Abolishing the Audit Commission: Framing, Discourse Coalitions and Administrative Reform

Abstract The abolition of the Audit Commission in England raises questions about how a major reform was achieved with so little controversy, why the agency lacked the institutional stickiness commonly described in the literature on organisational reform and why it did not strategise to survive. In this paper, we apply argumentative discourse analysis to rich empirical data to reveal the pattern and evolution of storylines and discourse coalitions, and the ways in which these interact with and affect the practices of Parliament, the media and the Audit Commission itself. Our analysis shows that the politics of administrative reform are as much about discursive framing and the ability of pro-reformers to gain discursive structuration and institutionalisation as they are about the material resources available to a newly elected government and its ministers. Questions of technical feasibility are unlikely to derail a reform initiative once its promoters gain discursive ascendency.


Introduction
The Audit Commission for Local Authorities and the National Health Service (NHS) in England, a UK government agency operating at arm's length to ministers, has played a key role in the audit, inspection, performance improvement and regulation of local authorities and other local public service providers in England since its creation in 1983. It developed a significant role in supporting the local government efficiency agendas of Conservative governments from 1983 to 1997 and was central to the 1997-2010 Labour governments' ambition to improve the performance of local public services. Following the May 2010 general election, however, the incoming Conservative-Liberal Democrat coalition government moved quickly to reduce the Audit Commission's budget, terminate some of its functions and require all its local audit work to be outsourced. Shortly thereafter it stated its intention to legislate to abolish the Commission and introduce a new locally managed system for auditing local authorities. The announcement of the proposed abolition emerged without prior consultation with the agency itself, or with stakeholders in the wider local government public service community. The proposal was justified by a narrative of improved accountability and efficiency, reflecting a combination of the 'localism' and 'austerity' agendas being pursued by this new government. Somewhat surprisingly for such a major and unexpected organisational change the announcement was not regarded as contentious. Debate during the consultation and legislative phases of the policy process focused largely on the feasibility of the new local audit system, rather than the end of the Audit Commission.
The literature on changes to the machinery of government typically emphasises the complexity and contestation that surrounds such proposals in the face of the tendency to path dependency (Dommett and Skelcher 2014;March and Olsen 1983;Yesilkagit and Christensen 2010). Further, the Audit Commission's functions were deeply embedded in the structure and psyche of central-local government relations, as well as through its wider role in relation to NHS bodies, police and fire authorities, social housing providers and a number of other local public service organisations. Thus it might be expected to exhibit 'institutional stickiness' (Pierson 2002). On the other hand, observers who are familiar with the history of the Audit Commission may point to the growing antipathy by local government and other stakeholders towards its ever-evolving data collection and inspection regimes, combined with a new government that wishes to remove centralised performance and audit regimes. In addition, we should not ignore the appointment of a populist, and avowedly anti-centralist, Secretary of State.
The purpose of this paper is therefore to explore why such a major institutional reform was able to be enacted in an apparently unproblematic way, and in doing so to provide insights for the literature on administrative reform. We show how narratives, stories, metaphors and other linguistic devices are an important medium through which change is negotiated (Hardy, Palmer, and Phillips 2000;Llewelyn 2001;Vogel 2012). Specifically, we employ 'argumentative discourse analysis' (ADA; Hajer 1993Hajer , 2005, whose central theoretical proposition is that policy change results from a process of normative framing expressed as storylines by a coalition of actors that potentially become hegemonic. This is a somewhat different approach to that adopted in much of the substantial literature on public sector administrative reform, where change is explained at the macro-level in terms of ideational, institutional and ideological factors (e.g. Christensen and Laegreid 2002;Pennings 2010;Pollitt and Bouckaert 2011;Yesilkagit and de Vries 2004). ADA highlights how narratives can translate complex ideological concepts into everyday 'common sense' and as such provides this micro-level of analysis (Lau and Schlesinger 2005;Laws and Rein 2003;Moynihan 2006;Smullen 2010aSmullen , 2010bVogel 2012). It helps to explain success or failure in the implementation of administrative reform by paving the way for the deployment of the material resources available to protagonists.
We gathered documentary data covering the period from mid-2009, when political party general election manifestos began to be formulated, until early 2014 when the legislation to abolish the Audit Commission received Royal Assent. Data were gathered from press releases, Hansard (the record of Parliamentary debates), Parliamentary select committee reports and evidence, and ministerial and Audit Commission statements. Utilising the LexisNexis database, we conducted a keyword search of Local Government Chronicle (LGC), the weekly professional journal for that sector, the Financial Times, Guardian and Daily Telegraph, to give a cross section of the quality daily newspapers; and the Daily Mail, which had run a large number of sensational articles about the Audit Commission. This produced 66 articles relevant to our study.
We supplemented these data with in-depth, semi-structured and transcribed interviews with five Audit Commission board members or senior officials. We also gathered data from a current Coalition minister, and ministers in former Labour governments, through participation in a small, by-invitation workshop to discuss the demise of the Audit Commission, organised by an independent policy think tank. The purpose of the interviews and attending the workshop was to gain insight into central government decision-making processes concerning the abolition of the organisation. We were interested in gaining insights on how the storylines related to the decision-making and implementation processes. We were unable to interview civil servants; however, we did have access to email correspondence between ministers and civil servants including briefing notes, submissions and reports, released under a third party Freedom of Information request.
In the next section, we provide a descriptive analysis of the development of the Audit Commission from 1983 to 2010. This is followed by a theoretical section explaining our approach to investigating the abolition process, and then we present a detailed analysis before concluding with our key findings and theoretical reflections.

The Audit Commission: 30 years of an expanding domain
The Audit Commission commenced its operations in 1983, having been established by the then Conservative government. It acquired the local government audit functions of the District Audit Service, originally established in the nineteenth century, in the context of a wider remit to ensure councils had arrangements to secure economy, efficiency and effectiveness in their use of resources (Campbell-Smith 2008). The functions and status of the Audit Commission expanded considerably (Table 1), one significant factor being the commitment of the incoming 1997 Labour administration to public service improvement. It was the view of that government (DETR 2001), The Local Government Association (LGAthe peak body for the sector, and at that point Labour controlled), and the Audit Commission that underperformance in local government could be solved by the application of technocratic systems that would expose problems, thus stimulating improvement to avoid 'naming and shaming'. This led to new and extensive systems for inspecting and reporting on local government performance, in particular the Best Value regime and its successor the comprehensive performance assessment (CPA) (Campbell-Smith 2008;Stoker 2004;ODPM 2003). Similar systems were developed for other local public services, thus extending the Audit Commission's reach.
As the Audit Commission's remit expanded, so concerns began to surface about the value judgements underlying the inspection regimes and their cost in terms of staff resources (Brown 1997;Cope and Goodship 2002;ETR 2000;Financial Times 2002;Humphrey 2002;Kelly 2003;LGC 2006). The select committees for Public Administration (PASC 2003) and Environment, Transport LGA was no longer Labour controlled and thus lacked political alignment to the then government, and centralised systems of inspection and regulation were being undermined by an emergent localist narrative which was supported by the now majority of Conservative-controlled councils (Walker 2011). Even so, the Audit Commission continued to acquire functions and powers, expanding its domain to include social housing providers and NHS bodies, and developing a key role in the prevention and detection of welfare benefit fraud (Table 1).

Into reverse: the Audit Commission and the 2010 general election
The Audit Commission anticipated that the May 2010 general election would lead to a change of government. They thought that they would have a continued, albeit reduced, role given the three main English political parties' desire to improve local government efficiency in the aftermath of the 2008 financial crisis (interview 1). The Commission had already introduced a programme of efficiency measures (Audit Commission 2010), but was not oblivious to further retrenchment given the Conservative Party pledge to end 'the bureaucratic inspection regime that stops councils focusing on residents' main concerns' (Conservative Party 2010, 76) and the Liberal Democrats proposed to 'scrap the nearly £1 billion of central government inspection regimes on local councils' (Liberal Democrats 2010, 90). Nevertheless, the Audit Commission anticipated that Caroline Spelman MP, then Shadow Secretary of State for Communities and Local Government, was a likely contender for this ministerial position should there be a Conservative government, and noted that she had spoken in favour of the new CAA system (interview 1). The Commission also attended the Conservative, Labour and Liberal Democrat party conferences in autumn 2009 to showcase the Oneplace website, a flagship component of the CAA, which it hoped would demonstrate the ability of this new inspection regime to provide politicians and the general public with easy access to performance data, thus being of value to whichever party formed the new government (Audit Commission 2009). Ending inspection was considered possible, but abolition was regarded as unlikely since it had not surfaced on party political agendas (interviews 3 and 4). Following the election, the Conservative-Liberal Democrat coalition agreement went further than the Audit Commission had expected, stating that the government would 'cut local government inspection and abolish the Comprehensive Area Assessment' (HM Government 2010a, 12). When Cabinet portfolios were announced, it was Eric Pickles rather than Caroline Spelman who was appointed Secretary of State for Communities and Local Government. The Coalition announced an emergency budget on 22 June 2010 which included a range of measures to reduce public expenditure. On 25 June 2010, Pickles formally instructed the Audit Commission to cease work on the CAA and associated inspection work, and to cease its research activities. In the light of these developments, the Audit Commission planned further workforce downsizing and outsourcing of more of its audit work (interview 3).
Finally, on 13 August 2010 Pickles announced his intention to 'disband' the Audit Commission and replace it with a localised audit system (DCLG 2010a), stimulating a Select Committee inquiry (CLGC 2011). 1

Explaining reform through argumentative discourse analysis
There was very little criticism of the proposed abolition of the Audit Commission, despite the body's significant and long-standing role in the relationship between central and local government (and other local agencies) and its origins in a previous Conservative government wishing to control local public expenditure. This is somewhat surprising when viewed from the perspective of historical institutionalism, in which institutional 'stickiness' is predicted on the basis that considerable political investment is needed to overcome the accumulated sunk costs of the status quo (Pierson 2002). On the other hand, neither did the Audit Commission take steps to challenge the decision and seek its survival, as rational actor and resource dependency theories would predict (Oliver 1991;Overman 2011;Dommett and Skelcher 2014).
It is also surprising given the challenges the government faced in achieving other proposed abolitions of public bodies. The Public Bodies Bill proposed abolishing 44 types of public body, with the option to extend this to a further 151, but its passage through Parliament was complex and lengthy despite all party support for its overall intention. Proposals for specific abolitions were opposed through high-profile stakeholder campaigns (Dommett and Skelcher 2014).
We turn to ADA (Hajer 2005) to explore this paradox. ADA draws on a policy analysis tradition that emphasises the importance of linguistic devices in framing the construction of policy problems and solutions (Fischer 1993;Rein and Schön 1993). As Hajer observes, 'language has the capacity to make politics, to create signs and symbols that can shift power-balances and impact on institutions and policy-making' (2005,300). By framing an issue in a particular way, language serves the political functions of persuasion and justification (Rein and Schön 1993) as well as coalescing a community of allegiants around this portrayal (Stone 1989). As a result, analysis of discourse -'an ensemble of ideas, concepts and categories through which meaning is given to. . . phenomena' (Hajer 2005, 300) is central to understanding how and why particular issues come to be defined as policy problems, the specific responses that are developed and the extent to which they are accepted or rejected.
The statements through which discourses are articulated take the form of narratives or stories, which can translate one construction of a policy issue into accepted 'common sense'. In UK public sector reform, these are often expressed in a 'from-to' form, in which the proposed change delineates a past failing from an anticipated future success (Newman 2001). Following Hajer, we see storylines as the statements actors employ to summarise these complex narratives. Actors who articulate similar storylines may develop 'discursive affinity' (Hajer 2005, 304). Thus, we can talk of a discourse coalition, 'the ensemble of story lines, the actors that utter these story lines, and the practices through which these story lines are expressed' (ibid.). A discourse coalition may be said to be dominant where other actors accept its position, and where it comes to inform the policy process itself (Miller 2012). This dominance is achieved not simply by forming a storyline, but rather through the deployment of discursive resources which comprise the different linguistic tools and media through which their understanding of the world can be presented.

The abolition agenda: discourses, storylines, practices and coalitions
We now apply the ADA approach to the politics of the Audit Commission abolition proposal. We map the pro-and anti-abolition discourse coalitions (Table 2) and then use these data to explain the lack of resistance to the proposal and how it came to be widely accepted in a relatively short space of time.

Pro-abolition discourse coalition
Pickles, the incoming Secretary of State responsible for the Audit Commission, held robust views on the need for there to be greater opportunities for citizens to challenge local councils' spending and performance. On taking office, and unknown to the Audit Commission, he instructed a small group of civil servants to establish how the Commission could be abolished and how this proposal could best be presented (DCLG 2010b(DCLG , 2010c. From this developed one of the most prominent storylines within the pro-abolition discourse coalitionthe Audit Commission's lack of local accountability when compared with the prospects offered by citizens becoming 'armchair auditors' of local authorities in an environment of extensive online data transparency (interview 2). Thus, the official announcement of the proposed abolition stated that 'Ministers believe that the work of the Commission has become increasingly less focused on accountability to citizens but on reporting upwards to government', with Pickles quoted as saying that 'rather than being a watchdog that champions taxpayers' interests, it has become a creature of the Whitehall State' (DCLG 2010a). Pickles was supported by grass-root Conservative criticisms of the Audit Commission's inspection regime and growing disillusionment with CPA/CAA, especially by Conservative-controlled councils (interview 2). A related storyline was 'promoting localism', removing '[e]laborate mechanisms of audit, inspection, targets and guidance [that] enable the centre to micromanage the public sphere to an unprecedented degree' (HM Government 2010b, 4). The localism storyline was also prevalent in evidence to the Select Committee inquiry into the government's local audit reforms. Private sector auditing firms such as Deloitte LLP and BDO LLP argued that the audit regime should be responsive and accountable to 'local taxpayers' (CLGC 2011, ev.194-205). Cornwall County Council's submission was characteristic of those local authorities that gave evidence: 'We believe that individual authorities are best placed to understand the challenges facing their communities' (ev.w8) and the LGA agreed that the plan to abolish the Audit Commission was 'a practical expression of the localism and devolution agenda' (ev.137-140), revealing again the critical role of predominantly Conservative-controlled local authorities in this discourse coalition. This was a widely supported storyline which also brought the County Councils Network, a peak sector group, and the Taxpayers' Alliance, an influential right-wing lobby group, into the discourse coalition.
A storyline about waste, inefficiency and self-interested lobbying at the Audit Commission was also prominent. Some sections of the press, especially the populist right-wing-oriented Daily Mail, gave regular, front-page coverage to claims that spending at the Audit Commission was out of control, linked to the Mail's more general campaigns against quangos on which it ran 150 articles between 2009 and 2013. Other newspapers ran similar stories, claiming days out at horse races, expensive dinners and dubious purchases of flowers, all funded by the Audit Commission (e.g. Telegraph 2011). Many of these stories were linked to particular PQs, and the Audit Commission replies claimed they misrepresented or exaggerated the position. Nevertheless, they gave popular credence to Pickles' claim that the abolition would 'save the taxpayer £50 million per year' (DCLG 2010a).
The final pro-abolition storyline concerns the merits of open competition for audit services. This was particularly emphasised by private sector auditing firms who stood to gain from the contracting out of the Commission's own audit work. For example, BDO LLP commented that 'the disposal of the Audit Commission's audit practice to a diverse range of large, medium and small sized firms in the private and third sectors presents the most effective way to create lasting competition and reduce the cost of providing audit and assurance services to public bodies' (CLGC 2011, ev.199). Government also emphasised the benefits of a fully competitive market, thus connecting the benefits of efficiency gains to the wider austerity discourse.

Anti-abolition discourse coalition
Only a small number of actors argued against the abolition decision, and most of those were connected in some way to the Audit Commission. One storyline employed was that the Audit Commission's provided beneficial and high-quality functions. This is implicit in the Audit Commission evidence: Over the years we have helped local authorities, health bodies, the police and fire services, and social landlords manage their money better, improve their performance, and work with other agenciesto the benefit of local users and local tax-payers. (CLGC 2011, ev.161) Some parts of the more pro-Audit Commission press also supported this; for example, the Guardian (2010) and LGC (2011) highlighted the Commission's key role in tackling poor quality local services and driving improvement.
The Audit Commission sought to utilise the austerity storyline to its advantage, noting that 'it is counter-intuitive to disband expertise on [value for money], data, analysis, governance and assessment when severe financial pressures are increasing the risk of financial or service failure in local public services' (CLGC 2011, ev.166). Some charitable organisations reflected similar concerns. For example, the Women's Resource Centre commented specifically on the abolition which, it argued, would negatively impact on gender equality at a local level (ev.179-181).
A second storyline portrayed the Audit Commission as an independent, publicly minded organisation fulfilling an important public function. The Audit Commission itself framed the proposals as a challenge to the principle of independent audit. Michael O'Higgins (then Chair of the Audit Commission) stated that '[i]ndependent appointment is one of the main safeguards of auditors' independence. It should not be set aside lightly' (Audit Commission 2010). The Commission stated that the proposals risked 'undermining the established safeguards over the stewardship and use of public money, which are essential for maintaining public confidence' (2011a, 3). This storyline was developed by others, for example, Prospect, the trade union representing Audit Commission staff, noted that It is commonly held that the most effective way of ensuring the independence of auditors is for them to be appointed wholly independently of the audited body. The current system [achieves this]. (CLGC 2011, ev.176) Eugene Sullivan, then Chief Executive of the Audit Commission, is reported as saying: . . .we should acknowledge that this is a pivotal point in the history of public service. We will be losing a distinctive, and publicly owned, local public audit service and its District Auditors who have helped to protect the public purse effectively for over 150 years. (Johnstone 2012) This statement reinforces the independence of the Audit Commission and frames its audit functions as an essential aspect of good public governance, as did the Labour Party, which claimed that the closure of the Audit Commission would 'lead to a cosy, incestuous relationship between councils and the firms inspecting their books' (LGC 2010). Similar sentiments were echoed by Clive Betts MP, Chair of the Select Committee undertaking the inquiry into the abolition, who asked 'If a council appoints an auditor, the auditor reports the public interest and the councils says "we don't like that, we're getting rid of you". Who is supposed to oversee this? It is a fundamental issue' (Kaffash 2011).
The Audit Commission also developed a third storyline countering the proabolition view that open competition for audit would bring efficiencies. In his response on the DCLG consultation, O'Higgins detailed potential extra costs that could stem from the changes to the audit function brought about by abolition (Audit Commission 2011b)something which was also highlighted by Prospect in their Select Committee evidence. This was based on the multiplicity of different bodies (private audit firms, local audit committees and the National Audit Office) required to discharge the Audit Commission's audit functions should it be abolished. The Audit Commission board also tried to dispel 'myths' about its spending and inefficiency by conveying its efficiency storyline as 'fact'. For example, shortly after the announcement of the abolition, O'Higgins wrote an open letter to the Secretary of State in which he highlighted that the Audit Commission had not been reluctant to disclose its spending, as some reports had claimed, that it had not wasted money and that it had not been attempting to lobby on behalf of specific causes (Audit Commission 2010).
Explaining the dominance of the pro-abolition discourse coalition Discourse coalitions affect the distribution of power when they are able to achieve discursive structurationdominating the way an issue is conceptualised and discursive institutionalisationsolidifying this understanding into practices or institutional arrangements (Hajer 2005). Both effects are evident. The number and mutually supportive nature of the pro-abolition storylines played a major role in achieving discourse structuration (Table 2), as did their affinity with the wider and broadly accepted discourse of the need for public sector austerity. The pro-abolition coalition was able to achieve multiple reinforcing articulations of the storylines due to its large membership (Table 3), which then provided a solid basis for the newly elected governmentaided by material resources including a popular mandate, legislative capacity and executive authorityto move forward with the abolition. The influential actors it contained were able to access various institutional settings that ensured their utterances would be reported in the media, and connections between particular Conservative MPs and right-wing newspapers had the effect of amplifying these storylines. Although the various parties in the discourse coalition did not necessarily accept all the storylines, there was enough to bind them together.
Our analysis shows that timing and time are important factors in discursive structuration. Pickles' abolition proposal was developed in secret, and he only notified the Audit Commission's chair a few hours ahead of his Commons statement (interview 3; DCLG 2010b, 2010c). Such timing prevented the Commission from formulating and seeking to build a discourse coalition around its own anti-abolition storyline. The strength of the pro-abolition coalition, informed by well-developed and strong storylines, may have meant that such an anti-abolition coalition would still not have saved the body. However, timing is important in understanding why there was so little of the contestation over abolition that, as we have noted, is typically found in such cases. Time was also important. Cumulative discursive structuration is apparent when the evidence to the November 2010 Select Committee inquiry (Table 3) is compared to that for the government consultation some 12 months later (DCLG 2012). In the latter case, none of the submissions suggested retention of the Audit Commission but all commented on the proposed new local audit regime. Actors had already accepted that abolition was inevitable and had turned their attention to the future. Clive Betts MP reflected this position on the completion of the Select Committee inquiry, when he noted that the demise of the Commission had gone by 'largely unlamented' and focused his attention instead on ensuring that the regime replacing the Commission would be robust and preserve the principle of independent audit (Betts 2011).
From this, it is a short step to discursive institutionalisation. This is illustrated by the practices for the appointment of a new Audit Commission chair. Despite the legislation to abolish the Audit Commission not yet having been considered by Parliament, the government's advertisement sought an individual with experience and skills to 'help oversee a process of significant downsizing leading up to its closure' (CLGC 2012, 7). At the pre-appointment hearing by a committee of MPs, the preferred candidate was asked whether he thought the decision to abolish was right. He responded: 'I applied for the job given that the decision was taken, and my responsibility is to make sure that that is what happens' (op cit., oral evidence 4). Questions then focused on his Discursive structuration and institutionalisation show us how the pro-abolition coalition was able to dominate the discursive field, but not why the Audit Commission refrained from mounting a defence. Here we turn to Bourdieu's (1977) concept of habitus -'a practical sense of the "the game", a set of dispositions to act' (Chouliaraki andFairclough 1999, 101), or what Miller (2012, 72) refers to as 'an almost unthinking form of social action'. Habitus, then, is constituted by the deep background norms and practices associated with particular positions. Thus, the Commission's response was to frame the proposals as a challenge to the principle of independent audit rather than to seek to preserve the organisation: Our strategy from the beginning was. . . Firstly, to respect the right of the Secretary of State to change the architecture of government. . . We shouldn't complain about being abolished. That's the privilege of ministers. Secondly, to defend the Commission and its staff from abuse and gratuitous inaccuracy of which there was a lot. . . [We] didn't oppose the decision in principle, but did argue quite strongly against much of the unnecessary abuse. (Interview 3) The habitus of the Audit Commission was one of public self-denial in respect of the future of the organisation. The Commission's board could have mounted a defence and opposed abolition, but instead almost automatically chose to acquiesce due to the deep norms that come with accepting appointed public office (Dommett and Skelcher 2014). Consequently, the discursive resources open to the Audit Commission were constrained, and without substantial stakeholder opposition to the proposals the abolition was relatively straightforward.

Conclusion
Our analysis of the abolition of the Audit Commission shows the importance of understanding the way in which the politics of administrative reform is a matter of discursive framing; the expansion of the Commission's role in the 2000s was framed in terms of the public service improvement imperative, while its demise was constructed in terms of the body being out of control. The determining factors in the latter case were the ability of the pro-abolition discourse coalition to achieve discursive structuration and institutionalisationabolition became a social fact even before Parliament had considered the complex legislation involved, and this was reflected in the new chair's appointment process. Instead, debate focused on the technical and ethical issues involved in reverting to the nineteenth-century practice of councils appointing their own auditors. This helps to explain why, contrary to the literature and parallel agency termination cases, the Audit Commission's abolition was relatively simple and unopposed. The ability of one discourse coalition to structure the high politics of reform to their advantage thus set the context for the low politics of technical debate about the new audit system, which suggests that questions of technical feasibility are unlikely to derail a reform initiative once its promoters gain discursive ascendency.
This analysis contributes to the literature on administrative reform because it demonstrates that major reforms are not just a matter of the material resources available to a newly elected government. Material resources should not be ignored -Pickles had the authority and political support to propose the abolition. However, they interact with the discursive realm and thus, in the case of the Audit Commission, the storylines of the various political, media, business and local government actors developed a discursive affinity and melded into a powerful pro-abolition discourse coalition. Framing the Audit Commission as wasteful and unaccountable enhanced the material resources available to government. Isolated and bound by institutional norms not to criticise its own abolition, the Audit Commission and its few supporters were placed in a weak position. By highlighting the interaction between material and discursive levels, the paper has demonstrated the merits of using the ADA approach to analyse administrative reform and therefore to help explain why the assumptions of institutional stickiness commonly found in the literature are sometimes not observed. Further research into the implementation of other reforms in other policy areassuch recent reforms to the NHS and to the social security system in the UKas well as in different circumstances with greater contestation over the reform or within the broader policy arenacould help to shed more light on the role of discourse coalitions in the implementation of such policy decisions. This paper has also added valuable insights to the ADA literature, with particular reference to interactions between the discursive and the material. Interpretive approaches such as are often critiqued on the basis that they fail to link the ideational and institutional contexts in which discourses are articulated (Gains 2011;Hay 2011), and that they lack a solid link between discourse and action (Wagenaar 2012). Yet in this study, we have shown how the institutional context both constructs the norms of behaviour which facilitate contexts for the application of actors' utterances and provides the material resources through which a reform proposal is realised. Discursive resources have been shown to be connected to material resources, with the parliamentary system giving a government minister considerable legislative capacity, and important material resources to underpin the discursive resources at their disposal.
Further, we have also demonstrated that the ability of a discourse coalition to gain advantage is affected by timing and sequencing in the deployment of discursive resources, placing opponents at a disadvantage since they have to respond to the initial framing. The case of other bodies proposing abolition under the Public Bodies Act shows that the ability to launch a more effective defence could have given rise to greater contestation over the decision, although this may have proved ultimately unsuccessful given the strength of the pro-abolition coalition which had built up over a long period of time and coalesced around numerous high-profile storylines.
Our study also contributes to policy and practice. Policy advice of administrative reform recognises the importance of winning 'hearts and minds', but our work shows that winning control of the way actors talk about an issue is at least as important. Talk frames an issue, locating it in relation to other issues, engendering it with normative flavour and pointing towards a solution. As Schmidt argues: 'speaking of change. . . rather than just thinking it, is key to explaining the actions that lead to major policy transformations ' (2011, 107). Discussion of discourses may seem abstract and obtuse, but as any effective politician and manager knows it is the ability to affect the way people talk about an issue that is the key to change.