Letting go and holding on: The politics of performance management in the United Kingdom

Governance-by-numbers and logics of performance management

Described as one of the ‘most widespread international trends in public management’ (Pollitt, 2006: 25), countries throughout the world have embraced performance management to deliver a range of policy or ideological benefits, symbolic benefits or direct electoral benefits (Hood and Dixon, 2010; see also Behn, 2003; James and Wilson, 2010). Indeed, performance management has been described as ‘among the most important tools by which governments structure relationships, state values, and allocate resources with employees, third-party providers and the public’ (Moynihan et al., 2011: 141). Yet whilst the vocabulary of performance management assumes a technocratic air, ‘the definition of performance in a public sector organization is often elusive’ (Smith, 1995: 278). The risk therefore exists that the focus shifts to the measurable at the expense of the important, resulting in a ‘naïve goal oriented model’ which ‘not only defines and therefore constrains what is measured’ but ‘also ignores the long-established understanding that public sector goals are ambiguous, multiple, complex and frequently in conflict with one another’ (Jackson, 2011: 15). Several scholars have also questioned whether public service performance can be measured with a meaningful degree of accuracy (e.g. Carter et al., 1992; Flynn et al., 1988; Jackson, 2011). Whilst outputs are tangible and more readily quantified, it is the outcomes of public services – their contribution to human welfare – that are of primary importance. Yet, outcomes are difficult to manage, as often government departments and service providers will not control all the levers necessary to change; and the achievement of an outcome could be coincidental to the effects of a target, or even in spite of it. Indeed, Boyne and Law (2005: 254) suggest that ‘in some circumstances, the external environment may be so influential that the outcome is felt to be outside the control of the organization’, which could have the unintended consequence of promoting complacency or inactivity: ‘whether the target is hit is likely to be beyond the control of the organization, so service managers may as well sit back and see if “their number comes up”’. Moreover, if outcomes focus on the important, then defining ‘the important’ is in itself an inherently political act (Stewart and Walsh, 1994), which in turn risks the neglect of non-priority areas and the demoralisation of staff that are not engaged in the delivery of priorities (Moynihan et al., 2011: 150).

Such perverse consequences have been widely recognised. Bevan and Hood (2006: 521), for example identify ratchet effects (‘bas[ing] next year’s targets on last year’s performance, meaning that managers … have a perverse incentive not to exceed targets even if they could easily do so’); threshold effects (‘providing a perverse incentive for those doing better than the target to allow their performance to deteriorate to the standard’); and output distortions (‘[a]ttempts to achieve targets at the cost of significant but unmeasured aspects of performance’). Jackson (2005) distinguishes between ‘definitional gaming’ (where the definition of a target’s subject or remit distorts what is being reported); ‘numerical gaming’ (where an organisation presents their data in a way that exaggerates their performance); and ‘behavioural gaming’ (where changes in behaviour that allows a target to be met have adverse effects on other areas of an organisation’s work). Building on this, Coulson (2009: 277) advances a fourth category of ‘fraud’ to capture instances of the knowingly false reporting or alteration of performance data. The empirical manifestation of such behaviour has been subject to particular attention; and the literature is replete with examples of how service providers manipulated the system to protect their interests (e.g. Bird et al., 2005; Hood, 2002; Propper and Wilson, 2003). In their study of the star ratings system applied to NHS hospitals in England and Wales between 2001 and 2005, Bevan and Hood (2006) reveal that many hospitals diverted resources to activities that would be more easily awarded good star ratings (e.g. waiting times and cleanliness), at the expense of less quantifiable aspects of performance (e.g. the quality of treatment provided). Moreover, within the field of healthcare, several studies have underlined the potentially fatal consequences of such instrumentality. The regular publication of unadjusted patient mortality rates by the New York State Department of Health, for example prompted reluctance amongst cardiac surgeons to take on high-risk cases, which subsequently led to an increase in deaths amongst the most vulnerable at-risk Medicare patients (Dranove et al., 2003). Similarly, in England, the stipulation that 75% of ‘immediately life-threatening’ emergency calls were responded to within 8 minutes encouraged ambulance trusts to concentrate their fleet in densely populated areas at the expense of patients in rural areas (Bevan and Hamblin, 2009).

It is clear that the measurement of public sector performance is beset by operational challenges, which are further exacerbated by the complex and uncertain environment in which public providers function. Against such a backdrop, important questions therefore remain regarding the added-value of performance management; and several scholars have highlighted the lack of ‘definitive evidence about whether performance regimes ultimately improve public sector capacity and outcomes’ (Moynihan et al., 2011: 151; see also Boyne and Chen, 2007; Jackson, 2011). That such questions persist reminds us that ‘management is not some neutral technical process, but rather an activity which is intimately and indissolubly enmeshed with politics, law and the wider civil society. It is suffused with value-laden choices and influenced by broader decisions’ (Pollitt and Bouckaert, 2004: 14). Such contextual variables are therefore key to understanding the design and implementation of performance management, and several comparative studies have sought to explore their effect thereupon (e.g. Bouckaert and Halligan, 2008; Pollitt, 2006; Pollitt and Bouckaert, 2004). Pollitt and Bouckaert (2004), for example compare performance management in 12 Anglo-American and European states, analysing their politico-administrative regimes in accordance with five factors: state structure, executive government, minister/mandarin relationships, administrative culture and diversity of policy advice. Elsewhere, Bouckaert and Halligan (2008) argue that variances in the ‘span’ and ‘depth’ of performance management are a product of a polity’s public administrative framework, and that reform trajectories follow different paths in (neo-) Weberain and (neo-) public management states. In a slightly different vein, Hood (1999, 2012) applies grid-group cultural theory to identify four predominant styles of public management organisation (fatalist, individualist, heirarchist, egalitarian), and the way in which these different social contexts encourage the uptake of certain forms of performance measurement.

Taken together, such comparative studies reveal an interplay between the institutional framework of a given polity and the performance management tools adopted by its executive; and in turn highlight the way in which the attendant governing norms of such institutions structure relationships between ‘managers’ and ‘the managed’. Institutions therefore matter: institutions ‘frequently have a profound shaping effect on what actually happens during the course of (public management) reform’ (Bouckaert and Pollitt, 2004: 23); and a dynamic relationship exists between the structural features and cultural norms generated by such institutions (Christensen and Lægrid, 2008: 138). More broadly, such studies have highlighted the effect of macro-level ‘patterns of democracy’ (Lijphart, 1999, 2012) on meso-level ‘patterns of public management’, with a discernable distinction existing between the majoritarian states of Anglo-America and the consociational states of Western Europe. Thus, the predominance of central government, the top-down governing culture, and the capacity for rapid policy change of majoritarianism is broadly aligned with the top-down imposition of indicators, the combative deployment of performance data, and an enthusiasm for the latest managerial trends. In contrast, the wider dispersal of power, the consensual governing culture, and the tendency towards policy stability of consociationalism is more likely to result in the local determination of priorities, the decentralisation of responsibility for delivery, and a more incremental approach to public service modernisation. Whilst such distinctions are undoubtedly more nuanced than this brief sketch allows, in the light of such findings, the limits of policy learning and transfer have been stressed. Christensen and Lægrid, for example argue that ‘[r]eforms that are culturally incompatible with political-administrative traditions will be more difficult to implement and will have less effect than those that are compatible because they are seen as culturally “inappropriate”’ (Christensen and Lægrid, 2008: 135; see also Jones and Kettl, 2003). Similarly, Pollitt suggests that ‘[p]olicies and practices are very frequently sensitive to contexts’, and therefore ‘it may be easier to learn reliable lessons from the past of one’s own organization or sector (where contextual features will remain reasonably constant) than from other jurisdictions, however “advanced” or exotic’ (Pollitt, 2008: 166).

Overall, such studies indicate the effect of institutional factors upon the dynamics of performance management within different politico-administrative contexts. Building on these observations, this article therefore adopts a (broad) ‘new’ institutionalist approach to explain the politics of performance management in the UK, and to account for the prevailing disjuncture between the ‘talk’ and ‘action’ described in the introduction (Brunsson, 1989, 1993). In broad terms, new instititutionalism understands institutions as constituting:

a relatively stable collection of rules and practices, embedded in structures of resources that make action possible – organizational, financial and staff capabilities, and structures of meaning that explain and justify behavior – roles, identities and belongings, common purposes, and causal and normative beliefs. (March and Olsen, 1989, emphasis in original)

Together, the strands of scholarship reviewed in this section combine to develop an important argument regarding the inescapable effect of the overarching politico-administrative context on performance management; and, in turn, the utility of adopting an institutional lens to examine the dynamics of this relationship. In case of the UK, this lens directs attention to the way in which Westminster majoritarianism gives rise to a set of institutional norms regarding the balance of power between core and periphery (see Lijphart, 1999, 2012; Matthews, 2013, 2015); and provides a potential explanation for the persistence of a top-down, highly interventionist approach to performance management. The following sections interrogate this further by focusing on the politics of performance management in terms of governing incentives and institutional norms; and in doing so highlights the prevailing disconnect that exists between the ‘talk’ and ‘action’ of both the Labour and Coalition governments.

Whole-of-system control-freakery: The Labour Governments, 1997–2010

‘For the first time’, the Prime Minister Tony Blair announced in 1998, ‘we are setting targets right across the public services for the modernisation and reform we need to meet the testing demands of the future’. Moreover, he stated, the approach of New Labour would be different, as ‘[t]oo often in the past governments have only made commitments for what they put into public services – money, manpower, and policies – and not for what the public will get out in return’ (Cm. 4181, 1998: foreword). Despite such rhetoric, attempts to foster effective performance within Whitehall and across the British state were not new. Policies such as the Financial Management Initiative of 1982 and the Next Steps programme of 1988 had sought to strengthen link between funding and outputs; the establishment of the Audit Commission in 1983 had brought the performance of local authorities under systematic scrutiny; and the creation of tools such as the Citizens’ Charter in 1991 had mandated standards for public service organisations. However, what did mark out Labour’s approach was the scope and scale of its interventions, which left not a single area of state activity untouched, and resulted in a brave new ‘targetworld’ redolent of the Soviet regime (Hood, 2006).

Within government and across the state, the Government established a multiplicity of performance frameworks intended to deliver its ambitious public service agenda. To drive key priorities across Whitehall, the Government introduced Public Service Agreements (PSAs) in 1998, which delineated the core policy objectives for individual departments, underpinned by a series of performance targets. Arising out of the 1998 Comprehensive Spending Review (Cm. 4011, 1998), several principles informed the PSA regime. Expenditure was fixed for a three-year period to enable departments to manage their resources more effectively, and became linked to the achievement of policy outcomes. Outcome-focused multi-year planning was also intended to promote greater co-ordination across all layers of government, and Departments were encouraged to work together to achieve policy goals that transcended Whitehall silos. Finally, the monitoring arrangements in place to oversee progress towards PSA targets reinforced the role of the centre as the driver of the new regime, its system of sanctions and rewards intended to ensure that the focus across government remained on the centre’s core objectives. Whilst the PSA framework initially established an eye-watering 600 targets across Whitehall, this gradually reduced and by 2007 a ‘streamlined set’ of 30 new PSAs were established to ‘articulate the Government’s highest priority outcomes’ and ‘span departmental boundaries, setting out a shared vision and leading collaboration at all levels in the delivery system’ (Cm. 7227, 2007: 36). As well as fostering strategic working across Whitehall, the PSA framework also sought to engage those responsible for delivering public services on the ground, uniting core and periphery around a set of shared goals. This was underpinned by the principle of ‘earned autonomy’, which would ‘ensure that public service providers have the discretion to innovate and improve the services they provide, constrained by the need to reach high minimum standards’ (Cm. 5570, 2002: 10).

On the ground, however, the mushrooming of performance mechanisms belied the Government’s professed commitment to ‘constrained discretion’. To formalise the relationship between top-level PSA priorities and on-the-ground delivery Local Public Service Agreements (LPSAs) were introduced in 2000. Initially piloted across 20 upper-tier local authorities, LPSA targets were intended as a ‘partnership agreement between individual local authorities and the Government, intended to accelerate or surpass key outcomes than would otherwise be the case, for people living in the authority’s area’ (ODPM, 2001: 1) . By 2002, LPSAs had been rolled out to over 60 upper-tier local authorities and by 2003, all but three upper-tier local authorities had negotiated an LPSA (Cm. 5571, 2002: 1). Yet in 2007, LPSAs were replaced by Local Area Agreements (LAAs), three-year agreements negotiated between central and local government which set a series of service improvements that authorities were committed to achieving, along with a detailed delivery plan. No longer explicitly linked to the PSA framework, LAAs entailed up to 35 priorities for each local authority, underpinned by ‘a single set of about 200 outcome based indicators covering all important national priorities like climate change, social exclusion and anti-social behaviour’; although the Government insisted that ‘LAA targets will generally be negotiated to balance local priorities and levels of performance with national improvement priorities’ (Cm. 6939-I, 2006: 6.39). At the same time, local authorities remained subject to the programme of inspection administered by the Audit Commission; and during this period, the Commission was tasked by the Government to implement several new audit regimes. In 2001, the Audit Commission launched the Comprehensive Performance Assessment, which rated councils against an excess of 1,000 performance indicators. This was replaced in 2009 by the Comprehensive Area Assessment (CAA), which was intended to streamline the process by assessing local public services against a total of 214 indicators. This period also witnessed the micro-management of specific public services via performance indicators of increasing granularity, such as the ‘balanced scorecard reflecting staff, patient and clinical focus’ applied to individual NHS hospitals in England and Wales between 2001 and 2005; and the aggregation of such information via published rankings and ratings served to underline the voracity of the Government’s ‘targets-and-terror’ ambitions.

Yet although impressive in scale, Labour’s approach to performance management was unplanned, ad hoc and often reactive. One senior civil servant described the PSA framework as little more than ‘an afterthought of the Spending Review … [t]he rationalisation and caveat creation’ (interview 35); and another acknowledged the way in which such machinery was ‘invented on the hoof’ (interview 39). More bluntly, one member of the Treasury Select Committee deplored the system as ‘a mess’ (interview 37). Unsurprisingly, this fragmentary ‘hyper-innovation’ (Moran, 2003) resulted in a lack of clarity regarding the roles of central and local government. The Public Administration Select Committee highlighted a ‘lack of proper integration’ between the simultaneous development of front-line organisational capacity and a centrally driven measurement culture, which had created ‘tension between those charged with centralised responsibility and those who are responsible for dispersed delivery of public services’ (HC 62-I, 2003: i). Such concerns were reiterated by several interviewees. Contra the rhetoric of flexibility and ‘earned autonomy’, local authority officers stressed the extent to which targets were ‘imposed’ (interview 12) and ‘fairly top-down’ (interview 15). This didactic approach to target setting was acknowledged by several within Whitehall. One senior civil servant in the Department for Health described the frenetic lobbying that occurred during the negotiation of LAA targets: ‘it was a bit of a bun fight to … persuade [local authorities] for your indicators to be included’ (interview 41); and another in the former Department for Children, Schools and Families acknowledged that whilst they sought to hold a ‘fairly open’ and ‘genuine two-way dialogue’, ‘if you were on the other side of the table you might have thought we were trying to enforce a top-down process’ (interview 22).

The feasibility of ‘earned autonomy’ was further constrained by the explosion of performance indicators further down the delivery chain. As one local authority officer put it, ‘we’ve had thousands of the flipping things. There’s more than you can shake a stick at’ (interview 11). Acknowledging the impact of this cascading, one former special adviser admitted ‘it’s just not possible, if you’re in a local authority, to treat 57 things as a priority. It’s absurd’ (interview 36). Indeed, many local authority actors suggested that they lacked the resources necessary to respond to the demands of the centre. ‘They’re tightening up our purse strings and then expecting us to do more and more’, complained one local authority officer, ‘and there just isn’t the capacity to do what they want’ (interview 15). Faced with ‘targets-and-terror’, the emergence of perverse consequences was inevitable, as local authorities and other services providers buckled under the burden of competing pressures. Evidence submitted to the Public Administration Select Committee illustrated the prevalence of output distortions or gaming in the field of healthcare, such as the cancellation of follow-up appointments to ensure that targets for new patients were met (HC 1259-i, 2002: Q. 2); and the diversion of resources towards target priorities at the expense of non-priorities (HC 62-iv, 2002: Qs. 431–434). Similarly, one local authority officer described the existence of ‘threshold effects’: ‘there’s no incentive to overachieve because there’s a limit of budgets, so you need to justify why you want to spend more money’ (interview 12). Such challenges underlined the limits of target-based performance management as a top-down mechanism of public service control; and several interviewees within Whitehall acknowledged the extent to which the machinery of targetry became overextended. One special adviser admitted ‘we ended up with far too many targets’ (interview 34); and one senior civil servant described many of these targets as ‘ridiculously ambitious and clearly beyond what government can control’ (interview 45). Flowing out of this, a senior Cabinet Office official vividly described the ‘illusion’ that was allowed to develop:

one of the big messages we were communicating was … a delusion that you can control things from the centre. The fact that we have a PSA board with all these levers that we yanked to deliver a strategy and a goal, created an impression that in Whitehall they have a dashboard that they sit at and control the world. (interview 44)

Certainly, the particular personalities at heart of the New Labour project were important. One special adviser described how ‘Blair didn’t trust local authorities and didn’t trust a lot of the deliverers of public services’ (interview 36); and apropos of Hood’s (2006) ‘targetworld’, one Labour backbencher described Blair’s ‘authoritarian tendencies … even Lenin was more flexible that that!’ (interview 28). Yet a focus on individuals offers only a partial explanation, and a range of evidence points towards the way in which Labour’s approach to performance management was intimately bound up with Westminster’s highly centralised governing culture. One former official in the Prime Minister’s Office described the Government as having ‘a “pushmi-pullyu” approach to local government, which debilitated any kind of understanding of the centre’s [relationship] with the local … and therefore its statecraft was very confused’ (interview 48). Moreover, a belief in the rightful predominance of the centre was explicit in several testimonials. As one former minister explained, ‘if government is responsible for handing money out, then it is absolutely entitled to lay down the parameters of how that is spent’ (interview 51). Similarly, another former minister stated ‘you can’t “earn” autonomy by negotiating Whitehall and delivering services … It denies the policy-making role of Whitehall. There’s no such thing as autonomy, as at the end of the day, Whitehall has the right to hand things out and take things back’ (interview 38). The sense that ‘government knows best’ was further reinforced by a clear distrust within Whitehall of service providers on the ground. One former minister stated that ‘when they [hospital consultants] were given the targets and the accountability, they still behaved in the same way and blamed the targets’ (interview 46); another former minister that the ‘vast underperformance in the education system’ rendered them unwilling to ‘just let [headteachers] get on with it’ (interview 51); and one senior civil servant bemoaned the way in which the ‘bitching about government [by] public sector workers’ rendered any attempts at self-improvement as ‘unsustainable’ (interview 22). Moreover, the combative adversarialism associated with Westminster politics, along with the concentration of accountability in individual ministers, encouraged a stranglehold over delivery. As one senior civil servant explained, ‘[a]t the end of the day, the Home Secretary has to stand up in Parliament and explain why [a target has been missed]. That’s not going to wash if he says “well, it’s the local authority’s decision”’ (interview 20).

Vocal localism and creeping centralisation – The Coalition Government, 2010–2015

As the previous section demonstrated, Labour presided over an era of performance management that rested on the antithetical principles of top-down priority-setting and bottom-up operational autonomy. Yet despite such tensions, Labour’s performance management yielded tangible results. Through the PSA framework, spending decisions became longer term in focus, and policy decisions were increasingly framed around societal outcomes; and whilst the Government fell short in meeting many of its targets, significant inroads were made into tackling many of society’s most intractable problems such as child obesity and teenage pregnancy (Matthews, 2013). Similarly, evidence on the ground reveals the relative success of LPSAs (Boyne and Chen, 2007), and the way in which ‘targets-and-terror’ forced results in fields such as healthcare (Bevan and Hood, 2006). Nonetheless, throughout this period, the Conservatives were vocally opposed to Labour’s government-by-numbers, publishing a range of highly partisan attacks (e.g. Conservatives 2002, 2003). Such themes were reiterated in successive manifestos, and in 2010 the Party pledged to ‘scrap the hundreds of process targets that Labour have imposed on councils’ and ‘end the bureaucratic inspection regime that stops councils focusing on residents’ main concerns’ (Conservatives, 2010: 76). Joining the Conservatives in their opposition were the Liberal Democrats, who in 2010 promised to ‘[s]crap nearly £1 billion of central government inspection regimes on local councils’ as part of a broader commitment to giving local people ‘the powers and the funding to deliver what they want for their communities’ (Liberal Democrats, 2010: 90).

It was therefore unsurprising that upon entering office, the Conservative-Liberal Democrat Coalition swept aside Labour’s machinery of performance measurement as part of its broader commitment to ‘promote decentralisation and democratic engagement, and … end the era of top-down government’ (HM Government, 2010: 11). As Cabinet Minister Francis Maude explained in July 2010, ‘there were many attempts to micro-manage delivery from the centre, through targets, PSAs, monitoring, auditing, endless guidance and regulation. This is both wrong and doomed to fail.’ ¹ Instead the Coalition would pursue what the Deputy Prime Minister Nick Clegg described as a ‘radical decentralisation’ of power’, including the ‘stripping away much of the top-down bureaucracy that previous governments have put in the way of frontline public services and civil society’ (HM Government, 2011: 1). Alongside the immediate abolition of the PSA framework as part of the 2010 Spending Review (Cm. 7942, 2010: 34), the Coalition dismantled the layers of performance management to which local authorities were exposed. In June 2010, the Secretary of State for Communities and Local Government Eric Pickles instructed the Audit Commission to cease all of its work on CAAs, a move welcomed by the Local Government Association. This was followed in August 2010 by an announcement of the abolition of the Audit Commission itself; and the Secretary of State explained that ‘ministers believe that the work of the Commission has increasingly become less focused on accountability to citizens and more on reporting upwards to Government, judging services largely against top down Government imposed targets’. ²

Through the Localism Act 2011, the Coalition abolished Regional Spatial Strategies (RSSs), along with their intermediate tier of regional targets, in order to afford greater local control over the development of their communities. The RSS framework was instigated in 2004 to co-ordinate regional planning activities around national strategic goals, such as infrastructure development and economic growth. Yet, whilst intended to bridge the gap between national planning aspirations and local planning policies, they had come under widespread criticism for imposing national targets without any sensitivity towards local conditions. Indeed, the Secretary of State for Communities and Local Government described RSSs as ‘mean in spirit and [based on] as assumption that things would always go wrong’. At the same time, he also alluded to the ‘measure fixation’ in which the RSS had resulted: ‘it was really about ticking a box … There were no consequences for not doing that other than that your target increased’ (HC 517, 2011: 6–8). The Coalition’s approach to decentralisation was also applied to service providers and local communities. Its ‘free schools’ policy sought to give parents and teachers the opportunity to set up state-funded schools outside the control of local education authorities. The Health and Social Care Act 2012 sought to transform the organisation of the NHS by abolishing Primary Care Trusts and Strategic Health Authorities, and establishing ‘GP commissioning’ to enable patient care to be commissioned in accordance with local clinical priorities. More broadly, the Coalition pledged to transfer additional competencies to local communities as part of its ‘Big Society’ agenda, which would afford ‘significant new rights direct to communities and individuals, making it easier for them to get things done and achieve their ambitions’ (Communities and Local Government, 2011: 8).

Yet, rather than attuning public services to local priorities, the eschewal of the machinery of performance management was instead associated with a reduction in services, reflected in evidence such as the upwards creep of NHS waiting lists and a shortfall against the Coalition’s goal of building 150,000 affordable homes by 2015. Indeed, in seeking to deliver high national standards within a framework predicated upon localism, there remained an inherent tension, as underlined by one senior official:

they’re pursuing very managerial instruments like performance appraisal, which require you to set performance expectations … But as soon as that gets down to programme level, they will walk away from that whole approach. It’s quite difficult to see how all of this marries through. (interview 52)

Yet, during a period of deep and wide-ranging spending cuts, the jettisoning of the machinery of performance management can be seen less as an ideological commitment to the empowerment of front-line service providers, and more as a pragmatic attempt to obscure the deleterious impact of cuts on service provision. Local government depends on central government for around 75% of funding, yet between 2010 and 2015 budgets were cut by over 25%, with other revenue streams shrinking; and councils and community groups were being asked to do increasingly more with increasingly less. One senior official described the ‘tension’ that resulted from ‘the desperate need to try to reduce budgets in a sustainable way’ (interview 49). More bluntly, another senior official stated that ‘the most striking difference between the Labour Government and the [Coalition] Government is: targets and money [versus] no targets, no money!’ (interview 29). Nonetheless, whilst targets may have disappeared in name, the practices of performance management reappeared by stealth. This was first apparent in July 2010 when the Coalition announced that all Departments would be required to publish Structural Reform Plans, which set ‘clear priorities and measurable milestones’. The emergence of target creep was highlighted by those working within Whitehall. ‘The message was very clear that we were not having targets’, explained one Department of Health official, ‘but actually we did keep measuring a number of important indicators’ (interview 41). Another official wryly observed, ‘the Government now refuses to use the word “target” and uses other words for similar things’ (interview 43). Indeed, the re-imposition of performance management was used by the Coalition as a stick to beat poorly performing local planning authorities; and in 2013 Blaby District Council became the first local planning authority to be placed in such ‘special measures’, in a move starkly at odds with a commitment to decentralised governance.

Despite divergences in rhetoric, it is therefore possible to detect a significant degree of continuity between the Coalition’s approach to performance management in practice and that of its Labour predecessor. As one former Labour minister argued, ‘if you look at [the Coalition] and talk about health and outcome frameworks replacing targets, you could almost see that one is morphing out of the other … Almost a natural development than a completely different way of doing things’ (interview 24). It is also evident that the Coalition, like its Labour predecessors, found it difficult to rein in its interventionist tendencies. The collection of waste in wintry weather, the salaries awarded to senior officers, and even the fate of a house in which Ringo Starr once lived were all matters on which ministers offered opinion! Reflecting on this, the Communities and Local Government Select Committee stressed that ‘[c]entral government cannot have it both ways—on the one hand giving local authorities the freedom to make their own choices, and on the other maintaining that only one of those choices is the “sensible” one’ (HC 547, 2011: 57). Yet, given the prevailing politico-administrative context and the norms of Westminster politics, such an unwillingness to ‘let go’ was unsurprising; and a former official in Number 10 stated that ‘[t]he actual revealed preference is to centralise power, certainly in education. The creation of free schools and academies that are directly contracted by the Secretary of State has resulted in quite a big centralisation of power’ (interview 48). Moreover, the concentration of accountability within a highly adversarial polity rendered the emergence of target creep as somewhat inevitable. ‘It’s very hard’, one official explained, ‘to run a big complex modern government without some quantitative and qualitative targets. It’s really difficult, otherwise, to hold people to account or indeed to give coherence to the whole’ (interview 39). Such evidence underlines the clear tension between the rhetoric of localism and the politics of accountability; and whilst the Coalition may have abolished centrally mandated targets, the way in which ministers assume ultimately accountability in principle (via individual ministerial accountability) and in practice (via the ballot box) provided few incentives to surrender control of the detail of delivery.

Conclusion – Decentralisation, autonomy and Westminster politics

This article has interrogated the politics of performance management in the UK, focusing on the extent to which a highly centralised majoritarian polity has encouraged the top-down control of public services. Through its comparison of the ostensibly different approaches to performance management that developed under Labour and the Coalition, the article has revealed a significant degree of continuity. In particular, it has demonstrated that despite various promises to ‘let go’, successive governments have instead sought to ‘hold on’ to the detail of delivery; a discrepancy between ‘talk’ and ‘action’ which has resulted in a disconnect between ‘managers’ and ‘the managed’. In principle, the Coalition’s approach represented a departure from the practices that developed under the previous government; and the dismantling of Labour’s machinery of performance management was intended to end what then Minister for Decentralisation Greg Clark described as the ‘inefficient [and] undemocratic … bureaucratic micromanagement of our public services’ (Communities and Local Government, 2011: 4–5). Evidence presented in this article lends a degree of credence to such charges, as Labour often proved unwilling to let go of the detail of delivery, and instead sought to control public services through a complex web of targets and indicators. Yet, the extent that the Coalition’s commitment to localism was a panacea should not be overstated. Certainly, the Coalition swept away much of the machinery of targetry and equipped local authorities with new competencies to tailor services in accordance with local priorities. Yet, this went hand-in-hand with swingeing budget cuts that limited the scope for discretion; and there was evidence that 13 years of Labour’s ‘control-freakery’ had further stymied institutional capacity for responsiveness. Apocryphal tales abound of local authorities being unable to act without a clear central steer; and whilst such stories may be exaggerated, the changing context of performance management under the Coalition necessitated new ways of working for many service providers on the ground. Finally, evidence suggests that the rhetorical commitment to localism was confronted by an innate desire to control delivery from the centre, which reflects the pervading norms of Westminster politics and the realpolitik of electoral accountability.

The last point is key. Rhetorically, the approaches of Labour and the Coalition were poles apart; and the Coalition’s retrenchment of performance management was ostensibly no less radical than Labour’s governance-by-numbers zealotry. Indeed, it has been argued such the capacity for ‘rapid reversals and alterations’ is a product of the unencumbered policy capacity of the executive associated with Westminster majoritarianism (Pollitt, 2006: 28); and that such rapid alterations are encouraged at critical junctures such as general elections as ‘a kind of virility symbol – letting everyone know that there is a new gun in town’ (Pollitt, 2007: 536). As one former minister admitted, ‘it’s the nature of being a pharaoh that you want to build a pyramid yourself’ (interview 42). Yet, this article has also demonstrated the extent to which the scope for transformative action is contingent upon wider institutional norms; and the way in which the context of Westminster creates a set of expectations regarding the balance of power between core and periphery, and the role of central government in public service delivery. The rhetoric of ‘earned autonomy’ and ‘localism’ thus belies the persistence of a highly centralised governing culture; and reiterates Talbot’s argument that institutional factors are key to explaining why performance regimes frequently do ‘not match the dominant principal-agent logic which was (rhetorically) supposed to be dominating policy in this area’ (Talbot, 2008: 1571). It is therefore clear that institutions matter, defining the overarching context in which managerial tools are adopted and implemented; and shaping the relationship between the ‘managers’ and ‘the managed’. It is also clear that institutions matter owing to the way in which they structure opportunities for change, and the extent to which changes are forged in the crucible of an institution’s ‘logic of appropriateness’. This article has therefore demonstrated the importance of locating public management reforms within their wider politico-administrative context, and has underlined the utility adopting an institutional perspective to examine the dynamics of this relationship. In turn, by highlighting the persistence of a top-down, highly interventionist approach to performance management, which endures despite rhetorical commitments to the contrary, the article has made a critical contribution to the ‘second generation’ of performance management scholarship.

The tensions identified in this article are likely to increase in salience over the course of the 2015–2020 parliament, as in May 2015, the Conservatives were returned with an outright majority and a clear manifesto commitment to continue the Coalition’s decentralisation agenda. Indeed, as the newly promoted Secretary of State for Communities and Local Government Greg Clark stated, ‘the question now is not whether we localise but how much and at what pace’ (Hansard, 3 June 2015). To this end, the Government published the Cities and Local Government Devolution Bill on 28 May 2015, which paves the way for cities and counties throughout England to assume wide-ranging powers. Already the Bill has been subject to criticism, in particular for its stipulation that only those areas opting for directly elected mayors will be given control of key policy areas such as local transport and housing. Defending this, Chancellor George Osborne stated, ‘I will not impose this model on anyone but nor will I settle for less.’ Yet, the Government’s refusal to countenance anything but its preferred model is at odds with its vaunted commitment to local discretion, and signals a continued unwillingness at the centre to truly ‘let go’. Government, it seems, still knows best; and the continued disconnect between ‘talk’ and ‘action’ means that the politics of ‘joke-alism’ is likely to raise few smiles on the ground.