicable way of applying for these documents electronically.
Some participants noted that recently, there have been some favourable changes in terms of digitisation, including self-service technology. For some, being able to submit forms online was a welcome step that significantly reduced the burden of waiting. Heli contrasted her relatively convenient experience of waiting at home with the highly burdensome waiting at the NII branch: ‘it’s better to focus, to study, to sit at home, do anything from home, rather than arrive at the National Insurance […]. It takes me a whole day to get to the center of town. I can’t take buses’.
Finally, as already implied, waiting also triggered high psychological costs. Being forced to wait, especially without obtaining reliable information about the causes for this waiting or its expected duration, led many participants to feel frustrated and angry with the NII and its staff. There were also cases in which avoidance, fear and negative feelings of the process triggered self-inflicted waiting. This was the case, for example, when participants postponed their claiming process to avoid the negative feelings involved. This is how Ruth (48, f) described her first attempt to claim GDA:
And all those stories in the news, about the people [who are told], dress like that and undress like that […] they don’t prepare you for a very pleasant experience. So I dragged it on for almost a year, with the forms all ready and waiting.
‘We will return your call within forty-eight hours’: communication breakdowns
Our participants’ accounts show how social security claimants do not only need to learn about the program and its rules, but also obtain more personalised information about what to do in their specific circumstances. Many described how failing to communicate with the NII made it difficult for them to receive such valuable information, which consequently left them uncertain in their claiming process. Moreover, struggling to receive such personalised information was highly related to the waiting experience (as well as to experiencing administrative errors, as we explain later). Some participants waited even months without receiving proper answers to their inquiries, such as which documents were needed to complete their application. This is how Ron described his frustrating experience with the NII’s call centre:
Some representatives don’t know anything about your problem […] you can’t communicate directly with the person who’s supposed to take care of you and give you an answer, so you wait for her to give you the answer. And then you get back to the representative […] “We will return your call within forty-eight hours, guaranteed”. They get back to you so you repeat your complaint to the welfare worker, and then she says, “OK, I’ll get back to you on that”. So it’s another forty-eight hours.
Communication breakdowns occurred in various contexts. Key of which were face-to-face meetings between clients and officials. Participants frequently described many meetings where the officials, for various reasons, did not provide them with adequate information. This was mostly evident in relatively complex cases where expert knowledge was required, the lack of which resulted in administrative errors and extended waiting times. Note, however, that not everyone experienced this attitude, certainly not all the time. Many even preferred face-to-face encounters to using digital communication. Such encounters made them feel certain about the information received or provided: ‘No, no… absolutely not, I go there and do everything manually. Direct contact. And even if I have inquiries, that’s the way I prefer’ (Seerin, forty-nine, f, IS).
Many participants noted the impossibility of contacting officials directly and immediately over the phone. For these participants, frustration and feelings of helplessness were the result of this unilateral communication breakdown. When the option for direct phone calls did exist, usually due to a private decision by caseworkers, this was perceived by the participants as a significant advantage: ‘I remember when the rehabilitation worker gave me her phone number. I wrote it down with four exclamation marks in my mobile, ‘cause I was so excited to finally have somebody to talk to’ (Moshe, twenty-six, m, GDA).
In recent years, the NII has established call centres to improve its communication with clients. Our participants were divided with regard to their benefits. Some considered them a convenient solution, enabling them to gain some control over the communication process: ‘I used them several times and they were wonderful’. Many others criticised them, however, particularly their outsourcing. Contracting out these services meant that their employees were not NII officials and therefore often did not have the authority to provide relevant information. All of this led to administrative errors and long-term waiting, and provided claimants with little control of the communication process.
Another cause of communication breakdowns leading to learning costs was outdated ICT. Many participants considered absurd the fact that ‘we’re in the laser age, but those running us are still the horse and carriage’ (Ron). Some even considered that evidence of the reluctance of the NII to see their rights taken up: ‘It makes no sense that they’re using the fax and regular mail. It feels like someone has an interest in things arriving more slowly and being more complicated’ (Maayan).
Indeed, a prominent example of outdated technology was the use of faxes, often the only way to pass information on to the NII. This outdated technology was problematic because most participants simply did not have a fax machine and because, again, it was unidirectional, increasing their uncertainty and taking the control out of their hands, being unable to ascertain whether their message had been received, what actions were taken, and when they would receive an answer. All of these exposed participants to burdensome consequences such as long waiting times and administrative errors.
In light of this criticism, the NII recently began upgrading its ICT systems to improve communication, enhance clients’ service experience, and above all promote take-up. Many participants were aware of and welcomed these efforts. A notable example was the upgrading of the NII website to enable them not only to receive generic information on eligibility, but also to convey and receive concrete, personal information, as well as to submit applications. Other examples were the use of emails and text messages. These technologies succeeded, among other things, in doing away with the cumbersome procedures of sending and receiving letters or arriving personally at the branches.
While recognising the benefits of new technologies, many participants criticised their implementation. A key criticism was that they often constituted a single, default channel, making life difficult for the technologically illiterate. David (66, m, IS), for example, admitted to ‘not managing with the mobile phone. And people… like you, tak-tak-tak… they click and things work out. I’m really ancient, primitive’.
Importantly, such lack of skills is not only an individual characteristic. A prime example is the Jewish ultraorthodox community, whose members often do not use the internet. Noga described it as follows: ‘I got most of the information from the website. But many people in our community don’t have internet at all.… For them it’s more difficult’.
The difficulty with a single, default channel was not only due to lack of digital skills. Some participants, despite having the right skills, noted the need for personal contact and the fact that these new technologies could not meet all their information needs. For example, when Noga was asked for advice about how to improve the NII’s service, she noted the difficulty in understanding the forms and the claiming process, and added that ‘maybe it would be good to have an official who will also talk, I mean […] not just a website’.
Finally, another context for communication breakdowns was forms and documents. For most participants, forms were presented in their mother’s tongue. However, for Arabic-speaking participants, Hebrew forms were sometimes experienced as communication breakdown leading to a learning burden. This burden was mostly felt in sections that required explanation or clarification:
There is a certain type of reading that we must master when completing the forms, and in my opinion, it is recommended that the explanations also be translated. I do not mean the form with the ID and the name… we all know that; I mean the explanations here and there – that’s something that must be accessible to all populations. (Rashida, twenty-four, GDA claimant).
‘They simply have this method’: administrative errors
Often, administrative errors caused the emergence of compliance costs, as participants were required to resubmit documents several times, either because they had provided incomplete information or because the NII officials had simply lost them:
The paperwork was simply insane […]. I […] obtained the final approval after about ten months. Each time it felt like they were inventing a new document that I had to submit and wasn’t told about in advance. So, I got all the money retroactively. It was one of the most frustrating […] bureaucratic processes (Maayan, twenty-seven, f, IS).
Importantly, even when these errors were the result of incomplete information provided by the applicants, the participants’ perceived them to be the NII’s responsibility. In particular, experiencing administrative errors was perceived by the participants to be highly connected to their difficulties in communicating with the NII about their error. These participants had to spend considerable time and effort to learn that an error had occurred, to notify the NII and to receive feedback about the measure taken to correct it:
I waited two years for retroactive approval of the funds due to one document I was not told about and not informed it was missing. […] I had to know it myself. How?… I didn’t receive a letter or a phone call […]. Until I found someone working in National Insurance who was my friend’s sister in-law and I received her phone number at work… And she could look for me and see, and tell me what was missing. That’s how I found out (Heli).
Documents were not the only context of administrative errors. One participant (Roslan, GDA claimant, sixty-three) for example, suffered the consequences of an error where his GDA claim resulted in only partial approval. This caused a significant loss of financial resources and services. He claimed GDA again after a few years and received the full payment. Tina, thirty-two, a GDA claimant, encountered an error in another stage of the benefit-claiming process. Interestingly, in her case the claim was fully approved, but in some months, she received less than what she deserved or nothing at all: ‘Sometimes they write to me, ‘You don’t deserve it. You make 20,000 NIS (4000 pounds) a month’. When did I make NIS 20,000 a month? ‘Our bad, we got confused and such’. But I need to go there to correct these errors.’
Accordingly, many participants experienced the NII and its officials as disorganised, chaotic and error prone. Paradoxically, this kind of routine disarray chaos was seen as consistent and even predictable and systematic. In other words, it was experienced as some kind of organised chaos:
They simply have this method. […] you bring the documents […] and you call after two or three days just to make sure they got them. And then they send you this text message. Usually it’s “the documents have arrived”. After a week or two, you expect some kind of development, something. They tell you, “No, this particular document is missing” (Ron).
The disarray was reflected in various aspects, from the physical space of the NII branches through the loss of documents to disinformation. In every case, the burden lay with the claimants, who needed to ‘solve the puzzle’ by themselves, adding to the already high costs of the process: ‘Even the front-desk workers don’t really know what exactly is required. This one has some of the info, and the other has another piece. So the client has to kind of solve a puzzle out of these pieces’ (Rita, forty-six, f, IS).
Discussion
Drawing on the Israeli social security case, the aim of this study was to further conceptualise administrative burden and its underlying triggering mechanisms. Our inductive analysis has unfolded three underexplored concrete elements that trigger administrative costs: waiting, communication breakdowns and administrative errors.
Waiting for the state
Time is a constitutive dimension of modern, liberal states (Cohen, Reference Cohen2018). The social policies they provide are highly structured upon an architecture of measured time, whether the need to reach certain age or accruing a qualifying period. Similarly, these social policies have a key role in (re)distributing work, care and leisure time, as well as in shaping people’s autonomy over their use of time (Burgoon and Baxandall, Reference Burgoon and Baxandall2004; Goodin et al., Reference Goodin, Parpo and Kangas2004; Lahat and Sened, Reference Lahat and Sened2020). Our study joins other bodies of knowledge in highlighting the importance of time and waiting also in the administration of these rights. In particular, according to our participants’ accounts, ‘waiting for the state’ (Carswell et al., Reference Carswell, Chambers and De Neve2019) has been found to be a crucial component of compliance, learning and psychological costs, in the simple sense that their encounters with state agency have been often time consuming.
Waiting can take different forms. It can be chronic, spanning a long period. In our case, the participants waited for months for feedback from the NII and, even then, many were asked by officials to produce new documents or ones they had already submitted. Two more short-term kinds of waiting evident in our case study are ‘on-the-day’ and ‘to-and-fro’ waiting (Carswell et al., Reference Carswell, Chambers and De Neve2019).
While waiting for the state can be experienced differently in different social contexts (Seefeldt, Reference Seefeldt2017), the experience is shaped by various factors other than location. Our findings suggest, for example, that another key factor in our participants’ waiting experience was their uncertainty. As also shown in studies on consumer and e-consumer behaviour (Ryan and Valverde, Reference Ryan and Valverde2006), some claimants were not only frustrated by waiting per se, but also by the fact that it involved uncertainty generated by poor communication.
Our findings also illustrate that as an ‘exercise of power’ (Turnbull, Reference Turnbull2016), waiting is often not distributed equally (Cohen, Reference Cohen2018). In our case, both GDA and IS are targeted benefits aimed to provide supplementary income to people who are relatively poor and marginalised, making the latter an easy target for being ‘patients of the state’ (Auyero, Reference Auyero2012). While more comparative case studies are needed to fully grasp such ‘temporal domination’ mechanisms (Reid, Reference Reid2013), our findings suggest that they play a crucial role in consituting adminstrative burden.
Communication breakdowns
Another decisive trigger is breakdowns in state-citizens communication that involve, not just general information, but crucially, personalised information enabling citizens to comprehend their specific circumstances. Our findings also indicate the two-sided nature of (good) citizen-state communication, as our participants often felt frustrated by lacking control of how, when and with whom they could communicate. This final point implies that communication breakdowns might trigger not only learning and compliance costs but also psychological ones, with the latter involving issues around autonomy loss during encounters with state authorities (Moynihan et al., Reference Moynihan, Herd and Harvey2015).
The risk of communication breakdown underscores the importance of ICT. The few studies exploring this aspect from the perspective of welfare clients have shown mixed results. This was also evident in our participants’ accounts. On the one hand, many pointed to the low-tech information technology used by the NII. Moreover, some praised the new technologies recently deployed by the NII for making it easier for them to control the communication, receive information, submit documents and claims, receive regular updates, and contact front-desk workers.
Conversely, this ‘electronic turn’ (Garrett, Reference Garrett2005) can also produce additional administrative burden (Herd and Moynihan, Reference Herd and Moynihan2019) – for example, by shifting responsibilities that in the pre-digital era lay with the bureaucrats. Further, when being the sole channel of communication, the penetration of ICT also means higher frustration for having to communicate, often unsuccessfully, with chatbots rather than with human beings (Bozeman and Youtie, Reference Bozeman and Youtie2020). This rejection of impersonal interactions (Hetling et al., Reference Hetling, Watson and Horgan2014) is especially evident when non-routine, complicated services are involved (Breit and Salomon, Reference Breit and Salomon2015; Buffat, Reference Buffat2015; Madsen et al., Reference Madsen, Lindgren and Melin2022).
Finally, a ‘cyberoptimist’ (Pollitt, Reference Pollitt2011) understanding of technology overlooks the fact that new technologies can reinforce pre-existing hierarchies, as specific people and groups struggle to make effective use of ICT (Sinclair and Bramley, Reference Sinclair and Bramley2011; Hansen et al., Reference Hansen, Lundberg and Syltevik2018; Madsen et al., Reference Madsen, Lindgren and Melin2022). This reminds us that adminstraive burden is also highly contingent upon people’s human and administrative capital (Christensen et al., Reference Christensen, Aarøe, Baekgaard, Herd and Moynihan2020), including their digital competencies. Importantly, inadequate technological skills are particularly evident in specific, vulnerable groups such as ultraorthodox Jews and elderly clients. All of this echoes the Catch 22 embedded too often in the administrative burden experience, in which the very factors that increase citizens’ need for social security, such as poverty or ill-health, also undermine their ability to cope with burdensome state actions (Christensen et al., Reference Christensen, Aarøe, Baekgaard, Herd and Moynihan2020).
Administrative errors
As recently identified by Widlak and Peeters (Reference Widlak and Peeters2020), administrative errors were a common trigger for our participants’ administrative burden. Interestingly, what stood out in our case were not only rare and critical errors, but more frequent types of errors, particularly what Krumer-Nevo and Barak (Reference Krumer-Nevo and Barak2006) call the ‘missing document syndrome’. Moreover, in line with Widlak and Peeters (Reference Widlak and Peeters2020), our case study shows that such burdensome consequences are often not only the result of the errors themselves, but often mainly the result of the tremendous efforts to correct them. Such a process, which requires learning what went wrong and how to solve it, is often hampered by communication breakdowns, leading to learning and psychological costs.
Interestingly, although Widlak and Peeters (Reference Widlak and Peeters2020) perceive administrative errors as informal and unintentional, the fact that they are common and due to the NII’s organisational incapacity has led some of our participants to experience them as an organised disarray. This experience echoes Herd and Moynihan’s (Reference Herd and Moynihan2019) idea of administrative burden as a ‘policymaking by other means’. Administrative burden in this sense is viewed as intentional, rooted in the political administrative system, and often aimed at restricting the use of state’s benefits and services. This also suggests that to understand administrative burden and its organisational causes, we need to take into account the subjective meaning citizens give to them (Seefeldt, Reference Seefeldt2017; Chudnovsky and Peeters, Reference Chudnovsky and Peeters2020), including their causal explanations (Barnes and Henly, Reference Barnes and Henly2018).
The multifaceted and context-dependent nature of administrative burden triggers
The discussion so far encompasses two important insights that cut across our inquiry and can contribute to the understanding of administrative burden. The first is the multifaceted nature of administrative burden and the elements that trigger it. For one, these elements often spill over from one administrative cost to another, often causing more than one type of burden. While administrative errors, for example, trigger psychological costs such as frustration and distrust in the system, they can also lead to additional administrative processes hence also triggering compliance and learning costs. More importantly, administrative burden is often the result of the interaction between these elements. Conversely, the presence of one element rarely causes burden in itself. The experience of waiting is a case in point. As shown above, this experience is shaped, inter alia, by the claimants’ uncertainty regarding the process. In turn, this uncertainty is generated, or at least enhanced, due to poor communication, leaving our claimants wondering how long they need to wait and whether their application is going the right way. In addition, knowing that administrative errors are always around the corner makes this waiting experience only more onerous.
Secondly, these elements are often context-dependent. This is apparent in this research in that triggers and the burden they cause sometimes varied depending on cultural aspects. An illustrative example was the differential impact of ICT on various Israeli sectors. Another example was the language of communication, as Arab citizens experienced the exclusive use of Hebrew as a site of potential communication breakdown. Further research is needed to deepen our understanding of the cultural aspects of administrative burden, including the deliberate, discriminative nature of such burden, the extent to which it is embedded into the program design, and its differential impact on citizens (Ray et al., Reference Ray, Herd and Moynihan2020; Mallinen, Reference Mallinen2021).
Combined, these two insights are particularly important when addressing the need to create a valid and agreed-upon set of measurements of how people experience administrative burden in different settings (Baekgaard and Tankink, Reference Baekgaard and Tankink2022). Developing these measures, for either scholarly or auditing purposes (Sunstein, Reference Sunstein2020), should take into account both the multifaceted nature of administrative burdens and triggering elements, which can create a complex environment in which they are hard to distinguish, and contextual factors that often operate together and change the experience of burdens. Relatedly, we see importance in measuring both these potential triggering elements as well as the overall (compliance, learning and psychological) burdens as experienced by claimants. This will allow us to better understand how these elements operate to cause administrative burden across different populations, seeing that inequalities are inherent to administrative burdens (see e.g. Chudnovsky and Peeters, Reference Chudnovsky and Peeters2020).
Limitations and implications for research and practice
Before presenting more concrete limitations, we should not lose sight of the specific context of our inquiry: encounters with welfare bureaucrats in the context of claiming social security benefits involves citizen proactively interacting with the state; limited face-to-face interactions (often complemented by digital interactions or call centres); and vulnerable clients. These interactions occur in ‘decision-making factories’ (Bovens and Zouridis, Reference Bovens and Zouridis2002), with standardised procedures and mass decision-making processes. Since the experience of administrative burden is highly context-specific, any effort to generalise from our findings should take these features into consideration.
Another related limitation derives from our use of convenience and snowball sampling, and specific tactics used to recruit participants such as social media groups and direct contact with advocacy organisations – all potentially leading to sample bias. While measures have been taken to increase the inclusiveness of the sample, note that this limits the generalisability of our findings. Similarly, the study did not address the perceptions of claimants who were eligible but did not complete the take-up process, who may have experienced administrative burden to an even greater extent.
Our analysis has some key implications for policymakers, who do wish to move from burden-enhancing to burden-reduction policy. First waiting times should be addressed by reducing both short-term waiting (e.g. by investing in the infrastructure of the system) and long-term waiting times (e.g. by providing immediate feedback on application status using ICT). Second, information dissemination should be improved, emphasising the importance of high-quality information, customised, case-specific information, and bilateral communication, increasing feedback between the claimant and state. Third, different channels of communication should be established to allow for a holistic response to fit the manifold needs of claimants from different backgrounds, Fourth, organisational mechanisms are needed aiming both to limit administrative errors and to communicate with claimants about these errors and enable them to easily identify and address them. Finally, efforts to reduce burden need to recognise the complex and often subtle interplay between administrative burden costs and their triggers. While this aspect means that costs sometimes create other costs, it also means that when actively intervening to reduce specific costs by addressing triggers, other triggers and costs could also be reduced.
Acknowledgements
This work was supported by Israeli National Insurance Institute [Grant number 0323807].
Open access
