Representing a Fading Welfare System that Is Failing Young People in ‘NEET’ Situations: a WPR Analysis of Swedish Youth Policies

The situation of young people who are neither in employment, education nor training (referred to in political, scientific and public discourses as ‘NEETs’) has received widespread attention during the last decade. However, while policy responses to young people’s work- and school-related marginalisation have been analysed by international scholars in a variety of contexts, to the best of our knowledge, no study to date has scrutinised problem representations of ‘NEET’ young people in youth policies in Sweden. To bridge the current knowledge gap and uncover taken-for-granted assumptions about the otherwise largely unchallenged Nordic welfare model, the aim of this research was to explore how the ‘problem’ of ‘NEET’ young people is represented in Swedish policies and policy proposals. To facilitate this, a discursive approach to policy analysis was adopted, following Bacchi’s ‘What’s the Problem Represented to be?’ (WPR) methodology. By focusing on the solutions that have been proposed to reduce the size of the ‘NEET’ group in two selected policies, four problem representations were developed. These connect the ‘NEET’ problem in Sweden, at the general level, to the ‘vulnerability’ of young people on the margins of education and employment (especially certain sub-groups) and, more specifically, to the failure of a fading welfare system to provide services and support for these ‘vulnerable’ subjects. Beyond representing the ‘problem’ along these lines, the identified problem representations may contribute to silencing young people’s agency and ignoring the consequences of a growing labour-market precarisation in Sweden, while failing to provide a basis for equity and social justice.


Introduction
The acronym 'NEET' ('not in employment, education or training') aims to capture a diverse group of young people, aged 15 to 29 years, who face challenges in their school-to-work transition. Increasingly used in political, scientific and public realms, this concept constitutes a good example of how policies (or policy proposals) give rise to 'new' problem representations and new social categories. First introduced in the UK, 'NEET' was adopted by scholars in the late 1990s as a way to monitor the extent of young people's disadvantage at a time when 16-to 18-year-olds were denied recognition as unemployed workers and removed from official statistics (Furlong 2006). In the current study, we follow international scholars who since then have criticised applications of the concept for assuming that this youth group is homogeneous in terms of abilities and experiences; that linearity and consistency characterise their life trajectories; and, ultimately, that they should be approached from the perspective of what they are not (i.e. in education, employment or training) (Thompson 2011;Maguire 2015;Simmons and Thompson 2013;Smyth et al. 2014;Yates and Payne 2006).
In line with the above criticism, studies focusing on the interplay between structure and agency have added many important nuances to our understanding of young people who are in 'NEET' situations. Specifically, to paint a picture that is less deterministic and degrading, research has indicated that the situation of this group may be characterised by low expectations of fulfilling 'normal' aspirations (Finlay et al. 2010); by sensed isolation and alienation from the authorities and communities that largely fail to account for (or attend to) their needs and aspirations Maguire 2018;Nairn and Higgins 2011;Russell et al. 2011;Haikkola 2018); and by pressure to follow standardised life trajectories in a bureaucratic welfare system that is difficult to comprehend (Görlich and Katznelson 2018). Other scholars have discussed how awareness, acceptance, avoidance and self-exclusion may all be ways through which these young people seek refuge from external pressures and exploitative or precarious conditions where opportunities for meaningful progression and participation are scarce Reiter and Schlimbach 2015;Russell 2013). Ultimately, as concluded by Rikala (2019 p. 14), the lives of these young people appear to involve 'everyday struggles of surviving, varying strategies of getting out, forms of everyday resistance, and the agency exercised for collective goals [which] all include negotiations over the ideals and norms of a work-centred society'.

Youth Policies and the 'NEET' Problem
As time has passed since 1999, when the 'NEET' term was first established in the UK (Eurofound 2016), several policies have been launched at European and national levels by the EU member states to reduce the proportion of young people who are neither in employment, education nor training. Concurrently, scholars have analysed these policy responses indicating, for example, that the recently reinforced Youth Guarantee (European Commission 2020) may be insufficient to support sub-groups in the Nordic countries that are 'hard to reach' (Mascherini 2012) while failing to tackle the 'NEET' problem more generally in Southern Europe (Pesquera Alonso et al. 2021). At the same time, the national implementation of the scheme has been criticised for being de-contextualised and for focusing on enhancing individual 'employability' rather than on creating opportunities for education and employment (Tsekoura 2019).
In the Spanish context, Strecker et al. (2021) have recently highlighted how policies like the EU Youth Guarantee build upon deficit discourses of 'NEET' young people, which underpin the provision of individualistic solutions that stigmatise them as lacking motivation and skills. Similarly, when analysing discrepancies between political discourse and lived experiences of 'NEETs', Gjersøe and Leseth (2021) have described how Norwegian policies portray the path to employment as speedy and linear which contrasts with the long and winding road that the young people experience. In the UK and Scotland, McPherson (2021) has also depicted how 'NEET' young people tend to be constructed as (risky) economic subjects within policies that foreground their personal responsibility for lacking the skills needed to meet market demands (i.e. 'employability'). Brunila et al. (2016) furthermore argue that individualistic government strategies in Finland may reinforce, rather than redress, the social exclusion of this youth group if underpinned by an intent to help them cope with normatively negative emotions and adversity. Additionally, Helms Jørgensen et al. (2019) have described how coercive methods have gained prominence in policy discourses within the universalistic Nordic regime during the last twenty years (albeit less so in Sweden than in Denmark and Finland), thus making young people increasingly responsible for the success of their schoolto-work transition.
In relation to the above body of European youth policy research, of particular relevance for the current study is Mertanen et al's. (2020) recent analysis. By applying Bacchi's (2009) 'What's the Problem Represented to be?' approach, the authors suggest that Finish youth policies and their implementation (re)produce two discourses-the lack of young people's participation in economic activities and the immaturity of young people-which are grounded in paternalistic and neoliberal governing practices that represent young people as 'vulnerable'. In this regard, vulnerability may imply certain subjectivities for 'NEET' young people. While it can emphasise their right to participate in society on equal terms and improve service delivery by legitimising well-meaning interventions, vulnerability discourses may also locate the grounds for vulnerability within the vulnerable subject, contributing to further stigmatisation while leaving structural causes and conditions unaddressed (Te Riele and Shelley 2021, Brunila et al. 2016, Brown 2017, Brown 2012, McLeod 2012. In Sweden, analyses similar to that of Mertanen et al. (2020) have so far been absent, despite the situation of 'NEET' young people becoming a priority among Swedish policymakers during the last decade. Considering that Sweden, similarly to Finland, exemplifies a largely unchallenged Nordic welfare model characterised mainly by publicly funded support systems and services, critical policy inquiries become especially salient since they may unearth taken-for-granted assumptions about government responsibility for equal opportunities and equity (Ecclestone and Brunila 2015;Mäkelä et al. 2021). In this regard, the current lack of Swedish policy analyses in the area is not only surprising, but also problematic. To address this knowledge gap, we have adopted a discursive approach to policy analysis with the aim of exploring how the problem of 'NEET' young people is represented in Swedish youth policies and policy proposals. Following the seminal work of Carol Bacchi (2009), two policies have been interrogated starting from the premise that policy proposals give shape to problems by containing implicit portrayals of what the 'problem' is represented to be. Subjecting these 'problem representations' to scrutiny is an important endeavour because problem representations can justify or legitimise certain policies or policy solutions. It is also essential because certain problem representations can contribute to our understanding of members of the targeted group (in our case 'NEET' young people), while also having direct or lived effects for them (Bacchi 2009).

Materials and Methods
The concept and classification of 'NEETs', referred to in Swedish as 'UVAS' (Unga som Varken Arbetar eller Studerar), was first introduced in Sweden in the late 2000s. Since then, no overarching legislation or act has been directed towards the group, despite the Government commissioning inquiries and the Swedish Agency for Youth and Civil Society to examine the situation of 'NEET' young people and propose solutions to address the 'problem'. After going through the six official reports (henceforth referred to as policies) produced as a result of these investigations, we decided to analyse two of them in more detail. Specifically, we chose to focus on the following: a) The Swedish Government Official Report titled 'Our common responsibility -for young people who neither work nor study' published in 2018 (SOU 2018:11) b) The report titled 'Nationally coordinated support for young people who are neither working nor studying' published in 2021 (Swedish Agency for Youth and Civil Society 2021) These policies were selected because they included the most recent official examinations of, and solutions to, the 'NEET' problem in Sweden by the two bodies of government (the Swedish Government Official Report and the Swedish Agency for Youth and Civil Society) that so far have been responsible for conducting investigations in the area.

The Analytical WPR Method and Its Theoretical Underpinnings
In order to scrutinise problem representations within the two selected policies, while reflecting upon assumptions that underpin these representations of the 'problem', the silences they contain and the effects they may have, we used Bacchi's (2009) 'What's the Problem Represented to be?' (WPR) approach. As adopted in this study, the WPR approach builds upon a Foucauldian view of problematisation: a concept that Foucault (1991) uses in a dual way to: (1) analyse how things are constructed as problems in specific ways and (2) scrutinise taken-forgranted assumptions in order to open them up for change. Based on this notion, the WPR approach considers it impossible to represent, or have contact with, a valuefree or interpretation-free reality. This means that policy proposals/solutions are seen as a set of practices that enter a terrain of competing constructions of truth and falsity (Bacchi 2009). The representation of problems in policies should therefore not be seen as objective descriptions, but as contested claims about the nature and existence of social problems (Bacchi 1999). While this means that policy development affects, and is affected by, dominant discourses, it does not imply that the authors or institutions behind the policies have particular intentions beyond the problem they have set out to address.
In contrast to conventional understandings of public policy as the government's way of solving fixed and identifiable 'problems' that sit outside of the policymaking process, at the heart of the WPR approach is the idea that policies give shape to 'problems'. Beyond providing a structure for scrutinising representations of problems in policies, identifying their underlying assumptions, reflecting upon the social developments on which they may be contingent (genealogy) and considering issues that are silenced or fail to be problematised, Bacchi (2009) explains three effects-discursive, subjectification and lived-that 'problem representations' can have. While discursive effects include notions about what can be thought or said in relation to the problems, subjectification and lived effects capture the subject positions that become available in, and material consequences of, the specific problem representations, respectively.

The Analytical Process
To facilitate the exploration of how a certain 'problem' (in our case 'NEET' young people) is represented in policies and policy proposals, combined with analyses of the underpinning assumptions, genealogy, silences and potential effects, Bacchi (2009 p. 2) provides a step-by-step approach to policy analysis comprising six questions, of which we have used questions 1-5 as outlined in Table 1. Specifically, following Bacchi's (2009 p. 3) idea of 'working backwards' from concrete proposals, we first individually read and re-read the policies in a first analytical step, while taking extensive notes focusing on identifying solutions proposed in the text. We then met to discuss insights from these initial readings. In the second step, all solutions identified in the policies were then compiled and analysed in detail using Bacchi's (2009) question 1 while considering explanation to, and descriptions of, the solutions outlined in the policies. This process resulted in four candidate problem representations.
In the third step, the candidate problem representations were then considered in relation to WPR questions 2-3. This meant that we discussed and analysed them in light of previous empirical and conceptual research to understand their underpinning assumptions and to trace a few societal developments (i.e. ones that exists outside of the specific policies) that may have contributed to their formation. Following questions 1-3, this iterative analysis of both readings and writings, resulted in the formation of four problem representations. These problem representations were then analysed in a fourth step using questions 4 and 5, which meant that we did a close reading of the problem representations while discussing, depicting and describing silences and effects that they might produce.

Findings and Discussion
By analysing the solutions proposed in two Swedish policies directed towards young people in 'NEET' situations, we identified four problem representations following questions one to three (Q1-Q3) of the WPR approach: 'a lack of systematic and scientific knowledge', 'weak welfare collaboration and coordination', 'a need for more direct and indirect welfare support' and 'inadequate welfare services available to certain sub-groups'. After presenting these problem representations in the first section of this combined findings and discussion section, we continue to describe the silences and effects that they might produce addressing WPR questions four and five (Q4-Q5).

The Problem Representations, Underlying Assumptions and Genealogy
Although the organisations responsible for the two policies differed, the proposed solutions and subsequent problems representations were very similar. This meant that government bodies that were more youth-oriented (i.e. the Swedish Agency for Youth and Civil Society) did not reflect contrasting or opposing discourses in comparison to discourses in the Swedish Government Official Report, as illustrated below. Table 1 Outline of the WPR questions and how they were applied in the analysis

WPR questions
How the questions were applied in the analysis 1. What is the 'problem' represented to be in Swedish policies about 'NEET' young people?
Solutions were identified through multiple readings of the policies. The solutions were then compiled, analysed and thematised into candidate problem representations using the first question 2. What presuppositions or assumptions underlie these representations of the 'problem'?
The candidate problem representations were analysed in light of previous empirical and conceptual research guided by questions two and three. This analytical process resulted in the identified problem representations 3. How have these representations of the 'problem' come about?
4. What is left unproblematic in these 'problem representations'? Where are the silences? Can the 'problem' be thought of differently?
The identified problem representations were scrutinised and discussed in light of questions four and five, resulting in silences and potential effects that they might produce 5. What effects are produced by these representations of the 'problem'?

A Lack of Systematic and Scientific Knowledge
Register-based statistics about young people who are not working or studying should be produced by Statistics Sweden to strengthen knowledge about the group. (SOU 2018:11, p. 20) The above quote provides an example of how the policies under scrutiny called not only for more information about various aspects related to 'NEET' young people, but also for information of certain types (or from particular sources), thus representing one of the 'problems' as being a lack of systematic and scientific knowledge (Q1). Specifically, beyond stressing the need to develop databases, models and systems for monitoring school absence, grades and other indicators, coupled with an emphasis on assessing the effects of initiatives and interventions, the policies drew attention to specific kinds of knowledge, namely the descriptive, statistical and standardised, in addition to the evidence-based.
Overall, this problem representation should be seen in relation to ongoing knowledge production in the area, given that the body of international research analysing the situation of 'NEET' young people has grown rapidly since the 1990s (Q3). At the same time, Nordic studies (Holte et al. 2019) and Swedish reports (Forslund and Liljeberg 2021) have contributed insights into risk factors for, and characteristics of, this demographic segment of young populations. However, beyond building on the assumed benefit of producing knowledge about young people in the process of cultivating their 'employability' (Mäkelä et al. 2021) (Q2), the proposed solutions show how the urge for more knowledge (or as Foucault calls it 'the will to knowledge') contains ambitions to categorise, monitor and govern. 1 These ambitions, especially monitoring and governing, were explicit in solutions stressing the need for statistics and register-based information about characteristics of the 'NEET' group (e.g. who, and how many, they are) and about risk factors for early school leaving (e.g. grades and school absences). Rather than representing a call to enhance young people's market-centred skills, the policy's focus on systematic knowledge thus seems to be more strongly underpinned by the vulnerability discourse (te Riele and Shelley 2021) indicating that a lack of descriptive data makes it difficult for welfare actors to identify young people in need of support (Q2). Specifically, by depicting governing practices focused on intervening in the present lives of these young people to align their future trajectories with a normative route of successful subjects (Haikkola 2018; Hodgson 2019; Helms Jørgensen et al. 2019), similarly to Mertanen et al. (2020), we interpret the policy solutions around (a lack of) knowledge as shaped by paternalistic and individualistic ideals that represent young people as vulnerable.
Adding to the above emphasis on systematic knowledge in areas related to 'NEET' young people, this first problem representation is also explicitly tied to a lack of the scientifically informed knowledge known as 'evidence-based practice' (EBP). In this regard, the policies stated that our understanding of 'NEET' young people has grown, while at the same time stressing that initiatives directed towards the group are not sufficiently guided by 'the basics of EBP' (SOU 2018:11, p. 168). This call can be interpreted as reflecting dominant ideas about the benefits of systematic and scientific information within the framework of national knowledge governance in social work (DS 2014:9) (Q2). However, it should also be seen in relation to ongoing developments in Sweden, where serious (and sometimes radical) attempts have been made to implement EBP to align with 'current trends of neoliberalism and managerialism, promoting efficiency and effectiveness in the delivery of care' (Jacobsson and Meeuwisse 2020 p. 279) (Q3). Following these movements, the policy emphasis on EBP can be interpreted as emerging from views that national levels of social work practice are 'riddled with flaws and carried out on whims or intuition' (Lauri 2016 p. 59) and thus (undesirably) reliant on experience-based knowledge rather than systematic or scientific knowledge (Q2).

Weak Welfare Collaboration and Coordination
The government should stimulate collaboration between regions and municipalities in order to strengthen regional-level coordination for young people who are neither working nor studying. This can be achieved with support from the Swedish Association of Local Authorities and Regions (SALAR). (...). The actors to be involved in such a collaboration include compulsory and upper secondary schools, as well as student health, leisure organisations, social services and healthcare. (Swedish Agency for Youth and Civil Society 2021 p. 24) As exemplified in the quote above, the policies also called for improved steering of government agencies, such as SALAR, in order to promote interaction between regional and municipal actors, while stressing in particular the need for coordinated support from national to provincial and local levels. In this regard, the policies also represent the 'problems' as weak welfare collaboration and coordination (Q1).
On the one hand, we interpret the emphasis on public actors collaborating and coordinating as reflecting a widespread belief in the value of largely autonomous organisations coming together to 'share risks, pool resources and allocate benefits evenly' (Costumato 2021 p. 250) as a way of avoiding unnecessary overlap, redundancy and fragmentation while fostering efficiency, effectiveness and accountability (Q2). In light of the growing decentralisation, combined with new public management reforms that have been widely implemented in Sweden since the 1990s, it is thus likely that these solutions are a response to the increased corporatisation (management by objectives and performance) and projectification (growing reliance on project organisations) of the Swedish public sector (Hall 2013; Fred 2019) (Q3). In relation to this, Löfström (2010) has described how demands for improved quality at lower cost, coupled with organisational structures that differentiate between (and within) the state, regions and municipalities, has contributed to a Swedish public sector that is governed by a body of largely distinctive, specialised and sometimes even competing units. As a result, policy proposals to address the 'NEET' problem through strengthened collaboration and coordination can be understood as both a consequence of, and a solution to, a complex Swedish welfare system that has become increasingly characterised by boundaries (cf. Löfström 2010) (Q3).
On the other hand, the policy call for collaboration and coordination can be seen as underpinned by views of the 'NEET' problem as a 'wicked problem'-or, more specifically, as a social issue that is open ended, uncertain and beyond the capacity of any single public actor to resolve (Head and Alford 2013) (Q2). In other words, moving towards integrated services that rely on partnerships across traditional organisational and professional boundaries (Valentijn et al. 2013) may reflect notions about the inadequacy of specialised approaches to welfare delivery for vulnerable groups with complex needs, such as 'NEET' young people (Flagship School to Work 2020).

A Need for More Direct and Indirect Welfare Support
A requirement shall be introduced for principals to ensure that students at risk of leaving upper secondary school prematurely or without a diploma are offered a closing guidance meeting. The meeting shall be offered before the student has formally left school. If a physical meeting cannot take place, the same information must be provided to the student in writing. (SOU 2018:11, p. 231) As the quote above exemplifies, this third problem representation builds upon policy proposals which, in addition to calling for coaching and personalised care, direct initiatives of varying types towards 'NEET' young people by emphasising their need for information and guidance. In addition, solutions were also proposed in the policies to strengthen support for actors who, in turn, would support young people, thereby representing the 'problem' as a need for more direct and indirect welfare support (Q1).
Similarly to Mertanen et al. (2020) and in accord with the vulnerability discourse, we interpret the call for more 'direct' support as emerging from assumptions that 'young people can overcome their struggles and make decisions and choices that are in line with the norms and values of wider society' (p. 10) if they are provided with sufficient information and guidance (Q2). Relatedly, we can understand this focus as an appeal to the young people themselves based on the idea that, once they have gained enough knowledge about what to do or how to act (e.g. in order to complete upper secondary education), they will realise what is in their own best interests and act accordingly. A contradiction arises from these interpretations, however, since it seems that young people are considered ignorant or uninformed, and thus in need of management and surveillance, while at the same time being expected to 'learn' or internalise normative aspirations for school-towork transitions. Nevertheless, this discrepancy aligns with Hodgson (2019), who argues that young people who are considered vulnerable tend to be surrounded by both paternalistic and neoliberal ideals. These ideals legitimise top-down surveillance and social control while simultaneously expecting individuals to take personal responsibility for behaviours, actions and attitudes that do not align with the norms of a work-centred society.
Beyond calling for initiatives to increase the 'direct' support to young people, the policies outlined various ways through which the role and responsibility of welfare actors primarily within, or close to, formal education, could be strengthened. In light of ongoing discussions about the disadvantages of an increasingly marketised education system in Sweden (Wennström 2020), one might have expected the 'indirect' solutions to be aimed at improving the functioning of compulsory and upper secondary schools. However, instead of developing such proposals, the emphasis was placed on strengthening the support to 'NEET' young people by calling for a greater distribution of resources to popular education. In line with the vulnerability discourse (Brown 2017;Brunila et al. 2016), we interpret this emphasis as a response to the assumed potential of folk high schools to reduce educational inequalities by creating learning opportunities for, and promoting the personal development of, groups whose aspirations and needs tend to be neglected within the formal education system (Rubenson 2013) (Q2).

The Inadequate Welfare Services Available to Certain Sub-groups
A national actor should develop models for strengthening the link between language learning programmes, further education and employment for young immigrants aged 15-24 years. This should be done in project format and in collaboration with a number of selected municipalities. (SOU 2018:11, p. 239) In contrast to the previous problem representations, which locate the 'problem' mainly within the realm of a fading welfare system that is failing vulnerable 'NEET' young people more generally, this fourth one was developed from (the few) solutions that positioned specific sub-groups, namely, young immigrants and young people with disabilities, at the centre. Specifically, the policies and proposed solutions highlighted the challenges faced by young immigrants in language-learning programmes, as exemplified in the above quote. By also calling for a national coordinator to 'improve society's joint ability to increase the proportion of young people with disabilities who transition from education to employment' (SOU 2018:11, p. 244), this last 'problem' was represented in the policies as the inadequate welfare services available to certain sub-groups (Q1).
The specific policy focus on young immigrants can be traced back to 2015, when a growing number of refugees sought protection in Europe (Q3). In line with Fejes and Dahlstedt (2017), we consider the call to enable transitions from language introduction to formal education and employment as a reflection of dominant discourses in which social integration in Sweden has become increasingly contingent upon certain abilities and (Swedish) language skills. Relatedly, this emphasis should be seen in light of language learning courses in popular education becoming a central policy tool used in Sweden to facilitate the social and labour-market inclusion of immigrants and refugees (Ministry of Education 2015). The fact that the proposal exemplified above is also situated within a project-based framework not only mirrors the growing public sector projectification in Sweden (Fred 2019) (Q3), but it also appears to rest upon the idea that complex long-term commitments, such as integration, can be resolved with short-term and task-focused solutions (Abrahamsson and Agevall 2010) (Q2).
With regard to the policy proposal directed towards young people with disabilities, a key element seems to be the idea that 'society' will be unable to help everyone as indicated by the emphasis on 'increase the proportion' (SOU 2018:11, p. 244). Hence, reducing the size of the group by supporting some will be sufficient (Q2). We interpret such a partial approach as a response to the fact that upper secondary school for young people with learning disabilities [gymnasiesärskolan] is not preparatory for higher education in Sweden (Q3). This means that the employment barriers for this group are high and the route to paid work is particularly long and winding (Germundsson and Runesson 2014). Notably, instead of approaching this issue in relation to the labour market, which perhaps should be modified to include a broader range of talents and abilities (see also Smyth et al. 2014), the policy focus on young people with disabilities seems to be underpinned by the idea that society can facilitate the inclusion of some by making them 'fit' within existing institutions (Q2).

Silences and Effects
Beyond providing a structure for scrutinising representations of problems in the policies, the WPR approach draws attention to issues that are silenced in specific problem representations (Q4), and different effects of those representations (Q5) (Bacchi 2009). Against this backdrop, we suggest that the above-described problem representations contribute to at least two main silences: one concerning the agency of 'NEET' young people and the second concerning the labour market.
Young people's ability to take ownership of themselves and their lives, regardless of whether this implies reproducing or resisting existing inequalities, is central to our understanding of fractured youth transitions (Rikala 2019;Coffey and Farrugia 2014). By locating the 'problem' primarily at the level of the welfare system-in the knowledge, collaboration and services required to provide support to the 'NEET' segment of young populations-we suggest that the problem representations disregard the young people's capacity to question these practices (Mäkelä et al. 2021) and talk back to 'the deafness of an unbalanced politics' (Smyth et al. 2014 p. 492). Through the above representations of the 'problem', the policies thus risk silencing the agency of young people in 'NEET' situations by reducing them to vulnerable welfare recipients with complex needs who 'require restorative or compensatory action to ensure equality of treatment and recognition' (McLeod 2012).
Additionally, the problem representations risk silencing alternative interpretations of how the 'NEET' phenomenon may be understood in relation to larger societal changes characterised, for example, by the growing labour-market precarisation in Sweden (Gauffin 2020). Here, we agree with scholars who have stressed that young people's labour-market engagement can neither be reduced to a question of their vulnerability (at an individual level) nor be seen solely as a responsibility of the welfare state to support these vulnerable subjects (at an organisational level) (Smyth et al. 2013;Mäkelä et al. 2021). Instead, employment-especially of those on the margins of labour markets and education systems-needs to be understood as a complex structural issue that is highly contingent upon the availability of ('decent' 2 ) work which, in turn, is shaped by a range of institutional, demographic, technological and economic factors (ILO 2018). Moreover, the lack of policy attention being directed towards the labour market may also contribute to silencing the situation of young people in the 'missing middle' who are employed (i.e. not classified as 'NEET'), but only as part of the precarious labour force (Irwin 2020).
Although the policies under scrutiny considered labour-market aspects to be key for young people's employment, the Swedish Government Official Report specifically stated that such structural aspects extended beyond the scope of their mission (SOU 2018:11). Nevertheless, in line with the WPR approach, the ways in which the policies represent the 'problem' still contribute to our understanding of 'NEET' young people, and will have different discursive, subjectifying and lived effects for them. Adding to the above silences, we thus acknowledge that the identified problem representations may have a number of effects, as discussed below.
Firstly, (lacking) knowledge is often depicted as an impartial and apolitical issue. However, Foucault (1980) has argued that knowledge is tightly intertwined with power, and with governing, stating that 'it is impossible to govern a state without knowing its population' (p. 124). Representing the 'problem' of 'NEET' young people as a lack of systematic and scientific knowledge is therefore not a neutral choice, and neither can the knowledge production itself be unbiased. Instead, the power/ knowledge creates statistics and categories in which (implicit) definitions of what is normal and deviant are integral. The policies we have scrutinised may thus, in and of themselves, have subjectifying effects, simply by focusing on 'NEET' young people (and especially those with immigrant experiences or dis/abilities) while calling for descriptive data to identify them. Relatedly, the policy concern with 'NEET' young people more generally, and the desire for increased knowledge about them in particular, may have discursive effects by representing them as a vulnerable population sub-group in need of targeted governing and control (Mertanen et al. 2020). This may also contribute to stratifying discursive effects via processes of othering as emphasis is placed on features that deviate from the norms of a work-centred society (McLeod 2012).
Secondly, with reference to the policy emphasis on identifying young people through their grades and school absences, the risk factors for ending up in a 'NEET' situation (i.e. early school leaving) become the problem to be solved, rather than the circumstances that underpin young people's detachment from formal education. The 'problem' can thus be seen as reductionist and individualised. Within a neoliberal discourse where engaged school completers are the normative standard, this may, in turn, have discursive and subjectifying effects by encouraging a stereotypical and simplistic view of 'NEET' young people as troubled and responsible for their own 'failure' (Smyth et al. 2014). From such a vantage point, the lived effects in terms of alienation, avoidance or self-exclusion may come to further shape the lives of these young people, who tend to be well aware of the discourses that label them problematic and at fault (Nairn and Higgins 2011;Simmons et al. 2014;Reiter and Schlimbach 2015).
Thirdly, while the solutions focusing on grades and school absences may be understood as calls to improve the employability of young people at risk of ending up in 'NEET' situations, most policy proposals (and thus the 'problems') were located within the realm of a fading welfare system that is failing vulnerable young people subjected to social hardship and (actual or potential) harm. In relation to the section above, these problem representations can thus have alternative or parallel discursive effects by painting a picture of 'NEET' young people as victims of circumstances over which they have little or no control. Rather than being held personally accountable for their situation, in a discourse of vulnerability, these young people may be considered neither (fully) responsible for their situation nor to have the agency necessary for managing the difficulties in their lives (Brown 2012). This rationale builds on the idea that vulnerable subjects who are 'in the hands of social structures which cause morally unacceptable social harms' (Brown 2017 p. 669) need protection via state intervention and control. The problem representations can therefore have lived effects for 'NEET' young people by directing initiatives and assessments their way, while legitimising the provision of, and their access to, welfare services and support (Mäkelä et al. 2021, Brunila et al. 2016, Hodgson 2019, Te Riele and Shelley 2021).

Concluding Remarks
To the best of our knowledge, this is the first study exploring how the 'problem' of 'NEET' young people is represented in Swedish youth policies and policy proposals. By adopting a discursive approach to policy analysis using Bacchi's (2009) WPR methodology, we identified four problem representations. These connect the 'NEET' problem in Sweden to the vulnerability of young people (especially certain sub-groups) at the margins of education and employment in general and to the failure of a fading welfare system to provide services and support for these vulnerable subjects in particular. The findings thus partially contrast with previous research, which shows that international youth policies-including those of other Nordic countries-tend to emphasise the role and responsibility of young people, rather than of the welfare state, when trying to solve the 'problem' of fractured schoolto-work transitions (Helms Jørgensen et al. 2019, McPherson 2021, Mertanen et al. 2020. Nevertheless, regardless of whether or not the scrutinised policies were developed with good intentions to include well-meaning solutions, when situated within a vulnerability discourse that can be both stigmatising and tied to strong(er) state control (Brown 2012;Brunila et al. 2016), the policy proposals (and thus the problem representations) may not necessarily contribute to equity and social justice. Instead, echoing Brown (2017), we believe that narratives through which the vulnerability of all people can be seen in less essentialising terms may provide the foundations for a fairer and more just society, especially when combined with views of vulnerable young people as 'agentic, just as other citizens are' (p. 678).
Funding Open access funding provided by Umea University. This study was supported by the Swedish Research Council for Health, Working life and Welfare (Forte) under Grant . The funding bodies had no further involvement in the research process.

Declarations
Ethics approval Not applicable.

Competing Interests The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.