Decentring social policy: narratives, resistance, and practices

Historicist explanations

Many social scientists adopt forms of explanation that reduce actors’ beliefs to formal axioms of rationality or to synchronic patterns associated with institutions, systems, or other social facts. Decentred theory begins instead with the everyday notion that explanations of actions refer to the reasons the agents have for those actions. Crucially, decentred theory then suggests, first, that social scientists explain these reasons by locating them in the agents’ webs of belief, and second, that social scientists explain these webs of belief by locating them in a historical context of traditions and dilemmas.

As early as the 1950s, philosophers were forcefully criticising positivism and its concept of pure experience (Quine, 1961, pp. 20-46). However, public policy scholars often failed to take seriously the consequences of rejecting a positivist notion of pure experience. Many clung tenaciously to the positivist idea that we can understand or explain human behaviour by reference to objective social facts about people rather than to their beliefs; they excluded the interpretation of beliefs from the discipline on loosely positivist grounds. Other public policy scholars reject positivism, distancing themselves from the idea of pure experience, but still abstain from interpreting beliefs. Often, they try to avoid direct appeals to beliefs by reducing beliefs to intervening variables between actions and social facts. Theories of path dependence, for example, seek to provide this kind of explanation (e.g. Greener, 2002).

We would suggest, however, that once scholars accept that there are no pure experiences, they undermine the positivist case against interpreting beliefs. A rejection of pure experience implies that we cannot reduce beliefs to intervening variables. When we say that a party has particular interests that lead it to adopt a policy, we rely on a particular theory to derive its interests from its position and role. Someone with a different set of theories might believe that the party is in a different position or that it has different interests. Divergent explanations of how professionals engage with markets in public services, for example, are based on different theories about agency and motivation (Pollock, 2004; Le Grand, 2009). The important point here is that how the people we study see their position and interests inevitably depend on their theories, which might differ significantly from our theories.

To explain peoples’ actions, we implicitly or explicitly invoke their beliefs. Once we reject positivism, we cannot identify their beliefs by appealing to allegedly objective social facts about them. Instead, we must explore the beliefs through which they construct their world, including the ways they understand all their position, the norms affecting them, and their interests (Béland, 2005). Because people cannot have pure experiences, their beliefs are inextricably enmeshed with theories. Thus, public policy scholars cannot “read-off” actors’ beliefs from objective social facts. Instead they have to interpret beliefs by relating them to other beliefs, traditions, and dilemmas.

The forms of explanation we should adopt for beliefs – and so for actions and practices – revolve around two sets of concepts. The first set includes concepts such as tradition, structure, and paradigm (Bevir, 2000; Kuhn, 1962/2012, pp. 43-51). These concepts explore the social context in which individuals think and act. They vary in how much weight they suggest should be given to the social context in explanations of thought and action. The second set includes concepts such as dilemma, anomaly, and agency (Bevir, 2003b; Kuhn, 1962/2012, pp. 52-65). These concepts explore how beliefs and practices change and the role individual agency plays in such change.

We define a tradition as a set of understandings an actor receives during socialisation. Although tradition is unavoidable, it is a starting point, not something that governs later performances. We should be cautious, therefore, of representing tradition as an unavoidable presence in everything people do in case we leave too slight a role for agency. In particular, we should not imply that tradition is constitutive of the beliefs people later come to hold or the actions they then perform. Instead, we should see tradition mainly as a first influence on people. The content of the tradition will appear in their later actions only if their agency has not led them to change it, where every part of it is in principle open to change. Within social policy, a tradition may relate to a particular conception of the welfare state in the Fabian or new left traditions (Clarke and Newman, 1997).

Positivists sometimes hold that individuals are autonomous and avoid the influence of tradition. They argue that people can arrive at beliefs through pure experiences, so we can explain why people held their beliefs by referring to those experiences. However, once we reject positivism, we need a concept such as tradition in order to explain why people come to believe what they do. Because people cannot have pure experiences, they necessarily construe their experiences using theories they inherit. Their experiences can lead them to beliefs only because they already have access to the traditions of their community. People’s beliefs about health, for example, are rooted in the distinctive tradition of the NHS in the UK, often described as “the nearest thing the English have to a religion” (University of Warwick, 2015).

A socially inherited tradition is the necessary background to the beliefs people adopt and the actions they perform. Equally, however, social contexts only ever influence – as distinct from define – the nature of individuals. Traditions are products of individual agency. This insistence on agency may seem incompatible with our earlier insistence on the unavoidable nature of tradition. However, our reasons for appealing to tradition allow for individuals to change the beliefs and practices they inherit. Just because individuals start out from an inherited tradition does not imply that they cannot adjust it. On the contrary, the ability to develop traditions is an essential part of people’s being in the world. People constantly confront at least slightly novel circumstances that require them to apply inherited traditions anew, and a tradition cannot fix the nature of its application. When people confront the unfamiliar, they have to extend or change their heritage to encompass it, and as they do so, they develop that heritage. Every time they try to apply a tradition, they reflect on it, whether consciously or not, to bring it to bear on their circumstances, and by reflecting on it, they open it to innovation. Thus, human agency can produce change even when people think they are sticking fast to a tradition they regard as sacrosanct. Constant reform of the NHS, for example, in the name of retaining its Bevanite tradition of being free at the point of use would fit into this notion of adaptation within tradition (Exworthy et al., 2016).

The concept of dilemma provides one way of thinking about the role of individual agency in changing traditions. People’s capacity for agency implies that change originates in the responses or decisions of individuals. Whenever someone adopts a new belief and action they have to adjust their existing beliefs and practices to make way for the newcomer. To accept a new belief is thus to pose a dilemma that asks questions of one’s existing beliefs. A dilemma arises for an individual or institution when a new idea stands in opposition to existing beliefs or practices and so forces a reconsideration of these existing beliefs and associated tradition (Alvarez-Rosete and Mays, 2008). Social scientists can explain change in traditions, therefore, by referring to the relevant dilemmas. Traditions change as individuals make a series of variations to them in response to any number of specific dilemmas.

It is important to recognise that social scientists cannot straightforwardly identify dilemmas with allegedly objective pressures in the world. People vary their beliefs or actions in response to any new idea they come to hold as true. They do so irrespective of whether that new idea reflects real pressures, or, to be precise, irrespective of whether it corresponds to a pressure social scientists believe to be real. In explaining change, there is no reason to privilege academic accounts of the world. What matters is the subjective and intersubjective understandings of policy actors, not scholarly accounts of real pressures in the world. The task of the social scientist is to recover the shared intersubjective dilemmas of the relevant actors (Newman and Clarke, 2009).

Dilemmas often arise from people’s experiences. However, we must add immediately that this need not be the case. Dilemmas can also arise from theoretical and moral reflection. The new belief that poses a dilemma can lie anywhere on a spectrum from views with little theoretical content to complex theoretical constructs only remotely linked to views about the real world.

A related point to make is that dilemmas do not have given, nor correct, solutions. Because no set of beliefs can fix its own criteria of application, when people adopt new ideas they change traditions creatively (Smith, 2017). It might look as if a tradition can tell people how to act; how to respond to dilemmas. At most, however, the tradition provides a guide to what they might do. It does not provide rules fixing what they must do.

Although dilemmas do not determine solutions, a social scientist can explain the solutions at which people arrive by appealing to the character of both the dilemma and their existing beliefs. Consider first the influence of the character of the dilemma. To hold on to a new idea, people must develop their existing beliefs to make room for it. The new idea will open some ways of adjusting and close down others. People have to hook it on to their existing beliefs, and their existing beliefs will present some opportunities and not others. People can integrate a new belief into their existing beliefs only by relating themes in it to themes already present in their beliefs. Change thus involves a pushing and pulling of a dilemma and a tradition to bring them together.

Meanings in action

A decentred approach to social policy highlights contests among diverse and contingent meanings rooted in different traditions. As a result, it privileges specific new empirical topics, including narratives, rationalities, and resistance (Bevir, 2010). These topics have a strong presence in social policy literature, particularly in recent decades, as the discipline has shaken off its (always caricatured) image as ‘‘an unacademic, atheoretical form of social book-keeping’’ (Baldock, 1991, p. 298). Baldock and others have highlighted that this kind of “blind empiricism” was only a very partial account of the discipline, even in its Titmuss heyday, with a strong critical and theoretical element to the subject. Recent scholarship in social policy has explored narratives, rationalities, and resistance in relation to topics including child poverty, the rise of food banks, homelessness, and the damaging consequences of low pay and benefit conditionality (Sefton and Stewart, 2009; Ridge, 2011; Shildrick et al., 2012; Hodkinson and Robbins, 2013; Garthwaite, 2016). Institutional change and reconfiguration, for example, the shifting role of the non-profit sector, and its co-option by the state has been another theme of social policy scholarship (e.g. Macmillan, 2013; Rees et al., 2014). In another strand of work, care and disability services have been framed in narrative terms, to offer critical engagement with the sorts of rationalities imposed by the state (e.g. Needham, 2011; West, 2013).

Perhaps a more accurate characterisation of the discipline than “blind empiricism” has been what Heidenheimer called its “evangelical moral uplift” (Heidenheimer, 1986; cited in Bulmer et al., 1989, p. 6). For a time in the post-war era this was located in an optimistic “Whiggishness” tradition of welfare state growth, succumbing since the 1970s to a “crisis mentality”, as the welfare state came to be subjected to decades of retrenchment (Bulmer et al., 1989, p. 11). Critical scholarship exposing the rationalities and contingencies of this retrenchment has been an ever present feature of the disciplinary journals. The title of a round table at the 2016 Social Policy Association conference “What can and should the social policy association do to resist radically and resolutely?” highlights the commitment to practices of resistance within the academy.

The contributions in this special issue all contribute to social policy scholarship understood through narrative, rationalities, and resistance. Kate Brown’s article focusses on vulnerability.

She argues that diverse narratives and practices concerned with “vulnerability” increasingly inform how a range of social issues are understood and addressed, yet the subtle creep of the notion into various governance arenas has tended to slip by unnoticed. The article explores the role of vulnerability in responding to long standing and ongoing dilemmas about social precariousness and harm. Drawing on in-depth qualitative research into how vulnerability was operationalised in services for “vulnerable” young people in an English city, prominent narratives of vulnerability are traced, which operate in relation to a variety of often-dissonant service user responses. The article shows the governance of vulnerability as a dynamic process; informed by policy developments and wider beliefs about the behaviours of “problem” populations, interpreted and modified by interactions between practitioners and young people, and in turn shaping lived experiences of vulnerability. Patterns in this process illuminate how vulnerability narratives re-shape long-running tensions at the heart of social welfare interventions between a drive to provide services that might mitigate social precariousness and an impetus towards regulating behaviour.

Jay Wiggan’s article draws together a decentred governance perspective that emphasises ideational tradition for understanding (re)construction of governance (Bevir, 2013, p. 27). The article uses Critical Discourse Analysis of Scottish Government documents to examine welfare interpretations/representations in Scotland. It casts light on how the narrative of austerity privileged by the UK Government in Westminster is challenged and disrupted by the Scottish Government through their articulation of a social democratic welfare state imaginary. It explains how the Scottish Government has reworked social democratic traditions to weave together a welfare imaginary that directly contests the problem-solution narrative of successive Conservative-led UK Governments.

Central to Jon Warren’s article is the notion of “Biographies of Place”. The core of this idea is that places have biographies in the same way as individuals and possess specific identities. These biographies have been shaped by the intersections between environment, history, culture, and economic and social policy. Warren argues for an understanding of the region through a biography of place approach, in which the region’s economic development, subsequent decline, and future development are explained through the alliance of Labour politics and industrial employers around a common consensus that sought economic prosperity and social progress via a vision of “modernisation”. The article argues that an appreciation of these spatial biographies has the potential to result in innovative and more effective social policy interventions with the potential to address issues that affect entire localities.

Needham and West’s paper explores personalisation within care services, and its claims to be a universalising approach. They explain its contribution as an empty signifier, contrasting the universalism of its claims with the relatively low take up of personal budgets by older people. The article draws on interview data from a project with a large national charity which sought to increase older people’s take up of personal budgets in order to illustrate the persistence of policy makers in seeking to address their lack of take-up to date, and to highlight the explanations that are typically invoked in response to that failure. The article also draws on critical social gerontology’s recent theorisation of ageism and the ways in which older adult social care has both played a central role in the development of this new ageism and is the arena in which it continues to be played out.

In his article, Robert MacDonald reflects critically upon current debates and tensions in the governance of research in the UK and more widely, particularly the imperative that social science research should demonstrate impact beyond the academy. Drawing implicitly upon Bevir’s theory of governance, the article positions discourses about “research excellence and research impact” as elite narratives that are rooted genealogically in forms of managerial audit culture which seek to govern the practices of social science academics. The article reviews relevant literature, draws upon key contributions that have shaped debate, and refers to the author’s own research and experiences of “research impact”. He argues that the “impact agenda” has both negative and positive potential: it can distort research priorities and can lead to overstatement of “real world” effects, but it can also provide institutional space for work towards social justice, in line with long-standing traditions of critical social science and “public sociology”.

Together these articles contribute to an understanding of the elite narratives, rationalities, and resistances which with social policy scholars engage.

Elite narratives

In explaining the emergence of policy ideas, decentred theory suggests that social scientists should pay more attention to the traditions against which elites construct their worldviews, including their views of their own interests. Moreover, the central elite need not be a uniform group, all the members of which see their interests in the same way, share a common culture, or speak a shared discourse. A decentred approach suggests that social scientists should ask whether different sections of the elite draw on different traditions to construct different narratives about the world, their place within it, and their interests and values.

Historically, in Britain, the several members of the central elite have been inspired by Tory, Whig, liberal, socialist, and other narratives, as discussed in Bevir and Rhodes (2003). Arguably, however, the dominant and shared elite narrative in the UK for much of the last decade has been “the austerity story” (Blyth, 2013; Johnson and Chandler, 2015). This story presents “structural current budget deficit” as Britain’s key problem. The incoming Conservative-led Coalition Government in 2010 explained the deficit as caused by the spending of the previous Labour Government, the secondary banking crisis in the USA, and world recession. The governing elite in Westminster agreed that the most pressing problem facing British government was the size of the public sector debt: government must do less and public spending must be cut. The budgets for health and education were ring fenced, so most cuts fell on welfare payments and local government. However, the combined impact of the Eurozone crisis, slow growth, high unemployment, and low productivity meant that deficit targets still were not met. Further cuts in public spending were planned over the life of the 2015 parliament. The deserving poor (older people) were protected. The undeserving poor (everyone else) paid for “fiscal consolidation”, helped along by a divisive language of “strivers” and “shirkers” (Jowit, 2013).

There are, of course, dissenters within academia and the media who flatly oppose the “austerity con” (Krugman, 2015; Toynbee and Walker, 2015). The key argument of these dissenters is generally about the impact of austerity on growth. The suggestion is that the cuts in spending go too far, too fast, and cause lower economic growth. Certainly, in the 2015 General Election, neither major party rejected fiscal consolidation. They disagreed only over the pace of the cuts. They have also challenged the ways in which the cuts have fallen hardest on the already disadvantaged. Austerity, and the policies it legitimates, has been a focal point for much social policy scholarship in the last decade. The disproportionate impact on already deprived parts of the country is evident in Warren’s article in this special issue, in which he highlights the impact of deindustrialisation on Teesside (Warren, 2017, pp. 655).

It is evident, however, that this combined scholarship does little to alter the elite narrative of essential austerity. MacDonald’s article in this special issue draws attention to the success of a narrative of intergenerational worklessness, endlessly repeated despite being unverifiable – a “zombie” argument which endlessly rises again (MacDonald, 2017, pp. 696). Garthwaite’s (2017) work on food banks similarly points to the pervasiveness of claims that such facilities perpetuate laziness and discourage work. Brown’s article (again in this special issue) highlights the ways in which a discourse of vulnerability masks conditions of poverty and exclusion, “arguably furthering traditional stereotypes about ‘problem groups’ and reinforcing unequal subjectivities in superficially more palatable ways” (Brown, 2017, pp. 667). She goes on, “Managing and addressing vulnerability appears often to be largely related to activating citizens away from ‘dependency’ and in the direction of more ‘entrepreneurial’ activities, suggesting increasingly narrow spheres of acceptability”.

Welfare retrenchment to achieve deficit reduction is increasingly an English policy, which is less evident in other parts of the UK (Birrell and Gray, 2014). As Wiggan argues in his contribution to this special issue, “New political opportunities have subsequently arisen for centre left nationalists to carve out welfare visions and practices that are distinct from the UK Government” (Wiggan, 2017, pp. 639). The Scottish National Party in particular has shown that the anti-austerity argument can win votes (Gamble, 2015). In Government they have been able to push for a “devo-max” agenda, which although falling short of full independence, has enabled them to pursue distinctive agendas in social security and employment. As Wiggan (2017, pp. 639) puts it:

The Scottish Government for example have contested the austerity and welfare reform discourse of successive UK Governments by developing an interpretation of the problem-solution dynamic that is rooted in social democratic traditions. This has involved drawing on concepts such as the social wage, once common in British social democratic discourse, weaving this together with contemporary European policy interest in social investment and the “Nordic welfare model” […] to promulgate a new social democratic social investment welfare governance imaginary for Scotland.

Rationalities

Modernist social science has inspired the public sector reform agendas that promoted markets and networks (Bevir, 2011). However, this general pattern hides wide variation across particular cases. Narrower, even different, traditions of social science have influenced social policy in ways that vary across time, space, and sector. Governance encompasses many clashing rationalities. “Rationalities” refer here to the social scientific beliefs and associated technologies that govern conduct. It captures the ways in which governments and other social actors draw on knowledge to construct policies, actions, and practices.

Britain, like much of the developed world, has witnessed the rise of neoliberal and managerial rationalities using a technology of performance measurement and targets that spread far beyond the central civil service to encompass the control of localities. Britain has witnessed a stream of reforms aimed at involving individual citizens in the production, selection, and assessment of social services, from the height of Thatcherism through New Labour’s narrative of the consumer-citizen to today’s spread of ideas about “self-management” and the personalisation of social services (Bevir and Trentmann, 2007; Needham, 2007, 2011; Mol, 2008). West and Needham’s article in this special issue draws attention to the continued relevance of a personalisation narrative within care for older people, which persists as a fantasy rather than a realisable programme. They argue, “For the older old majority, then, the gap between an idealised narrative of transformation, autonomisation, personhood and empowerment and the situation of frail older people, approaching care services for the first time at a moment of crisis, seems unbridgeable […] a fantasy that places out of sight this radical gap between the conditions of crisis and abjection afflicting the older old majority of people in need of care and the ideal of universal transformation through personalisation” (2017, pp. 683).

A case can be made that all of these reforms and also the austerity narrative show the continuing resonance of the neoliberal ambition for the minimal state. As Bale (2014) argued “the right – free-market, small-state, low-tax, tight-borders, tougher sentences, eco- and Euro-sceptical – is where the solid centre of the [Conservative] party now comfortably resides”. The 2016 referendum on European Union membership and the Brexit result, further intensified this wing of the Conservative Party, albeit by limiting the access to the European market.

However, the Brexit vote also highlights the limitations of seeking to interpret the austerity narrative as a high point of neo-liberalism. The recent Coalition and Conservative Governments may have tried to restrict public spending, but they have shown nothing like the high neoliberal faith in markets and quasi-markets to solve all problems. On the contrary, they have drawn at least as much on communitarian and institutionalist concepts and technologies such as big society, social capital, networks, partnerships, and joining-up. The “little England” narrative deployed so successfully by UKIP and parts of the Conservative Party in the run up to the Brexit referendum, derives from communitarian nostalgia rather than a faith in markets.

The limitations of neo-liberalism as an explanatory framework are apparent (Bevir, 2016). Recourse to a dominant neo-liberal logic is too coherent and purposive to adequately explain recent social welfare change (Newman and Clarke, 2009; Bevir, 2013). In their contribution to this special issue West and Needham, argue against a reductionist explanation of personalisation within care services as an expression of neo-liberalism. Elsewhere Needham (2016) has called for an alertness to “the extent to which neo-liberalism is itself deployed as a hegemonic position with an affective dimension”. A totalising neo-liberalism risks becoming both a catch all explanation, but also an academic silencer – closing down positions which are not sufficiently adept at distancing themselves from the neo-liberal taint.

Other rationalities have accompanied the austerity narrative, of which two prominent ones are evidence-based policy making (EBPM) and nudge. Both of these are survivors from the earlier New Labour era which have endured and claimed universalist relevance in very different fiscal times. Recent UK governments have made an explicit commitment to EBPM as part of a stronger commitment to “what works” (Sullivan, 2011; HM Government, 2013). The discourse of EBPM has created a shift within the evaluation communities of government about what kinds of evidence are admissible, with a reassertion of the positivist empiricism which has waxed and waned within government in the post-war era (Rhodes, 2011; Sullivan, 2011).

In the 2010s, this reached its apogee as randomised control trials (RCTs) became the fad of the day. In brief, RCTs involve identifying the new policy intervention, determining the anticipated outcomes, and specifying ways of measuring those outcomes. Then, the investigator chooses control groups, whether comprised of individuals or institutions. The policy intervention is randomly assigned to the target groups with a designated control group. Using a randomly assigned control group enables the investigator to compare a new intervention with a group where nothing has changed. The next step is to measure the impact of the intervention and adapt the intervention because of the findings. The catchphrase for the approach is “test, learn, adapt” (Haynes et al., 2012, p. 5). The strongest statements of the enthusiasm for RCTs in particular are contained in the “Treasury’s Green Book” and civil service “Magenta Book”, which set out for other government departments and agencies guidelines on policy assessment and evaluation (HM Treasury, 2014). A Ministry of Justice (NOMS) publication bluntly pointed out that “case studies [and] process evaluations will not tell you about impact” (NOMS/MoJ, 2013).

Behavioural economics is another variant of modernist social science in favour with recent British Governments (Dolan et al., 2010; John et al., 2011; Leggett, 2014). It uses research on how people make decisions – on the heuristics they use and the biases to which they are prone to identify ways of “nudging” people towards behaviours the government prefers. The government’s acronym “MINDSPACE” (Dolan et al., 2010) is based on seven such nudges. For example, “N” is for “Norms” and refers to the idea that we are influenced by what others do. Like other initiatives, the behavioural turn has differential impacts on different groups within society (Whitworth, 2013; Harrison and Hemingway, 2014). Brown’s article in this special issue discusses how the “new behaviourism” targets the least well-off and least well-educated groups with injunctions to behave “appropriately” (Jones et al., 2013). What is also evident is the patchiness of the narrative. The nudge approach has shaped policies (such as the auto-enrolment pension scheme) on the basis that people cannot make sensible long-term choices about money (Hardcastle, 2012), at the same time that other parts of government have demanded that poor people take more responsibility for their own financial literacy (Finlayson, 2009; Price and Livsey, 2013).

Resistance

When social scientists neglect agency, they can give the impression that politics and policies arise only from the strategies and interactions of central and local elites. However, other actors can resist, transform, and thwart the agendas of elites. Our decentred approach draws attention to the diverse traditions and narratives that inspire street-level bureaucrats and citizens. Policies are sites of struggles not just between strategic elites, but between all kinds of actors with different views and ideals reached against the background of different traditions. Subordinate actors can resist the intentions and policies of elites by consuming them in ways that draw on their local traditions and their local reasoning (Lipsky, 1979; Maynard-Moody and Musheno, 2003; Vinzant and Crothers, 1998). In Brown’s article in this special issue she draws on interviews with young people to write about “the governance of vulnerability as a dynamic process; informed by policy developments and wider beliefs about the behaviours of ‘problem’ populations, interpreted and modified by interactions between practitioners and young people, and in turn shaping lived experiences of vulnerability” (Brown, 2017, pp. 667).

Social policy has long been alert to these challenges, and for the need for critical scholarship to provide ways to access and share the stories of resistance. However, writing in the context of the Research Excellence Framework creates a new set of dilemmas and raises the stakes of resistance. The six-yearly REF process in which academic outputs are scored by a panel of peers on the basis of rigour, originality, and significance – and the high institutional and career implications of internal scores – can discourage articles which do not demonstrate rigour using the conventional markers of modernist social science (Gill, 2014; Stahl, 2015). MacDonald’s article in this special issue quotes Scott’s claim that universities are in the grip of “a monster, a Minotaur that must be appeased by bloody sacrifices”, first among these being academic principle and freedom (Scott, 2013).

The REF impact agenda poses distinctive challenges here, although there is an ambivalence around whether it opens up or closes down space for resistance. As MacDonald puts it, “REF requires researchers to explain and demonstrate the impact their studies have had in ‘the real world’, with multimillions of pounds of funding resting on assessments of this. Because of its obvious connections with social improvement and reform through policy action, social policy is regarded as one disciplinary area that is well placed to meet and benefit from this ‘impact agenda’” (2017, pp. 696). MacDonald suggests that there is a space for the impact dimension of REF to “allow spaces for progressive work towards social justice and thus warrants a cautious welcome. This suggests the opportunity for social science academics to develop local practices to resist the elite narrative”. However, MacDonald acknowledges the limitations of his own work – a high scoring REF case study – to deliver impact, if that is understood as achieving material change, or shifting the elite discourse around worklessness. He concludes: “The author’s experience here would seem to demonstrate the simultaneous failure of the research to have real impact and the flawed nature of the REF exercise (in that it judged that impact as ‘world leading’)”.