Autonomy, accountability, and ambiguity in arm’s-length meta-governance: the case of NHS England

ABSTRACT

Meta-governance involves orchestrating the ‘rules of the game’ in public management. Arm’s-length bodies are particularly important vehicles for this. We consider the case of an arm’s-length body (NHS England) created to oversee the English NHS’ day-to-day operation, and remove ‘political interference’. Although mandated by the Department of Health it has increasingly operated as policy-maker, developing policies in tension with existing legislation, while Ministers have faded from public-facing accounts of service operation. This suggests NHS England operates as a meta-governor, insulating government whilst pursuing its own agenda, and raises crucial questions about governmental accountability whilst simultaneously making answers harder to obtain.

KEYWORDS:

• Meta-governance
• accountability
• autonomy
• healthcare
• NHS

Introduction

A central concern of public management scholarship is to examine and explain how governments can enact their policy agenda. In recent years, attention has turned to difficulties associated with implementing policy in the context of what Rhodes (1997) called a ‘hollowed out state’. Others refer to ‘steering, not rowing’ to explain contemporary approaches to government (Barlow and Röber 1996); both metaphors invite the analyst to consider the control mechanisms available to governments in a more plural, networked state. Meta-governance is a relevant concept in this regard. Concerned with the ‘governance of governance’ (Sørensen and Torfing 2009, 245), or the government of governance (Jessop 1997), meta-governance entails consideration of the mechanisms by which governments seek to control distal networks and hierarchies to enact their agenda. This is particularly relevant to the operation of arm’s-length bodies (ALBs) in the modern state. Common across jurisdictions (Moynihan 2006), ALBs are a wide-ranging category of organizations fulfilling public activities, which might otherwise be undertaken by government departments, operating at ‘arm’s-length’ from them. Criticized as unelected, inefficient, and unaccountable (Pollitt and Talbot 2004), ALBs have proved remarkably resilient despite repeated calls for their abolition (Flinders and Skelcher 2012). The delegation of some political authority to an ALB allows actions taken to be presented as technical and driven by objective expertise, effectively distancing such decisions from the political realm (Durose, Justice, and Skelcher 2015). As such, ALBs are profoundly ‘useful bureaucratic animals’ (Flinders and Skelcher 2012, 334).

There is thus a disconnect in public management between public authorities’ statements about the problems associated with ALBs, and their continued survival and indeed proliferation in some fields. In this paper, we use a case study of an ALB to consider these issues. We use the literature relating to meta-governance as a theoretical lens, and consider issues of autonomy and accountability, answering the research question: how does the concept of meta-governance help to understand the relationship between ALBs and government? Our case study ALB – NHS England (NHSE; an executive non-departmental public body) – was created as a result of the 2012 Health and Social Care Act (henceforth 2012 Act). In a significant deviation from previous arrangements, NHSE was vested with oversight of the day-to-day operation of the NHS in England and responsibility for its budget. The government presented this as desirable because it would reduce the significant powers that the Secretary of State held over the health service, thereby reducing ‘political interference’ in its operation (DH; Department of Health 2010, 30). NHSE’s emergence occurred during a period of concerted effort to reduce the number of ALBs in operation in England due to concerns about opacity and the extent of their autonomy. The 2011 Public Bodies Act led to the abolition and amalgamation of many ALBs and extended the Cabinet Office’s degree of control over many of them through new governance arrangements (Dommett and Flinders 2015).

Drawing on policy statements and government reports, we explore the developing role of NHSE from 2010 to 2017. We argue it has moved beyond the scope of its original remit, operating as a meta-governor itself, steering national policy and re-shaping the organizational constituents and governance processes of the English NHS in a manner that runs counter to attempts to limit the prominence and power of ALBs (Dommett and Flinders 2015). This has increased ambiguity around accountability and responsibility for the health service. At a time when attempts are being made to tighten, and render more transparent, arm’s-length governance, NHSE provides an instructive example of how attempts to remove ‘political interference’ from the management of a significant area of public service can operate as a mechanism for de-politicizing that service (Hay 2007), limiting governmental control but also limiting the extent to which government might be held to account for its failures. It also raises important questions about the realities of meta-governance processes, highlighting a ‘slipperiness’ whereby an organization has been able to assume a meta-governor role without such a role apparently being willed or explicitly ceded by government.

First, we establish our theoretical base with a discussion of meta-governance, autonomy and accountability. Drawing on this, we provide a detailed examination of NHSE’s emergence and evolving operation within the NHS. Finally, we highlight a number of questions regarding NHSE’s role that warrant critical attention, consider the utility of applying meta-governance to this context, and consider the lessons of this case study in addressing our research question and for broader public management research that engages with meta-governance.

Methods

In this paper, we use meta-governance theory, supplemented by attention to issues of accountability and autonomy, in order to explore the relationship between NHSE and Government, and by extension the relationship between ALBs and governments in contemporary Western public management. To do this, we conducted a systematic analysis of a number of policy and other documents produced by either the Department of Health or NHSE between 2010 and 2017. These can be categorized into three groups. This first comprises the ‘Equity and Excellence’ White Paper (Department of Health 2010) and ‘Developing the NHS Commissioning Board’ (NHS 2011). These are Government policy documents setting out the rationale for and details of the intended role, governance and accountability processes associated with NHSE as an arm’s length body. The second group contains the first three iterations of NHSE’s mandate (Department of Health 2013, 2015, 2017), which is the primary mechanism for the Government, via the Secretary of State, to hold NHSE to account, setting out objectives against which NHSE is to be assessed. We analysed whether and how the mandate changed over time. In group three is the Five Year Forward View (NHS England, Care Quality Commission, Health Education England, Monitor, Public Health England, & Trust Development Authority 2014), which establishes a direction of travel for the NHS over the subsequent five years and is correlated with significant changes to the 2016–2017 mandate (Department of Health 2015); The Forward View into Action (NHS 2014); and Next steps on the NHS Five Year Forward View (NHS 2017). These are NHSE’s own policy documents, setting out detailed plans for the NHS up to 2020 and beyond.

We thus explored Government statements of policy, NHSE statements of policy, and the mandate, which is supposed to link the two. Our thematic analysis proceeded iteratively, cross referencing between the three groups of documents in order to understand how the different organizations articulated their own and others’ roles, and how accountability relationships were described.

We also considered evidence from a number of UK Parliamentary committee reports and records of oral evidence, and media sources that report public utterances of, in particular, senior Government and NHSE figures. Our objective here was to understand how both the Government and NHSE describe their respective roles and objectives publicly, notwithstanding the versions set down in the various policy documents. The collation and analysis of these additional sources was not conducted systematically but those included feature on the basis that they reflect the views or official statements of key actors or provide additional context. We focus on developments over the period 2010–2017, which encompasses the genesis of NHSE and some notable moments in its evolving operation.

The strength of using a documentary analysis approach is that these policy documents represent the officially stated views of the relevant organizations, whilst the mandate is the legal mechanism by which NHSE operates. Documents can, of course, be subjective, politically influenced, and project a particular vision for the future that serve certain groups rather than reflecting reality. However, even when heavily sanitized or produced with specific implicit aims they still offer insight into the version of reality that someone, or some group, perceive as desirable (Shaw, Elston, and Abbott 2004). Our testing of these documents against public statements by senior officials thus provides a further level of analysis.

Theoretical framework

Meta-governance

Meta-governance refers to the ways in which different modes of governance – that is hierarchy, networks, markets (Thompson 1991) – are selected or combined for specific purposes, and how government might (amongst other things) ‘provide the ground rules for governance…’ and ‘ensure the compatibility of different governance mechanisms and regimes…’ (Jessop 1997, 575). Meta-governance approaches are ultimately concerned with examining how ‘control mechanisms’ are employed to enforce ‘the rules of the game’ (Dommett and Flinders 2015, 3). Our interest in meta-governance lies in Jessop’s (2004, 52) contention that a state’s ‘…capacity to steer is most successful when exercising power through hegemonic practices and third parties that regulate organizations on the behalf of the state’. Yet the state’s capacity to steer, or even accurately conceive of the structure and operation of the system, may be hamstrung by the sheer complexity of emergent (meta-)governance arrangements, thereby raising questions about the control capabilities of any government (Jessop 2015).

Issues of complexity and control are prominent features of the contemporary meta-governance literature (e.g. Larsson 2017; Qvist 2017) and raise concerns relating to autonomy and accountability of public services. Voets, Verhoest, and Molenveld (2015) suggest that effective meta-governance involves sound judgement about when to switch between exerting, or threatening to exert, hierarchical control over a network of organizational actors, and providing the conditions for those actors to act with greater autonomy. The difficulties associated with the exercise of this ‘sound judgement’ are illustrated by what has been called the ‘pendulum swing’ (Axelsson et al. 2000) between decentralization and central control suffered by many health systems (De Vries 2000). Such attempts to redistribute power from central government to agencies and localities provide fertile conditions for ‘blame games’ (Hood 2010), ‘buck-passing’, and accountability ambiguity when objectives are not met. Jessop (2015) argues that complex public service systems render the administrative landscape difficult to comprehend, in turn rendering meta-governance problematic for a government seeking to exert control. This gives scope for ‘mission creep’ and bureaucratic agenda expansion, as network constituents prioritize their survival and accumulate power where possible. Finally, Sørensen and Torfing (2005) argue that the day-to-day realities of meta-governing involve politicians allocating practical responsibilities for liaising with governance networks to public administrators. Consequently, such actors inhabit influential entrepreneurial roles, and their actions have great potential significance, which raises concerns around democratic legitimacy.

Jessop’s (2004) conceptualization of meta-governance resonates with the findings of Kelly (2006) who examined the UK’s Audit Commission (statutory corporation, 1983–2015), and showed how the independent agency operated as a ‘vehicle of meta-governance’ (608) by providing indirect regulation of local government to specifications established in legislation by government. As such, a meta-governance lens supports the articulation of the Audit Commission as a tool employed by government to enable it to steer local government even when changes to the power and capacity of the state made maintaining central command and control arrangements increasingly difficult (Rhodes 1997). The concept helps explain ‘how and why there has been reconfiguration of the traditional bureaucratic institutions resulting in the growth of governance arrangements’ (Kelly 2006, 605–606), and how the power of central government is thus obscured through the actions of an agency, best characterized as operating in the ‘shadow of hierarchical authority’ (Scharpf 1994, 41) rather than through a central command and control model. This work highlights meta-governance’s capacity to facilitate exploration of the orchestration of governance rather than more direct investigations of governance processes and arrangements. Whitehead (2003) notes that the concept helps articulate the dynamics of government and governance, rather than entrenching a sense of separation between them – enabling ‘the political and economic changes associated with governance to be positioned within the context of changing patterns of state power, strategy and intervention’ (7). Identifying and understanding how particular meta-governance arrangements have emerged provides a window into such manifestations of power in the modern state.

Next we set out in more detail the focus of our study: the operation of ALBs.

Autonomy and accountability of arm’s-length bodies

The movement to ‘agencification’, or decentralization from ministries to arm’s-length agencies, is common to many Western public administrations (Moynihan 2006), and is associated with New Public Management reforms beginning in the late 1980s (Wettenhall 2005). Although an enduring feature of the political landscape in the United Kingdom and elsewhere (Dommett and Flinders 2015), the perceived legitimacy, desirability, and value of ALBs remains under debate (Overman and van Thiel 2016). They are simultaneously framed as unaccountable, undemocratic, and inefficient yet also potentially more effective than government, capable of providing ‘external’, objective perspectives and expertize that can positively mediate public perceptions of governmental operation. This reputational tension lies at the heart of what Gash and Rutter (2011) refer to as the ‘quango conundrum’ (quasi-autonomous non-governmental organization; ‘quango’ encompasses ALB). ALBs are identified as barriers to efficient and effective government, yet also serially deployed by governments as policy solutions in specific contexts – ‘hated as a class, but individually seen as valuable devices for improving effectiveness and increasing public confidence in government’ (Gash and Rutter 2011, 95). In this paper we address the utility of this paradoxical position in relation to practices of meta-governance, exploring what this apparent contradiction might enable different actors to accomplish through the deployment and operation of ALBs in public management. For example, the inculcation of expertize within ALBs, and their distance from government, has potential to insulate government ministers from unpopular policies, thus shifting, or obscuring, the locus of accountability.

Accountability and autonomy are crucially important to the discussion of ALBs, particularly as they pertain to the relationship between an ALB and its sponsor governmental department. Characterizing this relationship, Flinders and Tonkiss (2015) offer a framework of ALB-Sponsor autonomy (see Figure 1), demonstrating different relational levels of strategic and operational control between government (i.e. the sponsor) and the ALB. These relations are typified by the tightness of the relationship between strategic and operational control (both intended and actual). Strategic control relates to the ‘broad policy framework’, whereas operational control relates to the working or administrative elements of delegation. Tight control of these elements is evidenced in ‘explicit, dense and regularly monitored “hard” control mechanisms’ (6) and loose control can be observed in the existence of ‘implicit, weak and rarely enforced “soft” control mechanisms,’ (6). A highly regulative arrangement might signal a tight form of hard control mechanisms (i.e. very little autonomy is offered through legislation). Conversely, a very loose form of soft control mechanisms might mean that at a practical level there are ample opportunities for ALBs to exercise autonomy.

Figure 1. Conceptual map of sponsorship between arm’s-length bodies and governmental sponsors (Amended from Flinders and Tonkiss 2015).

Accountability is a broadly interpreted and frequently employed term necessitating upfront clarification. We draw from Bovens’ (2007, 450) definition of accountability as the ‘…relationship between an actor and a forum, in which the actor [individual or organisation] has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgement, and the actor may face consequences’. In assessing the adequacy of public management accountability arrangements in a given context two evaluative perspectives are particularly relevant: a democratic perspective considers ‘[t]he degree to which an accountability […] regime enables democratically legitimized bodies to monitor and evaluate executive behaviour and to induce executive actors to modify that behaviour in accordance with their preferences’ (Bovens 2007, 465); a constitutional perspective asks whether accountability arrangements are sufficiently robust to ensure that public officials/organizations are disincentivized from abusing their authority. Underlying both perspectives is a recognition that the legitimacy of governance depends on accountability processes that are considered adequately effective (Bovens 2007).

Analysis of the accountability of ALBs has examined concerns that their operation might be associated with ‘…cronyism, clientelism, nepotism, sleaze or corruption…’ (Flinders and Tonkiss 2015, 492). Effective meta-governance requires proper oversight to prevent such issues, but there is a risk that government actors benefiting from the insulatory effects of ALBs’ operation may be incentivized to ignore them. Furthermore, delegation of responsibility and control over public spending to ALBs is not always associated with robust or explicit governance arrangements and requires clarity about the relationship with the sponsor department (Skelcher 1998), which relates directly to Flinders and Tonkiss’ (2015) sponsorship map above. In the United Kingdom, a 2016 House of Commons Committee report argued that the increasingly complex approaches to public management and delivery, including the operation of ALBs, have not been matched by mechanisms to ensure proper meta-governance and associated accountability to Parliament for the spending of public funds (Committee of Public Accounts 2016). This speaks clearly to concerns over democratic accountability.

Taken together, consideration of accountability and associated meta-governance mechanisms focuses attention on the details of how ALBs operate and how they are governed. An alternative theoretical lens might have been to consider the literature relating to the regulation and oversight of public bodies (e.g. Baldwin, Cave, & Lodge, 2012). However, this would have led to an analysis focussing on one dimension – the degree of control by one entity over another. Meta-governance, by contrast, provides an opportunity to consider the nuances of the two-way relationship between Government departments and ALBs. We now turn to our case study ALB, NHSE, and use a meta-governance lens to explore its creation and subsequent operation. In particular, we explore the associated architecture of accountability, and draw attention to its increasing role as meta-governor of the system, reflecting upon the conditions which have allowed this to occur and the associated implications.

NHS England as an arm’s-length body

The history of UK public administration features numerous official reviews of ALBs and oft-repeated arguments to abolish them (Flinders and Skelcher 2012). The Public Bodies Act 2011 marked a commitment to ‘reduce the number and cost of quangos’ by the UK Coalition Government (2010–2015). Approximately 900 were catalogued, with a third recommended to be abolished or merged. Simultaneously, the Cabinet Office was awarded new powers over ALBs (Dommett and Flinders 2015), including a new Controls Framework. It was empowered to enforce spending controls over a range of domains and administer a triennial review process to assess whether the functions undertaken by ALBs were being appropriately discharged or whether their functions should be transferred to government departments. In parallel, several new ALBs outside of the remit of the reform programme emerged, of which NHSE was the largest and most significant. The fact NHSE – ‘the mother of all executive Non-Departmental Public Bodies’ (Rutter 2014, 150) – emerged in the context of a drive to check ALB autonomy and enhance mechanisms for their accountability is notable and an illustration of the quango conundrum in operation (Gash and Rutter 2011).

The health and social care act 2012 and the role of NHS England

NHSE (originally, still legally, NHS Commissioning Board) was created as part of the 2012 Act’s numerous and far-reaching changes. The most significant include abolition of 152 Primary Care Trusts as the main health care service commissioning (i.e. planning, purchasing, monitoring) bodies and their replacement with newly created Clinical Commissioning Groups (CCGs; 207 at the time of writing), which are membership organizations comprising GP (family doctor) practices. Strategic Health Authorities, 10 regional bodies that held Primary Care Trusts to account which were in turn directly accountable to the DH, were abolished. NHSE was created to authorize and oversee CCGs, and commission several areas of care, most notably, primary care (i.e. from members of CCGs for their GP services), and specialized services. From April 2013, NHSE took responsibility for the vast majority of the £116.5 billion budget for the NHS (Department of Health 2014), which by 2016/17 had risen to over £122.5 billion. Crucially, however, the split between purchasers (i.e. commissioners) and providers (i.e. NHS Trusts and others) and the quasi-market within the English NHS was unchanged by the 2012 Act and, in fact, competition law was more fully embedded into the health service through the legislation.

In the ‘Equity and Excellence’ White Paper preceding the 2012 Act, NHSE was explicitly framed as being intended to operate with a significant degree of autonomy – ‘…a lean and expert organization, free from day-to-day political interference’ (Department of Health 2010, 30), establishing for the first time a ‘statutory division between ministers and the Department of Health on the one hand, and the commissioning and provision side of the NHS on the other’ (Timmins 2013, 3). Indeed, the White Paper states the reforms would curtail the ‘extraordinarily wide powers’ held by the Secretary of State over the NHS by limiting his/her ability ‘to micromanage and intervene’ (Department of Health 2010, 33). These rhetorical policy features related to another explicit objective of the 2012 Act: to give health professionals more autonomy over the commissioning of care so they could use their professional expertize to make better decisions about what patients needed, allied to increased accountability for their performance (Speed and Gabe 2013). For CCGs, accountability arrangements have proved significantly more complex than those applied to Primary Care Trusts (Checkland et al. 2013). Far from experiencing NHSE as a ‘power decentralizer’, as officially intended (Hunt 2013), many CCGs have portrayed their experiences with the ALB as one of constraint rather than empowerment, characterized by intense, burdensome, and sometimes unnecessary assurance demands (Checkland et al. 2017).

Following the publication of the White Paper, NHSE published their initial proposal detailing the organization and operation of the ALB (NHS 2011). This document was the first and only one setting out in detail how the new system could work, focussing on the culture, style and leadership of NHSE (then still NHS Commissioning Board). NHSE’s approach would be ‘characterised by its developmental relationship with clinical commissioning groups, the promotion of a single, evidence-based model of change, and its positioning of quality at the heart of the commissioning system’ (17). The key framework for assessing the performance of the NHS and, therefore, NHSE is the NHS Outcomes Framework, which consists of nine ‘domains’ (e.g. preventing people from dying prematurely; enhancing quality of life for people with long-term conditions) and a list of ‘indicators’ within each. This is supported by various other frameworks such as the CCG Outcomes Indicator Set which NHSE uses to hold CCGs to account for their performance.

NHS England accountability and the mandate

The 2012 Act established that the Secretary of State and NHSE share a duty to promote a comprehensive health service. The formal relationship between them is defined through a mandate issued (and annually renewed) by the DH. This establishes the government’s aims, requirements, and budget for NHSE. It is the primary mechanism for the government, via the Secretary of State, to hold NHSE to account. The first mandate, published in 2013, was structured around the NHS Outcomes Framework (Department of Health 2013). The objectives set for NHSE are notably general and focus on overarching aims. For example, NHSE is required to ‘make significant progress in ensuring people have access to the right treatment when they need it’ (7), and ‘make measurable progress towards making the NHS among the best in Europe at supporting people with on-going health problems to live healthily and independently’ (9). The document’s preamble makes a number of statements about how the mandated dynamic between NHSE and government will establish transparency, accountability, and autonomous clinical leadership over the health service (see Figure 2).

Figure 2. Extract from the 2013 mandate from the Department of Health to NHS England (our emphasis; Amended from Department of Health 2013, 4).

In the context of NHSE’s publication of the Five Year Forward View (NHS England, Care Quality Commission, Health Education England, Monitor, Public Health England, & Trust Development Authority 2014) (discussed in the following section), significant changes were made to the 2016–17 mandate (Department of Health 2015). The seven objectives in the refreshed mandate were underpinned by specific ‘deliverables’ (Department of Health 2015, 6) to be achieved in the short-term (for 2016–17) and long-term (by 2020 or beyond). These were more specific than in previous iterations but remained mostly process-based (e.g. ‘agree a trajectory’, ‘produce a plan’), and are still ‘relatively broad brush’ (Specialised Services Commission 2016, 34). For example, objective one requires NHSE to ‘demonstrate improvements against the NHS Outcomes Framework’ (Department of Health 2015, 8). In a significant development, the mandate explicitly refers to NHSE’s ‘freedom to develop its own plan’ (Department of Health 2015 6), which relates to the content of the Five Year Forward View and suggests a shift to a greater degree of strategic autonomy. Indeed, the evolution of the mandate between first and second iterations shows that NHSE’s responsibilities have changed from overseeing the delivery of NHS services in the original mandate (Department of Health 2013, 4) to arranging the provision of health services in England in the 2016–17 mandate (Department of Health 2015, 6). This shift suggests a more comprehensive allocation of responsibility to NHSE for orchestration of the healthcare system. This stands in contrast, however, to certain deficiencies in specificity around NHSE’s explicit areas of accountability established in the 2012 Act. For example, specialized commissioning is NHSE’s most significant direct spending responsibility (approximately £15 billion in 2015–16) yet it does not feature specifically in the mandate, with only a broad intention expressed to hold NHSE accountable for its direct commissioning outcomes (Specialised Services Commission 2016).

The 2017–18 mandate continues the approach of the 2016–17 iteration, with some minor changes. ‘Metrics’ are introduced to ‘measure progress […] in delivering the Five Year Forward View’ (Department of Health 2017, 8), and an explicit expectation is established that NHSE will support organizations in ‘…local areas in developing credible, financially balanced operational plans…’ (Department of Health 2017, 10). This is to ensure that NHS performance at a local level can be linked more explicitly to ‘national accountability’. Lastly, there is an explicit reference to NHSE’s requirement to comply with responsibilities as set out in the Framework Agreement (Department of Health, & NHS England 2014) and Managing Public Money (HM Treasury 2015) documents, focusing on ensuring the overall financial balance in the NHS.

The DH presents the mandate as a means of ensuring that NHSE is accountable to Parliament and the public (Department of Health 2017). Beyond NHSE’s accountability to Parliament through the Secretary of State, the organization may also be asked to give an account via the Commons Health Select Committee, Public Accounts Committee, and National Audit Office. Accountability to patients and public is via various self-assurance mechanisms, but these are fairly limited (Specialised Services Commission 2016). While our focus thus far has been on NHSE’s accountability ‘upwards’ to the DH and government, it is important to also recognize there are also accountability arrangements ‘downwards’ that continue to evolve. CCGs are accountable to NHSE, yet non-legislated policy developments post-2012 Act have resulted in collaborative commissioning arrangements for primary care and specialized services, both of which necessitate new relational dynamics between NHSE and CCGs, NHS hospital Trusts and others, which are not necessarily covered by existing ‘upwards’ or ‘downwards’ formal accountability arrangements (McDermott et al. 2018). This represents a layering of hierarchical and lateral relations between NHSE and CCGs, resulting in a more nuanced and locally variable set of governance arrangements. The picture is further complicated by the fact that while CCGs are accountable to NHSE, NHS Trusts and Foundation Trusts are accountable to another ALB: NHS Improvement. Consequently, accountability arrangements are more fragmented than they were pre-2012 Act and these conditions prevent any one organization exercising authority to steer local arbitration processes, particularly between commissioners and providers, and health system decision-making (Checkland et al. 2017). If the intention of these reforms were to provide a context whereby government could control the service at a distance, this more fragmented set of accountability relationships would suggest a failure in implementing an effective meta-governance mechanism.

This section shows that, while the mandate mechanism purports to provide a clear line of accountability between the DH and NHSE, and forms the main formal accountability link between the two, the reality of NHSE’s operation suggests that the accountability relationship is significantly more complicated in practice. We explore this in the following section, showing how NHSE has gradually adopted a position of primary policy maker in the context of a nuanced dynamic between government and ALB, suggesting a degree of autonomy and control over the health service that was not originally specified, and raising important questions about accountability.

The Five Year Forward View and NHS England as primary policy maker

In 2014, NHSE published a document called the Five Year Forward View (NHS England, Care Quality Commission, Health Education England, Monitor, Public Health England, & Trust Development Authority 2014). It set out a diagnosis of what ailed the NHS, and proposed a solution – the rapid adoption of new approaches to delivering services, focussing upon the integration of care across primary, secondary and community settings. The document called for the identification and funding of so-called ‘Vanguard’ sites to test ‘New Care Models’. Five types of new service model are identified, including: integrated primary and acute care systems, and multi-speciality community providers. Exemplars of each type were subsequently identified and offered funding to develop. Strikingly, given the content of the 2012 Act, no attention was paid to quasi-market competition in the NHS in the document. Indeed, the word competition does not feature once.

More recently, as a means of delivering the objectives of the Forward View, local health and care organizations across England have been encouraged to come together to form ‘Sustainability and Transformation Partnerships’ (STPs) (NHS England, NHS Improvement, Care Quality Commission, Health Education England, National Institute of Health and Care Excellence, & Public Health England 2015). Forty-four geographic STP areas have been demarcated, each with a designated leader. Every STP was required to ‘create its own ambitious local blueprint for accelerating its implementation of the Forward View’ (NHS England, NHS Improvement, Care Quality Commission, Health Education England, National Institute of Health and Care Excellence, & Public Health England 2015, 4), setting out how organizations will collaborate within their footprint in order to increase efficiency and improve patient care. As an architect of the policy noted, STPs are intended as a means of getting local places to ‘own’ a portion of the NHS deficit in the hope of addressing it more effectively (Hammond et al. 2017). STP areas whose plans are adjudged to be acceptable can access £2.1 billion of sustainability and transformation funding to support the local health and care economy and pay down deficits. STP areas whose plans are not acceptable have no such access.

STPs represent a significant change in national policy and a move further from the model of quasi-market competition which the 2012 Act re-emphasized as a statutory requirement for the NHS’ operation. As the STP guidance states: ‘planning by individual institutions will increasingly be supplemented with planning by place for local populations. For many years now, the NHS has emphasized an organizational separation and autonomy that doesn’t make sense to staff or the patients and communities they serve’ (NHS England, NHS Improvement, Care Quality Commission, Health Education England, National Institute of Health and Care Excellence, & Public Health England 2015, 4). This is a policy presented by an ALB as a means of both diagnosing and addressing the problems with the health service. It is a ‘workaround’ instituted by an ALB (Alderwick et al. 2016, 6), encouraging the creation of local inter-organizational collaborative networks which have no statutory authority, and largely operate in tension with existing government policy and legislative frameworks.

Further communications from NHSE stated that all STPs will become Accountable Care Systems (ACSs) which will ‘bring together local NHS organisations [providers as well as commissioners], often in partnership with social care services and the voluntary sector’ to create locally joined-up services and reduce system fragmentation (NHS England 2017). These are major changes to the way the NHS operates, indicating a shift away from competitive commissioning and a move from a market-type governance arrangement to one based upon networks. This has occurred without any significant policy debate about the benefits or risks of doing so.

Accountable Care Organizations (ACOs), announced alongside ACSs (NHS 2017) and related to some New Care Models types (NHS England, Care Quality Commission, Health Education England, Monitor, Public Health England, & Trust Development Authority 2014), would typically involve a group of providers contracted to work together to provide health and care services for a population to a specific budget. For commissioners, this means having a single contract with the ACO rather than multiple contracts with multiple providers. ACOs represent a key plank in the objective announced in the ‘Next steps on the NHS Five Year Forward View’ ‘…to make the biggest national move to integrated care of any major western country’ (NHS 2017, 31). However, the British Medical Association expressed concern about ‘the lack of clarity and accountability surrounding their development so far’ (British Medical Association 2018), clearly illustrating the issues that we have raised with the ambiguities surrounding NHSE’s accountability for its actions. Concerns among sections of the public about the potential for ACOs to increase privatization and care rationing resulted in two High Court challenges, one of which names the Secretary of State for Health and NHS England as joint defendants, and the other, which names only NHS England (Dyer 2018).

In February 2017, Simon Stevens (Chief Executive, NHSE) gave evidence to the Public Accounts Committee and stated that the purchaser-provider split (of the quasi-market) would ‘effectively end’ as a consequence of these changes (House of Commons Public Accounts Committee 2017). It is highly notable that the Chief Executive of an ALB is making such significant statements about the reform of a health service, without apparent concomitant co-ordination with government in terms of the legislative context, particularly given the scale of the proposed change (and the amount of public money involved).

At the same time as this shift in policy direction, ambiguity about NHSE’s role and status, and that of its Chief Executive, has increased. A letter from the Secretary of State for Health in 2013 to the previous Chairman of NHSE emphasized that changing the name of the organization from NHS Commissioning Board to NHSE ‘…does not mean that NHS England will now become the headquarters of the NHS in England’ (Hunt 2013, 2), yet this is precisely the impression that is now commonly propagated by both NHSE and government. For instance, the Five Year Forward View stated that an additional £10 billion increase by 2020–21 would be necessary to ensure the continued successful operation of the NHS. In referring to this, the government stated that it was backing ‘the NHS’s own plan for the future’ (Campbell 2016). Across a range of media, Stevens is commonly referred to as ‘the Chief Executive of the NHS in England’, ‘the NHS boss’, and ‘the head of the NHS in England’ (e.g. Campbell 2017). This conflation between NHSE and ‘the’ NHS is compounded by the fact that official publications now often carry the NHS logo without additional specification of authorship.

This has allowed the Secretary of State to promulgate a degree of ‘othering’ of the NHS not previously possible. The Secretary of State has, consequently, subtly adjusted his role ‘from chief defender of the providers to principal patient advocate – the man holding the service to account on our behalf – rather than himself being held to account for how it performs’ (Rutter 2014, 150). Conversely, Stevens has at times been the most prominent figure in the media providing an explanation for NHS performance problems when previously it would fall to the Secretary of State to take on such tasks. For example, in March 2017 Stevens went on national radio to announce that the target on elective routine surgery waiting times would be dropped. He presented this as necessary to reduce pressures on accident and emergency departments and cancer treatment, and was even lauded in sections of the British press for coming up with a plan to address problems with the health service and being sufficiently brave to pursue it (Orr 2017). It is questionable whether the Secretary of State would receive a similar response for delivering the same message. However, the fact that Jeremy Hunt (now Secretary of State at the Foreign Office), was the longest serving Secretary of State for Health since the NHS was founded, is taken as evidence by some that the dynamic with NHSE has proved politically protective (Timmins 2017).

NHS England: autonomy, accountability, and meta-governance

In this section, we call on the theoretical concepts set out previously to explore the case of NHSE. Returning to Flinders and Tonkiss (2015) conceptual map of ALB-sponsor autonomy (Figure 1), NHSE’s promulgation and apparent orchestration of national policy which deviates from existing legislation and government policy can be interpreted as a situation where ALB is operating with a high level of autonomy and little accountability: a loose-loose neglectful parenting model. Unfortunately, reliable insights into the dynamic between government and NHSE are scarce. In 2014, Timmins (2014) highlighted the difference in approach between successive Secretaries of State for Health. Jeremy Hunt was appointed in 2012 after Andrew Lansley, the architect of the 2012 Act, was dismissed from his ministerial post. On one hand, Lansley’s vision of a self-correcting, highly autonomous NHS suggests an aspiration of a loose-loose (neglectful parenting) approach with very minimal government involvement; on the other, Hunt reportedly arranged twice weekly meetings with NHSE and other national NHS organizations where current and forthcoming delivery problems were discussed (Timmins 2014). This suggests, at least at that time, that Hunt was exerting tighter operational control than previously. If one assumes that NHSE was indeed driving national strategy at this time then a loose-tight (indulgent parenting) approach, which Flinders and Tonkiss (2015) note is somewhat anomalous, was potentially in effect. This approach results in loose strategic control coupled with tight operational control (little control in terms of policy aims, but significant oversight of day-to-day practice). For example, Hunt was reported to have personally called hospital chief executives whose accident and emergency department performance had dropped to enquire what action was being taken (Timmins 2014). Whilst this undermines the 2012 Act’s framing of the reformed NHS as free from ‘political interference’, this activity might also be construed as behaviour from the Secretary of State that is consistent with a role of chief patient advocate. The current de facto dynamic between ALB and sponsor, however, remains largely opaque beyond the publicly transmitted vision whereby NHSE exercises significant autonomy in regards to the strategic direction of the NHS.

Revisiting Bovens (2007) democratic and constitutional accountability perspectives, whilst NHSE is explicitly identified as accountable to the public for its performance in fulfilling its statutory duties, clear processes for the public to hold NHSE to account are limited and where specific responsibility lies for driving national policy has become more uncertain (Specialised Services Commission 2016). The conflation of NHSE and ‘the’ NHS contributes to this, and represents an expression of a common criticism of ALBs. In this regard, on one hand, it seems that NHSE is explicitly, and overtly, exercising considerable influence over the orchestration of the NHS. On the other, when we push to identify hard evidence to substantiate this, we find that we cannot say, with significant certainty, that this is actually the case. This creates ambiguity over the contours of accountability. This is illustrated clearly in the way that NHSE commissions specialized services, something that the ALB has been criticized for due to a lack of transparency (Specialised Services Commission 2016), including failure to publish minutes of commissioning meetings. As noted, specialized services commissioning represents NHSE’s most significant direct spending responsibility and the lack of a robust regime for holding NHSE to account for it is a concern.

A constitutional accountability perspective requires us to consider whether accountability arrangements are suitably strong so as to ensure that public officials/organizations are sufficiently disincentivized from misusing authority. ALBs’ potential utility to government comes from their capacity to ‘[o]ccupy a niche whereby ministers can remove themselves from direct responsibility for policy implementation and quasi-judicial decisions, and at the same time gain the informational and political benefits of incorporating outside expertize in the policy process’ (Flinders and Skelcher 2012, 327). In the framing of the creation of NHSE, and its operation, we see how public confidence in the health service might logically be improved by having ‘a lean and expert organisation, free from day-to-day political interference’ (Department of Health 2010, 30) overseeing the NHS’s operation. Yet, we have shown that NHSE offers a politically useful insulation to government ministers. What are the limits of such usefulness? How much latitude might ministers be minded to offer in return for this insulatory effect, particularly given the broader political context of austerity and the dominance of macro-issues such as the United Kingdom leaving the European Union? These questions are important when considering the case from a constitutional accountability perspective, and the level of information currently available about the dynamics of the relationship between NHSE and government is insufficient to develop an informed judgement.

Such concerns are not particularly new. A 2014 Public Administration Select Committee report examining accountability in the relationships between government departments and ALBs identified an enduring lack of clarity in the way that different types of ALBs are governed by government departments, precisely where accountability resides, and the underlying rationale for any such arrangements. It advocated the production of a taxonomy for public bodies that makes clear their governance arrangements and the nature of sponsorship. Of NHSE, the report states:

“Up to date, plain English statements of statuses, roles and relationships are needed even if the underlying arrangements are complicated. This is far from the reality in many cases, particularly in the NHS. With a budget of £95.6 billion NHS England is now by far the largest public body in England and its accountability should not be in any doubt, but it is still evolving” (House of Commons: Public Administration Select Committee 2014, 3)

There is little to suggest that significant steps towards addressing these recommendations have been taken since the report’s publication, and our analysis strengthens the pressing case for such action.

Exworthy, Powell, and Mohan (1999) characterized the NHS as a hugely complex organization, which exhibits evolving, and locally specific, combinations of market, hierarchy, and network. Post-2012 Act, the NHS continues to exhibit evolving strata of these governance modes with, for example, the continued use of competitive tendering (market), national outcome frameworks and associated assurance processes (hierarchy), and STPs (network). NHSE has taken on a key role in determining the balance between, and content of, hierarchical, networked, and market governance activities within the NHS, making it a very powerful actor. For instance, the introduction of New Care Models, STPs, ACOs and ACSs demonstrates both an engineering of network constellations involving a broad range of organizations and the use of mechanisms that shape market dynamics. This is achieved through ‘soft’ means, such as the propagation of narratives articulating the problems that the NHS faces and presenting new collaborative arrangements in order to solve them. Again, the intentional conflation of NHSE with ‘the’ NHS renders these particularly compelling, and, as noted, benefits both NHSE and Government. Furthermore, NHSE commonly presents the policies that it spearheads as reflecting ‘wide consensus’ in the NHS (NHS England, Care Quality Commission, Health Education England, Monitor, Public Health England, & Trust Development Authority 2014, 16) about what is required. There are also ‘hard’ means employed, such as pressure exerted on organizations to participate in an STP and produce a plan for making efficiency savings in order to access the much needed additional funds via the ‘sustainability and transformation fund’.

Crucially, aspects of these inter-organizational collaborations and partnerships may exist in tension with existing competition regulations (Sanderson, Allen, and Osipovic 2017). In one case in 2017, the Competition and Markets Authority approved a merger between two Foundation Trust hospitals, despite it having recognized an adverse effect on market competition. In part, the approval was justified because it was in harmony with a national policy established by NHSE (Dunhill 2017). In this context, it appears that NHSE policies, which encourage integration and collaboration rather than competition, have superseded the statutory position set out in the 2012 Act. In this sense, it might be viewed as an example of a failure in the governance of competition (Jessop 2015).

Kelly’s (2006, 608) Audit Commission analysis framed the ALB organization as a ‘vehicle of meta-governance’ because it operated as a governance tool for government to purposes specified in legislation. This is consistent with assertions that it is government that most commonly functions as meta-governor because it establishes certain boundaries to the operation of networks and markets as well as exerting, or threatening to exert, hierarchical control, i.e. governance forms occur in the ‘shadow of hierarchy’ (Fawcett and Marsh 2014). NHSE’s (apparent) orchestration of governance configurations across the NHS through its shaping of policy and practice goes beyond this. We argue that NHSE is itself wielding and operating a range of tools and thus merits classification as a meta-governor in its own right. The nature and extent of NHSE’s role as meta-governor, however, should be considered in relation to Flinders and Tonkiss (2015) framework of ALB-Sponsor autonomy presented earlier, because the actual degree of strategic and operational control NHSE enjoys determines the extent to which the meta-governor label is warranted. The uncertainty of this dynamic – of the distribution of meta-governing activities between ALB and government – obscures where power is situated and how it is being deployed to drive change in an area of public administration of extreme import. This may yet be the intended consequence of the relation between NHSE and government. Fawcett and Marsh (2014, 181) argue that ‘depoliticisation, as a process, usually takes place in the shadow of statecraft’, and our analysis suggests that depoliticization of the English NHS is occurring in the shadow of meta-governance processes, yet it is uncertain to what extent it is NHSE or government casting the shadow itself.

Discussion

We have shown that developments in England post-2012 Act have created a situation where, because of the operation of an ALB, the policy direction and orchestration of the healthcare system is harder to discern than previously. Using a meta-governance lens, we have shown how political benefits associated with agencification of responsibility for healthcare delivery have enabled a policy entrepreneur (Simon Stevens) to apparently assume for his organization a far-reaching meta-governor role, with associated power to set the agenda and determine the direction of policy. In particular, NHSE under Stevens has introduced policies (STPs, ACOs and ACSs) with the potential to alter the balance between networks, hierarchies and markets in a complex system. The ostensible accountability mechanism – the so-called mandate – is vague and non-specific, providing little leverage for the Government in managing the performance of NHSE. The fact that it is an ALB – framed as the health service itself – that has been spearheading these changes means that scope for debate about the desirability and potential dangers of such approaches is diminished – as are opportunities to hold the relevant actors accountable (both directly and indirectly).

We have explored the potential benefits accruing to government from this arrangement, but significant issues remain which are relevant to meta-governance more widely. It is clear from our analysis that the strong meta-governor role assumed by NHSE has evolved rather than being designed. This suggests further study of meta-governance might benefit from what Bevir has called a ‘decentred’ approach (Bevir 2013), in which focus is maintained upon the enactment of governance rather than its formal structures. Our conceptualization of the task of meta-governing has proved valuable in elucidating nuances of this case study, the issues it raises for the legitimacy of governance arrangements, and the location and deployment of power in public management.

Meta-governance theory has provided a framework within which to consider the bilateral relations between an ALB and its sponsor department. This leads on to a discussion of accountability, and we have shown that a lack of transparency around the relationship between ALB and Government means we cannot sufficiently assess the degree of autonomy exercised by the ALB or the contours of accountability. NHSE’s operation as an ALB thus far suggests accountability deficits from both a democratic and constitutional perspective. Democratically, it is unclear who should be held to account for policy decisions, whilst constitutionally the uncertainty that we have described raises questions about the legitimacy of current governance arrangements (Bovens 2007). Flinders and Tonkiss (2015) framework of ALB-sponsor autonomy provides an instructive means of articulating this relationship. We highlight the lack of clarity over the extent and strength of sponsor oversight of the ALB, and concur with the House of Commons: Public Administration Select Committee (2014) that the nature and dynamics of sponsorship for NHSE, and the extent and nature of autonomy that NHSE enjoys, should be made clearer as a matter of priority. Whilst our documentary analysis approach has allowed us to interrogate the espoused policies of both the ALB and associated Government Department, and compare these with public utterances, more detailed exploration of these dynamics would require primary data collection within the organizations themselves.

These changes have occurred under the rubric of reducing ‘political interference’ (Department of Health 2010, 30) in the operation of the NHS. The consequence has been to introduce an additional link in the chain of accountability, which obfuscates where power and accountability lie, thus de-politicizing further the development of a significant public service. While de-politicization might have short-term advantages for those in government, future research might focus on the longer term consequences of such a strategy. De-politicization of policy in an attempt to reduce potential conflict can ultimately be counterproductive and serve to increase conflict over the long term, in addition to reducing some of the constructive benefits that can be associated with such conflict (Wolf and Dooren 2018). Typically, the actions of an ALB are taken to be more objective and less partisan than those of Government (Mulgan 2007). However, NHSE’s establishment as a meta-governor and de facto policy maker should be considered alongside Stevens’ (as NHSE Chief Executive) tendency to leverage politically sensitive issues to make a point and a case for increased NHS funding (e.g. in a 2017 speech appealing for Government to ensure the NHS received the £350 million per week that the Vote Leave suggested would be available if the UK voted to leave the European Union in the 2016 referendum (Stevens 2017). The potential tensions between Government and ALB arising from such actions might feasibly incentivize Government to opaquely decide future appointments on the basis of candidates’ political views, and expectations of acquiescence, as well as their experience. This possibility makes clarification of sponsor-ALB dynamics, as noted above, all the more pressing.

How the role of NHSE as an ALB evolves will be of interest for its own sake and because of the potential insights it offers on the interplay of meta-governing forces in a modern Western context. Our case study suggests it is possible for an ALB to assume a meta-governor role beyond that which was initially articulated in its creation. One explanation lies in the complexity of modern public service delivery. NHSE can be understood as a response to the contemporary environment in which many governments find themselves: attempting to cope with rising healthcare demand from a consumerist, increasingly multi-morbid and elderly population, with technological advancement (and associated expense), budget deficits, and an overarching austerity agenda. In other words, struggling to control a complex and hugely politically significant public service area, and exposed to blame and the political consequence of problems within it. The attraction for governments of having an ALB oversee the operation of such a politically charged area of public service is clear, and likely to manifest in similar ways in other country contexts. However, our case study highlights dangers inherent in this approach.

Conclusion

Our use of a meta-governance lens has allowed us to ask questions about autonomy and accountability within and between an ALB and its sponsor department, moving beyond a consideration of regulation and oversight to examine the bilateral relationships at work and highlighting the assumption by an ALB of a meta-governing role. In an increasingly complex world of public service orchestration and delivery, with evolving combinations of pressures relating to market competition, collaboration, and integration, our case study suggests that asking questions of the details of the meta-governance arrangements present, sketching the contours of these arrangements as far as possible, and highlighting deficiencies of transparency where they exist is a useful approach. In particular, given that the appropriate balance of governance approaches in public services between markets, hierarchies and networks is a profoundly political question, the meta-governance lens provides a valuable tool through which to interrogate – and potentially enhance – democratic accountability arrangements.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Jonathan Hammond

Jonathan Hammond is a Research Fellow in the Division of Population Health at The University of Manchester. His research interests are in health policy, health service orchestration and history, and health care governance.

Ewen Speed

Ewen Speed is Senior Lecturer in Medical Sociology in the School of Health and Social Care, at the University of Essex. He has researched on-going changes in the policy context of the UK NHS, with a view to understanding the changing role of the state in the provision of public services.

Pauline Allen

Pauline Allen is Professor of Health Services Organisation at London School of Hygiene and Tropical Medicine. Her research interests include socio-legal theory, institutional economics, organizational theory, governance, and the structure of the NHS.

Imelda McDermott

Imelda McDermott is a Research Fellow in the Division of Population Health at The University of Manchester. Her research focus is on health policy, commissioning, governance, and the organization of health services. She has an interest and expertize in the use of linguistic methods for policy analysis.

Anna Coleman

Anna Coleman is Senior Research Fellow in the Division of Population Health at The University of Manchester. She worked in a variety of policy and research roles in local government before joining the University where she has specialized in health and care policy, partnership working, commissioning and governance.

Kath Checkland

Kath Checkland is Professor of Health Policy and Primary Care at The University of Manchester. Her work focusses upon the impact of national health policy on primary care organizations, and she works part-time as a General Practitioner.