Money, competences or behaviour? On the many worths of the unemployed

The article investigates how ordinary citizens establish the worth of unemployed persons when discussing politics. It analyses data from two deliberative forums held in Copenhagen, Denmark and Birmingham, UK at which citizens discussed welfare policies. It uses core concepts from French pragmatic sociology in an attempt to grasp the moral complexity of such discussions. Contrary to the moral singularity found in the majority of existing analyses—distinguishing ‘negative’ from ‘positive’ perceptions of the unemployed—the article emphasises the plurality of coexisting normative standards that people turn to when establishing the (lack of) worth of the unemployed. It finds that three normative standards are most important in both countries but being applied in quite different ways: an industrial standard of worth, measuring worth in terms of competences; a domestic standard of worth, measuring worth in terms of behaviour; and a market standard of worth, measuring worth in terms of money.

This article analyses how ordinary persons in Denmark and the UK respectively evaluate the worth of ‘the unemployed’ when discussing politics. What scales are applied when citizens measure the worth of the unemployed during discussions and draw distinctions between deserving and undeserving?

The literature on discourses, images and social constructions of the unemployed and the poor in the Western world is large, nuanced and methodologically diverse. Still, the majority of existing studies portray what I term ‘singular moral universes’: they depict one-dimensional moral worlds where discourses are categorised with recourse to just one evaluative criteria, leading to a mapping of, for instance, negative versus positive discourses (e.g. Larsen & Dejgaard, 2013), stigmatising discourses versus nonstigmatising discourses (e.g. Harkins & Lugo-Ocando, 2016; Jensen, 2014; McArthur & Reeves, 2019) or individualist as opposed to structuralist discourses (e.g. Bauer, 2014).

The article seeks to demonstrate how a pragmatic sociological approach (Boltanski & Thévenot, 2006) to analysing discourses of unemployment might add something of importance to the existing field of research. In particular, the analysis demonstrates the usefulness of the central notion of a plurality of coexisting ‘orders of worth’, each representing a specific normative scale used by actors to assess and rank the relative worth of things and persons (Boltanski & Thévenot, 2006). Simply put, the application of core pragmatic sociological concepts adds three things of importance to the conceptual grid. Firstly, rather than asking only if the unemployed are perceived ‘negatively’ or ‘positively’, the pragmatic sociological analysis seeks to locate the underlying evaluative criteria according to which such moral assessments are based. Secondly, the analysis looks for a potential plurality of coexisting orders of worth used by actors to distinguish the worthy from the less worthy unemployed. And thirdly, it pays close attention to the ways in which people ‘put their arguments to the test’ (Lamont & Thévenot, 2000), that is, the material proofs people mobilise and lean on in order to justify themselves.

To exemplify, take the two introductory quotes from the deliberative forum participants in Denmark and the UK respectively into consideration. We can say that the first quote from a Danish participant portrays the unemployed positively, in the sense that they are depicted as deserving, and that the second quote from a British participant depicts the unemployed negatively, in the sense that they are depicted as undeserving (e.g. Larsen & Dejgaard, 2013). We can also say that the first quote utilises nonstigmatising discourse, whereas the second one, in contrast, uses stigmatising rhetoric (e.g. Jensen, 2014).

But such forms of analytical classification reveal little about the different moral criteria underpinning citizens’ moral evaluations of the unemployed. The participant in the first quote uses dignity as a normative lens through which he assesses the unemployed. The participant in the second quote, on the other hand, refers to market failures in terms of perverted financial incentives in his moral evaluation. The first citizen speaks about ‘sickness’ and ‘coercion’, whereas the second participant mentions very different objects: ‘benefits’, ‘massive houses’ and ‘work-shyness’. The moral criteria underpinning the two quotes and the ways in which the two citizens justify themselves are essentially different.

A pragmatic sociological perspective on unemployment discussions offers a helping hand as it opens up the analytical grip and allows us to locate such evaluative criteria—orders of worth in the terminology of Boltanski and Thévenot—and to map their potential plurality, thus revealing not only ‘positive’ as opposed to ‘negative’ evaluations but also a potential rich plurality of normative standards according to which normative evaluations are made. Thus, a pragmatic sociological approach strives to encapsulate what Thévenot elsewhere terms the ‘moral complexity’ (Thévenot, 2002) of such discussions.

I demonstrate the fruitfulness of the approach through a qualitative analysis of transcripts from democratic forums at which ordinary citizens in Denmark and the UK respectively discussed what the welfare state ought to look like in the year 2040. Citizens were compelled to justify their opinions, and the transcripts contain numerous justificatory statements.

I have chosen to analyse material from exactly Denmark and the UK for a number of reasons. Following the arguments presented by Thévenot and Lamont for doing in-depth comparative studies of France and the US, it also goes for Denmark and the UK in relation to unemployment policies that they are ‘defined in opposition to one another and hence make for an especially rich contrast’ (Lamont & Thévenot, 2000). Commonly, the UK is used as an illustrative example of a liberal welfare state, whereas Denmark is said to illustrate a social democratic one (Esping-Andersen, 1990).

Whereas both countries have implemented so-called active labour market strategies and developed new regimes of conditionality to battle unemployment during recent decades (Jørgensen, Klindt, & Rasmussen, 2020; Patrick, 2017), UK policies tend to focus on punishment and discipline (e.g. Wright, 2016), epitomising a so-called work-first approach (Soss, Fording, Schram, & Schram, 2011), while Danish ones tend to focus much more on developing the human capital of the unemployed (Torfing, 2004). Patrick (2017, p. 36) emphasises how the dominant British political representation of the problem of ‘welfare’ suggests that benefits for working-age adults are in themselves inevitably problematic since benefits are believed to promote laziness and welfare dependency. The representation of ‘welfare’ in Denmark tends to be less negative (Larsen, 2008).

Thus, the two countries seem to represent very different ways of thinking morally and politically about the ‘problem of unemployment’ (Lødemel & Trickey, 2001). This should provide the case for a fruitful investigation of the very diverse and multifaceted ways in which citizens apply standards of worth to morally evaluate the unemployed.

The remaining article falls into four parts. Part I contains a brief sketch of the concepts applied from French pragmatic sociology, introduces previous pragmatic sociological research on unemployment and discusses the value added by this article. Part II explores recent state-of-the-art literature on discourses on poverty and unemployment in the UK and Denmark. Following this, part III outlines the empirical data and analytical strategy pursued and presents the analysis of the material. Part IV summarises the findings and discusses the implications.

A few words about what exactly is meant by order of worth will be beneficial at this point. Boltanski and Thévenot (1999, p. 364; see also Boltanski & Chiapello, 2005, p. 22) pedagogically explain the practical functioning of orders of worth by turning to a hypothetical trivial example—the order in which a host chooses to distribute food amongst guests who are present at a meal. In order to decide whom to serve first and whom to serve last, the host encounters the problem of the relative status of the guests. The host realises that different standards can be used to measure the relative worth of the people seated around the table. The elderly can be served first and the children last. The most productive and efficient worker among the guests can be served first, or the most social and popular person among the guests can be served first. It depends upon the order of worth applied to allocate relative status to the seated guests.

The main conceptual idea of this article is similar. The relative worth of the unemployed might be defined and articulated with recourse to a plurality of different ranking principles, leading to evaluations of the unemployment on radically different moral scales. Using the vocabulary of Boltanski and Thévenot (2006), the evaluation of the worth of the unemployed can be based on a plurality of different orders of worth, each representing a specific principle of equivalence used to rank the entities of a situation from the most worthy to the least worthy.

Boltanski and Thévenot (2006) identify six orders of worth, which include: (1) a market form of worth valuing free competition among individuals and ascribing worth to the opportunistic and wealthy; (2) an inspirational worth ascribing greatness to truly inspired beings—the creative, artistic beings who are independent of recognition from others; (3) a domestic form of worth based on family values and valuing strong, kinship-like relations between people and ascribing worth to the humble who acknowledge their place in the hierarchy and display good manners and respect for traditions; (4) civic worth, valuing solidarity and ascribing worth to persons who give up their particular personal interests and direct themselves exclusively towards the common good; (5) industrial worth valuing efficient production and ascribing worth to the efficient, productive and operational; and finally (6) an order of worth based on fame, in which your worth depends on other people’s opinions of you.

Utilising justification theory to analyse welfare or unemployment discussions is not exactly a new idea. Borghi (2011) gains inspiration from the order-of-worth framework to develop a macro model of different ‘institutional regimes of justification in EU welfare capitalism’, representing different ways of establishing and justifying the state-citizen relationship. Others apply the order-of-worth framework to categorise different moral ideas underpinning rhetoric in recent public debates (e.g. Nielsen 2014; Nielsen, 2015; Patriotta, Gond, & Schultz, 2011). An analytical framework, justification analysis (Ylä-Anttila & Luhtakallio, 2016), has been developed for doing exactly that.

Even others have applied the order-of-worth framework in comparative analyses of justifications of welfare policies. Frederiksen (2018) draws on a large number of qualitative interviews to analyse why and how respectively Swedes and Danes support universal welfare policies. He concludes that while citizens in both countries provide three similar overall forms of justifications for universalism—revolving around ‘collective obligations’, ‘efficiency’ and ‘equitable access’—important differences occur within these overall forms: Swedish informants tend to rely on compromises between civic and industrial forms of worth, whereas Danish participants rely on compromises between civic, industrial and market forms of worth. Hansen (2017, 2019) studies four major labour market reforms in respectively Denmark and France. Empirically, he turns to a combination of newspaper articles and policy papers to analyse the ways ‘in which public opinion sediments into the policies and instruments shaping practices in the jobcentres’ (Hansen, 2019, p. 190). He uses pragmatic sociology to map how coexisting moral ideas concerning unemployment are mobilised during the reform processes to justify certain policy ideas while criticising others. To do so, he rethinks the conceptual grid of justification theory and constructs seven new orders of worth reflecting competing moral ideas underpinning the development of unemployment policies. He documents how recent reforms are driven by forms of justification based on ‘paternalism’, ‘investments’, ‘financial incentives’ and ‘mobility’, whereas forms of worth connected to ‘redistribution’, ‘insurance’ and ‘demand’ are rejected as useless or even referred to as root causes of unemployment (Hansen, 2019, p. 177).

The analysis presented in this article takes inspiration from the works of Frederiksen and Hansen but differs from them in some important respects. First, it presents a most different comparative case design (comparing material from Denmark and the UK), rather than a most similar one, following the arguments already presented in the introduction (cf. Lamont & Thévenot, 2000). Second, it curiously analyses conversations and discussions amongst ordinary persons who retain a high degree of control over the conversation (Taylor-Gooby, Leruth, & Chung, 2019) rather than elite discourses of newspapers and policy documents (Hansen, 2019) or researcher-guided one-on-one interviews that follow a standardised interview guide (Frederiksen, 2018). This allows for an investigation of the ways in which ordinary citizens openly apply their critical competences during actual unfolding disputes. Thirdly, the analysis presented here applies the six ‘original’ orders of worth constructed by Boltanski and Thévenot rather deductively. From my perspective, the main advantage of doing so is that I conceptually recognise that people ‘over a wide variety of issues and across different political contexts […] tend to justify their arguments using a relatively limited set of moral principles’ (Ylä-Anttila & Luhtakallio, 2016, p. 1). Thus, operationalising the original order-of-worth framework—rather than inductively developing new orders or focusing on context-based compromises—opens up precious possibilities for transferability and comparability to other pragmatic sociological studies of very different topics, as it uses the same concepts.

As documented by Larsen and Dejgaard (2013), stigmatising rhetoric is more dominant in British media articles on poverty than in Scandinavian ones. They investigate how poor people are depicted in stories and images in newspapers in the UK, Denmark and Sweden from 2004 to 2009 by coding articles into two main categories, depending on whether the vulnerable person in the story is presented ‘negatively’ (in the sense of being undeserving) or ‘positively’ (in the sense of being deserving) and find that a larger share of the British news stories were negative (43%) in comparison with the Scandinavian ones (27% in Sweden and 26% in Denmark). This confirms their main hypothesis that selective welfare regimes (such as in the UK) are more likely to produce othering and stigmatising media discourses on the poor than universal welfare regimes (such as in Denmark and Sweden).

Quite a few studies have analysed such stigmatising rhetoric in British news media stories covering poverty and unemployment. McArthur and Reeves (2019) quantitatively investigate a large archive containing newspaper articles spanning more than one hundred years in order to investigate the correlation between stigmatising rhetoric and unemployment rates. They find that the media use of stigmatising rhetoric, describing poverty and unemployment as primarily self-induced symptoms of moral decay, generally rises when unemployment rises, except at the peak of very deep recessions.

Harkins and Lugo-Ocando (2016) similarly draw clear distinctions between stigmatising and nonstigmatising discourses in their study of a sample of articles from four British tabloid media, containing the word ‘underclass’ from 2007 to 2012. The authors conclude that so-called Malthusian discursive regimes, defining vulnerable people as ‘others’ who are themselves to blame for their own situation, have ‘remained the most important paradigm in defining the way poverty is reported by the tabloid press in Great Britain’ (Harkins & Lugo-Ocando, 2016, p. 88). Generally, they note, such discourses displace responsibility from structural conditions to the individual and pejoratively depict the underclass as, for example, ‘feral’, ‘white chavs’ or as ‘freeloaders’ ‘scrounging on the dole’ (p. 85).

Jensen (2014), using comparable conceptual tools, finds a similar rhetoric in British TV reality shows depicting the lives of unemployed people (e.g. ‘Benefit Street’) and notes how such shows ‘serve to transform precarity into a moral failure, worklessness into laziness and social immobility and disconnection into an individual failure to strive and aspire’ (p. 4). Garthwaite (2011) notes that groups receiving sickness-related benefits—conventionally seen as ‘deserving’ (Van Oorschot, 2000)—are increasingly met by stigmatising rhetoric from the media and politicians that portrays their lack of employment as a mere result of idleness and workshyness and contains no consideration of why people are on such benefits in the first place (Garthwaite, 2011, p. 371).

Qualitative research has explored how such rhetoric, portraying vulnerable citizens as ‘others’ and describing their situation primarily in terms of self-responsibility, is handled by the very citizens who are the targets of this rhetoric. Lister (2004, 2015), having worked extensively with the concept of poverty, studies relations between dominant discourses on poverty on the one hand and the lived experiences of poverty on the other, focusing particularly on what she defines as the ‘poverty-shame nexus’. The nexus emphasises how feelings of stigma, guilt and a lack of sense of entitlement is felt by those described as ‘the poor’ or ‘the underclass’ and how such feelings are directly connected to these dominant discourses. Some dominant discourses take a sympathetic stand towards the marginalised and call for the pity of the surrounding world, whereas others evoke feelings of ‘fear, contempt, disgust and hostility’ (Lister, 2015, p. 144). Either way, people living in poverty are described in ways that strongly distinguish the majority from the poor, and they are therefore subjected to processes of ‘othering’ (Lister, 2015, p. 140).

Shildrick and MacDonald (2013) investigate such processes based on qualitative interviews with citizens moving in between low-paid part-time jobs and periods of unemployment and find that informants who themselves belong to ‘one of the most economically marginalised and impoverished fractions of the working class’ (p. 299) refuse to describe themselves as being poor. Rather, they ‘normalise’ their situation and describe the hardship of their lives as a general condition shared by most people living in the area (Shildrick & MacDonald, 2013, p. 290; see also Melrose & Dean, 1998).

Turning to the case of Denmark, recent quantitative studies (e.g. Esmark & Schoop, 2017) indicate that the depiction of the unemployed in the Danish media might be moving towards portraying the unemployed more and more in negative terms of being underserving. Investigating how Danish politicians frame the unemployed in the public debates surrounding two large reforms of the Danish social assistance system in 2005 and 2013 respectively, Esmark and Schoop find that the use of frames presenting the unemployed as underserving is much more prevalent in the case of the latter than in that of the former. Such findings are backed by other media content analyses. Bauer (2014), for instance, finds that the use of discourses portraying unemployment as self-inflicted has risen markedly from 1993 to 2013, and Frederiksen (2018), as already touched upon, finds that the market order of worth plays a prominent role when Danes justify welfare universalism—contrary to what he finds in neighbouring Sweden.

Studies working with the perspective of deservingness theory have concluded that Danes, like others, are much more likely to support work-first initiatives targeting certain groups of unemployed people deemed to be in control of their situation (such as young people) (Petersen, Slothuus, Stubager, & Togeby, 2007). Analyses of the Danish public debate on unemployment and poverty indicate that the most dominant discourses tend to depict unemployment and poverty as either something that is self-inflicted or as a structurally conditioned phenomenon—and discourses depicting unemployment as self-inflicted seem to be on the rise (e.g. Sørensen, 2009). In similar ways, Bauer (2014) draws distinctions between articulations of unemployment as self-inflicted on the one hand and articulations of unemployment as structurally produced on the other.

Attempts on the part of welfare claimants to deal with pejorative discourses by passing the stigma on to others is fairly well documented for the Danish case as well. For instance, Pultz (2018) investigates how young Danish recipients of unemployment benefits govern themselves in order to handle the potential shame ensuing from not having a job (see also Dencker-Larsen & Lundberg, 2016, p. 7).

From the perspective of pragmatic sociology, we can say that the majority of recent research on discourses on unemployment and poverty in the UK as well as in Denmark depicts singular moral universes, many of which are characterised by dichotomist discourses of unemployment. As shown above, numerous analyses contrast ‘negative’ versus ‘positive’ discourses, ‘stigmatising’ versus ‘nonstigmatising’ rhetoric, ‘individualist’ versus ‘structure-focused’ discourses or depictions of the unemployed as ‘deserving’ versus ‘undeserving’.

Some of these studies do offer alternatives to such a binary conceptualisation of unemployment discourses (e.g. Esmark & Schoop, 2017; Melrose & Dean, 1998) but quite rarely exhibit much interest in the different moral principles underpinning the variety of discourses.

Deservingness theory (Larsen, 2008; Van Oorschot, 2000, 2006), to take one such example, makes a substantial contribution to understanding moral evaluations, as it theorises five different criteria which are decisive for the level of deservingness ascribed to subgroups (such as the unemployed), according to popular opinion.

One of these is a criterion of control (the more control a person has over his or her situation, the less deserving), and another is a criterion of need (the needier a person is, the more deserving). However, the theory does not acknowledge that all of the five criteria can be qualified in very different ways, depending on the moral world in which they are brought to life. In the terminology of Boltanski and Thévenot, control and need—as well as any other criteria of deservingness—can be qualified in very different ways, depending on the order of worth mobilised by the evaluator.

For instance, when ‘control’ or ‘need’ are qualified within the contours of a market logic, they will manifest as closely connected to market value: a poor person will be said to have a high level of need and a low level of control of his or her own situation, and the rich person will—vice versa—be said to have a low level of need and a high level of control. If, on the other hand, ‘control’ or ‘need’ are qualified within an industrial logic, the truly needy, who is without control of his or her own situation, is presumably no longer the poor but the one who lacks competences and skills. The scale for measuring neediness and control changes from weighing the market value of a person to weighing his or her competences. In a moral world based on fame, measuring ‘need’ will rather be about measuring popularity: The unpopular is the truly needy, and so on.

The point is that deservingness criteria are not at all stable but manifest themselves differently according to the moral world in which they appear: Need and control can mean several different things. Criteria of deservingness are dependent upon the contextual normative logic to which they are inscribed and can manifest in a range of different ways (see Nielsen, Frederiksen, & Larsen, 2020). One needs a conceptual grid that recognises the possible plurality of coexisting normative logics to detect such different manifestations.

In the remaining sections of the article, I concentrate on my empirical data and investigate what normative standards participants mobilise in discussions about unemployment in our two deliberative forums.

Deliberative forums are large group discussions where people gather to deliberate on an issue deemed to be of broad concern for their community. Often, such forums are established as elements in the realisation of a deliberative democracy, allowing ordinary citizens to have a say, and for ideas and solutions to be products of genuine democratic discussions (Fishkin, 1997; Goodin & Dryzek, 2006; Lafont, 2015).

Recently, democratic forums have attracted awareness from qualitative social researchers since they offer a valuable alternative to qualitative interviews and focus-group interviews (Taylor-Gooby et al., 2019). In contrast to qualitative interviews, deliberative forums offer insights into lightly moderated conversations among peers, and contrary to focus groups, deliberative forums offer larger and less moderated plenary sessions over a longer period of time in which participants prepare arguments and strive to present them convincingly in front of others.

The forums analysed in the article were set up by a team of researchers (including the author) and held over two Saturdays in the fall of 2015. Thirty-four people participated in both days of the Birmingham forum and 35 Danish citizens participated in both days in the Copenhagen-based forum.¹ Participants were recruited in order to achieve some socio-economic diversity, thereby increasing the probability of capturing a variety of different perspectives, and participants were given a financial incentive, around €200 each, for showing up. As they were part of the same research project, the deliberative forums in the two countries followed the same pattern, with the two days involving a similar mixture of plenum discussions including all participants and discussions in smaller group constellations.

Prior to the forums, participants were informed that the overarching question to be discussed was ‘What should the priorities of the government in this country be for benefits and services in 2040?’ They were encouraged to think about the challenges their society currently faces and to come up with specific proposals on how to solve those problems. Participants were thus assigned the role of policymakers and encouraged to think about how ‘we’—society as such—can act on the challenges that face us.

The two days involved a mixture of lightly moderated discussions on different welfare topics, with most discussions taking place in three predefined focus groups in each country. On the second day, participants in the three groups were to agree upon a number of specific policy proposals and to present these proposals in a concluding plenary session. The deliberative forums offered a setting that highly encouraged participants to develop ideas about their preferred policy and to construct arguments for their preference. To borrow a phrase from Boltanski and Thévenot, participants were directly ‘subjected to an imperative of justification’ (Boltanski, 2012, p. 38). This makes the material well-suited for an analysis of how people establish worth during discussions.

All plenary sessions and all focus group sessions were video and audio recorded. In total, this gave more than 30 h of recorded deliberative discussions from each of the two countries. The material was transcribed in its entirety, and the Danish material was translated from Danish into English. All text has added layers of codes, using the computer software NVivo 11. In an initial first round of coding, all statements in the material were person coded and thematically coded by the author and a team of colleagues. For a second round of coding, done solely by the author, all parts of the material referring to unemployment were isolated.² I coded what has been defined elsewhere as ‘units of meaning’—phrases and arguments that are bound by a clear ending that expresses at least one clear idea (Patriotta et al., 2011). Nonsense and mere descriptive comments were thus not coded in this process. In quantitative terms, the British material contained more relevant units of meaning than the Danish material (around 350 British statements as opposed to around 140 Danish ones; cf. Table 1).

Through careful rereadings of the isolated units of meaning, I developed a coding system to classify the normative statements, following the six orders of worth elaborated in the original framework of Boltanski and Thévenot (2006).

Operationalising the six orders of worth in the field of unemployment discourses was ultimately an abductive process, shuffling between central theoretical texts on the one hand (primarily Boltanski & Thévenot, 1999, 2006) and the transcribed material on the other in order to develop the best possible translation of the abstract conceptual work to contain the empirical data. Orders of worth are best operationalised as potentially mutually inclusive (Boltanski & Thévenot, 2006), and some statements were assigned more than one code. For instance, a statement such as ‘unemployed people are generally lazy and are unemployable due to their lack of skills’ would be coded as being simultaneously industrial and domestic. I concentrate the analysis on exploring the three orders of worth that turned out to be the three most prominent in both countries: the market order of worth, the domestic order of worth and the industrial order of worth.

I found almost no statements applying fame or inspirational orders of worth in unemployment discussions. I did, however, find some civic statements in the Danish material but almost none in the British material. Such statements would, for example, invoke the need for ‘solidarity’ with the unemployed or argue against recent reforms that have increased financial inequality. Compromises between respectively civic and industrial forms of worth and civic and domestic forms of worth occasionally occurred. The former was the case when Danish participants argued for solidarity by emphasising, for example, how the unemployed are already skillful and doing all they can to get a job and the latter when participants argued for solidarity with the unemployed by pinpointing the unjust indecency of the unemployment system.

In a third and final round of coding, I distinguished between negative and positive perceptions of the unemployed while following the general trends in the literature, with negative meaning that a person is depicted as undeserving and positive meaning that the person is depicted as deserving. The following analysis unfolds the moral complexity of the discussions about unemployment.

When applying the market order of worth, human actions are perceived to be motivated by opportunistic desires of people, driving them to possess objects deemed valuable (Boltanski & Thévenot, 2006). Insecurity about the relative worth of persons or entities can be settled in terms of price and money, which are believed to express their relative importance.

Data from both countries contain numerous statements assessing the worth of the unemployed within the contours of such a market world. Participants on several occasions im- or explicitly refer to the unemployed as rational beings in a context consisting of objectively desirable material goods. A market sense of fairness problematises access to goods that are independent of market coordination and pinpoints how unemployment results directly from the availability of such ‘free goods’. ‘Welfare benefits’ are described as potentially disturbing for the natural market order of competition, as they allow an income disproportionate of market value. The most important objects are objects of consumption, and they are therefore closely related to money. Participants who use market standards of worth to justify themselves speak of ‘benefits’, ‘incentives’, ‘income level’, ‘the amount of money’, ‘minimum wage’ and typically mention a range of specific desired material goods, such as ‘a massive great house’, ‘a nice car’, ‘a PlayStation’ and ‘cigarettes’. In such market statements, there generally is a longing for what Thévenot (2014) has defined as a ‘liberal grammar of commonality’: a social world where actions are depicted as ‘individual choices’ made to obtain personal ‘interests’ or ‘preferences’ and where good government is a matter of providing options to choose from to rational individuals.

When applying market standards of worth, the unemployed are typically afforded a low level of worth in cases where the individual is described as a cynical speculator who deliberately chooses unemployment because of its financial pay-off. The UK material contains numerous vivid examples of such negative market evaluations:

With recourse to the market order of worth, the unemployed individual is thereby depicted as a cynical, calculating merchant, who weighs the financial pros and cons and who finds unemployment benefits and social assistance to be more attractive than labour market participation. Unemployment is, by some interviewees, described as an individual ‘choice’, made attractive by high unemployment benefits:

When making such moral evaluations based on market standards of worth, income levels stand out as the most important object since the reference to income levels makes it possible for informants to compare the income of the unemployed directly with salaries of workers:

While being generally less visible in the discussions from Copenhagen, Danish participants do cling to comparable evaluations based on market worth. For example, numerous Danish interviewees state that ‘work must pay’, that some ‘take advantage of the benefit system’ and ‘can’t be bothered to work’ because of the generous benefits or that some people cynically and deliberately commit benefit fraud and thus ‘cheat’ the system. In particular, Danish participants pinpoint the lack of ‘responsibility’ of the unemployed, ‘demanding to be supported when having himself to blame for the situation.’ The situation of a person, in this view, is ‘a consequence of former actions’ as ‘you need to organise your life so that you can say ok, well, I can accept the risk of becoming unemployed’:

Participants in both countries do, however, also use market standards of worth to evaluate the status of the unemployed more positively. Rather than portraying the unemployed as deliberatively speculating cynical merchants, these more positive evaluations tend to portray unemployment as a natural and logical result of imbalanced financial incentive structures for which the unemployed individual is not to blame. Several British participants refer to poor working conditions in the labour market—‘low pay’, ‘insecurity’ and ‘zero-hour contracts’ with ‘no guaranteed income’—as being so unattractive that it logically keeps people out of employment. As stated by one participant presenting a policy proposal during the plenary sessions: ‘The current system effectively encourages people to stay out of employment.’ In such statements, the blame is not placed on the unemployed but on the system, which is believed to effectively encourage rational citizens to stay out of employment:

When applying domestic standards of worth, participants implicitly tend to think of society in terms of a familiar community where well-off members protect the less well-off (Nielsen, 2018). Thus, the familial analogy refers less to blood ties than to the fact of belonging to the same household as a territory in which the relation of domestic dependence is inscribed (Boltanski & Thévenot, 2006). Worth is ascribed to the humble and obedient unemployed individual, who respects the norms, values and traditions of the community to which they belong and understands their specific rank within a larger hierarchical chain.

Low worth, particularly prominent in the data from Britain, is allocated to the unemployed who misbehave and fail to respect the community within which they are placed. Status is less concerned with money and varying income levels and is more concerned with decency and attitudes, as illustrated in the following exchange of words between two British informants:

Numerous statements from British interviewees stress the bad behaviour of the unemployed as ‘lazy’, completely lacking ‘discipline’, ‘lacking a focus in their life’, having ‘a bad attitude’, a ‘weak mentality’, no ‘work ethic’ and ‘just wanting to stay on the dole’. Young unemployed people are generally referred to as a particularly badly behaved group, and they are paternalistically described as subjects in need of discipline:

Danish participants use domestic standards of worth to assign low worth to unemployed people as well, although less prominently. In particular, concern is expressed that some able-bodied unemployed, due to a lack of discipline, might actually not be at the disposal of the labour market. ‘Lazy Robert’—a social assistance recipient who stated that he would rather receive social assistance than having a ‘crappy job’ (see Hedegaard, 2014) —is used by some participants as the epitome of such a character: a ‘lazy’ and ‘work-shy’ scrounger’ who ‘takes advantage’ of the community of which he is a part. Some participants propose ‘some form of control, since control is better than trust’. In some discussions, it leads to consideration of whether or not to introduce stricter demands than we already have:

Danish participants, furthermore, pinpoint the ingratitude of those who apply for a disability pension, even though they possess resources that might be applicable to the labour market:

Resentment towards granting a disability pension is occasionally even paternalistically justified as being best for the unemployed: Participants do rely on domestic standards of worth to portray the unemployed more positively as well. British participants particularly tend to do so during combined discussions on unemployment and migration. Such discussions tend to transfer the label of ‘the other’, who breaks with established norms of decency, from the unemployed to immigrants. In such discussions, the boundary separating ‘us’ from ‘them’ is no longer drawn between employed and unemployed but between British nationals and non-British nationals. The unemployed individual is even described as ‘one of ours’ who is incapable of getting a job because foreigners jumped the queue in the labour market. The boundary separating ‘us’ from ‘others’ is thus redrawn, and in order to solve the problem of unemployment, some British participants propose different variants of ‘stricter border control’ or ‘deportations’ (‘if they come here and they are naughty, send them home’). A group of participants even developed a policy proposal, making it mandatory for British firms to interview all or at least a quota of British national applicants when recruiting new employees.

Among the Danish participants, the more positive domestic evaluations of the unemployed are somewhat different. Rather than transferring the label of ‘other’ to groups situated below the unemployed on the ladder of deservingness (see Van Oorschot, 2000), such positive evaluations tend to question whether the indecency stems from the system rather than from the unemployed individual. During some of these discussions, the unemployed individual is described as a tormented member of the flock, wrongfully exposed to the indecency of the unemployment system. As stated by one:

Worth is still measured in terms of decency and respect, but disrespectful and indecent behaviour is no longer associated with the unemployed but with the unemployment system. In particular, Danish participants pinpoint the indecency of the system when it comes to granting disability pensions, clearly pointing to an indecent treatment of those who are legitimately ‘needy’.

The third order of worth which is dominant in the empirical material is industrial logic, which perceives society as an ‘organised machine’ whose parts are organs that fulfil different ‘functions’ (Boltanski & Thévenot, 2006). Society is thus understood in terms of a ‘general factory’, and persons and entities are evaluated in terms of efficiency. Relying on such industrial standards of worth, participants understand and evaluate the unemployed neither in terms of financial incentives nor in terms of decency but according to their skills. Worth, when perceived by industrial standards of worth, is ascribed to the skilled and employable worker. British participants, relying on industrial standards of worth, perceive the unemployed as unworthy when stating how they are characterised by a distinct lack of useful skills. As stated by one:

The major problem of the unemployed has nothing to do with money or decency but with the fact that an individual lacks competences, skills and experience: Thus, unemployment is constructed as a result of a mismatch between qualifications demanded by vacant labour market positions on the one hand and the qualifications possessed by the unemployed on the other. An important distinction is drawn between having skills generally and having ‘the right skills’ to make yourself employable. British participants mention the ‘smart academic’ who has studied for many years but who is still unemployed and ‘doesn’t become functioning in society’. As stated by one: Thus, the unemployed individual is ascribed low status when they fail to display the right skills needed to be employed. One apparent tool to solve this problem of unemployment is to make the right investments in human resources to avoid wasting too much money and to get as large a pay-off as possible:

Although less prominently, some Danish participants rely on similar negative industrial moral evaluations that depict unemployment primarily as a result of a lack of skills or a wrongful investment in useless skills. In particular, Danish participants discuss whether we, as a society, invest too much money in granting people university degrees that lead them directly to unemployment. As illustrated by the following quote,

A couple of Danish participants mention the need for ‘upgrading skills’ and for ‘offering people help with creating a CV and so’. However, a more positive industrial evaluation of the unemployed individual as a person who is employable and who has sufficient competences is quite prominent in the Danish material. A common concern is that the existing competences and skills of unemployed Danes are threatened by ‘pointless courses’ and training programmes at the job centres: These courses are described by several participants as investments that do not deliver any pay-off: ‘I was unemployed last year, and I honestly think that what I got out of it wasn’t worth anything’. Job centres, in such a world, are denounced as institutions that clearly lack efficiency—being characterised by ‘red tape’, ‘Kafkaesque bureaucracy’, ‘too many rules, akin to ‘a circus’ and ‘a waste of resources’ that prevent the unemployed from achieving the skills required by the labour market. A similar narrative cannot be located in the British data material.

The analysis have investigated how ordinary citizens establish the worth of unemployed persons when discussing politics and argued that the pragmatic sociological order-of-worth-framework provides a fruitful conceptual framework for doing so. Particularly, the article demonstrates how the framework can encapsulate moral complexity, contrary to the moral singularity present in many existing studies of the discourses of unemployment. Rather than simply distinguishing positive from negative (or stigmatising from nonstigmatising) discourses, the order-of-worth framework makes it possible to analytically grasp how discussions of unemployment unfold between a plurality of qualitatively different moral worlds, each depicting and assessing the unemployed and the concept of unemployment in relation to its own specific moral standards of worth.

In both of the two country settings analysed in the article (the UK and Denmark) participants predominantly evaluate the status of the unemployed in terms of market worth, domestic worth and industrial worth, leading to three quite different ways of describing the unemployed positively as deserving and, contrarily, three different ways of depicting the unemployed negatively as undeserving. Table 2 sums up these three different understandings of unemployment.

The analysis portrays overall discursive similarities as well as clear differences between the material from the two countries (see also Frederiksen, 2018). Thus, market, domestic and industrial evaluations were dominant in the two cases. Despite these overall similarities, the three orders of worth were applied quite differently by participants in the two countries. This overall finding echoes results from pragmatic sociological comparative studies of quite different topics. For instance, Thévenot, Moody, and Lafaye (2000) sum up their comparative case study of environmental conflicts in France and the US by stating that evaluations based on market standards were more prominent in their US case than in the French one and (more interestingly) that orders of worth are applied differently in the two countries.

Something similar can be concluded from the present analysis. Thus, British participants were generally much harsher and more negative in their depiction of the unemployed as undeserving than our Danish informants and the orders of worth were applied differently in the material from the two countries.

For instance, different objects were mobilised by participants in the two countries to justify their statements. For each country, when certain topics and certain objects entered the discussion, participants were somehow inclined to think of the unemployed less as an undeserving other and more as a deserving but tormented member of the flock. For British participants, this was particularly evident when discussing immigrants in relation to unemployment, leading to the blame being placed on immigrants rather than on the unemployed as such. For Danish participants, this was particularly evident during discussions of the Danish jobcentre system, leading to the blame being placed on ‘the system’ rather than on the unemployed individual.

This connotes an important difference in the application of industrial worth between the two countries: whereas Brits tended to define the unemployed as persons who lack competences, Danish participants quite commonly spoke of the unemployed as persons who already have competences that are threatened by the incompetence of the Danish jobcentres. Connecting to the theory of deservingness we can furthermore say that British participants tend to speak of the unemployed as persons in control of their situation whereas Danish ones are more inclined to speak of unemployed as persons who have no or little control of their situation. The most ‘negative’ statements in the UK case follow market logics – and portray unemployed people as cynically speculating scroungers, in control of their own situation, who deliberately take use of the system – whereas the most ‘positive’ statements in the Danish case follow industrial logics and portrays unemployed people as competent and well-intentioned beings who have lost control of their lives due to inefficiency of the jobcentre system.

The Danish case additionally contain some civic arguments –emphasising solidarity and forming compromises with domestic and industrial statements – whereas such a form of argumentation was almost completely absent from the UK material. These findings are in line with previous comparative research, stating that perceptions of the unemployed and the poor are harsher in the ‘selective’ welfare regime of the UK than in the ‘universal’ Danish one (Larsen, 2008). They add, however, to the insights of such previous research as they nuance the picture and shed light on some of the moral evaluations leading to the positive and negative depictions.

Finally, the different orders of worth described in the analysis seem to echo the complexity of modern-day unemployment policies. Rather than being simply a struggle regarding whether to raise or lower a certain benefit, workfare polices are fundamentally about how to understand the complex problem of unemployment (see Hansen, 2019). Present policies are woven mixtures of, for example, incentive policies, disciplining policies and investments in the human capital of the unemployed subject. Everyday disputes of citizens over the problem of unemployment seem to mirror this complexity (Caswell & Larsen, 2020).

In order to capture, analyse and compare the moral complexity of such disputes, we need two things in particular. Firstly, we need methods that enable us to grasp and analyse what people say during such disputes. I argue that deliberative forums have something to offer in this regard as they are much less researcher-controlled than the most commonly applied methods (e.g. surveys and qualitative interviews) and faces participants with an imperative of justification. Secondly, we need concepts allowing us to somehow recognise the rich moral diversity inherent in such discussions while still making it possible to compare results to previous and forthcoming findings from other studies. In that regard, pragmatic sociology certainly has something to offer with its limited plurality of different forms of moral evaluations.

No potential conflict of interest was reported by the author(s).