Financial Accountability & Management
Volume 31, Issue 1 pp. 1-22
Original Article
Full Access

Performance Management in the Public Sector: The Ultimate Challenge

Michela Arnaboldi, 

Michela Arnaboldi

Search for more papers by this author
Irvine Lapsley, 

Corresponding Author

Irvine Lapsley

Address for correspondence: Irvine Lapsley, IPSAR, University of Edinburgh Business School, Edinburgh EH8 9JS. e-mail: [email protected]Search for more papers by this author
Ileana Steccolini, 

Ileana Steccolini

The Special Issue Editors are respectively from Politecnico di Milano; IPSAR, University of Edinburgh Business School; and Bocconi School of Management.

Search for more papers by this author
First published: 06 January 2015
https://doi.org/10.1111/faam.12049
Citations: 242

Abstract

Performance Management is the challenge confronting public service managers. However, the enduring research focus on performance measurement in public services, without resolution, does not offer neat solutions to performance management in public services. This drawback of measurement difficulties has not abated interest in performance management. But there are significant adverse outcomes associated with the clumsy use of performance management systems in public services, particularly negative effects on staff morale. The lack of ready-made answers to performance management makes this task complex and demanding for public service managers. This paper identifies critical dimensions of effectiveness in performance management systems.

INTRODUCTION

There are relentless pressures on managers in public services to act on the quality of their services. The idea of ‘more with less’ has become a slogan, as managers seek to maintain or improve the quality of service delivery. This phenomenon is pervasive – an international trend from which there is no escape for public service managers. This global interest has attracted the attention of key world institutions such as the OECD (see Perrin, 2003; and Curristine, 2005, 2007 and 2008) on the fostering of performance budgeting and monitoring systems) and the World Bank (see the Talbot, 2010, study of UK performance management for the World Bank as the promulgation of best practice). And yet there is no single solution as the public sector has many variations in scope and features in the 196 countries of the world, all shaped by economic performance, political philosophy, the involvement of external agencies and demands for public services (CIMA, 2011). These layers of complexity complicate, but do not lessen the interest in, and need for, performance management systems in public services.

The financial crisis of 2007–2008 is still unfolding and we do not yet know its outcome. The fiscal pressures have intensified the need for making best use of reduced resources in public services. This is in the midst of an uncertain environment in which traditional paradigms for public policy have experienced policy reverses (Coen and Roberts, 2012). However, the genesis of performance management systems is the global impact of three decades of the new public management (NPM) phenomenon which drives the focus on results oriented public services. The global financial crisis has accentuated the longstanding need for effective performance management of public services. There has been considerable research activity on performance management systems across a range of services to the extent that we may have a performance measurement industry (Johnsen, 2005). Nevertheless, that research effort is diffuse and has not been consolidated into a coherent body of thought (Broadbent and Guthrie, 2008). Indeed, the activity of performance management has been characterised as risky for public service managers (Cuganesan et al., 2014).

This paper contributes to the debate on performance management by offering a nuanced interpretation of the nature of this activity in public services. First, this paper discusses how complexity in public services may be theorised. Then it addresses the topic of performance management in public services by examining three dimensions of processes and impacts: (1) The key pitfall of performance management (2) Performance technologies: An accounting problematic and (3) Performance management in a complex setting. The paper concludes with closing comments on the challenges facing performance management in public services and with a future research agenda.

THEORISING COMPLEXITY

The public sector is widely recognised as a complex setting for study. The public sector has been described as an area of inherent complexity (Lapsley and Skærbæk, 2012), stemming from the location of managerial culture in a sector which experiences many political influences. This, in turn, may confound managerial discretion and complicate levels of accountability, especially in a sector which is repeatedly reformed, with uncertain outcomes, and where expectations are high on the delivery of social justice, social responsibility, equity in society, democratic entitlements and pressures for social change.

So, the concept of complexity abounds in the public sector. But how can we study this phenomenon for public services? There is now an increasing focus on the idea of the development of complexity theory which has captured the imagination of many researchers in public policy settings. The early work of Axelrod (1997) seeks to build a theory of complexity of cooperation based on the classic case of the Prisoner's Dilemma. In this work, Axelrod stresses its difference from conventional inductive or deductive theorising, in which he describes his work as a modelling to ‘aid intuition’ (Axelrod, 1997, p.4). Fundamentally, the Axelrod approach is based on simulation and attempts to study the actions and interactions of individual agents in society and then observe patterns which occur at the total society level (Axelrod, 1997, p.3). In subsequent work, Axelrod collaborated with a public policy specialist, Cohen, to address how ideas of complexity could be harnessed by organisations (Axelrod and Cohen, 1999). They saw diversity of effort in the study of complexity, but also identified recurring themes which informed their framework which they describe as a unified view of complexity studies for the analysis of complexity in organisations (Axelrod and Cohen, 1999, p.18). In this elaboration of their approach they made the following observation (Axelrod and Cohen, 1999, p.19):

…complexity research does not make detailed predictions. Rather it is a framework that suggests new kinds of questions and possible actions.

The complexity theory framework as elaborated by Axelrod and Cohen (1999, passim) had the following elements:
  1. Complex adaptive systems
  2. Co-evolution (and self organisation?)
  3. Dynamic systems
  4. Non-linear relationships
  5. Emergent properties
  6. Edge of chaos (and strange attractors?).

The concepts in italics refer to common elements in contemporary complexity theory which were not explicitly mentioned by Axelrod and Cohen (1999). The complex adaptive systems are those in which agents and or populations actively seek to make adaptations (Axelrod and Cohen, 1999, p.7). This is a building block in their analysis. The idea of co-evolution is the way in which the actions of some change agents may spill over and affect the practices of other agents. This is a process which may never settle down (Axelrod and Cohen, 1999, p.8). The related concept of self organisation refers to the manner in which agents within organisations can shape the nature of the organisation by their actions and interactions. While this concept was not explicitly included in the Axelrod and Cohen framework, it is entirely consistent with their elaboration of dynamic agents in systems. The dynamic systems feature of complexity often arises because of the varieties of agents at work in organisations and in total systems (Axelrod and Cohen, op.cit., p.32). The non-linearity of systems means that small actions can lead to disproportionate larger differences in subsequent actions (Axelrod and Cohen, op.cit., p.36). The idea of emergent properties is that systems may exhibit traits which are distinct to the system and which are not a characteristic of the individual parts of the system (Axelrod and Cohen, op.cit., p.15). The idea of the edge of chaos refers to interactions between organisations and their environment, which exist in a delicate balance between order and chaos. This phenomenon occurs as evolutionary systems structure interaction patterns to achieve a balance between exploration and exploitation (Axelrod and Cohen, op.cit., p.72). A related idea which features in current thinking on complexity theory is the strange attractor. This idea is often attributed to the work of Lorenz (1995). In complex systems the initial starting conditions can influence the dynamics of interactions, but as systems evolve they may be attracted to other properties (or attractors). The strange attractor is particularly complex: chaos may be present, but strange non-chaotic properties may also coexist at the same time and point.

This set of concepts in complexity theory underlines its systemic approach to the study of phenomena. This approach to the study of complexity is an endeavour to take a holistic look, not only at the system dynamics, but also at the constituent parts of systems. This systems thinking is a distinctive feature of modern complexity theory. While this theorising originated in the natural sciences and has been used in, for example, forecasting weather patterns, there is a movement recommending its use in the social sciences. This use includes public administration (Klijn, 2008), public management (Teisman et al., 2009; and Rhodes et al., 2011) and public policy (Geyer and Rihani, 2010; Room, 2011; and Haynes, 2012).

While there are advocates of the merits of complexity theory in the investigation of the management of public services (Lin and Lee, 2011), reservations can be expressed over its mobilisation in this study setting. In the first instance, this thinking originated in the natural sciences and there remain questions over the applicability of models derived from natural to social sciences. The translation of ideas from different disciplines introduces the likelihood of misrepresentation or misinterpretation or ambiguity. This seems particularly likely with a theory which introduces concepts with names like edge of chaos, strange attractors, co-evolution, all of which look particularly susceptible to managerial capture and reinterpretation. Another dimension of this is whether there is indeed a unified theory of complexity theory. Klijn (2008) identified complexity theory as a collection of different theories: complex adaptive systems; dissipative structures; autopoiesis theory; chaos theory; path dependencies. However, Klijn did acknowledge the commonalities of systems thinking and non-linearity. The significance of Klijn's observations is that they intensify the potential for misrepresentation or misinterpretation of complexity theory. The Klijn critique has attracted observations that the underlying model of complexity theory offers a theory which needs to be tested empirically and which does not lend itself to such testing (Pollitt, 2009). There are proponents of complexity theory who see the idea of self organisation as a basis for enhanced democracy in public organisations (Blackman, 2001), but this does not sit well with the realpolitik of centralised control and a results focus in many governmental settings. There is also concern over the tensions between complexity theory and the need for evidence-based policy to inform policy making in public services (Parsons, 2002). More importantly, there is a deterministic tendency within systems thinking which does not sit well with the highly influential view of postmodernism, which emphasises social construction, the frailty of causality, the importance of power relationships, stresses the relational and interpretive nature of human activity, the proliferation of the contested nature of knowledge and ideas (see Best and Kellner, 1991, passim).

Therefore, the application of complexity theory is not so self-evident or straightforward as it might appear. While complexity theory has not been advocated specifically to examine performance management in public services, it has been recommended for the study of the closely related phenomenon of NPM (Lin and Lee, 2011). This is an important area of research which we mobilise to test it out below in the challenges facing performance management in a complex public service setting and which we reflect on in our conclusion.

THE KEY PITFALL OF PERFORMANCE MANAGEMENT

The single largest pitfall for performance management systems in public service organisations is a negative side-effect which undermines the motivation, morale and behaviour of human resources. The key resource in many public services is their human capital – the staff employed, their expertise, their capacity for problem solving and policy implementation. There are distinct, adverse outcomes for the human dimension of performance management. This paper examines research on this facet of performance management, which traces human resource issues arising from its implementation. The most trenchant critique of the adverse impact of NPM on public service staff is made by Diefenbach (2009). Diefenbach (2009, p.905) made the following observation:

NPM's impact on employees and corporate culture of public sector organisations comprises a whole range of negative psycho-sociological and organisational effects, such as: increase in occupational stress, illness, low morale, decline in job satisfaction and motivation, alienation, fear, resentment, the distorting intellectual effects of writing for audit, a competitive, adversarial and punitive ethos, as well as wasteful, stressful, over- bureaucratic, and expensive audit procedures, increased tensions, more distrust between people, forms of symbolic violence and institutional bullying, a rougher working climate, an invisible net of managerial power and domination.

There is a body of literature which supports this critique of NPM impacts on employees. The primacy accorded to audit in the NPM world has led to Power's famous observation that we now live in an audit society (1997) in which the need to respond to audit controls, procedures and practices can shape the day-to-day life of organisations. This has the deleterious consequences of additional work and the subsequent displacement of the primary purpose of organisations so affected. In this NPM world, organisations grapple with the bureaucracy of the audit society, in which boxes must be ticked to demonstrate compliance with diktats whether this compliance is for real or merely legitimating activity (Arnaboldi and Lapsley, 2008; and Lapsley, 2009). The roles and expectations of performance auditors continue to grow (Funnell, 2015).

Within a variety of public services there is evidence of the adverse impact of NPM reforms on employee welfare. One such service is policing where new managerialism has increased workload and pressure on officers (Butterfield et al., 2004, p.176). This has led to a significant majority of UK police officers reporting stress and overwork (UNISON, 2014). The corporate language of mission, commitment and strategy in public services is claimed to sit uneasily with, and promote disillusion in, public services which experience cost reductions with increased job insecurity, larger class sizes, fewer nurses (Hoggett, 1996). In health and social services, NPM reforms have resulted in increased intensity of work practices, with fewer staff, increasing levels of stress, demoralised staff, high absenteeism and labour turnover (Kirkpatrick et al., 2005, p.98, p.120). Although Kirkpatrick et al. (2005, p.150) do report that workers within social housing have more readily embraced the managerialism (its language, its logic) of NPM.

Within universities, the advent of NPM has resulted in a more adversarial working environment (Newton, 2003, p.434) in which tolerance for independent thinking by academics and collegiality is marginalised by management agendas (Saunders, 2006). The NPM focus on quantification and results has spawned a wide range of metrics within universities including measures of quality (Pettersen, 2015). It has been suggested that these are mobilised to manage and control academics (Burrows, 2012). These metrics include citations, workload models, research assessments, teaching quality assessments and league tables for university activities and performances, including commercial activities. Gill (2010, p.228) has observed that the impact of this managerial regime has had wide ranging adverse effects:

Many academics are exhausted, stressed, overloaded, suffering from insomnia, feeling anxious, experiencing feelings of shame, aggression, guilt, hurt and ‘out of placeness’.

It has been suggested that this state of affairs is widespread in universities as part of a ‘deep, affective somatic crisis which threatens to overwhelm us’ (Burrows, 2012, p.355). It has been observed that the strength of this phenomenon is such that academics have been unable to challenge a universal focus on calculative practice and audit culture in universities stemming from NPM practices (Shore, 2008). This primacy accorded to performance management systems is attributed to NPM thinking, in which managerial elites have a vested interest in legitimising an audit culture focussed on constant assessment at the expense of collegiality and scholarship (Craig et al., 2014).

The above observations demonstrate the potential negative outcomes when public organisations implement clumsy performance management systems which are not geared to the key human actors engaged in the delivery of services. On balance it should be noted that the policy design and implementation intensity of NPM varies between countries, and that much of the literature and research is scarce when it comes to studies of net effects of NPM reforms including performance management (Bouckaert and Halligan, 2008; van Helden, Johnsen and Vakkuri, 2012; and Pollitt and Bouckaert, 2011). Next we comment on the range and quality of tools available to public service managers.

TECHNOLOGIES OF PERFORMANCE MANAGEMENT: AN ACCOUNTING PROBLEMATIC

For the managers there are significant problems in the selection of management tools to mobilise in the name of performance management (see, for example, Hope and Player, 2012). The quality of available techniques is a particular problem for public service managers. There have been various attempts at devising models of performance management. These include the idea of a global measure of performance. The failure to achieve this and the difficulties of operationalising the resultant partial performance indicators are discussed below. The classic Anglo-American NPM approach of seeking appropriate private sector models has led to the introduction of Benchmarking, the Balanced Scorecard and Lean Management into many public services, with mixed results. These initiatives resonate with the classic phenomenon of following the latest managerial fads and fashions (Abrahamson, 1996). The early attempted solution of managerial checklists as a reductionist treatment of complexity is revived here and considered as a potential way forward for public services. The range of commonly used tools in public services performance management is set out in Table 1. We identify five such possibilities for public service managers and as Table 1 shows there is no obvious choice from the set of available technologies. Each of these approaches to performance management is discussed, next.

Table 1. Performance Management Technologies
Technology Key Attributes Comment
1. Budgetary Control Traditional Accounting Crude, limited
2. KPIs & Benchmarking Partial performance indicators in comparable settings What gets measured gets included
3. Balanced Scorecard The Harvard model of performance management Identifies multiple dimensions but is over specified
4. Lean Management The Toyota Production model Negative side effects
5. Managerial Checklists An exercise in reductionism Susceptible to box ticking legitimation

Budgetary Control as a Performance Tool

The first technology identified in Table 1 is budgeting. This inclusion is recognition of the significance of the accounting function within public service organisations. It is also recognition of the longstanding centrality of the budgetary process in public sector organisations (Henley et al., 1992, p.56). However, while tradition, the size of public sector organisations and the level of sophistication in financial management may lead public service organisations to rely on budgets for performance control, this option has many disadvantages. This technology ignores non-financial performance indicators and can therefore be seen as very narrowly focussed on financials. In this way, the budget can be seen as offering crude cost control. However, the reluctance to refine the historical budget setting practices of public sector organisations by adopting activity-based measures undermines the rigour of budgets. Furthermore, the achievement of budgetary equilibrium may be seen as a kind of success, but it is limited. The achievement of balancing the books does not equate to meeting all service demands nor does it mean the organisation has operated efficiently. However, the downside of all this is that whatever performance management system is implemented within any public sector organisation, it will have to relate to the budget system given its centrality in the life of the public sector organisation, a facet of performance management which is often overlooked and is neither simple nor straightforward.

Private Sector Technologies for Performance Management

The next three technologies in Table 1 all fit the ‘fads and fashions’ (Abrahamson, 1995) idea of management diffusion. All of them have private sector origins: Benchmarking came from Rank Xerox (Cross and Iqbal, 1995); the Balanced Scorecard came from Analog Devices, an electronics company (Schneiderman, undated); and Lean Management came from Toyota, the Japanese car manufacturer (Womack et al., 1990). The practice of looking to the private sector for management ideas is a fundamental element of NPM. The appropriateness of this practice is contestable, given the difficulties of translation from the private to the public sector. In the following sections each of these private sector imports is discussed and then we revisit the idea of managerial checklists.

In this paper we link Key Performance Indicators (KPIs) and Benchmarking as technologies of a similar vintage, which are used in concert. They can both be seen as technologies which have emerged in response to the lack of the commercial bottom line for performance assessment in public services. Efforts to mobilise value for money or the 3Es of economy, efficiency and effectiveness as a bottom line have failed to deliver a global model of performance measurement (see, for example, National Audit Office, 1988; FEE, undated; and Mayston, 1985). Efforts to devise a public sector bottom line have degenerated into the generation of many partial performance indicators (Lapsley and Pong, 2000). The relationship of these partial indicators is not self-evident in many cases, without a hierarchy of performance indicators which form an articulate statement of performance. The decision to identify KPIs for performance assessment focuses on what can be measured rather than necessarily capturing key dimensions of organisational performance (Bevan and Hood, 2006). The forced nature of this selection process and the arbitrary manner of choices made is captured by the following statement by an elected member of a legislative assembly (cited in Ezzamel et al., 2007):

We are fabulous at firing arrows at walls, drawing targets around them and then saying it was a brilliant shot.

The shift from organisations devising their own KPIs to a benchmarking arrangement in which comparable organisations are identified in which performance information is exchanged creates further complexity. While public service organisations are not in competition and therefore open, in principle, to sharing performance information, there have been difficulties from the beginning in identifying appropriate comparators, focussing on performance and embracing the concept of benchmarking (Bowerman et al., 2000; and Bowerman et al., 2001). This concept of the contested field in which actors’ own interests prevail over a common or shared purpose remains a major obstacle to effective benchmarking (Siverbo, 2014).

The Balanced Scorecard has achieved widespread interest in the public sector in the last decade of the 20th century and the first decade of the 21st century. This technology was claimed to be first devised in the electronics firm, Analog Devices, by Schneiderman (Schneiderman, undated) and subsequently developed by Kaplan and Norton (1992 and 1993). The attraction of this technology was its focus which was broader than narrow financials by including consideration of financials, internal processes, customers and learning. This configuration appeared to offer a comprehensive basis for linking corporate financial planning and strategic planning, unlike the proliferation of partial and so called key performance indicators. However, this particular management fashion has started to encounter considerable criticism. Notably, the Balanced Scorecard has been criticised because its four dimensions understate the complexity of many organisations (especially in public services) and it offers little more than lists of metrics (Norreklit, 2000). Also it has been argued that the key dimensions in the Balanced Scorecard are not as tightly coupled as they appear in the writings of Kaplan and Norton (Modell, 2004). A scrutiny of 20 years of research studies of Balanced Scorecard systems reveals implementation difficulties with little integration with accounting information systems (Hoque, 2014). While the Balanced Scorecard has had immense popular appeal, it looks like this management fad is now, or may be becoming, out of fashion.

Lean Management, however, has become the choice of performance management technology of many public services. This is somewhat surprising as Lean Management is an earlier technology than both Benchmarking and the Balanced Scorecard as it was developed by Toyota in the 1960s (Womack et al., 1990) as the Toyota Performance System (TPS). Within the UK the pressure of ‘more with less’ has led to widespread attempts at Lean Management in universities, hospitals and health care, and local government. At the centre of the TPS are the concepts of the standardisation of work into repeatable processes and the elimination of unnecessary stages in the production process to eliminate waste and reduce costs in a process of continuous improvement. A report commissioned by the Scottish Government argued that Lean Management was applicable to the public sector (Radnor et al., 2006). Government ministers have identified the potential of Lean Thinking in reducing public sector waste (Murden, 2006). Universities UK (2011, p.37) has acknowledged what it regards as the successful use of Lean Management in universities and advocates its wider use. Health care has been at the forefront as a suitable case for Lean Management. It has been argued that there is no lack of resources in health care and the core issue is the uniform application of best practices to reduce costs and increase quality. The TPS is ideal for this purpose and the UK NHS Institute for Innovation and Improvement has developed guidance on Lean Thinking for health care managers (NHS Institute for Innovation and Improvement, 2013).

However, the rapid spread of Lean Management has been questioned on a number of levels. There have been reservations about the efficacy of Lean Management at Toyota. In particular, concerns have been expressed that the constant focus on cost reduction has affected the safety of their vehicles. On February 3, 2010, Chris Lastrella, an off duty Highway Policeman in California was driving his Toyota Lexus with his family on board when the accelerator pedal jammed (Frean and Lee, 2010). All of the car occupants were killed in the ensuing crash. There were a further 30 reported deaths in the US from sudden unintended acceleration on Toyota vehicles at this time (Frean and Lewis, 2010). At a subsequent hearing of the US Congressional Oversight and Government Reform Committee, Roy LaHood, the US Transport Secretary said Toyota was ‘safety deaf’ and Akio Toyoda, the grandson of the Toyota company founder apologised for the accidents (Frean and Lewis, 2010). However, a senior US executive challenged Toyota's handling of safety issues as it expanded its operations (Lewis, 2010). These outcomes undermine the application of Lean Thinking as a relentless driver for cost reduction without due consideration for other important issues such as safety.

There are other concerns over the application of Lean Management in public services. There are issues over the portability of lean – how many public services are like car factories? Also many public services exhibit high levels of interdependence in service delivery which may confound the standardisation of public services. By focussing on segments of public service organisations, local star optima may be devised which do not yield overall improvement. There is evidence of a high failure rate for Lean Management in the National Health Service (NHS) which may be attributed to the lack of supportive information systems at project level, inter-unit level and organisational level (Kinder and Burgoyne, 2013). More fundamentally, it has been suggested that the Lean Management approach adopted within the UK public sector is doomed to failure, in theory and practice, given its piecemeal application without an overarching service model to inform its adoption and design (Radnor and Osborne, 2013).

Managerial Checklists

The final strand of performance technologies which we discuss here, is the use of managerial checklists. At one level this might be seen as a relatively unsophisticated management tool. For example, an early attempt at devising a managerial checklist for performance management was Jackson (1988). Jackson introduced nine key concepts for managers to manage performance. Some of these are beyond criticism (consistency, comparability, clarity, controllability) but others are in contradiction (comprehensive versus bounded). While this contribution is well intended it lacks precision, is not readily operational and, most importantly, it looks susceptible to the tick box mentality of the Audit Society (Power, 1997).

A more promising offering is the managerial checklist from Likierman (1993). This is a somewhat neglected study. Likierman's work is based on a three year research project in which 500 middle and senior public services managers were interviewed. All of those interviewed were managers who used performance indicators. This paper looks at performance from multiple dimensions, looking at the trajectory of performance systems through the concepts underpinning the system, the preparation of the system, the implementation and the use of performance information. This study was not fully exploited as its completion coincided with the author becoming a senior member of the UK Government Civil Service. However, this study offers a potential way forward. It points to the need for evidence based studies rather than the fragmented policy of snatching at the latest management fashion as the solution to performance management.

PERFORMANCE MANAGEMENT IN A COMPLEX SETTING

The complex setting analysed in this paper is the UK NHS. The NHS has an immense workforce of many different health care professionals and sits firmly within those public services which have human capital as a key resource, as outlined in the third section of this paper. In terms of technologies of performance management, budgetary control remains a central mechanism for the NHS. There has been some limited use of the Balanced Scorecard in the NHS, but there has been more sustained use of KPIs and benchmarking and for the past few years there has been a strong use of Lean Management. In terms of the NHS as a study setting, Lapsley and Schofield (2009) highlight the complexity of the UK's NHS, which makes it of particular interest for public service performance management. This includes the distinctive origins and ethos of the NHS and their financial implications. The density of NHS organisation and the scope of the services which it delivers, also present difficult challenges for financial planning and control. The sheer scale of NHS operations, with it often being considered one of, if not the, largest organisation in the world, presents significant challenges for financial accountability and control. There is also the fundamental issue of the efficacy of a national health care system funded from general taxation. These general contextual issues heighten the complexity of the NHS but other factors at work impinge directly on NHS performance management, as Lapsley and Schofield (op.cit., p.367) express it:

Furthermore, and particularly in times of financial restraint, the existence or otherwise of formal or informal rationing systems in health care presents significant ethical and financial dilemmas. Most importantly, the enhanced visibility and significance of accounting numbers and performance measures in NHS management processes and accountability mechanisms over the course of its life continues to attract the attention of researchers.

This is the context which we use to illustrate the challenges confronting performance management in public services. It is interesting to note that a number of complexity theory advocates have identified health care as a suitable context for study (Arndt and Bigelow, 2000; Geyer and Rihani, 2010; and Rhodes et al., 2011) and performance management in health care is as complex and elusive as ever (Chang, 2015; and Kelly et al., 2015).

It was noted above that in many public services, what can get measured gets counted (Bevan and Hood, 2006) and this can form the basis of performance management systems and this is the case with the NHS. The key performance measure in recent years has been the use of waiting lists for patients in need of health care treatment. Successive governments have set targets for achieving reductions in waiting lists. The waiting list is a capacity utilisation indicator, which may appear sensible in a service which is free at the point of use and for which there is an excess demand in the absence of a pricing system. However, there is a longstanding critique of waiting list indicators as soft performance targets which are easy to manipulate. In his critique, Williams (1985) raised questions over the criteria by which patients are admitted to lists, the uniformity of such criteria across the NHS, the frequency with which they are updated, the discretion in the alteration and admission of patients in targets.

In this section of this paper we examine the experiences of two health care organisations in the NHS: (1) Lothian Health, and (2) Mid Staffordshire NHS Foundation Trust. The experiences of these health care bodies resonate with the Diefenbach (2009) critique of NPM type performance management systems.

(1) Lothian Health

In the National Health Service in Scotland, Health Boards have targets that specify that 90% of their patients must be treated within 18 weeks of referral by their general practitioner. At Lothian Health, there was a major inquiry initiated by the Cabinet Secretary for Health over claims in the press from a whistleblower that Lothian Health was routinely manipulating its waiting list numbers. This report was prepared by PWC for the Scottish Government (Scottish Government Directorate, 2012). Its findings confirmed the systematic manipulation of waiting list information. As a public entity the governance mechanisms at Lothian Health included both a management and a board (with lay persons appointed by the Cabinet Secretary) to oversee its activities. The Board also had a Finance and Performance Review Committee (FPRC) drawn from its membership, which reported to the board. Within the management team there was an Executive Management Team (EMT) at the highest level, which reported to the Board and this top management tier had a Senior Management Team which reported to EMT. Beneath the Senior Management Group there are the departments and hospital specialties. Each month a Performance Management Report was prepared by NHS Lothian as a standard agenda item for all four parts of Lothian Health governance structure as outlined above. The PWC report stated (op.cit., p.17) that:

The EMT, the Board and the FPRC were not presented with a comprehensive picture of waiting time management or data, for example, as there is an absence of any detail on periods of unavailability data or full waiting list size. In addition there is no trend analysis of performance.

The absence of this information hindered the EMT and the FPRC in making informed decisions on waiting list issues and the Board may not have been able to identify that there was an issue (op.cit., p.17).

The lack of information presented to the top was based on routine manipulation of waiting list figures. The PWC investigation (Scottish Government Directorate, 2012, p.5) highlighted this:

…highlighted excessive and inappropriate use (and apparent misuse) of periods of patient unavailability, in particular retrospective creations and changes, which removed patients from waiting times breach (i.e., missing target) reports. This inappropriate use has masked the number of breaches (failures to meet targets) reported at a number of month ends and has resulted in certain patient journeys being longer than have been formally reported.

These inappropriate adjustments involved offering patients treatment in England at short notice and if they refused they were removed from waiting lists. There was also lots of retrospective alteration, manually, of patient records. For example, this report (Scottish Government Directorate, 2012, p.6) identified examples of this:

On 30 May, 2011 (just before breach or target failure reporting) a member of staff made 124 amendments to periods of unavailability, retrospectively, and then on 1 July, 2011 (just before target failure reporting) another member of staff made 154 amendments to periods of unavailability, retrospectively, between 0800am and 0900am.

It is also noteworthy that many managers in Lothian Health received detailed waiting list information from weekly waiting time position reports but this information was not included in formal reports to the EMT, the FPRC or the Board.

How did this happen? Interestingly, given the Diefenbach (2009) critique of bullying resonates with other findings in this report. It observed that many of these staff members were put under unacceptable pressure to manipulate waiting list figures in a culture of ‘no bad news’ about waiting list figures (Scottish Government Health Directorate, 2012, p.21).

However, senior figures within Lothian responded strongly to these suggestions (Lothian Health, 2012):

Bullying and harassment have never been tolerated in NHS Lothian and we will be following up immediately any claims of this through our own internal inquiry, which is already far advanced (Chief Executive),

and:

We do not tolerate any form of bullying and harassment and this is monitored by annual surveys. This is a message sent out to all employees when they join us and any time such behaviour is proven we are robust in tackling it (Director of Human Resources).

However, despite such protestations, the Cabinet Secretary for Health initiated a further report into the management culture at Lothian. This report confirmed a bullying management culture at Lothian which manifested itself in an extreme form around the performance management of waiting list targets (D.J. Bowles and Associates, 2012). In this report there are comments by members of staff which illustrate the nature of this bullying culture (D.J. Bowles and Associates, op. cit., p.19, p.23):

Some senior managers bully us with constant targets, targets, targets… shouting and relentless pressure.

A macho culture that has lasted for some time.

Shocking… an atmosphere of fear.

There is a blame culture, particularly for senior managers and I see it cascade and leak out to the lower graded staff.

If you don't reach your targets, you can collect your P45 (i.e., lose your job).

Those of you with mortgages and career aspirations had better be afraid.

The report by Bowles and Associates (op.cit.) described this as a dominant culture (p.21) in which the general leadership and management style is based on ‘command and control’ (p.19) in which there is an almost total concentration on targets and the tasks required to achieve them (p.19) and in which there is an ‘emphasis on targets and predominantly autocratic leadership’. This report also observed that the ‘obsessive focus’ on targets for performance management meant that the organisation had lost sight of policy objectives (op.cit., p.36).

Despite these reports, Lothian Health has just reported yet more difficulties over waiting list targets (Pickles, 2014). Indeed, Lothian Health has longstanding difficulties over the achievement of waiting list targets for many years. There are distinct circumstances which accentuate this problem. There are the general issues of a system which is stretched by no pricing or rationing systems. There are the increasing numbers of elderly patients with multiple medical conditions. Perhaps most importantly Lothian Health is a regional centre of excellence with world leading medical specialists which always seem to attract more referrals than they can handle. The major issue here is the way complex systems are reduced to simple performance systems with all manner of deleterious consequences, particularly for staff implementing them.

(2) Mid Staffordshire NHS Foundation Trust

There have been two major, highly critical reports published on Mid Staffordshire NHS Foundation Trust, the report of an Independent Inquiry (Francis, 2010) and the report of a public inquiry (Francis, 2013). These follow other critical reports by oversight bodies in health care. The initial concerns over the management and quality of care at Mid Staffordshire were aroused by mortality rates which were significantly worse than at other hospitals (Health Care Commission, 2009), which observed (p.137):

In this report, we have drawn together the different strands of numerous, wide-ranging and serious findings about the trust which, when brought together, we consider amount to significant failings in the provision of emergency healthcare and in the leadership and management of the trust.

Concerns expressed by patients and patients’ families led to the Secretary of State for Health establishing the first Independent Inquiry (Francis, 2010). A subsequent Secretary of State for Health commissioned a second, public inquiry to be led by Robert Francis (Francis, 2013). In addition to the concerns over mortality rates, these reports were extremely critical of the poor quality of patient care. The following instances (Francis, 2013, para 23, p.13) illustrate the woeful standards of patient care:
  1. Patients were left in excrement in soiled bed clothes for lengthy periods.
  2. Assistance was not provided with feeding for patients who could not eat without help.
  3. Water was left out of reach.
  4. In spite of persistent requests for help, patients were not assisted in their toileting.
  5. Wards and toilet facilities were left in a filthy condition.
  6. Privacy and dignity, even in death, were denied.
  7. Triage in A&E was undertaken by untrained staff.
  8. Staff treated patients and those close to them with what appeared to be callous indifference.
How did an NHS hospital ever sink to such low standards of care? The Francis Reports (2010 and 2013) identify a culture in which the needs of patients were, at best marginalised, and in which the management of this hospital gave primacy to performance targets and to good news stories. The following extract illustrates this point (Francis, 2013, p.43):

Those with the most clear and close responsibility for ensuring that a safe and good standard of care was provided to patients in Stafford, namely the Board and other leaders in the Trust, failed to appreciate the enormity of what was happening, reacted too slowly, if at all, to some matters of concern of which they were aware, and downplayed the significance of others. In the first report, this was attributed in large part to an engrained culture of tolerance of poor standards, a focus on finance and targets, denial of concerns, and isolation from practice, elsewhere. Nothing I have heard in this Inquiry (i.e., the Public Inquiry) suggests that this analysis was wrong. Indeed the evidence (from the Public Inquiry) has only reinforced it.

The above critique of the negative impact of performance management on the central mission of this hospital trust is accentuated by its ‘good news’ culture which sought to negate or play down setbacks and celebrate whatever minor successes were experienced. A further extract from the Francis Report (2013, p.44) underlines the insensitivity of management in its handling of challenges:

The Trust culture was one of self promotion rather than critical analysis and openness. This can be seen from the way the Trust approached its Foundation Trust application, its approach to high Hospital Standardised Mortality Rates and its inaccurate self declaration of its own performance. It took false assurance from good news, and yet tolerated or sought to explain away bad news.

Despite the management preoccupation with financial matters to the exclusion of standards of patient care, this hospital trust managed to survive numerous reports from a variety of oversight bodies (Francis, 2013, pp.48–64), which challenged its fundamental practices of health care. Yet this hospital trust managed to stagger on like a zombie organisation, still focussing on its performance targets, before it was closed. This adverse impact of performance management systems goes beyond the Diefenbach (2009) critique which identifies adverse impacts on staff rather than the totality of the organisation and those it has a duty of care for. It could be discussed whether the adverse effects were the results of a flawed performance management system designed and used too much for financial matters and too little for performance, an incompetent management, a dysfunctional organisational culture, or a combination of all these issues. Using the lenses of complexity theory one could describe this case as a complex adaptive system organising itself top-down into a system with emerging properties crossing the edge of chaos. In effect, the Mid Staffordshire NHS Foundation Trust may have had too little performance management rather than too much.

The initial starting point in this section was the implicit reference by Lapsley and Schofield (2009) to health care as a complex adaptive system. Rhodes et al. (2011, p.112) deploy complexity theory to demonstrate how health care information systems can be modelled as a complex adaptive system. Also, Geyer and Rihani (2010, pp.101–3) see the above kinds of adverse outcomes as an inevitable consequence of policy makers viewing the complex NHS as a mechanistic system which can be controlled from the centre. This approach fails to recognise the modernisers proposing cycled and recycled reforms through a single attractor, which limits the organisation in self organisation and co-evolution.

CONCLUSION

There is a clear need for effective performance management systems in public services. The NPM modernisers see this as a way of providing public services more efficiently. The era of austerity has reinforced the need for more effective management of performance. Yet, despite the very significant efforts by key actors within public services and by related agencies and experts, there is no unanimity on what constitutes best practice. Indeed, this paper has documented significant failures of existing practices when managers become so obsessed with managerial targets and appearance that they lose sight of the more fundamental issues of organisational mission and performance. This paper suggests that the underlying dimension which makes performance management so difficult is the sheer complexity of, and the often over-simplistic approach to performance management in the public sector.

This paper has elaborated upon the potential of complexity theory to be mobilised by researchers within public services management to address the nature of public sector complexity and its impact on performance management.This conclusion is not without caveats as the origins, aims and expectations of complexity theorists may differ from the realities of policy making in everyday life. However, this perspective offers a definite research agenda for empirical studies of performance management practice, as set out below:
  1. For performance management to be effective it must mitigate or eliminate the negative side effects on the key resource of human capital. There is scope for exploratory research in study settings where successful performance managements appear to operate to determine where and how this advance might be made.
  2. Much of the extant literature on performance management tools and technologies seems to be based on specific applications of particular practices. There is much to be gained from closely grained case studies of practice. But it would be interesting to see studies which undertook a more holistic evaluation of performance management by paying attention to the details and instability of systems. Such investigation potentially opens the way for developing a new, specific, framework for examining performance management in the public sector.
  3. In their writing, Axelrod and Cohen (1999) refer to complexity theory as an aid to thinking. It would be interesting to see if managers of public services can relate to and make sense of the ideas in complexity theory in their everyday tasks. There is scope for experimental work in this area.
  4. Within an era of rapid technological development the role of social media which gives free access to global web communication to every individual thereby opening opportunities and risks for the public sector. The role of social media in enlisting, enrolling and mobilising different perspectives on how public services are delivered and performance managed offers a fruitful area of research.
  5. There is a longstanding issue within public services of remuneration, reward and incentives for exceptional performance. This is an important element of performance management and needs careful evaluation in work on public services.
  6. Within many public services there are many short-term pressures to deliver. There is scope for the receptivity and feasibility of longer planning horizons in future performance management modelling.
  7. Within performance management in public services there is silence on ‘management’ – what it is and what it means. This paper suggests that performance management in public services is more complex than in other settings such as non-profit or private sector activities. Yet there is a presumption within existing literature that the ‘effective public manager’ exists and will deliver. This is a key part of performance management which merits careful research and analysis.
In addition to the above agenda focussed on performance management practices, there is scope for further theoretical research in this area. In particular we suggest the following issues:
  1. Complexity theory builds on the idea of agents within systems who detect the need for change and cluster to make change happen. Within this thinking the idea of the agent appears to presume homogeneity. Research on the manner in which agents act in complex adaptive systems merits careful evaluation.
  2. The raison d`etre of complex adaptive systems in complexity theory is an aversion to what is perceived as a mechanistic concept of how organisations function. However, the top down, command and control model reflects the need to deliver successful policy outcomes. It may be argued that this does not achieve its objective. But the self organisation, bottom up thinking in complexity theory needs to be reconciled with realpolitik for policy makers to endorse its use.
  3. There are strands of complexity theory which resonate with other ideas within the social sciences which have been used widely in the public sphere. This includes the idea of the street level bureaucrat which mirrors the purposeful agents depicted in complexity theory. Another example is isomorphism which has parallels with ideas of co-evolution. The idea of unintended consequences features too. There is scope for the study of this phenomenon to see if a melding of different approaches may be feasible.

This paper suggests that performance management is a big challenge facing public services. It is the intention of this paper not only to encourage a rethink of existing practices to avoid the negative side effects documented in this paper, but to encourage researchers to undertake more nuanced research in this most difficult, complex, testing area for researchers and practitioners alike.

    The full text of this article hosted at iucr.org is unavailable due to technical difficulties.