Introduction
Political trust increases when citizens have positive opinions about a state’s success in ensuring the fair distribution of public goods. There is a good deal of scholarship to explain the causal, though not unconditional, relationship between perceived distributive justice and trust in public servants, public institutions, and the government (Kumlin, Reference Kumlin2002; van Ryzin Reference Van Ryzin2011; Zmerli and Castillo, Reference Zmerli and Castillo2015; Berg and Dahl, Reference Berg and Dahl2020; Schnaudt et al., Reference Schnaudt, Hahn and Heppner2021; Zhang et al., Reference Zhang, Li and Yang2022). Numerous experimental studies and opinion surveys explain how people perceive inequalities, which justice principles people adhere to when evaluating those inequalities, and how these perceptions impact political trust. Much less is known about how people actually evaluate the distributive justice implemented by public institutions – that is, how they assess who in practice receives what from the state. Relevant knowledge on these issues, which often indirectly addresses the problem of perceived distributive justice, comes from scattered studies on beneficiaries’ welfare experiences and deservingness clues (Fersch, Reference Fersch2016; Nothdurfter and Hermans, Reference Nothdurfter and Hermans2018) or psychological analyses on processing information on redistribution (e.g. Lind, Reference Lind2001). As pointed by Berg and Dahl, Reference Berg and Dahl2020 and Zhang and coauthors (Reference Zhang, Li and Yang2022), there is an even greater paucity of knowledge on the role of various institutional embeddings – that is formal structures and processes taking place within them (Edelenbos et al., Reference Edelenbos, Klok and Van Tatenhove2009) in shaping the link between perceived distributive justice and political trust.
This article seeks to contribute to filling these two gaps by, first, deepening our understanding of justice principles and assessments, and the relationship between perceived justice and trust, as experienced and narrated by users of social assistance. Second, it shows how this relationship operates in a specific institutional embedding of highly complex and self-contradictory policy assemblages (Newman and Clarke, Reference Newman and Clarke2009; Clarke and Bainton, Reference Clarke and Bainton2015) of social assistance in two post-socialist countries, the Czech Republic and Poland. Analysis of in-depth interviews with parents receiving social assistance in these countries showed us that this particular context is perceived by beneficiaries as what we call in this article an ‘institutional enigma’. This enigma refers to the perception of a policy assemblage that includes the users’ assertion that the (welfare) system is illogical or difficult to understand, that it sends mixed messages and often applies contradictory solutions. This ‘enigma’ is also to some extent unpredictable, opaque to users, as they suspect hidden internal administrative tensions relevant to service delivery, which in their view result in a certain degree of bureaucratic burden. As explained below, such a complex and self-contradictory policy assemblage of social assistance in the Czech Republic and Poland is parallel to (and partly results from) the coexistence of highly contradictory welfare discourses due to the multi-layered welfare legacy of the countries under study (Saxonberg and Sirovatka Reference Saxonberg and Sirovatka2006, Reference Saxonberg and Sirovátka2018; Lendvai-Bainton and Stubbs, Reference Lendvai-Bainton and Stubbs2021).
We expect this type of institutional context to become (increasingly) relevant in different countries and policy areas. Similar ones are already pointed out, for example, in labour activation policies or migration and migrant integration policies (e.g. Lister et al., Reference Lister, Smith, Middleton and Cox2003; Dubois, Reference Dubois2014) and are also discussed, for example, in the literature on administrative burden (e.g. Halling and Baekgaard, Reference Halling and Baekgaard2024 for a literature review).
In the following part of this article, we present the theoretical underpinnings of how incoherent institutional embedding can be relevant for the link between distributive justice and trust. We then show how social assistance in the Czech Republic and Poland meets the criteria of incoherent policy embedding. The next section presents the methodology of our qualitative research. This is followed by the presentation of findings based on interviews and conclusions on the justice-trust link in the context of ‘institutional enigma’.
Distributive justice criteria, perceptions of redistribution, and the justice-trust link in the complex context
The context, including the institutional embedding, matters a great deal for the distributive justice criteria people adhere to. Equity is found to be particularly relevant in work-related contexts, while equality is more pronounced in welfare and political domains and need comes to the fore in informal family settings (Tyler, Reference Tyler and Sarat2004: 436). Research has shown that context is relevant not only for the choice of justice criteria to evaluate (re)distribution, but also for the way people apply them. They prove that justice principles operate in a highly nuanced pattern. In particular, qualitative contributions on welfare users’ lived experiences demonstrate that, when people are asked about what would constitute distributive justice in their welfare system, they discuss nuanced and comprehensive criteria. Willen and Cook (Reference Willen, Cook and Thomas2016) summarise the specific distributive justice criteria of deservingness, arguing that they are relational and conditional – that is target- and situation-specific; syncretic, as they are simultaneously grounded in various moral insights; infused with affect; and mutable. Thanks to work such as Lavee’s (Reference Lendvai-Bainton and Stubbs2021), deservingness criteria are understood to function in a multilayered pattern. In her study, Israeli welfare beneficiaries’ views on the fairness of the welfare state combine the ‘new’ values of market citizenship with the ‘old’ ethics of state responsibility. The latter permeate the former, resulting in a hybrid perception of entitlement: the assertion that people should only be entitled to help if they are active, but it is the state’s responsibility to enable them to do so.
The contextual factors are also proven to impact the perception of redistribution. This is relevant, because for distributive justice evaluations to be made, people need to apply more than justice criteria, such as an assessment of who actually gets what. From the literature we know that when making these assessments, people compare not only what they receive to what others get, but also compare different groups of other people. Both types of judgements – that is egocentric justice concerns (Kluegel and Mason, Reference Kluegel and Mason2004) and sociotropic ones (Zmerli and Castillo, Reference Zmerli and Castillo2015; Bobzien, Reference Bobzien2023) are found to impact political trust. However, our knowledge is scarce regarding which groups people actually zoom in on when assessing the effects of redistribution; however, we know, for example, that public discourses pointing to ‘undeserving’ groups play an important role (Ennser-Jedenastik, Reference Ennser-Jedenastik2018; Busemeyer et al., Reference Busemeyer, Rathgeb and Sahm2022; Halling and Baekgaard, Reference Halling and Baekgaard2024).
Moreover, the psychology literature informs us that in a heterogenous policy context, where people face mixed messages about how public institutions work, evaluations of distributive justice are far from rational and rarely involve detailed analysis. Instead of analytically juxtaposing the norm and perceived redistribution outcomes, people tend to rely on all available justice-relevant information: those about the procedures of granting or receiving benefits, as well as about individual and general outcomes of those procedures. Theory suggests the judgement process is rather rapid, automatic, and holistic (Greenberg, Reference Greenberg2001; Ambrose and Arnaud, Reference Ambrose, Arnaud, Greenberg and Colquitt2005) and, as such, distributive justice is, in practice, evaluated through the prism of procedural justice.
The impact of perceived redistributive justice on political trust also depends on contextual factors (Berg and Dahl, Reference Berg and Dahl2020; Zhang et al., Reference Zhang, Li and Yang2022). Overall, the literature discusses three interrelated mechanisms relevant to welfare policy that link perceived injustice and (dis)trust. They are: (a) people’s experience of material deprivation as a result of not receiving benefits or services; (b) feelings of insecurity; and (c) concerns that the state is unresponsive and/or incapable of doing what it should (Bobzien, Reference Bobzien2023). However, the salience of these mechanisms varies according to their institutional embedding. For example, means-tested policy design, typical of social assistance, is found to be detrimental to the emergence of political trust, not only because of increasing perceived distributive injustice, but also because of its negative effect on social trust (Kumlin and Rothstein, Reference Kumlin and Rothstein2005; Betkó et al., Reference Betkó, Spierings, Gesthuizen and Scheepers2022). Low social trust, reflected in a sense of mistrust towards others (welfare recipients), is in turn found to be reciprocally related to low political trust, which is also the case in post-socialist countries such as the Czech Republic and Poland (Bargsted et al., Reference Bargsted, Ortiz, Cáceres and Somma2023).
The heterogenous context of policy assemblage conveying mixed messages to citizens renders the justice-trust relationship even more entangled. First, policy feedback literature, and in particular the findings on when that effect is absent (Patashnik and Zelizer, Reference Patashnik and Zelizer2009; Dupuy et al., Reference Dupuy, Teuber and Van Ingelgom2022), shows that the effects of policy on people’s attitudes and behaviours, which should include political trust, may not appear at all in a context riddled with self-contradictory and opaque solutions, poorly visible policies, and cacophonic policy discourses. Second, when it is difficult to assess whether the policy promotes distributive justice or not, the reverse mechanisms of perceived justice-trust should be more prominent. It is already well established in the literature that this relationship is in fact bidirectional – not only does the contention that a system that promotes justice increase political trust, but citizens’ trust in political institutions also causes them to perceive those institutions as working fairly (Abdelzadeh et al., Reference Abdelzadeh and Ekman2015; Schnaudt et al., Reference Schnaudt, Hahn and Heppner2021: 5). Third, in a similar fashion, the literature shows that both education and class play a strong role in stabilising perceptions of inequalities, partly irrespective of their actual level (Foster and Frieden, Reference Foster and Frieden2017; Bobzien, Reference Bobzien2023). This suggests that we still do not know whether factual experiences with public institutions may change perceived justice and, potentially, trust – and if so, how.
The presented theoretical underpinnings help us to anchor the research problem, which addresses the following conundrum. On the one hand, a small “fairness gap” (Bobzien, Reference Bobzien2023), i.e. the coherence between people’s preferred and perceived redistribution, has been shown to have a positive effect on political trust. On the other hand, there are institutional contexts, such as social assistance policy assemblages of post-socialist countries, where it seems very difficult for users to assess fairness and thus to develop trust in welfare institutions. Consequently, we are trying to understand how people in these contexts make judgements about justice and how their trust in welfare can be fostered.
Mixed policy assemblages and discourses cacophony: social assistance in Czechia and Poland
The socialist legacy of Czechia and Poland, together with the neoliberal changes that have followed in both countries, has resulted in rudimentary/neoliberal social assistance policies (Saxonberg and Sirovatka, Reference Saxonberg and Sirovatka2006; Lendvai-Bainton and Stubbs, Reference Lendvai-Bainton and Stubbs2021). Low replacement rates of social assistance benefits, a failure to automatically adjust benefits, and social assistance that has fallen short in reducing poverty are all hallmarks of both systems. However, both are also classified as following a ‘European’ social model path (more than their CEE counterparts), with relatively high social spending, broad coverage of social protection, and ongoing political support for welfare (Lendvai-Bainton and Stubbs Reference Lendvai-Bainton and Stubbs2021: 209). The social assistance systems of both also meet the criteria of policy assemblages (Newman and Clarke, Reference Newman and Clarke2009) shaped within context-specific processes of bringing together and (re)combining heterogenous elements (Clarke and Bainton, Reference Lendvai-Bainton and Stubbs2015). They include Bismarckian social insurance schemes, flat-rate benefits, and strong conservative principles in family policies (Saxonberg and Sirovatka, Reference Saxonberg and Sirovatka2006).
A social assistance pattern of austerity remained unchanged in Poland even during the COVID pandemic (Aidukaite et al., Reference Aidukaite, Saxonberg, Szelewa and Szikra2021). In terms of governing minimum income schemes, Marchal and Cantillon (Reference Marchal and Cantillon2022) place Czechia and Poland in the same cluster (Marchal and Cantillon, Reference Marchal and Cantillon2022: 136) where eligibility conditions and benefit levels are centralised at the national level while municipalities, running social assistance centres (in Poland) and labour offices (in Czechia), remain (partially) responsible for the financing of social assistance benefits (Kazepov, Reference Kazepov2010). Marchal and Cantillon also show that Czechia and Poland each have a low share of population using social assistance (close to 2 per cent, according to OECD data) while displaying high (70-80 per cent) within-country variation in the level of benefits obtained by couples with children (Marchal and Cantillon, Reference Marchal and Cantillon2022: 142).
In terms of recent changes to their policy assemblages, Poland and Czechia display growing differences. In Czechia, the strong presence of a work-enabling dimension and workfare dimension in minimum income schemes (Natili, Reference Natili2020) and relatively higher replacement than Poland (39 per cent vs. 23 per cent, OECD, 2022) suggests that the Czech model leans more toward the workfare/protective type. In Poland, on the one hand, the minimum income scheme is closer to protective and workfare measures (Soler-Buades, Reference Soler-Buades2024). However, austere social assistance in Poland coexists with very generous, universal family benefits (the 500+ programme). At the same time, the social assistance system has been strongly shifted towards community centres that provide integrated social services (Karwacki and Rymsza, Reference Karwacki and Rymsza2025).
Both countries show medium levels of social trust (24.1 per cent of people in Poland believe that most people can be trusted, while 27.3 per cent of Czechs say the same) (Integrated Values Surveys, 2022) and relatively low levels of political trust, with 34.2 per cent of Poles and 34.1 per cent of Czechs trusting their government (OECD, 2022). The cacophony of welfare discourses is reflected in low levels of support for universal, means-tested, and activation programmes. However, the Czech Republic stands out as having the lowest overall support for redistribution in Europe (ESS, 2020).
Research methods
The comparative logic of our analysis is based on the idea of analysing the perceptions of distributive justice and its relation to political trust in two very similar national and policy contexts marked by incoherent policy assemblages, a relatively low level of social and political trust and self-contradictory and relatively strongly anti-welfare attitudes of two different countries. Although, as noted, the structure of policy assemblages in the fields of social assistance in Czechia and Poland bears some differences, they have a similarly confusing impact on welfare users, who are thus prompted to be suspicious of state institutions and other welfare claimants.
Our findings are based on the analysis of thirty-three individual in-depth interviews conducted with social assistance users in Poland and Czechia, between March 2020 and February 2021 within an Entrust research project. Our discussion partners were welfare users living in large cities. All were parents who had had encounters with social assistance frontline workers due to their having received social welfare. The Czech sample consists of fourteen women and one man; and the Polish of twelve women and six men. In both groups about half of interviewees have used social assistance for at least seven years. The main reasons they sought social assistance included living in poverty; serious housing difficulties (a few interviewees had experienced homelessness or housing insecurity e.g. due to living in hostels, including ones for victims of domestic violence); single parenthood, including being a parent of children with disabilities that need constant care; parenting issues, including experiences of children being removed to foster care; addiction; and domestic violence.
Research participants’ recruitment was based on their availability and was handled by means of local and social media advertisements and elements of snowballing. Detailed anonymisation procedures were applied during recruitment, handling of data, and in quoting the interviewees. The names used herein are aliases. The research teams in both venues obtained permission from their respective universities’ ethics commissions to conduct the research.Footnote 1
During the interviews, participants were asked detailed questions about their experiences of applying for social welfare, (in)justice assessments, and trust. The analysis presented in the next section is a qualitative content analysis of the interview transcripts in the national languages, following Saldaña’s (Reference Saldaña2021) approach. In most cases, the reflection on users’ perceptions of distributive justice and trust in the welfare system were presented in response to the questions on perception-changing moments (‘Can you remember any moments that made you trust or distrust social assistance?’, ‘Were there any moments when your trust was broken? How did this happen?’).
By ‘trust’, we mean political trust in a welfare system, as part of the public administration of countries. However, in a way that is typical of multi-level perceptions of trust (Schneider et al., Reference Schneider, Theiss, Gędek and Zschache2025), our respondents often spoke of ‘(not) trusting them’, bringing together the system and frontline workers. It needs to be underlined that the notion of ‘them’ is very different from (not) trusting a specific frontline worker, which represents personal trust and is beyond the scope of our analysis. In terms of distributive justice and its evaluations, we coded those parts of the text that represented the idea of ‘who should get what’ and ‘who gets what’.
Thus, we inductively coded those parts of the interviews where a notion of distributive justice was present, looking for instances where the relation to trust was made directly by the interviewee. There were not many cases where this relationship was made within a sentence, but the quotation of Rafał’s speech (as follows) is a clear example of how this relationship could be explicitly narrated by interviewees. Accordingly, we present the relationships between perceived justice and trust from the emic perspective of welfare users (Dubois, Reference Dubois2020), acknowledging the described problem of a de facto bidirectional relationship between perceived justice and trust, and users’ large number of diverse policy experiences (Dupuy et. al, Reference Dupuy, Teuber and Van Ingelgom2022).
Data analysis focused on answering three overarching research questions:
-
1) In the context of inconsistent and mixed-message policies, what distributive justice principles do social assistance users use to evaluate social assistance, and how do they use these principles?
-
2) How do social assistance users assess who actually receives what from social assistance – who do they compare and how when making justice assessments?
-
3) In a given context, how exactly do these assessments of the fairness or unfairness of the social assistance system translate into trust and mistrust in the system?
In the next section, we present the four combinations of (in)justice perceptions and trust in social assistance that we reconstructed from the data.
Findings
Perceiving the context of institutional enigma
The overall perceptions of social assistance ranged widely in both Czechia and Poland – from opinions, on the one hand, that the system works very well and is helpful to stances that it is a complete failure. The general assessment that social assistance is a complex and incoherent institutional setting was very salient in the interviews. Even in narratives among interviewees who were satisfied with social assistance, there was a striking tension. On the one hand, they emphasised highly formal, precise, and mundane procedures for applying for some benefits or services, on the other hand they had experiences of high discretion of frontline workers and random granting some other forms of help. The logic of procedures was very often described as ‘messy’ or ‘incomprehensible’, with interviewees in both countries repeatedly saying something like ‘I just don’t get how it works’. Some interviewees said that frontline workers ‘have their tricks’, whilst others felt that frontline workers are overall nice and helpful people but ‘the system’, including managers, demands they work ‘like that’. The perception of the system being opaque was compounded by the general conviction that one can be through all the procedures only if the frontline worker assigned is good. Frontline workers’ role was most often characterised as that of leading others through the system, though there was also conviction that you never know whom you will get. Though infrequent, there were a few salient references praising communist times. They referred not to the ideal of equality, but to that of very clear rules of who gets what, even if interviewees emphasised that those rules had in the past been breached. Overall, the participants’ perception of the welfare system, which we term ‘institutional enigma’, was based on experiences and reflections that the system was inscrutable, unpredictable, and unknowable.
Trust combined with perceived system’s rationality
The individuals we interviewed believed that helping those in need should be the main principle according to which a social welfare system operates. The interviewees underlined that a social assistance system is meant to help people in need. They were thus evaluating that system by judging how it actually fulfils this goal in general and whether it helped them when they themselves were in need. Overall, they believed that social assistance institutions properly served the goal of helping people in need, as expressed by one individual from Czechia:
Our social system is really the best, compared to other countries. When you need help, the system helps you so you don´t end up being a beggar. (Lucia, CZ)
More specifically, the criterion of whether social assistance properly targets people in need was discussed mostly in relation to procedures of eligibility testing. Our interviewees largely believed that eligibility- and means-testing are fair overall, as social assistance has to set clear boundaries between those who are in need and who are not. The conviction was also expressed that those truly in need have to comply with the rules not only by accepting eligibility checks, but also by appropriately spending welfare benefits. Thus, the stance prevailed that fair targeting of welfare entails conditionality and the withdrawal of assistance if beneficiaries fail to cooperate.
In addition, some believed an individual should be able to prove their need. They argued that since the welfare system is open and transparent, so too should individuals be by undergoing eligibility testing. The typical expression of the logic was that ‘it suffices not to cheat’, ‘keep documents in order’, ‘follow the rules’, ‘do what has been agreed’ to receive what one should because those truly in need have nothing to hide:
So, if you are a person who has things [the documents needed for financial help] organised and all right and you show that you pay rent and utilities, there´s no problem. The system is set up that way. If you have everything in order, you show it and the system works. (Denisa, CZ)
The goal of serving the needy is one pillar of the welfare system’s rationality. Others include, first, eligibility tests meant to distinguish between those truly in need from those who are not, and second, beneficiaries facing some degree of struggle. As Denisa said (see above), ‘the system is set up this way’ and that ‘the system works’, the notions of institutional rationality and predictability are visible. That perspective was connected to users’ trust in the welfare system. A meaningful example of the connection is clear in the following observations from a Polish beneficiary:
Q: Do you trust social assistance?
A: I’d say I have no reason not to trust a worker because I submit an application, I know what I write is true, and I know what I get, so I get it. Because they have their rigid calculations (…) With permanent benefit, they calculate it (…) and with temporary benefit it depends on how much money they have and how many people apply (…) I’m interested in this stuff and I know how it works. (Rafał, PL)
Rafał links his stance of granting moderate trust, or, as he puts it, ‘having no reasons not to trust’ the system’s predictability and its rational mechanisms. Rafał understands that ‘they have their procedures’ and repeats that he knows well how the system works. This example of experienced distributive justice is thus linked to the perceiving of institutional procedures as rational and understandable. Rafał describes the welfare offices’ practice of arbitrary deciding on the amount of temporary benefits as depending on ‘how much money they have and how many people apply’, and he sees it as a solid and legitimate procedure. This, he believes, justifies his argument ‘not to not trust’ frontline workers.
Distrust combined with perceived abuse of power and lack of sympathy
As noted, our interviewees strongly believed that the principle of need is a primary purpose of the social assistance system and accepted means-testing procedures. However, they recognised the denial of help when the eligibility threshold is exceeded by a few pennies as a sheer sign of distributive injustice. For many welfare users that rule of denial was clear proof that the social assistance system fails to function according to the principle of helping people in need. This is because, as our interviewees argued, a person claiming welfare benefits whose income is just slightly above the eligibility limit is still very much in need.
Similar examples of injustice were cited, including institutions’ denial to, first, classify an individual as living in poverty by arguing that s/he has family members who can be asked to provide help, or, second, to confirm someone has a disability and thus also special needs. Although this argument appears more related to procedures than outcomes, eligibility testing was perceived by interviewees as a cornerstone of distributive justice, because it is decisive of whether those who should be granted support for actual need do, in fact, receive it. Perceptions of the system being unjust because it has regarded citizens in need as ineligible for help referred not only to financial means-tests but also need assessments and expectations that those receiving help should participate in activation measures.
This is a typical fashion in which the injustice of the system not recognising factual need was portrayed:
The social worker denied us a new car for me to drive around my disabled son. They[the case workers] said that he is not disabled enough for a car to be provided. (Jana, CZ)
Jana’s expression that her son with a disability was recognised by the system as ‘not disabled enough’ is an example of absurd and cruel social assistance system. Moreover, according to interviewees, the system often fails to operate according to the principle of equity: it differentiates poorly between clients in need (from whom less of a contribution should be expected, such as parents of little children with severe disabilities) and those from whom more effort should be expected (such as capable adults failing to take care of dependent family members). The interviewees perceived single parents of babies or disabled children being classified as ‘not in need’ because they were able to work as a violation of justice.
There was a person denying help to us, mothers of little babies because ‘you could go to work’. Seriously? We were in that hostel for victims of violence and could not reveal our address (…) for security reasons. (Irena, PL)
Although, as noted, the general rule of eligibility testing tends to be recognised by our interviewees as a rational implementation of the rule of providing help according to need, whenever the system fails to follow that rule, it is interpreted as more than a lack of rationality. The two quotations above allude to an abuse of power by frontline workers demanding the impossible. A lack of trust here is the result of perceiving the welfare system as insensitive and even cruel to people seeking help.
Apart from interviewees’ experiences of social assistance failing to respond to the principle of need, our interviews revealed a salient thread in which the violation of both neediness and equality principles was explicitly linked to the abuse of power in the welfare system. This sub-theme revolves around interviewees’ conviction that there is an implicit distributive rule ingrained in the welfare system, which assumes that the living standard for welfare beneficiaries should never be equal to that of non-beneficiaries. Our respondents did not argue that benefits should be as high as salaries, but they underlined that in many cases the distribution of benefits or services is deliberately designed to make them feel like second-class citizens. This is how one individual linked these issues:
Q: So how do you assess those procedures – are they just or unjust?
A: Well, I’d like to give this kind of help to a manager, or to a social worker [from the social assistance centre], so that they [need to] survive with a kid with that amount of money. That’s how I assess it. (Mikołaj, PL)
Mikołaj here links the unfairness of being granted a small benefit to the conviction that social assistance workers impose on him a condition they would never impose on themselves. He underlines the gap between welfare beneficiaries and non-beneficiaries. Thus, the concept of equality is applied here to equality of welfare users and non-beneficiaries in due dignity and a decent living standard.
In a similar fashion, Polish interviewees often quoted frontline workers’ opinions, some of which included that they have astonishingly well-kept flats and furniture or TV sets of too high a standard; that they shouldn’t buy pizza because it’s too expensive; and that they should wash cotton diapers for babies instead of buying disposable ones. Low-quality food, which was recommended to them or provided them as in-kind assistance, was particularly often perceived as a sign of unfairness and intended to humiliate. The interviewees believed that frontline workers would never use, eat, or give their own children the products they were providing to clients. Interviewees often explicitly criticised the system’s unjust assumption that beneficiaries deserve something considerably worse than other people.
Although interviewees did not address trust issues explicitly when talking about injustice they had experienced related to system’s abuse of power, the way they framed their experiences left little room to doubt that they distrusted the system. Mikołaj’s sarcastic comment: ‘I’d like to give this kind of help to the manager, or to a social worker’ suggests that grievance and conflict define his relation with the welfare system.
Distrust and the perception that the system has a limited capacity to withhold pressures from ‘undeserving’ groups
Joining the principle of need, equality is a salient category according to which our interviewees evaluated the welfare system’s distributive justice. What various groups of welfare recipients allegedly receive and what beneficiaries themselves receive compared to members of other groups can be understood in empirical terms. Some of our discussion partners in Poland and Czechia maintained that there are specific groups that systematically receive much more or much fewer benefits than they should.
The pattern of how our interviewees perceived injustice on this issue was similar in both countries, although there were differences in the target groups they identified. The systemic injustice entailed interviewees’ suspicion or conviction that some beneficiaries cheat, abusing the system or pretending that they are unable to work, mostly by hiding their actual income. Those welfare users were described as ‘scheming’, ‘working on the black market’, ‘taking their salary under the table’, etc. In some cases, cheating entailed gaining falsified medical statements about a beneficiary’s long-term sickness or disability, all just to receive welfare benefits. According to interviewees, those welfare users typically misuse welfare benefits by spending money on things unrelated to children’s welfare, such as cigarettes, vodka, or drugs. Our interviewees had a range of opinions on who is to blame for this state of affairs. It was frequently argued that it is not the caseworkers’ fault, but the system is constructed that way. Yet, some interviewees believed the blame should be shared –by beneficiaries who make unjust claims and frontline workers who accept those claims for the sake of their own comfort.
Two patterns of comparison were made. First, the vague descriptor ‘those people’ was applied to those who were also described as undeserving, cheating, overusing the system, and receiving more than they should. Second, there was a group of welfare users twhohat systematically received more or less than they should. Thus, the comparative scheme juxtaposed ‘them’ with all other beneficiaries. In Poland these were mostly alcoholics, while in Czechia it was the Roma who were singled out for receiving ‘too much’ welfare, thus violating the principle of equality. A typical example of speech describing this phenomenon in Czechia, in which an interviewee argues that young Roma people (twenty, thirty years of age) have never worked in their adult life, but still receive unemployment payments from the Labour Office:
Gypsies, who are twenty, thirty years old get maximal possible financial help. (Kristyna, CZ)
Other groups that were singled out in both countries were single mothers and, mostly in Czechia, immigrants:
A couple of migrants come to Czechia and get everything, so tell me, where is the justice in that?! (Barbora, CZ)
Interviewees’ conviction that some groups of beneficiaries receive more welfare than others is salient for the perceived injustice of the welfare system. It entails emotions, as seen in the above quotation, where Barbora was upset by the notion that immigrants struggle less than Czech people. She also suggested that since ‘they get everything’, much less is left for others.
However, in both countries and foremostly in Czechia some interviewees claimed that the social assistance system actually discriminates against the aforementioned groups, particularly single mothers. This is a typical expression of such a conviction:
The government is not able to help single mothers at all. (Verona, CZ)
Although the brunt of interviewees’ concerns, which we’ve reconstructed in this theme, is on ‘undeserving’ groups (immigrants in Czechia and alcoholics in Poland), when the welfare system grants them assistance, it affects how the system is perceived. Verona presents this perspective in the quote above, pointing out that the state ‘is not able’ to do what it should. Such an inequality and systemic bias or favouritism was perceived as a breach of moral values as the system appears to collude with some groups, gives in to the pressures those groups exert, or chooses to work ‘the easy way’.
The link between such a perception of the system and distrust is especially evident in cases of experienced injustice. Many of our interviewees compared what they were denied by the welfare system to what ‘undeserving’ group members allegedly receive. Feelings of personal harm and humiliation were often expressed in this context. Notions of harm, unfairness, and being treated worse than “undeserving” welfare beneficiaries are visible in this explanation by a Polish interviewee, who compares what she receives to what, in her opinion, alcoholics get from welfare system:
They give to those to whom they shouldn’t, that’s how it seems to me. Because then a tipsy comes to them and he says he is sick. So, they help him. I saw such a guy standing in front of me in the cashier there (…) and he got more money than me. So here I have this grievance towards them. (Katarzyna, PL)
Here, Katarzyna uses the phrase ‘grievance towards them; to describe her attitude towards the welfare system. In the following exchange, Izabela expresses similar views:
Q: Overall, would you say you trust social workers?
A: Not really (…) and nor the whole institution. As I said, my friends who have families and also use social assistance keep telling me they are treated worse than those drinkers in front of a liquor shop. (Izabela, PL)
Here, a sense of collective harm against ‘us who have families’ being treated, in the opinion of the interviewee, ‘worse than drinkers’ results in her distrusting social workers and, as she adds, ‘the whole institution’.
(Dis)trust resulting from experiencing a surprise
The last theme reconstructed from our data concerns the relationship between what beneficiaries anticipated receiving from social assistance and what they have actually been granted. Thus, it is about positive and negative surprises in one’s experiences of distributive (in)justice. It is also directly related to trust. The following narratives drew our attention to this aspect of comparisons and evaluations made by the interviewees:
…So I completely lost my trust in them. I started to observe how they work and realized it’s not going to happen, [namely] that the help I was expecting from everyone there turned out to be a failure, right? (Aldona, PL)
I went there with no hope for any kind of help in that month. And then, a couple of days later she [the social worker] called me and said she had managed to push something through and got something for me. I don’t know how she did it, but she got the money for me. So, I received it and managed to pay my bills. And I have to say, it fostered my trust in her (…) Because with that previous lady, I had known she would always do something for me and with this one – I hadn’t thought so, but she did it. (Sylwia, PL)
Here we see how Aldona’s strong expectation of help she thought she should have received was followed by disappointment and feelings of being abandoned, leading her to distrust the welfare system. Sylwia, on the other hand, trusts her caseworker and, as she explains later on, the welfare system as well. The trust she feels grew out of a positive surprise from receiving help, which she needed but expected to be denied. Both examples stand out as somewhat unique, as in many cases it seemed that our interviewees didn’t have strong expectations of the social assistance system. While they hope to receive help according to procedures, at the same time they anticipate a denial. This is a result of their experiences with the ‘institutional enigma’. Our interviewees had long-term experience with dozens of decisions, both granting and denying them welfare benefits. As is evident in Sylwia’s reflection, they know that frontline workers vary in their approach, leading her to compare ‘the new lady’ to ‘the previous lady’. Our interviewees emphasise that there are numerous unexpected factors, such as the institution’s financial situation at the end of the year, when there is ‘big money left that needs to be spent immediately before they open a new budget’, or just the opposite – ‘the situation in December when the money is already just gone’. Other factors include changes made to legal rules, interpretations, case-workers, and managers resulting in new modes of cooperation with the welfare institution.
We can see a path to trust built on a user’s strong negative expectation or expectation of refusal, followed by a highly positive surprise (being granted help). The stories of Sylwia, quoted earlier, and Adam, a homeless man from a Polish city, serve as examples here. Adam had been sure he would need to wait ages for social housing, but his caseworker surprised him by finding him one and by saying that everyone deserves a flat. Adam explicitly presents this as a trust-building experience. He told us:
And then I told her [social worker] I don’t take[drugs] anymore. She was happy and told me I’m eligible to a flat. So, I said ‘so many people in [name of the city] are waiting for social housing, how come I’m eligible to a flat?’. So, she said ‘everyone is eligible to have a flat!’ (…) So, she started to look for a flat for me. That was really cool. (Adam, PL)
Adam explicitly linked this experience to the perception that the welfare system can be fair and explained how it enabled him to build trust in it.
The opposite mechanism reconstructed within this theme is based on a strong positive expectation of receiving welfare followed by a highly negative outcome and disappointment (not being granted help), which has a strong trust-detrimental effect. Many of our interviewees, including Aldona quoted earlier, explained that they had lost faith or stopped trusting in social assistance when they didn’t get what they found just and had strongly hoped for. Being promised a benefit that they never received was a common cause of lost trust. They said that they felt utterly disappointed, had been mistreated by the system; they felt upset and helpless. The notion of not getting ‘what I should have’, typical for research on distributive justice, has two meanings for this group of people – they feel they are deprived of something they deserved and of something that was suggested they would receive.
Conclusions and discussion
Discussion of findings
For this article, thirty-three interviews with parents who receive social assistance benefits and/or services in institutional embedding of incoherent policy assemblages of Poland and Czechia were analysed. We have confirmed that such an incoherent assemblage is a relevant context for evaluating the system’s distributive justice and is perceived by our interviewees as an ‘enigma’. The research helped us unveil users’ compound distributive justice understandings and their impact on (dis)trust in the countries’ welfare systems.
The first research question we sought to answer concerned how welfare users actually apply distributive justice criteria. In line with the conclusions of other researchers’ (Fersch, Reference Fersch2016; Nothdurfter and Hermans, Reference Nothdurfter and Hermans2018; Van Hootegem et al., Reference Van Hootegem, Abts and Meuleman2020), we found that welfare users have a combination of need, equality, and equity, all highly intertwined. Also, as suggested in the literature on deservingness criteria being relational and conditional (Willen and Cook, Reference Willen, Cook and Thomas2016), distributive justice judgements bore features similar to deservingness criteria, which often makes those judgements self-contradictory. An example of this is welfare users perceiving the means-test as a just solution while at the same time viewing the denial of assistance when their eligibility threshold is slightly exceeded as a sheer injustice.
Specific mixes of need, equality, and equity criteria of justice were revealed by our interviewees, whose experiences were typical of welfare users. One was on display when welfare beneficiaries feel they are expected to rely on poor-quality food or goods that frontline workers would never use themselves. This was perceived as a violation of both equality and need principles (need was also understood by interviewees as the need of a decent standard of living). Another example was granting allegedly generous support to ‘undeserving’ groups, such as the Roma or alcoholics. A good number of interviewees perceived this as simultaneously violating the three principles of equality, need, and equity. These findings are in line with those of Lavee (Reference Lavee2021), which showed how welfare users often merge contradictory expectations from the state.
Thus, highly complex and incoherent policy settings result in highly complex and nuanced principles of distributive justice. A distributive ideal can be drawn from our interviews of social assistance being granted according to need, measured in a way that clearly distinguishes the needy from the non-needy whilst acknowledging that need is not a binary variable. Those assessments, and the help granted following them, should be free of bias and acknowledge people’s factual and possible contributions. Clearly, such a distributive justice ideal would be impossible to meet in any system.
Our second research question concerned how people actually assess distributional effects of social assistance and how, in consequence, they evaluate that justice.
The distinction between experienced (egocentric) distributive justice perceptions and sociotropic ones (Kumlin, Reference Kumlin2002) was blurred in the narratives we collected. Users apply various, cross-cutting scopes of comparison when evaluating a welfare system’s justice. The interviewees spoke of various groups of ‘us’ (welfare users, parents, Polish or Czech citizens, etc.) and of ‘them’ (mostly ‘undeserving’ welfare users or frontline workers comparing their own situation to specific target groups).
We also found that in the incoherent institutional context that provides users with mixed messages on who actually receives what from social assistance, there are blurred lines between users’ notions of distributive and procedural justice. While distributive justice refers to judgements about the fair distribution of resources, procedural justice concerns judgements about whether fair procedures were used to decide the outcome (Tyler, Reference Tyler and Sarat2004). Thus, the latter is understood as a perceived quality of procedures by which distributive decisions are reached and includes, among other criteria, a moral, unbiased, and transparent process (Schnaudt et al., Reference Schnaudt, Hahn and Heppner2021). In our results, the focus on means-testing could be regarded as the expression of procedural justice. For our interviewees, means-testing (such as Frontline workers ignoring assets of ‘undeserving’ group members) was recognised as a litmus test of how the distribution of benefits according to the desired principle of serving the needy poorly works in practice.
‘Substitutability’ between different types of justice – including distributive and procedural justice – can be explained by Lind’s (Reference Lind2001) fairness heuristic theory, which holds that people are motivated to form their justice judgements of individuals, groups, or institutions rapidly, at initial stages of their relationships. Such a motivation can be explained by the fact that judgements are used as a heuristic to determine whether to trust others and behave cooperatively in mutual interactions. In line with this tendency, people rely on all presently available justice-relevant information: those about the procedures of granting or receiving benefits, as well as about individual and general outcomes of those procedures, with no preference for either. Thus, the theory suggests a rather rapid and automatic holistic judgement process (Greenberg, Reference Greenberg2001; Ambrose and Arnaud Reference Ambrose, Arnaud, Greenberg and Colquitt2005), yet one which is relatively persistent over time. We interpret our findings that it is the context of incoherent policy assemblage that makes holistic judgements and replaces distributive justice concerns with assessments of procedural justice salient.
Our third question asked how evaluations of the social assistance system being just or unjust translate into trust and distrust in that system.
In our analysis we reconstructed four patterns of how the perception of the welfare system’s distributive (in)justice is linked to trust in that system. The perceived justice-trust relationships may be based on:
-
1) The welfare system being perceived as meeting the principle of helping the needy, being rational and predictable (that perception being strengthened by users’ acquiring information on the welfare system’s procedures), and being trusted
-
2) The welfare system not meeting the principle of helping the needy; being perceived as abusing power and lacking sympathy for clients in need; and being distrusted
-
3) The welfare system being perceived as failing to meet the principle of equality, lacking the capacity to withstand pressures from groups of ‘undeserving claimants’ and thus being distrusted
-
4) A strong positive surprise by experiencing justice, and thus leading to trust, or a strong negative surprise by experiencing injustice, leading to distrust
The first three relationships are bi-directional between the perception of justice and trust in the fashion suggested by Schnaudt and coauthors (Reference Schnaudt, Hahn and Heppner2021: 5). Based on what interviewees said, we conclude that those who trust the system think that it rationally applies means-testing criteria, while those who don’t trust it find the social assistance system oppressive and/or weak in resisting unwelcome pressures. Thus, we would argue that only the fourth relationship, which is far different than the other three, reveals a one-directional causal mechanism. The surprise-based mechanism of (dis)trust building that our interviewees spoke of shows the role that strong expectations play in a causal chain between justice evaluations and trust in the system. We observed that an increase or withdrawal of trust was often preceded by, respectively, positive and negative surprises in the justice experienced in the welfare system. In other words, whenever the ‘fairness gap’ – the difference between the latent idea that one is in need and should be helped by the welfare institution and the conviction that the system actually does, or does not, provide help – rapidly and unexpectedly closes or widens, the user’s trust in the welfare system increases or collapses, respectively.
This finding brings us again to Lind’s (Reference Lind2001) fairness heuristic theory. Although global judgements of fairness tend to be stable, according to Lind (Reference Lind2001) they can be reconsidered on some occasions. Termed phase-shifting events, these occasions typically involve a considerable alteration of one’s expectations, either negative or positive. The contrast between expected fairness and instances of indisputably unfair treatment or vice versa can return people to the ‘judgemental’ phase and allow them to update their global judgement (Lind, Reference Lind2001). Later formulations of the theory suggest that this shift towards reconsideration is more likely in situations of uncertainty and insecurity and less likely when people are cognitively busy or depleted (Proudfoot and Lind, Reference Proudfoot, Lind, Cropanzano and Ambrose2015). In the context of social welfare, phase-shifting events can be represented by either considerable institutional changes, during which it is evident that the relationship in question also changes (e.g. when rules or frontline workers are changed), or by very unexpected treatment that can concern both distributive and procedural justice. The fairness heuristic theory helps us to understand that small violations of users’ expectations are hardly able to alter their justice judgements as it is quite easy to assimilate these experiences to existing beliefs (e.g. by framing these violations as accidental or driven by a specific situation). Only when the violation is large enough to elicit surprise, and perhaps a certain level of uncertainty and insecurity, can the initial justice judgement be reconsidered. This mechanism also shows the salience of the egocentric scope of justice judgements and their impact on trust, in contrast to dominant stances that show how trust is influenced by sociotropic notions of distributive justice (Kumlin, Reference Kumlin2002; Zmerli and Castillo, Reference Zmerli and Castillo2015; Bobzien, Reference Bobzien2023).
Contributions to literature
Our analysis adds to research on the relation between distributive justice and political trust (Kumlin, Reference Kumlin2002; Zmerli and Castillo, Reference Zmerli and Castillo2015; Berg and Dahl, Reference Berg and Dahl2020; Schnaudt et al., Reference Schnaudt, Hahn and Heppner2021) and the experiences of citizens in encountering, coping with, and making sense of welfare state institutions (Auyero, Reference Auyero2011; Fersch, Reference Fersch2016; Wright and Patrick, Reference Wright and Patrick2019; Lavee, Reference Lavee2021). It contributes to filling a recognised gap on the role of institutional embedding in shaping perceived justice-trust relationship (Berg and Dahl, Reference Berg and Dahl2020; and Zhang et al., Reference Zhang, Li and Yang2022).
The added value of our study stems from the analysis of how the link between perceived distributive justice and political trust in the welfare system operates in the context of the self-contradictory, complex policy in two post-socialist countries – Poland and the Czech Republic. This context translates into users’ experience of facing what we have called here ‘institutional enigma’ – an institutional setting in which rules and procedures may be known, but they hardly make sense to users, are puzzling, and seem self-contradictory or random.
In line with the differences in policy assemblages presented in previous section (Saxonberg and Sirovatka, Reference Saxonberg and Sirovátka2018; Natili, Reference Natili2020; Aidukaite et al., Reference Aidukaite, Saxonberg, Szelewa and Szikra2021), there were differences in ‘institutional enigmas’ discussed by our interviewees in Poland and the Czech Republic. For example, activation requirements and issues of poverty traps due to failing means-test after starting employment are more pronounced in the Czech Republic. In Poland, users’ reflections on the procedures of eligibility tests and ceasing some forms of help due to universal family benefits were raised by respondents. Another area of difference presented was different groups perceived as representative of ‘undeserving groups’, i.e. Roma in the Czech Republic and alcoholics in Poland. This difference rather represents local variations in anti-welfare discourses in both countries (Bell and Valenta, Reference Bell and Valenta2024). Overall, however, we believe that regardless of such differences in policy assemblages and deservingness discourses, their impact on the perceived link between justice and trust was similar in both countries studied.
Our study goes beyond the state of the art in two ways. First, we found that in the presented context of complex and self-contradictory policy assemblages (Newman and Clarke, Reference Newman and Clarke2009; Clarke and Bainton, Reference Lendvai-Bainton and Stubbs2015; Saxonberg and Sirovatka, Reference Saxonberg and Sirovátka2018; Lendvai-Bainton and Stubbs, Reference Lendvai-Bainton and Stubbs2021), perceived by welfare users as a ‘institutional enigma’, the path to political trust in welfare system (as opposed to flourishing trust at the interpersonal level of relations with frontline workers) via perceived distributive justice is very difficult. From previous analyses, we know that institutional embeddedness can increase the impact of perceived injustice on political trust, e.g. due to high ascribed responsibility of some institutions (Berg and Dahl, Reference Berg and Dahl2020). In our study, on the other hand, the negative impact of context on the justice-trust relationship was due to other issues. Namely, the justice principles held by welfare recipients were very complex and in some cases self-contradictory. As a result, there were numerous occasions when welfare users experienced their violation. The importance of these experiences was reinforced by feelings of individual and collective grievance and negative emotions (Bobzien, Reference Bobzien2023), as seen in the second and third mechanisms presented in our findings. Moreover, one of the typical ways in which welfare users made sense of the ‘institutional enigma’ was by applying anti-welfare discourses typical in the country (Ochsner et al., Reference Ochsner, Ravazzini, Gugushvili, Fink, Grand, Lelkes and van Oorschot2018; Adriaans et al., Reference Abdelzadeh and Ekman2020; Bell and Valenta, Reference Bell and Valenta2024). Overall, therefore, it seems that incoherent policy assemblages, accompanied by public anti-welfare discourses, are to some extent responsible for users’ low trust in welfare institutions.
Second, our study reveals the paradox that political trust can still be nurtured in such an unfavourable institutional setting. Thus, the context of our research, which functions as a critical case study, contributes to theory development (Flyvbjerg, Reference Flyvbjerg2004). We have found two pathways that can lead to trust in circumstances where trust is unlikely to occur. The first is that citizens gather information about the procedures and develop an assertion that these procedures are indeed rational and lead to the desired outcome, i.e. to help people in need. This path also shows potential positive spill-over of some types of administrative burden (Tarshish and Holler, Reference Tarshish and Holler2024). The second one, which is less affected by risk of reverse causality typical in trust-formation mechanisms (Abdelzadeh et al., Reference Abdelzadeh and Ekman2015; Schnaudt and Hahn, Reference Schnaudt, Hahn and Heppner2021) and is consistent with Lind’s (Reference Lind2001) fairness heuristic, is through welfare users’ strong positive surprise about a positive redistributive outcome. An example of this mechanism was the experience of a homeless person who was quickly given a decent flat in a social housing scheme, although he had never expected this to happen. However, both mechanisms of fostering trust in an unfavourable context show rather ‘demanding’ ways to trust, but not impossible.
The limitations of this study stem from its research design. As mentioned earlier, we follow the user perspective (Dubois, Reference Dubois2020) in analysing the relationship between perceived justice and political trust, which also applies to our reasoning about causal mechanisms. In addition, this design does not allow us to answer a relevant question to which further research may contribute, namely why some welfare users ‘invest’ in gathering information about the procedures and conclude that the system, although flawed, is still rational and serves the desired goals. Finally, we lack information to assess the extent to which the mechanisms discovered are specific to a context such as the one under study. In fact, we interpret this context as extreme example of a widespread situation of confusing institutional settings, in which the citizens’ way to trust or mistrust institutions is to make sense of what they encounter.
Acknowledgements
We would like to thank the anonymous reviewers for their helpful comments on earlier versions of this article, and Llorenç Soler Buades for providing detailed information on the comparison of national social assistance schemes.
Funding statement
Results presented in this article have been obtained within the project ‘Enlightened trust: An examination of trust and distrust in governance – conditions, effects and remedies’ (EnTrust). This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 870572. The contents of this article are the sole responsibility of the authors and do not necessarily reflect the opinion of the European Union.
Open access
