Problem figurations of ‘troubled families’ and interventions

Van Wel's (Reference Van Wel1992) studies of the history of intensive intervention with families in the Netherlands develops the concept of problem figuration to show the socially constructed nature of policy rhetoric and proposed interventions in defining problems, their causes and the mechanisms to be deployed. Van Wel identifies successive waves of problem figurations, with each figuration based on a critique of the failings of its immediate predecessor (as evidenced in the Coalition's framing of New Labour policies). This process has been explicitly recognised by Hayden and Jenkins (Reference Hayden and Jenkins2014) in their depiction of ‘troubled families’ as a ‘wicked problem’ that is continually reconceptualised and ‘re-solved’ depending on changes in government. Van Wel also highlights the essentially ahistorical nature of any contemporary problem figuration, which, for example, fails to acknowledge the long genealogy of state intervention with vulnerable families, such as the Family Service Units that developed during the twentieth century (see Welshman, Reference Welshman2012). Van Wel (Reference Van Wel1992) argues that the frameworks of intervention (for example, Family Intervention projects or the Troubled Families programme) may be coherent and rational as an aligned governmental response to the narrative political construction of ASB and families (supported by the claimed scientific and empirical basis of government's use of research evidence).

However, this rationality is also underpinned by assumptions and prioritisations that represent fictional, historically embedded images and representations of the nature of the problem and the targeted subjects of intervention (i.e. ‘troubled’ families) that reflect the ‘structure of bias’ in a given period and the wider mechanisms stigmatising troubled families in contemporary discourse (see Bond-Taylor, Reference Bond-Taylor and Pickard2014; Lister, Reference Lister2014). Van Wel (Reference Van Wel1992) and Welshman (Reference Welshman2012) argue that governmental responses to ASB and family vulnerability are not based on cumulative insight or the rational utilisation of increased knowledge. Indeed, evaluation research, despite its controversies and the government's claim to scientific authority, has yet to robustly establish the precise effectiveness or mechanisms of various forms of intervention (Van Wel, Reference Van Wel1992).

The problematic nature of the policy knowledge base related to ASB and the efficacy of interventions to address it have long been recognised. Prior (Reference Prior2009) highlights how the relationship between power and knowledge in this field is manifested through an official governmental empirical ‘reality’ that denies the ambiguities and gaps in knowledge and fails to account for the disjuncture between policy rhetoric and even officially constructed evidence. Family intervention projects have been a site of particular contestation between government and researchers, and indeed between researchers themselves, with critics such as Garrett (Reference Garrett2007) and Gregg (Reference Gregg2010) critiquing the methods used to undertake evaluations and the analysis and presentation of findings that exaggerate positive and progressive outcomes for households (see Batty and Flint, Reference Batty and Flint2012). Similarly, Hayden and Jenkins (Reference Hayden and Jenkins2014) have critiqued the use of ‘policy-based’ evidence in governmental justifications for the Troubled Families programme.

The government continues to imbue its rationales for intensive interventions through a claimed empirical science of quantitative estimation. The rhetorical claim by Iain Duncan Smith (HM Government, 2012: 1) that there were ‘hundreds of thousands of individuals and families living profoundly troubled lives, marked by multiple disadvantages’, was operationalised in the identification of 120,000 troubled families underpinning the Troubled Families programme (Communities and Local Government, 2012), with local authorities provided with indicative numbers of troubled families in their own localities. Lister (Reference Lister2014) challenges the construction and methodology of this estimate, and it is not clear how it relates to other estimates, for example the 500,000 ‘forgotten families’ identified by the Riots, Communities and Victims Panel (2012). The government also constructs the definition of families to be targeted through its specific eligibility criteria of families with no adults in paid employment, a family member in or at risk of offending or a child with poor school attendance (HM Government, 2012). This constructs the private troubles of families that are to be the public troubles of governmental intervention. In doing so, it emphasises reduced problematic personal conduct, and improved education and pathways to employment that both New Labour and the Coalition framed personal responsibility within. These also constitute the priorities for intervention and the indicators that will be deployed to measure ‘success’.

The government articulates definitive costs associated with these families of £9 billion (or £75,000 per family), within a context of a reported 3.2m incidents of ASB in 2010–11 (Home Office, 2012), and claims to know that this figure ‘is likely still to be the tip of the iceberg’ (Home Office, 2012: 8). The government stated that its analysis showed that the Troubled Families programme should successfully ‘turn around’ 20,000 families by March 2015 (Department for Communities and Local Government, 2012: 10). Governmental updates state the success of the programme, with 62,000 families worked with and 22,000 families ‘turned around’ (Department for Communities and Local Government, 2013a, b). Lister (Reference Lister2014) identifies the power of the initial estimates of troubled families to frame the policy and intervention context, revealing how no local authority identified more troubled families than the central government estimate and how almost a quarter of local authorities identified the exact number of families estimated by central government. As Lister argues, these estimates appear to define the contextual ‘reality’ of the number of families needing support, driven by the financial imperatives built into the payment by results mechanism within the Troubled Families programme.

The government has claimed an authority to define the scale and nature of troubled families in England through the science of these statistics and through the establishment of the Early Intervention Foundation and publication of a good practice guidance (Department for Communities and Local Government, 2012) which purport to collate and disseminate the empirical evidence base and accumulated knowledge of effective practice. But there has been a striking reluctance to open up this ‘scientific’ analysis and interpretation to transparency and scrutiny. Freedom of Information requests to establish how the £9 billion/£75,000 per family figures were calculated failed to establish the methodology (Lister, Reference Lister2014).

These grey areas between research, and its social scientific basis, and the governmental problem figuration of troubled families, are epitomised in the report produced by Louise Casey, the Head of the Troubled Families Unit (Casey, Reference Casey2012), which was classified by the government as ‘dipstick/informal information gathering’ rather than ‘formal research’ (Bailey, Reference Bailey2012; Ramesh, Reference Ramesh2012). Bailey critiqued a shift from ‘evidence-based policy’ to ‘policy by dipstick’ (Bailey, Reference Bailey2012; Ramesh, Reference Ramesh2012), echoing a similar claim by Gregg (Reference Gregg2010) that evidence-based policy had been replaced by ‘policy-based evidence’. Casey's report presents an explicitly articulated ambiguity about its social-scientific status. It states that her interviews with sixteen case study families ‘do not pretend to be an exact science’ and that ‘this is not formal research and these interviews and the information they gave us is not representative’ (Casey, Reference Casey2012: 5). However, it also directly contradicts these caveats, stating ‘it was thought important to listen directly to troubled families in order to get a true and recent understanding of the problems they faced, their histories and what the real challenge of “turning around” thousands of such families nationwide would entail’ (Casey, Reference Casey2012: 4, emphasis added). The Secretary of State for Communities and Local Government, Eric Pickles, validated the report, claiming that it provided ‘real insights into these families’ lives’ and offered ‘a true understanding of the challenges local authorities face’ (Ramesh, Reference Ramesh2012). Casey claimed that she ‘wanted to get to know these families’, ‘when probed [answers] were clear’, that she ‘spent a long time listening to [the families]’ and that she had attempted ‘to [get] underneath the skin of these families’ (Casey, Reference Casey2012: 1–3). Leaving aside the volume of empirical and detailed research already undertaken with families (rather than one-off short interviews in an inappropriate and pressurised setting), Casey's report is significant for the narratives, prioritises and assumptions it articulates about troubled families. While Casey claims that ‘no judgements are made on these families’, this is precisely what her work serves to do.

Casey (Reference Casey2012: 1) identifies the centrality of ‘family’ as the causal site of problems and explicitly downplays wider structural factors: ‘several families talked of needing a bigger house from the council as a cause of problems for them, or of not getting enough free childcare, or they blamed teachers and schools for failing their children – when it was clear that their troubles were arising from their home life’ (Casey, Reference Casey2012: 51, emphasis added).

Structural factors are definitively dismissed (see Williams, Reference Williams2012, for a critique), despite the body of empirical work on intensive interventions identifying their prominence. Similarly, the complexity of interactions and vulnerability are summarised as ‘many of the people we interviewed were just not very good at relationships’ (Casey, Reference Casey2012: 48). The report highlights sexual and physical abuse; arson; a ‘majority’ of domestic abuse; incest and large numbers of children. It contrasts these families’ inability to ‘recover from and cope with’ episodes such as bereavement (Casey, Reference Casey2012: 3), failing to cite evidence showing that ‘other families’ are ‘not completely derailed’ by bereavement, and directly juxtaposes the case study families with ‘normal’ individuals (Casey, Reference Casey2012: 50). Casey's diagnosis confirms Bond-Taylor's (Reference Bond-Taylor and Pickard2014: 153) identification of the politics of the Troubled Families programme as manifested through a discourse of families as dysfunctional, inadequate, irresponsible and anti-social rather than disadvantaged, excluded and vulnerable. But, while Bond-Taylor (Reference Bond-Taylor and Pickard2014) also argues that the Troubled Families programme demonstrates more continuity than divergence with New Labour, the rhetorical construction of troubled families actually needs to be placed within the longer historical pervasiveness of these representations.

The report's argument that ‘intergenerational transmission of problems . . . was rife’ (Casey, Reference Casey2012: 46) locates this work in the long and still influential transmitted deprivation theory (Welshman, Reference Welshman2012) and tropes of deviant sexuality and domesticity consistently present since Victorian times (Wise, Reference Wise2009; Welshman, Reference Welshman2012). A focus on malingering, household squalor, inadequate parenting and financial management and failure to engage ‘appropriately’ with welfare services significantly influenced medical, social work professions and voluntary organisations’ engagement with families from 1940 onwards (Welshman, Reference Welshman2012).

Casey's report may be located within the broader problem figuration that spanned both New Labour and the Coalition, in which the family was identified as a site where ‘personhood’ is created and the complex inter-related patterns of vulnerability and poverty are acknowledged (Millie, Reference Millie2009a; Parr, Reference Parr2011; Gillies, Reference Gillies, Wagg and Pilcher2014), while simultaneously a focus on lifestyles, family dynamics and parenting are disassociated from structural determinants and lived realities (Morris and Featherstone, Reference Morris and Featherstone2010). Although Van Wel argues that problem figurations are often internally logical and rational, it is interesting to note that the gendered figuration of much anti-social behaviour discourse, focusing on the lack of maternal instinct or lone female parents, was not accompanied by gendered forms of supportive intervention that addressed the challenges of mothers or realigned the roles and responsibilities of fathers (Churchill, Reference Churchill2007; Holt, Reference Holt2009; Evans, Reference Evans2012).

From national problem figuration to the dilemmas of localised practice

The article now examines the importance of local policy interpretations in constructing meaningful practice within national problem figurations and policy frameworks (Bond-Taylor, Reference Bond-Taylor and Pickard2014: 142; Hayden and Jenkins, Reference Hayden and Jenkins2014). The findings in this section are drawn from an ESRC-funded study exploring the delivery of ‘Whole Household Interventions’ in a large northern English city. The research, conducted in 2012–13, comprised interviews with over fifty individuals involved in the delivery of the model being used in the city, including Multi Agency Support Teams (MASTs) managers and leaders, intervention workers from a range of agencies and organisations and seven families subject to whole household interventions. The city has been instrumental in pioneering family intervention initiatives and approaches since the early 2000s, including Family Intervention projects, resulting in a legacy of projects with different genealogies being delivered. This patchwork of projects is now framed within the architecture of the Troubled Families programme, but with a specific localised emphasis on a key worker–whole household approach as the main mechanism for delivering services to the most vulnerable families in the city, scaled up and mainstreamed through the work of MASTs.

In contrast to historic approaches, which typically involved multiple agencies working with families without sufficient coordination, shared knowledge of underlying issues or the resources needed to make a real difference, the whole household approach is delivered through three MASTs based across the city. These teams work with the whole family and are built on the principle of one key worker for each family. MASTs work with children, young people and their families to provide a range of services that aim to improve well being, school attendance, learning, behaviour and health care. Their additional remit is to signpost families to other services, and support and assist their engagement with these services.

The model adopted in the city appeared, therefore, to draw upon key learning from national research evidence about the importance of a key worker model, pooled resources, agency coordination and holistic whole family interventions that were closely aligned with the rationality of the national problem figuration advocating key elements of family intervention models. Specific cumulative learning, such as the need to enable and resource workers to spend increased amounts of time with families, getting to know them and building trust and rapport, delivered positive outcomes through facilitating an understanding of the complexity of family situations and a holistic assessment of their needs in order to identify causal factors and underlying issues which had often been hidden from other services. This ‘getting to know’ period proved central to enabling intervention workers to develop a support and action plan meeting the needs of the family:

Similarly, the centrality of addressing families’ immediate issues was also prioritised by intervention workers. Providing direct support to families mainly through emotional, practical and financial assistance was valued highly by families:

However, the case study also revealed the ambiguities, complexities and limitations of localised practice which are negated in national level problem figurations and policy guidance. Firstly, a genealogy of learning based on cumulative practice in the city was often in tension with the particular genealogy of different agencies and organisational practices with their own histories, challenging the extent to which a new localised regime of practice could be implemented. Although at a strategic level there were agreed principles and a clear delivery model, ‘buy in’ at an operational level, embracing new ways of working, was slow to filter down, and dislodging traditional working practices and challenging silo approaches were difficult, limiting the extent to which families may be ‘gripped’ and bound within the seamless and coordinated mechanisms of intervention as conceptualised in national policy rationales.

Existing services still tended to take an isolationist approach when addressing individual family members’ issues. It was not common for services to identify interdependencies and interlinked problems between family members by working with all the family, and inter-agency working between individuals in both MASTs and existing service providers was patchy. In cases where inter-agency working was stronger, this was often based upon positive relationships between key actors that had developed over a long period of time:

Furthermore, there was no shared, city-wide understanding of key working

Numerous agencies reported adopting a key worker approach, but these descriptions were rarely consistent. Roles presumed (by some services) to encapsulate key working did not fulfil all aspects of the approach and, despite the growing attempts by national and local policy and practice guidance to define and describe key worker roles, there remained a need to articulate what the whole household approach across the city would look like. The lack of knowledge and understanding about MAST and the key worker role and remit varied within and between services and limited the engagement of some services. For example, some social workers worked closely with MAST workers and valued their contribution to a family case, whilst other services were unclear about the role of MAST and the whole household approach taken by its workers. Knowledge and awareness of education welfare and parenting classes were widespread, but knowledge of the early intervention and prevention work of MAST was often more limited. This limited awareness, and a lack of understanding of the benefits flowing from preventative work, meant that officers in some services were unclear about when or why they would refer a client to MAST.

Assessing the issues faced by the whole family is time consuming, requires a cross-cutting multiple domain skill-set and involves an understanding of adult and child focused issues. Levering in appropriate services to deal with the more varied presenting issues for multiple family members requires close linkages with a wider range of services, and a broad commitment to greater multi-agency working. While services, such as MAST, are establishing closer working relations and referral protocols with a host of child focused services, it was clear that there remained work to be done across the city to better join-up adult and children's services.

Beyond the need for enhanced shared learning and definitions and improved coordination, there were very significant limitations imposed by limited resources and expertise, which are seldom acknowledged in national policy discourses. One of the functions of key workers is often to identify the range of vulnerabilities impacting upon families which are often excavated as relationships develop:

But this mapping of the range of problems does not necessarily translate into the necessary scale and diversity of interventions in response. It was often the case that support enabled families to function on a daily basis but, in line with the problem figuration of policy identified above negating structural explanations, did not address the underlying causes of the families’ problems. Signposting and referral are one of the key components of key working. However, intervention workers reported a number of difficulties escalating cases to specialist services: ‘We’ll have it escalated to social care when it escalates to social care, social care bat it back down and send it to intervention work and it feeds back to us.’ Whatever the reasons for the problems in referral, the intervention workers had to ‘fill the gap’. Intervention workers reported that they very often did not have the specialist expertise relevant to the case, but felt they needed to continue to offer support and assistance wherever possible:

This inevitably placed stress on intervention workers, and in some cases added to their already full workload. Additionally, the role and status of intervention workers was a source of tension. Often intervention workers were in a position to identify causal factors and action some form of preventative measures, often necessitating the involvement of other specialist services. However, their knowledge of the families’ issues was often overlooked by specialist services who themselves were governed by access thresholds. There was a reported unwillingness of some officers in some services to cede responsibilities to intervention workers. In some cases, this could be linked to professional status and grade. This reflects the misunderstanding of the key worker role and the differentiation between the role of key worker and a specialist agency, but this form of tension is likely to become increasingly prevalent as the Troubled Families programme and wider welfare reform seek to enhance the role of private, voluntary and community organisations in the delivery of interventions.

Rolling out a ‘whole household approach’ in the city raised a number of challenges and revealed a number of disjunctions with the proposed or presumed links within national policy frameworks between mechanisms and outcomes. The localised differentiation of a Troubled Families programme model and how this is interpreted and delivered is evident (not least in the case study city not articulating the terminology of ‘troubled families’). It is also clear that practice knowledge is unevenly shared across organisations (and cumulative knowledge is further threatened by cutbacks in specialist services and the precarity of many experienced practitioners and organisations’ funding status). The case study illustrates how key pillars of the Troubled Families programme – short time scales and explicit linear exit routes; clear cut identification of eligible/appropriate targeted families; measurable quantitative outcome indicators; the financial incentive apparatus (in the context of substantial reductions in budgets); and the perceived positive influences of multi-sectoral delivery – are all challenged by the complex reality of localised dilemmas and experiences of practice. This confirms previous findings (Van Wel, Reference Van Wel1992; Welshman, Reference Welshman2012; Hayden and Jenkins, Reference Hayden and Jenkins2014) that good practice and cumulative learning within problem figurations and governmental mechanisms are always selective: highlighting the efficacy of particular elements (such as key workers and multi-agency partnerships) while simultaneously denying the primacy of other elements (such as time, political power and status, resources and the complicated definitions of positive outcomes).