Introduction: The Measurement-Performance Link

As international economies continue their recovery after the 2008 global economic crisis, growth rates are regarded as being ‘weak and uneven and reflective of different evolutions across different countries and regions’ (IMF 2014). In such challenging competitive environments organisations tend to focus very strongly on the management and measurement of performance in order to meet the task of staying in business. Harris & Mongiello (2001) see a company’s emphasis on performance measurement (PM) as a route to competitive advantage, and turbulent business environments, such as those currently being experienced worldwide, can be seen as a key driver of both organisational and research interest in performance measurement.

However despite almost 30 years of research into performance measurement the discipline still has, what could be referred to as significant, foundational problems. There has been a lack of development of key concepts or a unified terminology in the subject area, this is mostly due to the multidisciplinary nature of the PM field which has “vast richness, but unmanageable diversity”(Neely 2007, p.2) .

One of the most significant issues in the PM field is that there have been contradictory findings as to whether or not the measurement of performance actually has an impact on the achievement of performance outcomes (Franco & Bourne 2004). In fact it is suggested that measurement has become such an accepted approach that few organisations genuinely challenge why they should measure in the first place, concentrating instead on what can be measured and how to measure it (Robson 2004). In many cases the relationship between the measurement of performance and the achievement of performance outcomes is often described simplistically with catchall phrases like “you can’t manage what you can’t measure”. These phrases imply a simple association between measurement and action in order to achieve performance. The reality of the relationship is however far more complex and the existence of a positive relationship between measurement and performance outcomes have yet to be definitively proven in the literature. Difficulties lie in the ability of researchers to capture the underlying factors that may mediate the relationship between measurement and performance and the considerable influence of organisational structures, culture or operating environments on the achievement of performance. The complex social structures at play in organisations can also have a significant impact on both measurement and performance and these are not sufficiently dealt with by many of the theories applied to contemporary performance measurement research.

The Performance Management literature is replete with investigations into the determinants of performance but the role of measurement in organisational outcome is rarely addressed. The following document will explore whether this fundamental gap in the research may be bridged by the adoption of a meta-theoretical perspective, in this case an emergent critical realist perspective, in the conduct of performance measurement investigations.

A brief review and critique of the theoretical evolution of research into the measurement-performance link will now follow and the case for the adoption of a critical realist perspective will subsequently be advanced.

The Measurement-Performance Link: Theoretical Evolution & Issues

A cursory review of the literature on performance measurement reveals 3 distinct phases, in terms of academic and practitioner focus. Starting in the late 80’s and early 90’s with the performance measurement (PM) revolution and the advent of more balanced and strategically focused measurement systems like the Balanced Scorecard (BSC), the literature evolved to focus on the design and implementation of Performance Measurement Systems (PMS) in the latter 90’s. More recently the performance measurement literature has become increasingly concerned with how organisations manage with measures and what the individual, team and organisational consequences of performance measurement are (Franco-Santos et al. 2012). This evolutionary development of the subject literature has been accompanied by significant theoretical development. A brief review of the literature and accompanying theory will now take place.

Robert Eccles’ “The Performance Measurement Manifesto” (1991) officially heralded a revolution in management thinking with regard to the measurement of business performance. It described a paradigm shift which would provide an equal (or greater) status to non-financial measurement, moving away from the traditional financial measurement focus, and the linking of measurement to strategic objectives. Eccles was by no means the first to advocate a more balanced measurement approach, academic heavyweights like Peter Drucker (1955) had been preaching the balanced measurement agenda decades previously, but Eccles work came at a time when the deficiencies of financial measurement were being exposed by an increasingly dynamic business environment and the pressing need to develop new measurement systems to cope with new strategies and competitive realities.

The performance measurement revolution, although changing the manner in which many companies focused on measurement, did not in fact alter theoretical perspectives in any significant way. Prior to the revolution, measurement was largely based around the classical management control theory of Anthony (1965) which assumed that managers had responsibility centres over which they could exert control, with measurement being used to influence employee behaviours and control the use of resources. Control theory was criticised for its overreliance on accounting measures (a key feature of the performance measurement revolution) and for its tendency to ignore the influence of the external environment and the complex role of power structures and human relations on performance within the organisation (Hewege 2012). The transition to multi-dimensional (financial and non-financial) measurement advocated by Eccles (1991) and Kaplan & Norton (1992; 1993) amongst others, did not remove control theory’s dominance from performance measurement, it rather moved the focus of control theory from a single (accounting based) perspective to a the inclusion of multiple factors. Measurement was still being used as a tool for control; however its scope was now extended towards controlling effectiveness (flexibility, timeliness and quality) as well as efficiency.

The main assertion of the performance management revolution was that a move from single (accounting based) measures to a multidimensional measurement system would improve control and therefore improve performance. The control perspective of the measurement – performance link asserted that measurement informed management action which in turn drove performance, implicit to the theory was the notion that the entire process was a feedback loop whereby performance outcomes (either positive or negative) were reported through the measures, which again prompted management action which led to performance outcomes. However, over time measurement and the control function of management became synonymous and this led to the often held, and erroneous, assumption that any application of measurement would result in improved performance, without any consideration of underlying causal linkages. Effectively the theory was conflated from a measurement-action-performance relationship to a simple measurement- performance relationship. This naïve simplification of the link between measurement and outcome(See fig 1 below) would have a lasting impact on research into performance measurement that is only recently being corrected.

Fig 1: Conflation of the Control Theory of the Measurement- Performance Link.

Immediately following the performance measurement revolution the focus of PM literature moved towards the design and implementation of PM Systems. This section of the measurement literature may best be described as focusing on process rather than theory as authors sought to describe best practice in design and implementation, all the while acknowledging the highly specific nature of this process for individual organisations.The necessity of focusing on process in terms of PM systems is explained by Neely (2005, p.1271) who claims it is an crucial element in the overall advancement of the subject area; that focus on ‘methods of application’, to be followed by ‘empirical investigation’ and ‘theoretical verification’ will advance knowledge and theory in the PM field. The focus on design and implementation processes did not mean that this phase of PM discourse was entirely devoid of theory. As interest in the design & use of systems moved to the evolution of PM systems and their ability to remain relevant over time, academics began to investigate the influences of contingent factors on PM systems.

Contingency theory states that an organisation’s structures and processes must fit with the organisational context (technology, culture, size, task and external environment) (Drazin & Ven 1985). In the case of performance measures, systems must be dynamic and have the ability to cope with changes in the competitive environment, another key criticism of financial measurement from the early 90’s. The adoption of contingency theory research has served to underline the concept that there is no one universally appropriate measurement system and the research has focused on particular contingencies that may impact on both the tendency to use PMS (Garengo & Bititci 2007; Jääskeläinen et al. 2012) and their successful implementation (Bourne et al. 2002). However contingency theory has been criticised for its failure to capture the social processes and subjective meanings of actors within the context of management control (Hewege 2012). Contingency theory also fails to take adequate note of the complexity of structures, or the interrelationships of people, that may underlie particular outcomes (Tosi & Slocum 1984).

The contingency theory perspective is only a slight improvement on the control theory dominated perspective of measurement, it takes a broader view that contingent factors influence tendency to measure, but retains the assumption that the presence of measurement and the achievement of outcome are guaranteed. Such was the focus on contingent factors that the initial work which focused on factors influencing PM systems was replaced with a significant body of research on contingent factors of performance, ignoring the involvement of measurement altogether. While the use of contingency theory does bring other factors into play in regard to adopting performance measurement systems and their influence on performance outcomes, its use, in the 2nd phase of PM research resulted in the maintenance of the conflated measurement-performance relationship assumption of the control theory perspective as described in fig 2.

Fig 2: Conflation of Contingency Theory of the Measurement-Performance Link

Further fusion of ideas based on the contingency theory concept resulted in the removal of measurement from the equation entirely, resulting in yet another erroneous assumption – that performance was based solely on contingent factors. The persistence of this assumption was aided by the concentration of the PM investigations on the large manufacturing sector where contingent factors like size, ownership and technology usage tend to have a significant causal relationship with performance; however the SME and services sectors were largely ignored in published research, until relatively recently.

Thepersistence of the control perspective and the conflation of performance outcomes to a dependency on contingent factors led to stagnation in knowledge about the measurement-performance link. This was most notably demonstrated in the results of a review of 99 studies on the impacts of measurement on organisational performance by Franco & Bourne (2004). This study concluded that findings were contradictory with most papers suggesting a positive relationship although this diminished with increasingly rigorous research methods. It would appear that the impacts (or consequences) of performance measurement is still a relatively open question amongst academics.

With the efficacy of performance measurement still not firmly established, practitioners and academics have, of late, concentrated their efforts towards what Neely (2005) refers to as the ‘empirical Investigation’ of PM. In particular research focused on the consequences of performance measurement focusing on three key areas; the consequences of measurement on people’s behaviour(Malina & Selto 2001; Henri 2006b), the consequences for organisational capabilities (Chenhall 2005; Cruz et al. 2011; Papalexandris et al. 2004)and the consequences for firm performance(Hoque & James 2000; Hoque 2004; Ukko et al. 2007). While research into the consequences of measurement in some areas seems to have reached relatively definitive conclusions research into the link between performance measurement and firm performance is inconclusive at best (Franco & Bourne 2004; Franco-Santos et al. 2012). This new focus of PM research seeks to go back to the basic notion of investigating the link between measurement and the achievement of performance outcome and is a significant step forward in terms of unpacking the conflated assumptions of previous theoretical perspectives.

With the move towards the examination of the consequences of PM, a broader pallet of theoretical perspectives has become apparent with 6 main theories coming to the fore of PM investigation (Franco-Santos et al. 2012). The appearance of a broader theoretical scope in investigating the measurement - performance link is welcomed but not without its own difficulties. One broad criticism that can be made about the utilisation of all of these theories is that they are applied very narrowly to the PM research discourse. Various theories are being employed in examining the impacts of measurement on organisational capabilities, individual behaviour and firm performance but the individual theories are being applied to specific elements of firm performance and individual behaviour, most notably around the area of performance related compensation. A synopsis of these theories and their application to the PM literature is supplied below (see table 1 over page). The final 6th theory suggested by Franco-Santos et al (2012)is contingency theory (discussed previously), which they suggest has the potential to resolve theoretical inconsistencies in an area that they believe is highly dependent on contextual factors.

Table 1: Theoretical Perspectives and their application to the PM Literature

Theory / Application to Performance Measurement Discourse
Agency Theory:
Outlines the relationship between principals (shareholders) and agents (managers) and the conflicts brought about by their differing goals and objectives. /
  • Performance measurement is seen as removing the information asymmetry issue which may exist between parties, thus improving performance (Dossi & Patelli 2010).
  • Describe the role of measurement in compensation and its ability to motivate agents and insure focus on principals’ goals.(Banker et al. 2000; Burney & Widener 2007)

Resource Based View of the Firm:
The resource-based view focuses on the individual resources of the firm. Competitive advantage is derived from the deployment of resources that are unique, valuable and difficult to emulate. / Firm capabilities can be enhanced through the use of measurement as a diagnostic and interactive tool in order to improve the capabilities upon which resource advantage is based (Bisbe & Otley 2004; Henri 2006a)
Cognitive & Information Processing Theories: / Examines how managers interpret and use measures in ways consistent with their preferences and may therefore focus on one type of measure (financial V’s non-financial) (Tayler 2010).
Goal-Setting Theory:
Specific and challenging goals produce greater performance effects / Used as a basis for the justification of valid measurement. Specific and clear measures will reduce ambiguity and positively impacts goal commitment and performance. This is often accompanied by discussions of equity theory (Lau & Sholihin 2005)
Equity Theory:
Individuals compare effort / reward in their own positions with perceived effort / reward of others and move to eliminate inequity. /
  • If equity is taken into consideration during system design and implementation the likelihood of its success is higher, as it is perceived to be fair by those using the system (Burney & Widener 2007).
  • Equity theory also explains reported dissatisfaction with PM systems which are perceived to be unfair due to inconsistent application or ambiguous measurement (Ittner et al. 2003)

Theory Definitions adapted from: The new Penguin Business Dictionary (Davis et al. 2003)

It can be argued that these theories are being applied to specific issues within the performance measurement lexicondepending on the problem being investigated or the disciplinary background of the researcher and that a single holistic view on the consequences of measurement for firm performance is lacking. In fact the utility of a meta-theoretical perspective to the investigation of the measurement-performance link is mooted by Franco-Santos et al. (2012) but they do not prescribe a particular perspective, while Hewege(2012) suggests that structural or anthropological theories may be a useful tool in the examination of management control systems. This article seeks to propose Critical Realism as a meta-theory which will fill many of the gaps in the existing knowledge of the measurement-performance link and be beneficial to closing the knowledge gap in this field of investigation.

To summarise, in theoretical terms, the performance measurement literature of the last 20 years is difficult. The ghosts of management control theory and its empiricist foundations are still strongly evident both from early research and from the widespread, but inconclusive, empirical testing of contingency factors in performance. Another ghost of the management control field is the acknowledged overdependence on research based in the large manufacturing sector (Neely et al. 2007). As the PMField has matured the application of more diverse theories has become apparent, but this may have clouded knowledge about the fundamental links between measurement and performance outcomes. There are increasing calls for the adoption of research using alternative meta-theory approaches, as opposed to scientific /empirically based perspectives, which may prove enlightening, in particular in regards to investigating the links between measurement and performance.

Critical Realism: A Meta-Theoretical Perspective of the Measurement-Performance Link

One of the prevailing criticisms of the existing body of PM research is that an empirically verifiable link between measurement and performance outcomes has yet to be satisfactorily established, this criticism holds equally for both qualitative and quantitative investigations. However an argument put forward by Hesketh & Fleetwood (2006) in discussing the link between Human Resources processes and performance argue that although empirical evidence may be inconclusive, the lack of empirical evidence:

‘…does not entail the non-existence of some kind of causal connection between them. It could be the case that a causal connection exists, but the nature of the causality is more complex than can be captured via the usual statistical techniques’ (Hesketh & Fleetwood 2006, p.678).