Reviewing and Theorizing the Unintended Consequences of Performance Management Systems

Abstract

Different design choices in the controls used to manage performance often lead to a range of unintended consequences, which have profound effects on individuals and organizations. This paper presents a mixed review (both systematic and eclectic) of the literature on the unintended consequences of performance management systems and develops a typology to explain how and why they occur. It finds that the most salient unintended consequences of directive performance management systems are gaming, information manipulation, selective attention, illusion of control and relationships transformation. It argues that these consequences exist due to limiting factors such as ignorance, error, short-term concerns, fundamental values, self-fulfilling forecasts and changes in social relationships. The emerging typology-based theory suggests that the choice of control mechanisms is based on two key assumptions concerning goal-alignment and goal uncertainty that relate back to ideas in agency theory and stewardship theory. It concludes that, in the design of performance management systems, the more the ‘assumed’ reality about the state of goal-alignment and goal-uncertainty diverges from the ‘real’ state of affairs, the more the resultant system is likely to create perverse unintended consequences, leading to poor organisational outcomes.

Full text

1
Reviewing and theorizing the unintended consequences of performance
management systems
Monica Franco-Santos
1
Cranfield University
School of Management
Cranfield, Bedfordshire
MK43 0AL
Tel. +44 (0) 1234 751122
monica.franco@cranfield.ac.uk
David Otley
Lancaster University Management School
Department of Accounting and Finance
Lancaster LA1 4YX
Tel: +44 (0)1524 593636
d.otley@lancaster.ac.uk
March 2018
1
Corresponding author.

2
Reviewing and theorizing the unintended consequences of performance
management systems
Abstract
Different design choices in the controls used to manage performance often lead to a range of
unintended consequences, which have profound effects on individuals and organizations.
This paper presents a mixed review (both systematic and eclectic) of the literature on the
unintended consequences of performance management systems and develops a typology to
explain how and why they occur. It finds that the most salient unintended consequences of
directive performance management systems are gaming, information manipulation, selective
attention, illusion of control and relationships transformation. It argues that these
consequences exist due to limiting factors such as ignorance, error, short-term concerns,
fundamental values, self-fulfilling forecasts and changes in social relationships. The
emerging typology-based theory suggests that the choice of control mechanisms is based on
two key assumptions concerning goal-alignment and goal uncertainty that relate back to ideas
in agency theory and stewardship theory. It concludes that, in the design of performance
management systems, the more the ‘assumed’ reality about the state of goal-alignment and
goal-uncertainty diverges from the ‘real’ state of affairs, the more the resultant system is
likely to create perverse unintended consequences, leading to poor organisational outcomes.
Keywords: Performance management, organizational control, management control,
unintended consequences, agency theory, stewardship theory.

3
Introduction
Considerable research has been focused on explaining the extent to which performance
management systems (PMSs), involving performance planning, measures, targets, incentives
and other means of control, deliver their intended consequences (Franco-Santos et al., 2012;
Melnyk et al., 2014; Smith and Bititci, 2017). For example, researchers have studied the
effects PMSs have on phenomena such as strategic alignment (Ahn, 2001; Dossi and Patelli,
2010; McAdam and Bailie, 2002), communication (Godener and Eric, 2004), corporate
control (Cruz et al., 2011) or accounting performance (Crabtree and DeBusk, 2008; Davis
and Albright, 2004; Ittner et al., 2003). These are some of the expected results most leaders
have in mind when promoting performance management in their organizations (Ferreira and
Otley, 2009). Less attention, however, has been given to the unintended consequences of
PMSs (Franco-Santos et al., 2012). A few studies (Bevan and Hood, 2006; Smith, 1995a)
have identified common negative side effects (e.g., measure fixation, myopia, gaming or
manipulation of data). Yet, whilst we know the general existence of unexpected results, we
still know little about how or why they occur. The aim of this paper is to address this gap in
our knowledge.
Research in this area is crucial as the unintended consequences of performance
management may often have a perverse impact on organizations. For example, a public
enquiry into the failings of a UK Hospital Trust (Francis, 2013) found that the overreliance of
senior staff on the hospital’s PMS led to an organizational culture focused on doing the
system’s “business” (e.g., hitting performance targets), resulting in patient neglect and high
mortality rates. The enquiry emphasized that well-intended decisions associated with the
process of managing performance produced consequences that were largely unintended
(Francis, 2013). Paradoxically, a system aimed at facilitating the delivery of high-quality
patient care and healthy lives, ended up creating the opposite results. Sadly, this event is not a

4
rare occurrence, and similar perverse outcomes generated by the use of performance
management mechanisms can be found in the literature (Dahler-Larsen, 2014; Lindsay et al.,
2014; Parkinson, 2012).
Sociologists have long fostered research on the unintended (Mica et al., 2011). They
argue that unintended consequences are actually the norm rather than the exception; and that
their existence requires theorising so these consequences can be curtailed (Elias, 1997;
Merton, 1936). It follows from these ideas that, if PMSs invariably generate unintended
consequences, it is vital for research in this area to go beyond their mere acknowledgement
and description. To advance our knowledge, we need a theoretical understanding of the
relationship between PMSs and their unintended consequences. This understanding will help
us to better assess their effectiveness and minimise their dysfunctional side effects. With that
in mind, the objective of this paper is to develop an organizational theory of the unintended
consequences of PMSs.
Theory development involves a number of tools and mental processes, some of which
are difficult to describe and validate (Bourgeois, 1979; Cornelissen, 2017; Delbridge and
Fiss, 2013; Doty and Glick, 1994; Shepherd and Suddaby, 2017; Snow and Ketchen David J.,
2014; Weick, 1989). Most scholars agree that a comprehensive review of relevant literature is
at the core of any theory construction process (e.g., Bourgeois, 1979). An immersion in the
literature can uncover tensions, paradoxes, and numerous puzzles that can lead to compelling
explanations (Shepherd and Suddaby, 2017). Thus, following the advice of theory building
scholars, our research thoroughly examines the literatures on both performance management
and the sociology of the unintended with the purpose of uncovering and explaining, through a
typology, the relationship between PMSs and their unintended consequences.
Drawing on the insights extracted from the literature coupled with ideas from agency
theory (Eisenhardt, 1989; Jensen and Meckling, 1976) and stewardship theory (Davis et al.,

5
1997a; Hernandez, 2012), we posit that the unintended undesirable consequences of PMSs
are likely to occur when a system is designed using false assumptions about people’s
behaviour and the incidence of uncertainty. False assumptions may be due to ignorance,
error, immediate short-term concerns, fundamental values and ideologies, or self-fulfilling
forecasts. Invalid assumptions will lead to the adoption of control mechanisms that do not fit
the “real” context, altering the existing social relationships that shape individual decisions
and actions and, over time, creating unintended effects with overall costs that may outweigh
the benefits.
Our work contributes to existing research in three important ways. Firstly, it provides
some words of caution regarding the uncritical adoption of agency theory assumptions when
designing PMSs. Secondly, it contributes to previous performance management contingency
research (Chenhall, 2003; Otley, 2016) providing additional evidence regarding the
importance of “fit” between control mechanisms and their environment. Thirdly, the study
enriches our understanding of the effectiveness of PMSs by illuminating the reasons why
unintended consequences may occur, so they can be minimized although never fully
eliminated.
Literature review methodology
We initially contemplated using a systematic review approach, consisting of a methodical,
transparent and reproducible literature search and synthesis process with “the twin aims of
enhancing the knowledge base and informing policy making and practice” (Tranfield et al.,
2003, p. 207). Systematic reviews have been praised for their usefulness in examining ‘what
works’ (Gough, 2015), but their value for theory building has been highly criticised (Gough,
2015; Gough et al., 2012). When the purpose of a review is not only to investigate ‘what
works’, but also to uncover the logic or mechanisms that explain ‘why something works’ (or

6
does not work), systematic reviews may not be sufficient. For theory building some authors
(e.g., Gough et al., 2012) suggest a systematic review can be complemented by a more
eclectic review, allowing for a broader search for insights and “trial and error thinking and
imagination” (Weick, 1989, p. 516). This alternative review method has been described as a
“mixed method review” (Gough, 2015, p. 3) because it integrates some of the transparent and
rigorous processes of systematic reviews (i.e., scoping study, selection of keywords and
search streams, exclusion and inclusion criteria and quality assessment) whilst keeping the
openness and scope of traditional reviews. Since our research has a theory building purpose,
we adopted a mixed method review, consisting of both an eclectic literature review and a
systematic review.
Our eclectic review had three key objectives, central to any theorising attempt
(Shepherd and Suddaby, 2017; Weick, 1989): to clarify the meanings of core terms; to
discover existing shared research assumptions; and to identify existing theories that could
help in our theorising work. To address these objectives, we analysed performance
management research and the sociology literature on theorising the unintended (Waddington,
2016). Our eclectic or traditional review process follows Ridley’s (2012) suggestions and is
outlined in Table 1. The insights from this review helped us to ascertain the language,
concepts, known relationships and previous considerations needed for our theory
development endeavour (Weick, 1989).
INSERT TABLE 1 AROUND HERE
Our systematic review focused on the unintended consequences of PMSs. We
addressed two specific review questions: (1) what are the most common unintended

7
undesirable consequences of PMSs? And (2) what is known about how and why they occur?
We followed the advice of Tranfield et al. (2003) and Briner and Denyer (2012), with a full
description of the process being presented in Table 1. We initiated the review with a scoping
study in which we examined classic works on the unintended consequences of performance
management. This helped to identify key search words and segments of the literature where
relevant knowledge was more likely to exist. We then investigated the research that had cited
these classic papers and also performed additional searches in the EBSCO library database
using search strings. We conducted our searches for the years 1950-2017 to give us good
coverage and to include the years when the emerging work on the unintended consequences
of performance management was published.
When analysing the documents identified, we adopted a set of inclusion, exclusion
and quality criteria. We included empirical and conceptual studies that provided insights
about the dysfunctional unintended consequences of PMSs published in English. We
excluded studies focused on the control of specific and narrow management processes or
single functions (e.g., supply chains, production or marketing). We also excluded
commentaries on books, erratum papers, introductory papers to special issues, and conference
summaries. To ensure the quality of the work, we focused on studies considered 3* and 4* by
the ABS Academic Journal Guide (see journal list in Table 1). To analyse the contend of each
selected article, we first extracted critical information about the field of research, research
method, context, control mechanisms studied, unintended consequences found, and
explanations provided. We then adopted a thematic analysis and provided a narrative
synthesis of our findings. The insights unearthed from both literature reviews are now
presented.

8
Findings: Eclectic review
We first synthesise the findings of our traditional review. These findings are presented in two
separate sections: one dedicated to the insights unearthed from the general performance
management literature; and another dedicated to the insights of the sociology literature on
theorising unintended consequences.
Performance management research
When reviewing performance management research, we paid special attention to three areas
considered to be critical for our subsequent theorising attempt: the meaning of PMSs, shared
research assumptions, and the organizational theories that can help explain PMS
consequences.
Definition of a performance management system
In any theorising endeavour, it is critical to clarify the meaning of core terms (Shepherd and
Suddaby, 2017). In the literature, a PMS has been broadly defined as a set of management
control mechanisms used by executives and employees with the overall purpose of
facilitating the delivery of organizational goals by influencing people’s behaviour and
performance (Broadbent and Laughlin, 2009; Ferreira and Otley, 2009; Malmi and Brown,
2008). At the very least, a PMS comprises: a planning element, which includes the goals that
reflect stakeholders’ expectations and thus defines performance; a measurement element,
which includes the metrics used to operationalize performance; a review element, which
refers to the evaluation and feedback of performance information; and a performance-related
reward element, which can be extrinsic (e.g., bonuses) or intrinsic (e.g., a clear sense of
achievement) (Ferreira and Otley, 2009; Flamholtz et al., 1985; Otley, 1999). A PMS can
also include further control elements, such as cultural and administrative controls (Malmi and
Brown, 2008; Ouchi, 1979).

9
It should be noted that the term ‘PMS’
2
is of relatively recent origin and has emerged from a
previous stream of literature that discussed what is known as ‘management control systems’
(Ferreira and Otley, 2009; Otley, 1999). We view these two terms as largely synonymous
whilst noting that earlier work tended to focus on systems of financial and budgetary controls
rather than on a wider range of organizational control mechanisms. This reflects the fact that
formal control systems were rudimentary and relied primarily on financial data until the later
1980s. In this paper, we use the term ‘PMS’. We view it as an overall system combining a
variety of specific mechanisms that will vary from organization to organization.
Shared research assumptions
Theory building emerges from the identification and questioning of existing shared
assumptions (Shepherd and Suddaby, 2017; Weick, 1989). The following assumptions of
PMS research are particularly relevant as they provide a foundation on which to base our
theoretical work.
Firstly, previous research largely regards each of the control mechanisms comprised
in a PMS as a subsystem in its own right (e.g. measurement system, reward system)(Ferreira
and Otley, 2009; Malmi and Brown, 2008; Otley, 1999). These sub-systems may have been
designed by different people (e.g., HR managers, strategy managers, management
accountants) and at different points in time. They may work differently within each of the
organizational functions (and possibly locations), and at different hierarchical levels (Ferreira
and Otley, 2009). It is often assumed that the overriding purpose of these separate subsystems
2
The literature in Human Resource Management often uses this term in the restricted sense of (annual)
employee performance appraisal of a primarily developmental nature.

10
is to act as a coordinated system to help the organization attain its goals, although such a
complete integration is rarely achieved in practice (see Otley, 2016).
Secondly, the control mechanisms forming a PMS are considered to be connected to
each other in a variety of ways (Ferreira and Otley, 2009; Malmi and Brown, 2008). Using
the concepts of “tightly” and “loosely coupled” systems (Weick, 1976, p. 4), it can be
suggested that some controls are tightly coupled (i.e., they are not only linked together in
some ways, but they are also highly dependent on each other so that changes in one control
lead to consequent changes in others) (Franco-Santos et al., 2014; Malmi and Brown, 2008;
Otley, 2016). Other control mechanisms may be loosely coupled (i.e., they are subsystems
that operate on an exchange relationship where little input is needed from each additional
subsystem – that is, subsystems may interact but they are not dependent on each other to
function adequately) (Franco-Santos et al., 2014; Malmi and Brown, 2008; Otley, 2016).
Thirdly, PMSs are constantly evolving (Cardinal et al., 2004; Ferreira and Otley,
2009). Control mechanisms that are seen to have deficiencies are improved or transformed
(Chenhall and Euske, 2007). Changes are made on an on-going basis, often with the aim of
aligning disparate subsystems with each other, or responding to new internal and external
pressures (Chenhall, 2003; Otley, 2016). However, significant lags in alignment (or degrees
of integration) may occur
3
; new control mechanisms may also be introduced that have their
own differences from the extant controls; and the speed of integration often lags the rate of
introduction of new elements (Otley, 2016).
Fourthly, the control mechanisms that comprise a PMS can be either formal or
informal (Broadbent and Laughlin, 2009; Ferreira and Otley, 2009; Malmi and Brown, 2008).
3
Pay for performance systems are particularly ‘sticky' in this respect as they often require renegotiation with
groups of employees and the measures used to drive performance-related pay are invariably slow to get updated.

11
Formal control mechanisms (e.g., measures, targets, etc.) are considered tangible objects that
can be controlled and easily changed. Informal control mechanisms (e.g., peer pressure,
routines) are less tangible social phenomena that are more difficult to investigate and change
(Chenhall and Morris, 1995; Kreutzer et al., 2016). In the literature, there appears to be a
general understanding that formal control mechanisms are likely to have dysfunctional effects
and these can be mitigated by the use of informal means of control (Kreutzer et al., 2016;
Malmi and Brown, 2008). Despite this general understanding of the likely existence of
unintended consequences, as suggested earlier, little appears to be known about how or why
these consequences emerge.
Finally, PMSs are based on implicit or explicit mental models that predict likely
outcomes (Otley and Berry, 1980). Individual managers each have their own mental models
by which they decide upon particular courses of action because they predict these will lead to
specific desirable outcomes. Different managers may often have markedly different
predictive models, which may lead to different uses of control mechanisms (Hall, 2011).
When organizations design central PMSs, their designers have cognitive schemas in their
minds that reflect their own interpretation of organizational goals, together with their view of
the drivers of organizational success and how these drivers relate to each other (Otley and
Berry, 1980). Designers’ models are often based on an array of assumptions about people, the
environmental conditions under which the organization operates, and about the control
mechanisms that best suit these drivers of organizational success (Broadbent and Laughlin,
2009).
Seminal theories on the consequences of performance management systems
For the development of new theory, it is important to review extant theories illuminating the
phenomenon under study (Shepherd and Suddaby, 2017; Weick, 1989). When reviewing

12
theories, we focused on those that provided general statements explaining PMSs existence
and predicting their consequences. As such, three well-known organizational theories were
identified: contingency theory of accounting (Otley, 1980, 2016), agency theory (Eisenhardt,
1989; Jensen and Meckling, 1976) and stewardship theory (Davis et al., 1997b; Hernandez,
2012).
The contingency theory of management accounting (Otley, 1980, 2016) argues that a
‘universal approach’ to the design and use of PMSs across and within organizations is
unlikely to be effective. Control mechanisms are to be tailored to fit the specific
circumstances of their use and context to deliver their intended consequences. Since the
1980s, an array of factors (or contingencies) has been shown to influence the design, use and
consequences of control mechanisms (e.g. environmental uncertainty, task complexity)
(Chenhall, 2003; Otley, 1980). Recent contingency research suggests that the number of
factors influencing PMSs has expanded together with the complexity of these systems (Otley,
2016). As a result, the provision of precise prescriptions taking into consideration the
increasing number of contingencies affecting PMSs is unlikely to be possible.
Most previous PMS research is underpinned by agency theory (Franco-Santos et al.,
2012; Otley, 1999; Otley and Berry, 1980). Agency theory (Eisenhardt, 1989; Jensen and
Meckling, 1976) explains the choice, use and consequences of organizational control
mechanisms. According to the basic model of this theory
4
, organizations seek to fulfil the
goal expectations of their owners and employees are assumed to be self-serving or
opportunistic, effort-averse, and risk-averse. This situation generates a goal-alignment
4
We acknowledge that there are different strands of agency theory and that over the years it has been extended
to accommodate numerous developments. However, we noticed in the review of the literature that most
performance management research interprets agency theory in terms of its basic model (Eisenhardt, 1989). This
version of the theory is also the most commonly recognised by managers and by management scholars.

13
problem and increases the uncertainty of outcomes as employees may “shirk” by putting in
less effort, or by focusing on their own goals at the expense of owners’ goals (Eisenhardt,
1989; Nilakant and Rao, 1994).
To address these issues, agency theory (Baiman, 1982; Eisenhardt, 1989), drawing
mainly on the discipline of economics, suggests the use of two tightly coupled control
mechanisms: monitoring systems including explicit goals, performance measures and targets
that provide information and help to assess employee behaviour and the results of their
actions; and incentive systems such as short-term bonuses or stock options to encourage
employees to focus on the realization of the organization’s goals. Previous research has
referred to these control mechanisms as directive (Franco-Santos and Doherty, 2017, p.
2319).
Stewardship theory (Davis et al., 1997b; Hernandez, 2012) is another theory
explaining the choice, use and consequences of control mechanisms; although it is often
positioned as contrary to agency theory. It has its roots in psychology and sociology and
suggests that the assumptions of agency theory are too restrictive to apply to many
organizations and employees, so exclusive reliance on its ideas and proposed controls may
lead to undesirable consequences (Davis et al., 1997b). Stewardship theory assumes
organizations have multiple and complex goals, and employees can behave as stewards rather
than agents, having “a shared sense of ongoing responsibility to multiple stakeholders, which
affects a focus on collective welfare over the long term” (Hernandez, 2012, p. 176).
Stewardship theorists (Davis et al., 1997b; Hernandez, 2012) argue that goal-
alignment is not an issue stewardship type organizations need to resolve. Uncertainty is seen
as the natural state of affairs, so organizations must learn how to respond and cope with it.
Consequently, if there is no goal-misalignment and uncertainty cannot be ‘managed’, the
adoption of monitoring and incentive controls is unnecessary; it may actually be

14
dysfunctional undermining the positive attitudes of stewards (Hernandez, 2012). Rather than
relying on agency type controls, stewardship theorists suggest that organizations may be
better off by “fostering relationship-centred collaboration through shared leadership practices,
promoting employees’ collective responsibility for work outcomes, enabling employees to
derive intrinsic benefits from working towards a valued end; and cultivating self-efficacy and
self-determination through ongoing employee development” (Hernandez, 2012, p. 177). Due
to the focus on facilitating performance rather than controlling it, this type of control has been
labelled as enabling (Franco-Santos and Doherty, 2017).
The different control approaches suggested by agency and stewardship theories
resonate with a series of apparent conceptual dichotomies that have been frequently
mentioned in different management literatures although rarely considered in combination.
For example, concepts such as ‘technical’ and ‘social controls’ (as in socio-technical
systems) (Smith and Bititci, 2017) focus on different aspects of the design and use of PMSs,
but suggest that both views need to be combined. ‘Transactional’ and ‘relational’ controls
have been distinguished (Broadbent and Laughlin, 2009), with the first approach based on the
assumption that business is conducted as a series of unrelated arm’s length transactions,
whereas the second recognizes the importance of continued relationships over time.
‘Coercive’ and ‘enabling’ (Adler and Borys, 1996) have also been used in previous works to
refer to two different approaches of organising work in bureaucracies. Finally, the well-
known ‘Theory X’ and ‘Theory Y’ (McGregor, 1960) has been highly influential, with
Theory X assuming that people must be coerced into behaving appropriately in an
organizational context, and Theory Y assuming that people will naturally behave in a
cooperative manner if treated appropriately. We do not see each of these pairs of contrasts as
identical but regard them as displaying a significant degree of commonality that allows us to
use the distinction between them as an underlying construct of considerable importance.

15
Building on recent research (Franco-Santos et al., 2017) and for the sake of clarity, we will
use the terms ‘directive’ and ‘enabling’ systems to describe this contrast whilst recognizing
the much wider literature that is embodied in this overall idea.
Having synthesised the insights extracted from the performance management
literature we now turn to examine the ideas acquired from the sociology literature on
theorising unintended consequences.
Unintended consequences research
To better understand how and why the unintended consequences of performance management
may be dysfunctional, we reviewed previous research focused on theorising unintended
consequences. The management literature is largely silent on this topic (Franco-Santos et al.,
2012) with most relevant research coming from sociology (Waddington, 2016). Sociology
has been described as the social science most active in examining the “unrecognised,
unintended and emergent consequences of goal-oriented action” or “the science of the
unexpected” (Portes, 2000, p. 1). We narrowed our focus to the most well-known
sociological work investigating core meanings, shared assumptions and theories of
unintended consequences.
Definition of unintended consequences
An unintended consequence has been defined as “a particular effect of purposive action
which is different from what was wanted the moment of carrying out the act, and the want of
which was a reason for carrying it out” (Baert, 1991, p. 201). In other words, it is the
outcome of a conscious action other than that foreseen and intended. The term ‘purposive
action’ was initially used by politicians and economists (e.g., Smith, 1759), but it was the
sociologist Robert Merton (1936) who popularised it and the concept is now widely use in
social science research. It is also clearly relevant to PMSs investigations.

16
Shared assumptions
Research focused on theorising unintended consequences often shares the following
assumptions (Baert, 1991; Mica et al., 2011). Firstly, it assumes there is a recognised
intention on the part of the person or actor carrying out an action. This intention is to improve
the state of things. Secondly, it accepts that any action will unavoidably have unintended
consequences. Some of these unintended consequences may be desirable or beneficial, whilst
others may be undesirable or dysfunctional with the possibility of reaching a point where they
become perverse (i.e., the opposite of the intended). Thirdly, it suggests that unintended
undesirable consequences can be minimized, but never fully eliminated. Fourthly, it supposes
that it is vital to theorise unintended undesirable consequences to avoid an overall result in
which an action, policy or practice ‘backfires’ (i.e., does more harm than good). Most
sociology research, therefore, focuses on the explanation of undesirable unintended
consequences and the situations under which these consequences may frustrate the
underlying intention of the actors (Baert, 1991). It is worth noting that most researchers
equate unintended consequences with unanticipated, unforeseen or unexpected effects (Baert,
1991, p. 201).
Existing theories of unintended consequences
Most research explaining unintended consequences (Mackay and Chia, 2013; Mckinley and
Scherer, 2000; Peattie et al., 2016; Perri, 2014) refers to the ideas introduced by Merton
(1936, 1968, 1996), and later extended by Elias (1998). Both authors provide the bedrock for
the theoretical developments of the unintended consequences of purposive action (de Zwart,
2015; Mica et al., 2011). Merton (1936, p. 894) contends that any “purposive social action”
inevitably has unintended consequences. Thus, some unintended consequences may actually
be expected or anticipated as they may have been taken into consideration when deciding on

17
the action. This point is critical for PMS research and, as suggested by Baert (1991), most
social research underplays it or completely ignores it.
Merton (1936) is the first scholar to provide a general theory of unintended
consequences. His work is widely recognised as the point of departure for any theory of
unintended consequences (de Zwart, 2015; Mica et al., 2011). He argues that unintended
consequences result from the interplay between actions and the situation in which actions
occur, which is an explanation that resonates with the idea of ‘contingency fit’ in
management research (Donaldson, 2001; Drazin and Van de Ven, 1985; Otley, 2016). He
further elaborates his theory suggesting that unintended consequences exist because actors
cannot anticipate all that can happen as a result of their chosen actions. Merton identifies five
factors that may limit actors’ possibilities to anticipate the full consequences of their actions:
(1) actors may not have sufficient knowledge or they may ignore the knowledge they have;
(2) they may err in the appraisal of the present and future situation, in the selection of a
course of action or in the execution of the action chosen; (3) they may have an acute concern
with the foreseen immediate or short-term consequences and give little consideration to
longer-term outcomes, which Merton (1936) labels as the ‘imperious immediacy of interest’;
(4) actors may have fundamental values, which justify the felt necessity of certain actions;
and (5) actors may take into consideration a preconceived idea suggesting particular
outcomes and this prediction in itself becomes a new element in the concrete situation which
subsequently makes the prediction accurate (‘a self-fulfilling prophecy’). Merton suggests
that these factors are not mutually exclusive and may reinforce each other.
The extension that Elias (1997, 1998) applies to Merton’s (1936, 1968, 1996) work is
crucial as he provides a clearer explanation of the way in which purposive social action
creates unintended undesirable consequences (Mica et al., 2011). Elias argues that unintended
consequences mainly occur due to the social relationships existing in any group of people.

18
People’s knowledge of the social relationships in which they operate is unavoidably
imperfect, incomplete and inaccurate. Therefore, their actions, which are based on their
inadequate knowledge, more often than not, will have consequences they do not foresee
(Mennell, 1977). Elias argues that the consequences of intended actions can only be
understood by investigating the dynamics of social relationships. Equally, people's behaviour
can only be understood in relation to the social relationships in which they operate. Intended
actions bring about changes to social relationships, and social relationships in turn influence
people's behaviour not always in the way intended. In short, Elias contends that unintended
consequences of intended actions will always occur and that these consequences, together
with the intended ones, create the basis for future action.
Bringing Merton’s and Elias’ ideas closer to our research, it can be argued that the
development and use of PMSs involves purposive social actions and that these actions will
have both intended and unintended consequences. As researchers, we need to strive to
explain the occurrence of these consequences; and, in particular, the occurrence of
undesirable or dysfunctional consequences, so we can minimize abnormalities. The use of
Merton's limiting factors leading to unintended consequences together with Elias’ focus on
the importance of social relationships dynamics are instrumental for the purpose of our
theory. However, as they recommend, further insights about the specific unintended
consequences of particular purposive actions are needed in any theorising attempt supporting
the need for a systematic review of the unintended consequences of PMSs. The next section
describes the key findings from our systematic review.
Findings: Systematic review
Our systematic review had two key objectives. Firstly, to identify the most common
unintended consequences of PMSs, and secondly, to analyse what is known about how and

19
why they occur as reported in the performance management literature. Performance
management research has identified a number of unintended undesirable consequences over
the years (see Table 2). A priori, it is interesting to note that this literature interprets the
concept of ‘unintended consequences’ as the negative unexpected effects of PMSs, which
implies that the analysis of unintended beneficial consequences as such has been largely
overlooked. The literature also seems to focus (not always explicitly) on the ‘directive’
control mechanisms endorsed by agency theorists (Eisenhardt, 1989; Jensen and Meckling,
1976); and it is mainly published in management accounting, general management, and
public-sector management journals. Most studies concentrate on how the use of performance
measures and targets linked to monetary incentives (i.e., directive or agency theory-related
systems) may have unexpected negative outcomes, especially in complex environments.
Little has been said about unintended consequences emanating from ‘enabling’ or
stewardship theory-related controls (e.g., peer pressure, cultural controls).
INSERT TABLE 2 AROUND HERE
The unintended consequences of performance management systems
From our literature analysis and thematic coding five salient dysfunctional unintended
consequences of PMSs emerged. Firstly, directive PMSs have been shown to produce
strategic behaviour or gaming as individuals alter the way in which they behave to fulfil
performance expectations, sometimes to the point of breaching ethical norms (Berliner, 1956;

20
Hood, 2006; Jensen, 2003). In total 81
5
percent of the literature selected identified gaming as
a major unintended consequence. Interestingly, not all studies presented gaming as perverse.
A reduced set (e.g., Berliner, 1956) treated it as a ‘necessary evil’ worth having for the effort
gains obtained from PMSs.
Secondly, directive PMSs appear to be associated with various forms of information
manipulation (Cardinaels and Yin, 2015; Hood, 2006; Jensen, 2003; Kalgin, 2016; Smith,
1995a). Performance information is often misrepresented, misinterpreted, reclassified or
made up in order to meet performance requirements. 74 percent of the literature reviewed
mentioned this unexpected consequence. The deliberate manipulation of data can range from
creative accounting to clear fraud. Concepts such as ‘fiddling’ (Mannion and Braithwaite,
2012), ‘managing the numbers’ (Jensen, 2003; Li, 2015) or plain ‘dishonesty’ (Hannan et al.,
2006) are common in this body of work.
Thirdly, directive PMSs are likely to generate selective attention with 55 percent of
our selected articles referring to this phenomenon. Selective attention occurs both in terms of
‘what’ is measured and in terms of ‘when’ is measured. Managers appear to become fixated
on performance measures and targets that are quantified or formally considered in PMSs,
causing those aspects of organizational goals that are not (or cannot be) measured to be
overlooked (Hood, 2006; Kerpershoek et al., 2016; Mannion and Braithwaite, 2012; Smith,
1995a). They also tend to focus on actions that focus on short-term goals (or what in the
literature is described as ‘short-termism’ or ‘myopia’ (Merchant, 1990)), rather than selecting
5
It should be noted that this percentage reflects the degree of attention that each consequence has received in the
literature. It does not necessarily reflect the importance of each consequence or its relevance in the process for
creating perverse outcomes for the organization.

21
actions which may have greater long-term benefits but smaller (or even negative) impacts in
the short-term.
Fourthly, directive systems often produce an illusion of control. Over time, the use of
these controls leads managers to believe their assumptions about performance, its
measurement, and accountability are applicable to their organization. These premises give
way to the idea that PMSs accurately and validly reflect ‘actual’ performance (Chwastiak,
2006; Cugueró-Escofet and Rosanas, 2017; Hood, 2006). This unintended consequence is
cited by 24 percent of the studies included in the review. The illusion of control is often
perverse, especially, in highly uncertain and complex environments.
Finally, 81 percent of our selected studies suggest that both directive and enabling
PMSs are likely to alter the social relationships operating in organizations. For instance,
Chwastiak (2006) and Conrad and Guven (2012) found the PMSs studied, which were
focused on monitoring through performance measures, targets and incentives (i.e. agency
theory related), promoted transactional relationships, diminishing trust as well as generating
inequalities and differentiation. By contrast, Segal and Lehrer (2012) show how the transition
from an agency-related performance management approach to a stewardship-related approach
in a Canadian school district led to a transformation in the social relationships experienced by
staff from mistrust and alienation to high trust and benevolence, decreasing corruption and
other unintended effects.
Other unintended consequences of directive PMSs less prominent in the literature are:
administrative overload and managerial time costs due to the design, implementation and use
requirements of the system (e.g., Cox, 2005; Hansen et al., 2003); de-professionalization or
movement away from professional values and standards (Agyemang and Broadbent, 2015;
Tan and Rae, 2009); decreased well-being and morale (Bonner and Sprinkle, 2002; Franco-
Santos et al., 2017); ‘ossification’ (i.e., tendency towards being in a rigid state, constraining

22
responsiveness and resisting change) (Mannion and Braithwaite, 2012; Smith, 1995b); stifled
innovation (Lindsay et al., 2014); and injustice, unfairness and inequalities (Cugueró-Escofet
and Rosanas, 2016).
In comparison with the amount of work conducted in the for-profit sector on the
intended consequences of PMSs (Franco-Santos et al., 2012), it is noticeable that a large
proportion of the research on unintended dysfunctional consequences has taken place in
public sector organizations. In these organizations, the unintended consequences of PMS
mechanisms were deemed to be perverse (rather than necessary evils), not just for the users of
the system, but also for other stakeholders (e.g. public, patients, students, society in general)
(e.g., Conrad and Guven Uslu, 2012). In for-profit organizations, the predominant
undesirable unintended consequences studied were related to information manipulation,
strategic behaviour and short-termism (Healy, 1985; Jensen, 2003; Lowe and Shaw, 1968).
Explanations of how and why unintended consequences occur
Most extant research describes undesirable unintended consequences of performance
management, but very few scholars provide in-depth explanations about how and why these
consequences emerge in the first place. Nevertheless, after carefully reading and interpreting
the literature selected, the following insights can be extracted. Most scholars relate the
occurrence of unintended consequences to an increase in managers’ perceived pressure (Li,
2015; Merchant, 1990); the particular design or use of the PMS (Hopwood, 1972; Mannion
and Braithwaite, 2012; Schwepker and Good, 2012); and environmental complexity (Conrad
and Guven Uslu, 2012; Cox, 2005; Tan and Rae, 2009). Each of these reasons is examined in
turn.
There is some consensus on the idea that managers perceived high performance
pressure when directive PMSs are in operation (Merchant, 1990). Managers may interpret the

23
presence of high pressure as a threat (more so when there are known penalties for not
reaching targets such as lower pay or reduced opportunities for promotion). If they feel there
are possibilities for non-compliance, they may choose to do things that in other circumstances
they would not do (e.g., gaming, lying) (Carmichael, 1970). Over time, these behaviours may
be internalised and rationalised, affecting trust and social interactions (Cardinaels and Yin,
2015). In some contexts, gaming and low-trust environments are considered a ‘price worth
paying’ (Berliner, 1956) or ‘part of business life’ (Jensen, 2003). In others, high-trust social
relationships are paramount (e.g., hospitals, schools, universities) and, when these relations
are changed, perverse effects may appear (McCann et al., 2015).
Many scholars argue that unintended consequences are due to issues associated with
the particular design or use of PMSs (Berliner, 1956; Chow et al., 1988; Hopwood, 1972;
Kalgin, 2016). For instance, the link between performance measurement and rewards is found
to be a critical choice that can lead to negative unintended consequences (Berliner, 1956;
Jensen, 2003; Kerpershoek et al., 2016). Scholars have noticed that people’s sense of gain (or
loss) associated with performance-related rewards can lead to strategic behaviour, affecting
coordination, collaboration and social dynamics (Lowe and Shaw, 1968). It can generate
feelings of unfairness and injustice pushing people in the wrong direction (Cugueró-Escofet
and Rosanas, 2017). Hopwood (1972) highlights that including performance information in
the performance management process does not necessarily lead to dysfunctionalities, rather
arguing that these are caused by the way in which such data is used. He finds that using data
in ways that enable learning rather than with the sole purpose of controlling individuals leads
to less data manipulation.
A few researchers suggest that unintended consequences are due to complexity and
uncertainty issues. For example, Merchant (1990) argues that managers operating in
uncertain environments were significantly more likely to react to budget pressure by gaming

24
the system. Berliner (1956) observes that the continuous change in leaders created increased
uncertainty and more opportunities for gaming. The work of Tan and Rae (2009) also
suggests that in highly complex and regulated organizations such as airports, hospitals or
education institutions the use of directive PMSs is more likely to create side effects. They
argue this is due to aspects such as measurement difficulties (i.e., not everything that matters
can be measured, and existing measures might be ‘noisy’ or distorted), diverse stakeholders’
needs which cannot always be reconciled, or conflicting targets.
Together with these common explanations, a few researchers assert the unintended
consequences of a direct PMS are associated with its underlying assumptions about people’s
behaviour (Cardinaels and Yin, 2015; Franco-Santos et al., 2017; Kerpershoek et al., 2016;
Lindsay et al., 2014). Rooted in agency theory, directive mechanisms are tools to help
managers (i.e., principals) control the performance of employees (i.e., agents), who are
assumed to be opportunistic (Kerpershoek et al., 2016). By choosing directive controls (i.e.,
measures, targets, individual reviews, incentive pay), managers are indirectly sending a
message to employees about the needs for control implying that employees cannot be trusted.
At the individual level, this choice of controls may indicate to employees that others are
behaving opportunistically, which in turn may influence their own behaviour. As suggested
by Cardinaels (2015), the belief that employees in general are opportunistic (i.e., agents) will
change social norms and, over time, people will conform or internalised these norms
increasing their self-interested behaviour. In ‘mission-oriented’ organizations (e.g., hospitals,
schools, universities), where professionals and professional values often dominate (i.e.,
people more likely to behave as stewards rather than agents), false theories about people’s
behaviour, the importance of relationships and the real complexity of the context, can be the
grounds for the unintended consequences of direct controls to become perverse (Chwastiak,
2006; Franco-Santos et al., 2017; Kerpershoek et al., 2016; McCann et al., 2015).

25
Notwithstanding these varied rationalisations, researchers appear to have shied away
from providing specific predictions (i.e., a theory) relating not just directive PMSs to
unintended consequences, but also enabling PMSs to their own unintended consequences.
The absence of theorising, may be partly due to the multidimensional and complex nature of
explanatory factors and their potential relationships. Based on the insights extracted from
both of our reviews (see Table 3 for a summary), we now attempt to provide such a theory.
INSERT TABLE 3 AROUND HERE
Towards a new theory of the unintended
Before explaining our suggested theoretical framework, we find it necessary to clarify our
style of theorising. We believe this clarification will help to improve the transparency of our
theorising process and, hopefully, avoid misunderstandings.
Style of theorising
As suggested by Cornelissen (2017, p. 2), “at its core, a style [of theorising] is a particular
form of argumentation that we use to structure our thinking and express our ideas about a
management or organizational phenomenon in a common idiom”. Such an idiom can take the
form of propositions, a narrative of a series of interconnected processes or a typology
(Cornelissen, 2017; Delbridge and Fiss, 2013; Shepherd and Suddaby, 2017). All three styles
of theorising have their benefits and limitations (Cornelissen, 2017). We have chosen to
derive our theory of the unintended consequences of PMSs following a typology style
because of the complexity and multidimensional nature of the mechanisms at play which
made the other two theorising options less suitable (Cornelissen, 2017).

26
A typology style of theorising proceeds by “categorizing and clustering ideas and
observations to offer a multidimensional take on a management or organizational subject”
(Cornelissen, 2017, p. 6). Typologies are “conceptually derived, interrelated sets of ideal
types. […] [Each ideal type] represents a unique combination of the organizational attributes
that are believed to determine the relevant outcome(s)” (Doty and Glick, 1994, p. 232). A
typology style allows us to move beyond traditional linear or interaction theories and to
incorporate the multiple contingencies that are likely to relate to the unintended consequences
of PMSs. This type of theorising assumes that the “way organizational factors fit together is
very important and that to understand organizations we must consider simultaneously
multiple characteristics” (Doty and Glick, 1994). Another important aspect of typological
theories is that they incorporate the idea of ‘equifinality’ (i.e., different ideal types can reach
the same outcome following different paths)(Drazin and Van de Ven, 1985).
Good typologies provide explanations at two different levels: At one level, they
provide a grand theory generalizable to all organizations; at another level, they include a set
of middle-range theories restricted to the individual ideal types (Doty and Glick, 1994). In
many cases, grand theories are implicit and state that particular configurations grouped into
ideal types maximise ‘fit’ and can explain the outcome(s) (e.g., Miles and Snow, 1978).
Middle-range theories have narrower boundaries and explain the internal consistency of the
underlying mechanisms operating within each ideal type (Doty and Glick, 1994). These
theories may be based on different assumptions and the pattern of relationships that explain a
particular ideal type does not necessarily need to generalize to all the ideal types or to all
organizations (Doty and Glick, 1994).
As described earlier, the process of theory development can be initiated with a
thorough review of the literature, but it needs to be accompanied by other less observable
cognitive processes occurring simultaneously as the theorist interprets the evidence and

27
makes sense of it, creating connections and making generalisations (Shepherd and Suddaby,
2017; Weick, 1989). We present the results of our theory development process next.
Theorising the unintended consequences of performance management
In this section we make explicit our grand theoretical assertion, we define the key dimensions
that form the basis of our typology, and we examine the middle-range theories that predict
each ideal type.
Grand theoretical assertion: The importance of fit
As shown in our review, the contingency theory of management accounting (Chenhall, 2003;
Otley, 1980, 2016) is an overarching theory underpinning most research in PMSs. This theory
is based on the premise that the control mechanisms co-existing to form a PMS need to be
tailored to fit the specific circumstances of the organization. The better the ‘fit’ between a
PMS and its contingencies, the greater its effectiveness will be (i.e., the degree to which its
intended results are achieved) (Otley, 1980, 2016). Building on these ideas, our grand
theoretical assertion is that an identified set of ideal PMSs types will maximize the fit
between the chosen control mechanisms and the existing organizational conditions
Knowing that a better fit will drive better results does not mean that, in practice,
organizations will design the systems that are meant to be best for their circumstances.
Managers may (inevitably) have imperfect knowledge of the organization, its context and its
people; they may make mistakes when analysing the organization or when choosing the
control mechanisms that best fit organizational conditions; they may have immediate or
short-term pressures, discounting the importance of long-term effects; they may be biased by
particular values or ideologies; or their own believes may lead to self-fulfilling prophecies
(Merton, 1936). Because of these limiting factors, managers will make specific choices
between alternatives resulting in PMSs that may be (to various degrees) not fully aligned with

28
the situations they are meant to deal with. Thus, managers cannot anticipate all that can
happen to avoid potential issues and, inevitably, unintended consequences will appear.
Following Merton’s (1936) logic, some of these unintended consequences can be predicted
and curtailed through careful theorising about the “purposive action” and the situation or
context in which the action takes place. Therefore, if we assume a PMS to be a “purposive
action” (i.e., a deliberate act intended to cause change, which in this case is to improve
performance): how can we predict the unintended consequences of a PMS, especially those
that can lead to dysfunctionalities making it a perverse system?
To address this question, we go back to the idea that there is variability in the type of
control mechanisms organizations design and use (Bedford et al., 2016; Broadbent and
Laughlin, 2009; Ferreira and Otley, 2009; Malmi and Brown, 2008). Some organizations
heavily rely on performance monitoring and extrinsic rewards whilst others rely on other
more enabling means to deliver their mission and performance (Franco-Santos et al., 2014;
Frey et al., 2013). We suggest that organisations relying on directive PMSs in highly
uncertain contexts are likely to experience more undesirable unintended consequences not
just for managers but also for other stakeholders (Mannion and Braithwaite, 2012; Smith,
1995a). The findings of our reviews can help us explain this phenomenon.
Key dimensions: goal-alignment and goal-uncertainty
Our conjecture is that when organizations develop their PMSs, managers base their decisions
about which control mechanisms to introduce on two key factors. As suggested by agency
theory (Eisenhardt, 1989; Jensen and Meckling, 1976), they make assumptions about the
existing level of goal-alignment (i.e., how aligned they believe the goals or interests of
employees are to the goals of the organization). Whilst it is recognized that goal-alignment
can never be perfect (for example, the trade-offs between conflicting goals need to be

29
continually re-evaluated and communicated
6
), it is likely that the degree of goal-alignment
will differ between and within organizations depending on their circumstances.
Managers also make assumptions about goal-uncertainty (i.e., how much uncertainty
exists in managers’ predictive models). Managers assess their environment and the
measurability of organizational goals, developing predictive models (explicitly or implicitly),
which define what the organization is aiming for, identify a plan to achieve it, and attempt to
estimate what is likely to occur when such a plan is implemented. Since complete
information is rarely available, managers need to take into account the uncertainty that exists
in their models. Based on their predictive models and assumptions about people’s behaviour,
managers make choices among alternatives of control mechanisms and decide how to act
(Broadbent and Laughlin, 2009).
Middle range theories
Based on the insights extracted from the literature (Cardinaels and Yin, 2015; Kerpershoek et
al., 2016; Lindsay et al., 2014; McCann et al., 2015), it can be argued that the unintended
consequences of PMSs occur in situations where managers’ ‘assumed’ reality differs from the
‘real’ states of affairs or observed reality. How this difference comes about can again be
explained by the limiting factors suggested by Merton (1936) (i.e., ignorance, error,
fundamental values or ideologies, an acute focus on short-term consequences, and the
existence of a self-fulfilling prophecy). This divergence affects the interplay between the
chosen control mechanisms and the situation in which these mechanisms operate, decreasing
6
For example, taking the well-known trio of objectives in a project management perspective, that is quality, cost
and time, the emphasis placed on each will vary considerably over the course of a project. Such changes in
priorities will need to be communicated to employees so that they can respond appropriately in the specific
situations they are dealing with.

30
its fit. How and why this divergence generates dysfunctionalities is discussed below. We
illustrate our overall conceptual framework using Figure 1.
INSERT FIGURE 1 ABOUT HERE
In situations where managers assume low goal-alignment (believing that employees
behave opportunistically) and low goal-uncertainty (in terms of the accuracy of their
predictive model of outcomes), we expect to find greater reliance on the control mechanisms
suggested by agency theory (i.e., directive PMSs which include specific goals, measures and
targets, incentive pay). These control mechanisms may be appropriate in highly programmed
and well-understood situations (although the varieties of human response may make even this
problematic), but they may run into severe difficulties when applied to situations where the
opposite contingencies exist. One example can be the R&D function of a large
pharmaceutical company, where the mechanisms used for managing performance may need
to be radically adapted to cope with managing long-term projects with highly uncertain
outcomes and staffed by committed professionals.
It may happen that the assumed situation reflects the premises of agency theory (i.e.,
self-serving behaviours, risk-aversion, maximisation of financial short-term goals) and the
‘real’ organization represents the conditions of stewardship theory (i.e., shared interests and
responsibilities, complex and long-term oriented mission and goals, many of which are non-
financial). In this extreme situation, the adoption of directive controls may significantly alter
the organizational social ‘fabric’ and relationships. The choice of directive controls will send
the message that people are not to be trusted, which, over time, may turn out to be true
(Cardinaels and Yin, 2015). The increased pressure coupled with the high uncertain

31
environment and feelings of injustice may encourage employees to retaliate and exhibit the
behaviours the system is intending to control. This situation will eventually transform the
social dynamics (Argyris, 1953; Tan and Rae, 2009) and, as suggested by Elias (1998), the
change in social dynamics will increase the likelihood of perverse unintended consequences.
In situations where managers assume high goal-alignment implied from their beliefs
of employees behaving as stewards (i.e., exerting long-term oriented, prosocial behaviours
(Davis et al., 1997b; Hernandez, 2012)) and high goal-uncertainty; we expect to find an
enabling PMS as suggested by stewardship theory. However, when managers’ taken for
granted assumptions do not correspond with the observable or actual reality, social
relationships will be altered leading to unintended undesirable consequences. For instance, in
a university when high autonomy and collegiality (i.e., enabling controls) are applied to parts
of the institution that have not been socialised into the academic profession and its
internalised values and standards, the consequences of this stewardship-related controls may
not be those intended and ‘free riding’ effects may appear.
In situations where managers assume low (high) alignment and high (low)
uncertainty, we expect to find hybrid PMSs. Hybrid systems will combine features of
enabling and directive PMSs with a stronger tendency towards one or other of them
depending on the presumed circumstances. Examples of hybrid systems can be the systems
suggested for third sector organizations (e.g., Lindsay et al., 2014) whose leaders (and
designers of the PMS) believe that employees and volunteers are highly aligned with the
cause but at the same time need to produce and report financial results and predictable
outcomes to demonstrate the value of their work to current and future donors or funders. If
the assumed situation corresponds to the actual situation, we expect that unintended
undesirable consequences will be minimized due to the use of a wide range of controls, which

32
will be more adapted to the needed diversity in social relationships. However, if it does not, a
set of unintended dysfunctional consequences will emerge.
For simplicity, we have so far assumed that reliance on directive controls and reliance
on enabling controls are independent dimensions so that a choice to emphasize one does not
imply a lower emphasis on the other, although there are probably constraints on the total
amount of control used by any single organization (Franco-Santos et al., 2014). Perhaps, a
more realistic position is to argue that how organizations choose to emphasize each of these
two ideal types of systems is an empirical issue that cannot be theorised in detail and in
advance, thus suggesting an important focus for future research.
Discussion
Performance management is an important phenomenon in organizational research and
management practice (Otley and Soin, 2015; Smith and Bititci, 2017). Most earlier research
has been devoted to the prediction of the intended consequences of a PMS, whilst limited
attention has been given to the explanation of its unintended consequences (Franco-Santos et
al., 2012). The aim of this paper is to develop a theory of the unintended undesirable
consequences of PMSs based on a mixed review of the literature.
Based on our mixed review of the literature, we propose that managers design and use
PMSs based on two key assumptions. They make their choices according to their beliefs
about goal-alignment (i.e., how aligned they believe the goals or interests of employees are to
the interests of the organization) and about goal-uncertainty (i.e., how much uncertainty exits
in managers’ predictive models). Based on these assumptions, managers introduce different
combinations of the controls available to them. As Merton (1936, p. 901) asserts,
management controls “predicated upon imaginary conditions must inevitably evoke
unexpected consequences”. The gravity and nature of the unintended undesirable

33
consequences will depend on how and on how much managers’ assumed reality differs from
the ‘real’ state of affairs. In sum, we argue that a false or misguided definition of the control
situation, leads to the selection and use of a set of control mechanisms that, over time,
generate unintended consequences transforming the existing social relationships, which can
ultimately lead to perverse outcomes.
We contribute to the literature on PMSs in various ways. Firstly, we question the
validity of the common agency theory related assumptions underpinning much performance
management research. People are not always opportunistic, the mission of an organization
might not be to maximise its financial performance, and uncertainty is not always
manageable through directive control mechanisms. Our typology suggests that using agency
theory assumptions as the default position when designing and using PMSs may often be
problematic. This has important implications for public sector entities focused on developing
complex missions such as hospitals, schools or universities and known to be guided by
professional values and standards. Further research could be devoted to the improvement of
the design and use of PMSs so critical assumptions in managerial mental models are surfaced
and questioned before key choices are made. Unintended consequences will always exist, but
improved design and use processes may avoid them doing more harm than good.
Secondly, the general idea behind our work connects with previous contingency
research (Chenhall, 2003; Otley, 2016), indicating that appropriate control systems are
unlikely to be universal. They need to be adapted to the circumstances in which they are
being used and to the objectives being sought. PMSs are inherently complex, with different
approaches to control being appropriate in different organizational functions, in varying
organizational environments, and at different hierarchical levels. This assertion is in line with
Ouchi’s (1977, 1980; 1978) ideas on the various configurations of controls according to
context; and with more recent management control (Bedford et al., 2016; Bedford and

34
Malmi, 2015; Malmi and Brown, 2008) and operations research (Smith and Bititci, 2017)
looking at the configuration of performance management packages.
Thirdly, previous research has tended to concentrate on the nature of unintended
outcomes rather than on the reasons underlying their existence. Based on our review, it is
likely that a richer explanation could be given by considering various aspects of the context
within which control is exercised. We would suggest that studies of how control works in
different parts of the same organization (e.g. different geographical sites and locations,
different business function, and different hierarchical levels) would give insights into this
important phenomenon. Given the complexity of both PMSs and organizational contexts, it is
suggested that studies which control for as much of this potential variation as possible (e.g.
by using different parts of the same organization, or analysing difference within the same unit
at different hierarchical levels) would likely give the most useful initial results. We hope that
the framework put forward here will assist in this endeavour.
Our work is not free from limitations. The purpose of this paper was to develop new
theory, but it was limited by being able to examine only a fraction of the considerable amount
of literature that exists across a range of disciplines. In our search for insights and potential
connections, we combined a traditional review of the conceptual performance management
literature and the sociological research on unintended consequences with a systematic review
focused on the identification of the most salient unintended undesirable consequences of
PMSs. This mixed review approach means that there may be parts of our work that may not
be reproducible, which is a common situation in theory building undertakings (Weick, 1989).
We restricted our review to research published in English using a narrow set of keywords due
to the volume of knowledge we were attempting to review. Based on this choice, relevant
research that may have been found using alternative concepts to the ones used (e.g., side
effects, performance appraisals) may have been omitted. Additionally, we took the decision

35
to limit the scope of our review to research focused on overall corporate PMSs, which means
that some unintended consequences of systems dedicated to the management of particular
functions or processes (e.g., supply chain, marketing, new product development or
production) have not been considered.
Going forward, it seems evident that further PMSs research investigating the
unintended will yield both new contributions to theory and prescriptions for practice. New
research could take the form of qualitative studies focusing on the performance management
practices and the unexpected consequences that can be observed to emerge in organizations
(Ashton, 1976). Real control systems attempt to cope with both complexity and uncertainty,
and to integrate a variety of control mechanisms. This implies that case studies of
performance management practices that attempt to take a holistic approach (i.e. attempting to
include all control mechanisms being used) will be an important way forward. The theory
outlined in this paper is intended to help provide a useful encompassing framework that will
assist such program of work, but it needs to be supplemented by empirical work that pays
attention to the idiosyncrasies of the situations in which different organizations and their
PMSs operate.
Conclusion
This research is intended to help explain how, why and when PMSs lead to unintended
consequences some of which can be perverse. All PMSs will have both intended and
unintended consequences. We found that directive systems are likely to produce gaming,
information manipulation, selective attention, illusion of control, and transform social
relationships. Most previous research has focused on the intended effects of these systems but
both types of consequences require researching and theorising as they both vary in how
beneficial they are for an organization. Borrowing from the sociology literature on the

36
unintended and well-known theories underpinning performance management processes
(contingency, stewardship and agency theories) our general conclusion is that unintended
undesirable consequences are likely to occur when a system is designed or used in
circumstances that differ from those originally assumed. At the extreme, we suggest that the
misrepresentation of a stewardship-type context as an agency-type context or vice-versa will
lead to unintended undesirable consequences and this situation can make the resultant PMS
perverse. Our main argument is that incorrect assumptions will lead to the design of ‘unfit’
control mechanisms, which will alter the existing social relationships influencing individual
behaviour and ultimately lead to undesirable consequences or ‘collateral damage’ that may
outweigh the intended outcomes.
Given that goal-alignment and goal-uncertainty conditions are likely to differ from
organization to organization, within different parts of a single organization, and over time, the
outcomes associated with the use of a particular PMS are also likely to differ. Studies on the
reasons for such difference in outcomes, rather than reports of their occurrence, are relatively
rare. Most research to date has concentrated on improving PMSs by examining how and why
intended consequences occur. Our work suggests that we need to pay similar attention to the
unintended undesirable consequences of PMSs, so the risks can be reduced. At present, due
to the complexity of most organizations, we speculate that the safest approach for
performance management designers might be one where balance is maintained between the
use of both directive and enabling systems. It is hoped that the framework proposed in this
paper will provide a stimulus and some guidance for conducting further research in this area.

37
References
Adler, P. S. and Borys, B. (1996). Two Types of Bureaucracy: Enabling and Coercive,
Administrative Science Quarterly, 41, pp. 61–89.
Agyemang, G. and Broadbent, J. (2015). Management control and research management in
universities: An empirical and conceptual exploration, Accounting, Auditing &
Accountability Journal, 28, pp. 1018–1046.
Ahn, H. (2001). Applying the balanced scorecard concept: An experience report, Long Range
Planning, 34, pp. 441–461.
Argyris, C. (1953). Human problems with budgets, Harvard Business Review, 31, pp. 97–
110.
Ashton, R. H. (1976). Deviation-amplifying feedback and unintended consequences of
management accounting systems, Accounting, Organizations & Society, 1, pp. 289–300.
Baert, P. (1991). Unintended consequences: A typology and examples, International
Sociology, 6, pp. 201–210.
Baiman, S. (1982). Agency research in managerial accounting: a survey, Journal of
Accounting Literature, 1, pp. 154–213.
Bedford, D. S. and Malmi, T. (2015). Configurations of control: An exploratory analysis,
Management Accounting Research, 27, pp. 2–26.
Bedford, D. S., Malmi, T. and Sandelin, M. (2016). Management control effectiveness and
strategy: An empirical analysis of packages and systems, Accounting, Organizations &
Society, 51, pp. 12–28.
Berliner, J. S. (1956). A Problem in Soviet Business Administration., Administrative Science
Quarterly, 1, pp. 86–101.
Berry, A. J., Coad, A. F., Harris, E. P., Otley, D. T. and Stringer, C. (2009). Emerging themes
in management control: A review of recent literature, British Accounting Review, 41, pp. 2–
20.
Bevan, G. and Hood, C. (2006). Articles What’s Measured Is What Matters: Targets and
Gaming in the English Public Health Care System, Public Administration, 84, pp. 517–538.
Bonner, S. E. and Sprinkle, G. B. (2002). The effects of monetary incentives on effort and
task performance, Accounting, Organizations and Society, 27, pp. 303–345.
Bourdieu, P. (1977). Outline of a Theory of Practice. Cambridge, UK.: Cambridge University
Press.
Bourgeois, L. J. (1979). Toward A Method Of Middle-Range Theorizing., Academy of
Management Review, 4, pp. 443–447.
Briner, R. B. and Denyer, D. (2012). Systematic review and evidence synthesis as a practice
and scholarship tool. In Rousseau, D. (ed),The Oxford Handbook of Evidence-based
Management. Oxford, UK: Oxford University Press, pp. 112–129.
Broadbent, J. and Laughlin, R. (2009). Performance management systems: A conceptual
model, Management Accounting Research, 20, pp. 283–295.
Cardinaels, E. and Yin, H. (2015). Think Twice Before Going for Incentives: Social Norms
and the Principal’s Decision on Compensation Contracts, Journal of Accounting Research,
53, pp. 985–1015.
Cardinal, L. B., Sitkin, S. B. and Long, C. P. (2004). Balancing and rebalancing in the
creation and evolution of organizational control, Organization Science, 15, pp. 411–
431+495.
Carmichael, D. R. (1970). Behavioral hypotheses of internal control, The Accounting Review,
45, pp. 235–245.
Chenhall, R. H. (2003). Management control systems design within its organizational

38
context: findings from contingency-based research and directions for the future, Accounting,
Organizations & Society, 28, pp. 127–168.
Chenhall, R. H. and Euske, K. J. (2007). The role of management control systems in planned
organizational change: An analysis of two organizations, Accounting, Organizations &
Society, 32, pp. 601–637.
Chenhall, R. H. and Morris, D. (1995). Organic decision and communication processes and
management accounting systems in entrepreneurial and conservative business organizations,
Omega, 23, pp. 485–497.
Chow, C., Cooper, J. . and Waller, W. S. (1988). Participative Budgeting: Effects of a Truth
Inducing Pay Scheme and Information Asymmetry on Slack and Performance, The
Accounting Review.
Chow, C. W., Kato, Y. and Merchant, K. a. (1996). The use of organizational controls and
their effects on data manipulation and management myopia: A Japan vs U.S. comparison,
Accounting, Organizations and Society, 21, pp. 175–192.
Chwastiak, M. (2006). Rationality, performance measures and representations of reality:
Planning, programming and budgeting and the Vietnam war, Critical Perspectives on
Accounting, 17, pp. 29–55.
Conrad, L. and Guven Uslu, P. (2012). UK health sector performance management: Conflict,
crisis and unintended consequences, Accounting Forum, 36, pp. 231–250.
Cornelissen, J. (2017). Editor’s Comments: Developing propositions, a process model, or a
typology? Addressing the challenges of writing theory without a boilerplate, Academy of
Management Review, 42, pp. 1–9.
Cox, A. (2005). The outcomes of variable pay systems: tales of multiple costs and unforeseen
consequences, The International Journal of Human Resource Management, 16, pp. 1475–
1497.
Crabtree, A. D. and DeBusk, G. K. (2008). The effects of adopting the Balanced Scorecard
on shareholder returns, Advances in Accounting, 24, pp. 8–15.
Cruz, I., Scapens, R. W. and Major, M. (2011). The localisation of a global management
control system, Accounting, Organizations & Society, 36, pp. 412–427.
Cugueró-Escofet, N. and Rosanas, J. (2017). The Ethics of Metrics: Overcoming the
Dysfunctional Effects of Performance Measurements Through Justice., Journal of Business
Ethics, 140, pp. 615–631.
Cugueró-Escofet, N. and Rosanas, J. M. (2016). The Ethics of Metrics: Overcoming the
Dysfunctional Effects of Performance Measurements Through Justice, Journal of Business
Ethics, 140, pp. 615–631.
Dahler-Larsen, P. (2014). Constitutive effects of performance indicators: Getting beyond
unintended consequences, Public Management Review, 16, pp. 969–986.
Davis, J. H., Schoorman, F. D. and Donaldson, L. (1997a). Toward a Stewardship Theory of
Management, Academy of Management Review, 22, pp. 20–47.
Davis, J., Schoorman, F. and Donaldson, L. (1997b). Toward a Stewardship Theory of
Management, Academy of Management Review, 22, pp. 20–47.
Davis, S. and Albright, T. (2004). An investigation of the effect of Balanced Scorecard
implementation on financial performance, Management Accounting Research, 15, pp. 135–
153.
de Zwart, F. (2015). Unintended but not unanticipated consequences, Theory and Society, 44,
pp. 283–297.
Delbridge, R. and Fiss, P. C. (2013). Editors’ Comments: Styles of theorizing and the social
organization of knowledge, Academy of Management Review, 38, pp. 325–331.
Donaldson, L. (2001). The contingency theory of organizations. Thousand Oaks, CA: Sage
Publications Ltd.

39
Dossi, A. and Patelli, L. (2010). You learn from what you measure: Financial and non-
financial performance measures in multinational companies, Long Range Planning, 43, pp.
498–526.
Doty, D. H. and Glick, W. H. (1994). Typologies as a unique form of theory building:
Toward improved understanding and modeling, Academy of Management Review, 19, pp.
230–251.
Drazin, R. and Van de Ven, A. (1985). Alternative Forms of Fit in Contigency Theory,
Administrative Science Quarterly, 30, pp. 514–539.
Eisenhardt, K. M. (1989). Agency theory: an assessment and review, Academy of
Management Journal, 14, pp. 57–74.
Elias, N. (1997). Towards a theory of social processes: a translation, British Journal of
Sociology, 48, pp. 355–383.
Elias, N. (1998). The Norbert Elias Reader, Goudsblom, J., Mennell, S., Jephcott, E.,
Krieken, R. van, Goudsblom, J., and Mennel, S. (Trans) (eds),. Malden: Blackwell
Publishing.
Evans, J. H., Hanna, R. L., Krishnan, R. and Moser, D. V. (2001). Honesty in Managerial
Reporting, The Accounting Review, 76, pp. 537–559.
Ferreira, A. and Otley, D. T. (2009). The design and use of performance management
systems: An extended framework for analysis, Management Accounting Research, 20, pp.
263–282.
Flamholtz, E. G., Das, T. K. K. and Tsui, A. S. (1985). Toward an integrative framework of
organizational control, Accounting, Organizations & Society, 10, pp. 35–50.
Foucault, M. (1977). Discipline and punish: the birth of the prison. London, UK.: Penguin.
Francis, R. (2013). Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry.
London, UK: The Stationery Office.
Franco-Santos, M. and Doherty, N. (2017). Performance management and well-being: a close
look at the changing nature of the UK higher education workplace, International Journal of
Human Resource Management, 28, pp. 2319–2350.
Franco-Santos, M., Lucianetti, L. and Bourne, M. (2012). Contemporary performance
measurement systems: A review of their consequences and a framework for research,
Management Accounting Research, 23, pp. 79–119.
Franco-Santos, M., Nalick, M., Rivera-Torres, P. and Gomez-Mejia, L. (2017). Governance
and Well-being in Academia: Negative Consequences of Applying an Agency Theory Logic
in Higher Education, British Journal of Management, 28, pp. 711–730.
Franco-Santos, M., Rivera, P. and Bourne, M. (2014). Performance management in UK
Higher Education institutions. Leadership Foundation for Higher Education.
Frey, B. S., Homberg, F. and Osterloh, M. (2013). Organizational Control Systems and Pay-
for-Performance in the Public Service, Organization Studies, 34, pp. 949–972.
Godener, A. and Eric, K. (2004). Use and impact of performance measurement results in R &
D and NPD : an exploratory study, R&D Management, 2, pp. 191–219.
Gough, D. (2015). Qualitative and mixed methods in systematic reviews, Systematic Reviews,
4.
Gough, D., Thomas, J. and Oliver, S. (2012). Clarifying differences between review designs
and methods, Systematic Reviews, 1.
Hall, M. (2011). Do comprehensive performance measurement systems help or hinder
managers’ mental model development?, Management Accounting Research, 22, pp. 68–83.
Hannan, R., Rankin, F. and Towry, K. (2006). The effect of information systems on honesty
in managerial reporting: A behavioral perspective, Contemporary Accounting Research, 23,
pp. 885–918.
Hansen, S. C., Otley, D. T. and Van der Stede, W. A. (2003). Practice Developments in

40
Budgeting: An Overview and Research Perspective, Journal of Management Accounting
Research, 15, pp. 95–116.
Harris, J. and Bromiley, P. (2007). Incentives to Cheat: The Influence of Executive
Compensation and Firm Performance on Financial Misrepresentation., Organization
Science, 18, pp. 350–367.
Healy, P. (1985). The effect of bonus schemes on accounting decisions, Journal of
Accounting and Economics.
Hernandez, M. (2012). Toward an understanding of the psychology of stewardship, Academy
of Management Review, 37, pp. 172–193.
Hood, C. (2006). Gaming in targetworld: The targets approach to managing british public
services, Public Administration Review, 66, pp. 515–521.
Hopwood, A. G. (1972). An empirical study of the role of accounting data in performance
evaluation, Journal of Accounting Research, 10, pp. 156–182.
Ittner, C. D., Larcker, D. F. and Randall, T. (2003). Performance implications of strategic
performance measurement in financial services firms, Accounting, Organizations and
Society, 28, pp. 715–741.
Jensen, M. C. (2003). Paying People to Lie: The Truth About the Budgeting Process,
European Financial Management, 9, pp. 379–406.
Jensen, M. C. and Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency
costs and ownership structure, Journal of Financial Economics, 3, pp. 305–360.
Kalgin, A. (2016). Implementation of Performance Management in Regional Government in
Russia: Evidence of Data Manipulation, Public Management Review, 18, pp. 110–138.
Kerpershoek, E., Groenleer, M. and de Bruijn, H. (2016). Unintended responses to
performance management in dutch hospital care: Bringing together the managerial and
professional perspectives, Public Management Review, 18, pp. 417–436.
Kreutzer, M., Cardinal, L. B., Walter, J. and Lechner, C. (2016). Formal and Informal
Control as Complement or Substitute? The Role of the Task Environment, Strategy Science,
1, pp. 235–255.
Kuhlmann, S., Bogumil, J. and Grohs, S. (2008). Evaluating Administrative Modernization in
German Local Governments: Success or Failure of the ‘New Steering Model’?, Public
Administration Review, September, p. 851–863.
Li, J. (2015). The paradox of performance regimes: Strategic responses to target regimes in
Chinese local government, Public Administration, 93, pp. 1152–1167.
Lindsay, C., Osborne, S. P. and Bond, S. (2014). The ‘new public governance’ and
employability services in an era of crisis: Challenges for third sector organizations in
Scotland, Public Administration, 92, pp. 192–207.
Lowe, E. A. and Shaw, R. W. (1968). An Analysis of Managerial Biasing: Evidence From a
Company’S Budgeting Process, Journal of Management Studies, 5, pp. 304–315.
Maas, V. S. and Van Rinsum, M. (2013). How Control System Design Influences
Performance Misreporting, Journal of Accounting Research, 51, pp. 1159–1186.
Mackay, B. R. and Chia, R. (2013). Choice, chance and unintended consequences in strategic
change: A process understanding of the rise and fall of Northco Automotive, Academy of
Management Journal, 56, pp. 208–230.
Malmi, T. and Brown, D. A. (2008). Management control systems as a package-
Opportunities, challenges and research directions, Management Accounting Research, 19,
pp. 287–300.
Mannion, R. and Braithwaite, J. (2012). Unintended consequences of performance
measurement in healthcare: 20 salutary lessons from the English National Health Service.,
Internal Medicine Journal, 42, pp. 569–74.
McAdam, R. and Bailie, B. (2002). Business performance measures and alignment impact on

41
strategy: The role of business improvement models, International Journal of Operations &
Production Management, 22, pp. 972–996.
McCann, L., Granter, E., Hassard, J. and Hyde, P. (2015). ‘You Can’t Do Both-Something
Will Give’: Limitations of the Targets Culture in Managing UK Health Care Workforces.,
Human Resource Management, 54, pp. 773–791.
McGregor, D. (1960). The Human Side of Enterprise. New York, NY.: McGraw Hill.
Mckinley, W. and Scherer, A. G. (2000). Some unanticipated consequences of organizational
restructuring, Academy of Management Review, 25, pp. 735–752.
Melnyk, S. A., Bititci, U., Platts, K., Tobias, J. and Andersen, B. (2014). Is performance
measurement and management fit for the future?, Management Accounting Research, 25,
pp. 173–186.
Mennell, S. (1977). ‘Individual’ action and its ‘social’ consequences in the work of Nobert
Elias. InHuman Figurations. Amsterdams Sociologisch Tijdschrift, pp. 99–109.
Merchant, K. A. (1990). The effects of financial controls on data manipulation and
management myopia, Accounting, Organizations and Society, 15, pp. 297–313.
Merton, R. K. (1936). The Unanticipated Consequences of Purposive Social Action,
American Sociological Review, 1, pp. 894–904.
Merton, R. K. (1968). Social Theory and Social Structure. New York, NY.: Free Press.
Merton, R. K. (1996). On social structure and science. Chicago: University of Chicago Press.
Mica, A., Peisert, A. and Winczorek, J. (2011). Sociology and the unintended: Robert Merton
Revisited. Peter Lang GmbH, Internationaler Verlag der Wissenschaften.
Miles, R. E. and Snow, C. C. (1978). Organization strategy, structure, and process. New
York, NY.: McGraw-Hill.
Murphy, W. H. (2004). In pursuit of short-term goals: anticipating the unintended
consequences of using special incentives to motivate the sales force., Journal of Business
Research, 57, pp. 1265–1275.
Nilakant, V. and Rao, H. (1994). Agency Theory and Uncertainty in Organizations: An
Evaluation, Organization Studies, 15, pp. 649–672.
Otley, D. (1999). Performance management: a framework for management control systems
research, Management Accounting Research, 10, pp. 363–382.
Otley, D. (2016). The contingency theory of management accounting and control,
Management Accounting Research, 31, pp. 45–62.
Otley, D., Broadbent, J. and Berry, A. (1995). Research in Management Control: An
Overview of its Development., British Journal of Management, 6, pp. 31–44.
Otley, D. T. (1978). Budget use and managerial performance, Journal of Accounting
Research, 16, pp. 122–149.
Otley, D. T. (1980). The contingency theory of management accounting: Achievement and
prognosis, Accounting, Organizations and Society, 5, pp. 413–428.
Otley, D. T. and Berry, A. J. (1980). Control, organisation and accounting, Accounting,
Organizations and Society, 5, pp. 231–244.
Otley, D. T. and Soin, K. (2015). Management control and uncertainty. Basingstoke:
Palgrave Macmillan.
Ouchi, W. G. (1977). The relationship between organizational structure and organizational
control, Administrative Science Quarterly, 22, pp. 95–113.
Ouchi, W. G. (1979). A conceptual framework for the design of organizational control
mechanisms, Management Science, 25, pp. 833–848.
Ouchi, W. G. (1980). Markets, bureaucracies, and clans, Administrative Science Quarterly,
25, pp. 129–141.
Ouchi, W. G. and Johnson, J. B. (1978). Types of Organizational Control and Their
Relationship to Emotional Well Being., Administrative Science Quarterly, 23, pp. 293–317.

42
Parkinson, J. (2012). Managing Police Performance in England and Wales: Intended and
Unintended Consequences. University of Cambridge.
Peattie, S., Peattie, K. and Newcombe, R. (2016). Unintended consequences in demarketing
antisocial behaviour: project Bernie., Journal of Marketing Management, 32, pp. 1588–
1618.
Perri, 6 (2014). Explaining unintended and unexpected consequences of policy decisions:
Comparing three British Governments 1959-74, Public Administration, 92, pp. 673–691.
Portes, A. (2000). The hidden abode: sociology as analysis of the unexpected, American
Sociological Review, 65, pp. 1–18.
Rankin, F., Schwartz, S. and Young, R. (2008). The effect of honesty and superior authority
on budget proposals, The Accounting Review, 83, pp. 1083–1099.
Ridgway, V. F. (1956). Dysfunctional consequences of performance measurements,
Administrative Science Quarterly, 1, pp. 240–247.
Ridley, D. (2012). The literature review: A step-by-step guide for students. London, UK.:
Sage Publications Ltd.
Salterio, S. E. and Webb, A. (2006). Honesty in accounting and control: A discussion of ‘the
effect of information systems on honesty in managerial reporting: A behavioural
perspective’, Contemporary Accounting Research, 23, pp. 919–932.
Schiff, M. and Lewin, A. Y. (1970). The impact of people on budgets, Accounting Review,
45, pp. 259.
Schwepker, C. H. and Good, D. J. (2012). Sales Quotas: Unintended Consequences on Trust
in Organization, Customer-Oriented Selling, and Sales Performance, Journal of Marketing
Theory & Practice, 20, pp. 437–452.
Segal, L. and Lehrer, M. (2012). The Institutionalization of Stewardship: Theory,
Propositions, and Insights from Change in the Edmonton Public Schools, Organization
Studies, 33, pp. 169–201.
Shepherd, D. A. and Suddaby, R. (2017). Theory building: A review and integration, Journal
of Management, 43, pp. 59–86.
Smith, A. (1759). The theory of moral sentiments. printed for A. Millar; and A. Kincaid and
J. Bell, in Edinburgh.
Smith, M. and Bititci, U. (2017). Interplay between performance measurement and
management, employee engagement and performance, International Journal of Operations
and Production Management, 37, pp. 1207–1228.
Smith, P. (1995a). On the unintended consequences of publishing performance data in the
public sector, International Journal of Public Administration, 18, pp. 277–310.
Smith, P. (1995b). Performance indicators and outcome in the public sector, Public Money &
Management, 15, pp. 13–16.
Snow, C. C. and Ketchen David J., J. (2014). Typology-Driven Theorizing: A Response to
Delbridge and Fiss., Academy of Management Review, 39, pp. 231–233.
Tan, K. H. and Rae, R. H. (2009). Uncovering the links between regulation and performance
measurement, International Journal of Production Economics, 122, pp. 449–457.
Tranfield, D., Denyer, D. and Smart, P. (2003). Towards a methodology for developing
evidence-informed management knowledge by means of systematic review, British Journal
of Management, 14, pp. 207–222.
Waddington, I. (2016). Theorising unintended consequences of anti-doping policy,
Performance Enhancement & Health, 4, pp. 80–87.
Weick, E. (1989). Theory construction as disciplined imagination, Academy of Management
Review, 14, pp. 516–531.
Weick, K. E. (1976). Educational organizations as loosely coupled systems, Administrative
Science Quarterly, 21, pp. 1–19.

43
Table 1. Literature review mixed methods overview
Research method
Objectives/
review questions
Process
1. ECLECTIC REVIEW
Area:
§ Performance
management
systems
§ Sociology
research on
theorising
unintended
consequences
Sources:
§ Papers, books,
unpublished
papers and
thesis (empirical
and conceptual)
To analyse…
§ The meanings of
core terms
§ Existing shared
research
assumptions
§ Existing theories
explaining
intended and
unintended
consequences
1. Selection of broad research areas: performance management systems,
sociology research on unintended consequences
2. Initial review of articles and books that have sought to describe the
research on performance management systems over the years (Berry
et al., 2009; Broadbent and Laughlin, 2009; Chenhall, 2003; Ferreira
and Otley, 2009; Franco-Santos et al., 2012; Malmi and Brown,
2008; Otley, 1999, 2016; Otley et al., 1995).
3. Conduct additional searches of literature online (EBSCO, Scopus,
google, google scholar), and university and personal libraries.
4. Subjective selection of relevant research
5. Qualitative synthesis addressing review objectives
2. SYSTEMATIC REVIEW
Area:
§ Performance
management
systems research
in management
accounting,
human
resources,
operations,
general
management and
public-sector
management
examining
unintended
undesirable
consequences
Sources:
§ Research papers
both conceptual
and empirical
and cross-
referenced
books
To analyse…
§ What are the
most common
unintended
undesirable
consequences of
performance
management
systems?
§ What is known
about how
and/or why they
occur?
1. Definition of review questions
2. Scoping study: Argyris (1953), Ridway (1956), Smith (1995a),
Jensen (2003), Mannion and Braithwaite (2012), and Segal and
Lehrer (2012).
3. Search strategy:
- EBSCO keywords in any field (21/12/2017): (management control OR
organi?ational control OR performance measure* OR performance
management) AND ((unintended OR dysfunctional OR undesirable OR
unexpected OR negative) AND (consequence* OR outcome* OR result*));
in peer-reviewed & academic journals; in English= 8,752papers
- In ABS 4*&3* publications in general management, operations,
HR, management accounting & public sector: ("Accounting,
Organizations & Society" OR "Management Accounting Research" OR
"Contemporary Accounting Research" OR "The Accounting Review" OR
"Journal of Accounting Research" OR "Academy of Management Journal"
OR "Academy of Management Review" OR "Administrative Science
Quarterly" OR "Journal of Management" OR "British Journal of
Management" OR "Journal of Business Research" OR "Journal of Business
Ethics" OR "Journal of Management Studies" OR "Journal of Business
Ethics" OR "Human Resource Management" OR "Human Resource
Management Journal" OR "Human Resource Management Review" OR
"Human Relations" OR "International Journal of Human Resource
Management" OR "Journal of Operations Management" OR "International
Journal of Operations and Production Management" OR "International
Journal of Production Economics" OR "Organization Science" OR
"Management Science" OR "Public Administration Review" OR "Journal
of Public Administration Research and Theory") = 839 papers
4. Selection of relevant research: (a) Title and abstract reading and
selection (based on exclusion and inclusion criteria) = 41 papers
5. Reading of full papers and selection = 14. Additional papers found
through cross-references and citations = 28. Total selected= 42
6. Data extraction: Date, field of research, context, research method,
control mechanisms investigated, unintended undesirable
consequences for actors and others, insights that explain how and
why unintended consequences occur.
7. Data analysis and qualitative synthesis: Based on our review
questions and adopting a thematic analysis, we created theme
clusters and sub-clusters.

44
Table 2. An overview of the unintended undesirable consequences of performance management systems
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Agyemang,
Broadbent
(2015)
ACC
Public sector
(Higher
Education –
REF process,
UK);
Conceptual/
literature review
Directive:
§ Performance
planning,
measurement, targets.
Enabling
§ Professional and
collegial values and
standards
Directive
§ Movement away from previously
held academic values
§ Resistance to external regulatory
systems originating controls that are
more restrictive than the systems they
are meant to resist
§ Gaming
§ Symbolic violence
§ Dysfunctional
consequences
for academics
§ Due to commensuration reactions and self-fulfilling
prophecy the inner working of organisations change.
These reorientation changes, over time produce other
changes in deeply held values, which ultimately affect
individual behaviour. The gaming that comes with the
reorientation processes ends up generating
dysfunctional consequences for individuals. Symbolic
violence (Bourdieu, 1977) is used to explain the
change process.
Argyris
(1953)*8
GM
For profit
(Manufacturing,
US); Case
studies
Directive:
§ Performance
measurement, targets,
review (budgets)
§ Grouping: employees ‘against’
management
§ Blame culture
§ Social relations problems (e.g.,
tensions and conflicts among
departments and against budget
setters)
§ Internalized pressure of line
managers resulting in overwork,
stress, increase monitoring of
employees, frustration, unhappiness
§ Self-interest and silo-mentality
§ Patterns of leadership (focus on the
numbers rather than people)
§ Illusion of control (implied)
§ No reference
§ Increase pressure leads to dysfunctions
§ Argyris mentions the use of budgets is based on the
believe that people do not want to work unless they
are pushed/controlled which may not always be a
valid assumption.
Ashton
(Ashton,
1976)
ACC
Various
contexts;
Conceptual
Directive:
§ Performance
planning,
§ Interpersonal tensions
§ Felt need of defensibility of
individual action
§ Rigidity of behaviour
§ No reference
§ Controller’s preconceptions about the abilities and
motivations of the people being controlled are
reinforced by the control system. The result is mutual
deception rather than learning. A circle develops in
7
Data field designated using abbreviations: Accounting (ACC), Human resources (HR), General management (GM), Public sector management (PSM), Operations (OP)
8
Studies with * indicate that they were selected for the scoping study.

45
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
measurement, targets
(budgets).
§ Use of trappings of authority
§ Difficulty with clients
§ Bifurcation of interests
§ Conflicts among organizational sub-
units
§ Illusion of control
§ Differences between organizational
goals and achievement
which each’s misconception about the other is
reinforced, potentially leading to perverse
consequences.
§ Time is crucial for understanding overall
consequences (i.e., beneficial short-term outcomes
may turn to be dysfunctional over time due to
deviation amplifying feedback).
§ Interventions can be conducted to change loops from
dysfunctional to functional. These interventions
operate by changing the pattern of relationships.
Relationships, not people, are the critical control
points in organizations.
Berliner
(1956)
GM
Public sector/
State enterprises
(Production,
Russia);
Interviews and
archival data
Directive:
§ Performance
planning,
measurement, targets
(budgets),
performance-based
rewards (individual
incentive pay,
promotions,
sanctions)
§ Gaming (storming) - Although side
negative effects will exist, these are
perceived to be a “price worth
paying” for the benefits obtained in
terms of managerial effort
§ No reference
§ The combination of production planning, measures,
targets and incentives lead to motivational and
behavioural responses from managers (ratchet
principle) which are dysfunctional
§ High complexity and uncertainty of state enterprises
(implied)
§ Agency theory related assumptions
§ Continuous change (in managers)
Bonner &
Sprinkle
(2002)
ACC
Various
contexts;
Literature
review
Directive:
§ Performance-based
rewards (individual
incentive pay
§ Employee well-being (Anxiety,
affect, stress)
§ No reference
§ Unintended consequences due to increased pressure to
perform.
Cardinaels &
Yin
(2015)
ACC
Management
students;
Experiment
Directive:
§ Performance
measurement, targets
(budgets),
Performance-based
rewards (individual
incentive pay vs.
fixed-salary)
§ Employees’ dishonesty and
misreporting
§ Distrust
§ Crowding out pro-social motivations
§ No reference
§ Principals with their choice of controls (incentive pay
vs higher fixed pay) leak information to agents about
what the principals have observed of other agents. The
choice of fixed wage contracts implies that the
principal believes agents have an inherent motive to
be honest (i.e., they are not opportunistic or self-
interested). The choice of incentive contracts sends a
message of distrust as it implies the principal assumes
agents are likely to be opportunistic and, if not
controlled, they will misreport.

46
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
§ Most importantly, the decision to introduce incentive
contracts may lead an employee to believe that other
employees in the organization are behaving
opportunistically, which in turn can affect his/her own
behaviour.
§ A belief suggesting that employees in general are
opportunistic, changes social norms. Over time,
people conform to this norm, increasing their
opportunistic/self-interested behaviour and
misreporting.
Carmichael
(1970)
ACC
Various
contexts;
Conceptual
Directive:
§ Performance
planning,
measurement, targets
(budgets).
§ Measure fixation
§ Gaming
§ Dishonesty
§ Misrepresentation
§ Implied (relationship problems
between management and
employees)
§ No reference
§ The assumptions made about people in organizations
operate as a theory which determines how employees
will be treated. If control designers believe that
employees have inherent mental, moral, and physical
weaknesses so they need internal controls to achieve
organizational goals; this assumption tends to be self-
fulfilling: Employees may retaliate by exhibiting the
behaviour the system was intended to control. Thus, a
perfect “control” system may stimulate violation of
control procedures.
§ Non-compliance tends to appear in the presence of
perceived thread (e.g., application of punitive
measures when targets are not met, pressure for higher
performance, the concept of accountability).
Chow,
Cooper &
Waller
(1988)
ACC
Students;
Experiment
Directive:
§ Performance
measurement, targets,
review (budgets),
performance-based
rewards (individual
incentive pay)
§ Misrepresentation of financial
information or ‘slack’
§ No reference
§ Design issues lead to unintended undesirable
consequences. Better designs reduce these
consequences.

47
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Chow, Kato,
Merchant
(1996)
ACC
Manufacturing
(profit centres,
US and Japan);
Interviews and
survey
Directive:
§ Performance
measurement and
targets, process
controls, directives in
formal meetings,
headcount controls
§ Data manipulation
§ Management myopia (short-termism)
§ No reference
§ Different cultural conditions (values & beliefs)-
Japanese under equally tight controls (although not
the same controls) respond with less dysfunctional
consequences.
Chwastiak
(2006)
ACC
Public sector
(Vietnam war,
US perspective);
Content analysis
of archival
information.
Directive:
§ Performance
planning,
performance
measures and targets,
reviews, performance
related rewards
§ Gaming
§ Misrepresentation
§ Selective attention (visibility of what
can be measured, invisibility of what
cannot be measured)
§ Restricting political change
(numbers, not people, as responsible
for decisions)
§ Illusion of control
§ Perverse
results for
civilians
§ The performance management system (PPB)
introduced a new discourse that equated truth focused
on the things that could be counted. Less tangible
aspects became invisible. Over time, this
‘instrumental rationality’ dismissed as irrational
decision making based on intuition and morality.
Meanings were lost and the focused moved from
human actors to structural and technical defence
issues. This process led to the belief that “every
problem could be managed” which affected leadership
perceptions and behaviours.
Conrad &
Guven Uslu
(2012)
ACC
Public sector
(NHS hospital,
UK); Case study
Directive:
§ Performance
planning,
performance
measures and targets,
performance reviews
§ Gaming
§ Short-termism
§ Increase
mortality
§ Decrease
quality of care
§ Adoption of private-sector like practices in a context
where these practices do not apply (different
rationalities)
§ Lack of control of factors affecting performance
Cox
(2005)
HR
For profit
(Engineering
SMEs, UK);
Case studies
Directive:
§ Performance
evaluation and
performance-based
rewards (individual
incentive pay)
§ Disrupted production
§ Administrative overload and
managerial time costs
§ Damaging social relationships
§ No reference
§ Increase pressure leads to dysfunctions
§ Environmental complexity
Cuguero-
Escofet &
Rosanas
(2017)
GM
For profit
(Banking,
Fashion, High
Tech, Spain);
Case studies
Directive:
§ Performance
measurement, targets,
reviews,
performance-based
rewards (individual
incentive pay)
§ Gaming (metrics “pushing” people in
the wrong direction)
§ Data manipulation
§ Double injustice (ethical dilemmas &
then unfair rewards)
§ Misreporting
§ Risk
transferred to
customers -
Banking case
(implied from
text)
§ Increase pressure leads to dysfunctions
§ Organizational complexity that cannot be reflected in
metrics.

48
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
§ Mission/Strategy drift (revisions
based on what can be counted rather
than what is relevant)
§ Measure fixation (people focused on
the measured and ignored what was
not measured)
§ Illusion of control
Evans,
Hannan,
Krishnan &
Moser
(2001)
ACC
Students
(Management
school, US);
Experiments
Directive:
§ Performance
measurement, targets,
review (budgets),
performance-based
rewards (individual
incentive pay and
profit-sharing)
§ Dishonesty
§ No reference
§ Agency theory assumes that people have a low
threshold for dishonesty; they will lie for little
payoffs. However, this research shows that people are
considerably honest
§ The degree of dishonesty or misrepresentation of
financial information is dependent on the type of
contract used: less sharing of wealth between
organization and employees; more dishonesty. Fixed
salary and profit-sharing prompt more honest
reporting. Performance related pay leads to more
dishonesty.
§ People care about more things that just wealth; they
care about equity and reciprocity.
Franco-
Santos,
Nalick,
Rivera-
Torres, &
Gomez-
Mejia
(2017)
GM
Public sector
(Higher
Education, UK);
Survey
Directive:
§ Performance
planning,
performance
measures and targets,
performance reviews
Enabling (Collegial):
§ Consultation,
communication,
provision of needed
job resources,
comprehensive
performance
recognition, focus on
development,
autonomy
Directive:
§ Decreased well-being (for
academics)
§ Decreased
benefits for
students,
science and
society at large
(not measured)
§ Wrong assumptions about academics’ motivations and
the context of academia leading to directive controls
which affect well-being and, over time, would have
negative effects on motivation, behaviours and social
relationships leading to further unintended
consequences not just for academics but also for
others.
§ Complexity of academic work

49
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Hannan,
Rankin &
Towry
(2006)
ACC
Students
(Management,
US); Experiment
Directive:
§ Performance
measurement, targets,
review (budget),
performance-based
rewards (individual
incentive pay
§ Dishonesty
§ No reference
§ The research assumes managers (agents) are
opportunistic (misaligned goals). They may
misrepresent private information in order to maximize
their own self-interest.
§ Using a measurement/information system influences
the ability of an agent to appear honest. It gives the
principal a signal about the agent’s private
information. By comparing this signal to the
managers’ report, the principal can form an
impression of the agent’s honesty.
§ Misrepresentation of financial information depends on
the design choices of the measurement/information
system
§ The use of an information system (vs the no use)
enhances honesty. A precise system is associated to a
higher level of honesty than a coarse system.
§ An assumption in this research is that the agent cannot
manipulate the measurement/information system.
Harris &
Bromiley
(2007)
GM
For profit sector
(Financial
restatements
with accounting
irregularities,
US)
Directive:
§ Performance
measures, targets,
review, performance
related rewards
(incentive pay)
§ Corporate misconduct: Financial
misrepresentation
§ Mentioned
negative
consequences
for society and
social
progress, but
not measured
§ Building on the behavioural theory of the firm, it is
argued that misrepresentation is caused by the use of
strong incentive compensation and poor financial
performance compared to expectations. This argument
is supported empirically.
Hasen,
Otley, Van
der Stede
(2003)
ACC
For profit and
non-for-profit
sectors;
Literature
review
Directive:
§ Performance
measurement, targets,
review (budgets)
§ Time costs
§ Constrain responsiveness and are
often a barrier to change
§ Strengthen vertical command-and-
control
§ Miss out on emerging network
structures
§ Encourage gaming and perverse
behaviours
§ Reinforce department barriers rather
than encourage knowledge sharing
§ Make people feel undervalued
§ No reference
§ Design issues. Better designs that improve the
budgeting process (activity-based budgeting
approach) or abandon it (beyond budgeting) reduce
unintended undesirable consequences.

50
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Healy
(1985)
ACC
For profit sector
(Multiple
industries, US);
Archival
Directive:
§ Performance-based
rewards (individual
incentive pay)
§ Gaming (earnings management –e.g.,
“taking a bath”)
§ No reference
§ Bonus plans incentivize managers to select accounting
techniques that enhance the value of their bonus
payments. It is implied that this relationship occurs
because managers are self-interested; the use of bonus
plans that use accounting measures does not limit their
opportunisms, it actually enhances it as they can
manipulate these measures.
Hood
(2006)
PSM
Public sector
(Various
agencies, UK);
Literature
review and
interviews
Directive:
§ Performance
measurement, target-
and-terror regime,
performance
benchmarks
§ Gaming or strategic behaviour
(ratchet effect, threshold effect,
output distortion
§ Manipulation of data (cheating)
§ Bulling and blame culture
§ Eagerness of central managers to
accept “good news”
§ Implied illusion of control
§ Decreased
benefits for the
public/society
(implicit in
text)
§ Not explicitly stated explanation. However, at the end
of the article, it is mentioned that most of the
measures taken to decrease gaming were based on a
low-trust approach that aimed to improve the design
of the measurement system (tighter definitions, targets
refinement, audit investigations, penalties, etc.). These
practices reinforced
Hopwood
(1972)
ACC
For profit sector
(Manufacturing
Departments,
US); Survey and
interviews
Directive:
§ Performance
evaluation (of
budgets)
§ Job related tensions
§ Less favourable relationships with
peers, distrust and rivalry, lack of
cooperation
§ Short-term orientation
§ Data manipulation
§ Illusion of control (false measure of
‘cognitive simplicity’)
§ Different approaches to the use of directive
mechanisms (budget constrained style, profit
conscious)
§ Problems with the design of accounting measures
(incompleteness, high complexity - lack of precision
knowledge of the organization’s economic cost
function, sole focus on outcomes, focus on short-term
financial performance, the cost of providing relevant
data, multiple purposes)
§ Hopwood argues that “accounting data do not in of
themselves pose a threat to members of an
organization” (p.174). The severity of unintended
consequences depends on the way in which the data is
used by managers (with the Budget constraints style
creating serious concerns).
§ The budget constraints style assumes the need to
control individuals by evaluating their performance

51
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
rather than interpreting the evaluation process as an
ongoing activity of organizational learning. It
becomes the primary source of control and motivation
becoming a constraint rather than a problem-solving
mechanism. It imposes a “false measure of cognitive
simplicity onto a complex and highly interdependent
series of activities” (p. 175).
Jensen
(2003)*
ACC
For-profit sector
(various
industries, US);
Literature
review
Directive:
§ Performance
measurement and
targets (budgets),
performance-based
rewards (individual
incentive pay)
§ Lying (misrepresentation)
§ Gaming (e.g., sale products to entities
that had not ordered them, move
revenues from one quarter to the
previous quarter by backdating sales
agreements, entered into secret side
agreements with clients, etc.)
§ Destroy value for organizations
§ Illusion of control
§ No reference
§ Relating pay and promotions to performance targets
leads to gaming and lying, which is disastrous for
organizations
§ Budget systems assume that managers should be
rewarded for meeting targets and punish if they don’t.
Because of this, managers will have incentives to set
easy targets, and do whatever it takes to meet them
(even if their actions destroy value for the company).
§ Gaming is inevitable, it is part of business life.
§ Jensen assumes that managers are self-interested.
Once a budget-target reward system is in operation,
managers have no interest in providing accurate
information in their budgets. This leads to critical
information being hidden and chaotic actions leading
to dysfunctionalities. Honesty and integrity are
eroded. “Managing the numbers” is considered a part
of management and these gaming behaviours become
undiscussable.
Kalgin
(2016)
PSM
Public sector
(Various
agencies,
Russia);
Interviews and
archival data
Directive:
§ Performance
planning,
performance
measures, targets,
performance reviews
including ranking
system
§ Misrepresentation
§ Data manipulation (prudent and
reckless)
§ Differences between measured
performance and actual performance
§ Decreased
benefits for the
public/ society
(implicit in
text)
§ The dysfunctionalities of unintended consequences
can be observable and theoretically predicted.
§ Initially, the study draws on agency theory ideas to
explain misrepresentation, suggesting that
misrepresentation occurs due to information
asymmetry and self-interested assumptions. However,
the results suggest that this model does not operate in
the Russian public-sector context.
§ The Russian public-sector context is characterized by:
negativity bias, blame avoidance and attention
aversion. The behaviour of individuals working in this
sector appears to reflect that of the ‘administrative

52
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
man’ model rather than the agency theory’s
‘economic man’ model.
§ Prudent misrepresentation emerged in intelligent
regimes (measurement system without targets and
rankings). This type of behaviour was driven by
individuals’ need for security and convenience (i.e., to
minimize attention). Prudent manipulated data are
inconspicuous to supervising authorities. If no audit is
done, the validity of the data for decision making is
undermined. However, it may be that the ultimate
purpose of the Russian government is to have a
‘meaningless system’ overloaded with data. This
system will reflect Foucault’s (1977) ‘bureaucratic
panopticon’, creating a normalizing effect. The
importance is not the quality of data, but the
conformity and power that the system provides.
§ Reckless manipulation was driven by the pressure
generated by the use of systems including measures,
targets and rankings.
Kerpershoek,
Groenleer &
de Bruijn
(Kerpershoe
k et al.,
2016)
PSM
Public sector
(Health care,
The
Netherlands);
Interviews
Directive:
§ Performance
planning,
performance
measures and targets,
performance reviews,
performance-related
rewards (incentive
pay)
§ Misrepresentation of data (or ‘up-
coding’) leading to misrepresentation
and suboptimal allocation of
resources
§ Strategic selection of expedient cases
(or ‘cherry picking’)
§ Negative
outcomes for
patients’ care
§ The justification for a new performance management
system was driven by an agency theory rationale.
However, the agency theory explanation of
unintended consequences (self-interest, opportunism)
is incomplete. Consideration of both financial and
professional motives is required to understand the
complex nature of this phenomenon.
§ Unintended consequences are related to design issues
(the link of measures to rewards and sanctions; a high
financial impact) and the level of professionalism
(specialized knowledge and competences).
§ Medical professionals assumed to be intrinsically
driven by professional or value-based motivations
(i.e., stewards rather than agents). Their motivation is
based on “making a contribution to their patients and
society”.
§ Managerial rationale in conflict with the professional
rationale: (1) performance management systems are
relatively static, which creates tensions with the

53
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
dynamic nature of professional environments; (2)
these systems are driven by management objectives
(transparency, efficiency, costs), while professional
processes are driven by shared standards, cognitive
beliefs and autonomy. Performance management is
perceived to undermine intrinsic motivation,
especially when incentive pay is used and undermines
professional autonomy.
§ Unintended consequences generate opportunistic
behaviour, but these may not be perverse (i.e., to
satisfy their own interests). This behaviour may be to
benefit patients or institutions
§ Based on the professionalism and cognitive evaluation
theory, unintended consequences can be explained
due to the underlying values that drive medical
professionals (competence and autonomy)
§ A third explanation is that the performance
management system inevitably lags behind the
dynamics of the professional process.
Kuhlmann,
Bogumil &
Grohs
(2008)
PSM
Public sector
(Local
governments,
Germany);
Survey and case
studies
Directive:
§ Performance
measurement, targets,
evaluation,
benchmarks,
performance-related
rewards (incentive
pay)
§ Neglect of the importance of the
relationship between administrators
and politicians
§ Time costs
§ Risk of corruption and personal
advantages
§ Decreased employee morale
§ No reference
§ The complex reality of political decision making
§ A “false theory” about the importance of relationships
and their interwoven nature
Li
(2015)
PSM
Public sector
(Local
government –
Police
department,
China);
Interviews
Directive:
§ Performance
measurement and
targets, performance
reviews/evaluation
§ Performance-based
rewards (implicit in
text)
Strategic behavioural responses
§ Gaming
§ Misrepresentation of data by local
officials,
§ Manipulation of data reported to
upper-level government officials
§ Measure fixation
§ Decreased
benefits for the
public/ society
(implicit in
text)
§ Increase pressure leads to dysfunctions

54
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Lindsay,
Osborne &
Bond
(2014)
PSM
Public sector
(Employability
providers, UK);
Case studies
(interviews &
focus groups)
Directive:
§ Planning (strategy
alignment),
performance
measurement, targets,
performance
reporting/reviews,
performance-based
rewards
§ Mission drift
§ Stifled innovation
§ Measure fixation
§ Increasing ‘contractualism’ (less
relational and more standardized à
less flexibility)
§ Gaming
§ Barriers to co-production of public
services
§ Short-termism crowding out long-
term focus
§ No reference
§ Reasons for unintended consequences associated to
the underlying assumptions of New Public
Management (i.e., agency theory) coupled with the
complexity of governing public services.
§ It suggests that for third sector organizations a better
approach would be a relational model of working
based on trust, open information, practice-sharing and
participation of stakeholders based on their expertise,
rather than contractual models, targets and other
market related mechanisms.
Lowe &
Shaw
(1968)
GM
For profit sector
(Sales, US);
Interviews and
archival data
Directive:
§ Performance
measurement, targets,
review (budgets),
performance-based
rewards (individual
incentive pay)
§ Gaming (slack)
§ Misrepresentation of financial
information
§ Decision making dysfunctionalities
§ No reference
§ Design issues (links to pay). People’s sense of gain (or
loss) associated with measured performance and
reflected in the various rewards they may or may not
receive may explain slack.
§ Influence of company practices and norms
§ Managers’ insecurity
Maas & Van
Rinsum
(2013)
ACC
Management
students;
Experiment
Directive:
§ Performance-based
rewards (individual
incentive pay and
profit-sharing),
performance reporting
§ Dishonesty - Misreporting
§ No reference
§ Affected by the design of the control system
§ Social preferences and social norms
Mannion &
Braithwaite
(2012)*
PSM
Public sector
(Hospitals -
NHS, UK);
Literature
review
Directive:
§ Performance
measurement and
targets, performance
reporting/review,
performance-based
rewards
§ Poor measurement (measurement
fixation, tunnel vision, myopia,
ossification, anachronism,
quantification privileging)
§ Misplaced incentives and sanctions
(complacency, silo-creation,
overcompensation, under-
compensation, insensitivity and
increased inequality)
§ Breach of trust (misrepresentation,
gaming, misinterpretation, bullying,
§ Decreased
benefits for the
public/society
(implicit in
text)
§ Unintended undesirable consequences cannot be
entirely mitigated. They can be minimized if design
issues and the way the measures are used are
improved.

55
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
erosion of trust and reduced staff
morale)
§ Politicisation of performance systems
(political grandstanding and creating
a diversion)
McCann,
Granter,
Hassard, et
al.
(2015)
HR
Public sector
(Hospitals –
NHS, UK);
ethnographic
study
Directive:
§ Performance
measures and targets
§ Targets culture
§ Workforce morale and dysfunctional
employee relations
§ Selective prioritization
§ Gaming
§ Ignoring
§ Working around targets
§ Negative
outcomes for
patients’ care
§ Conflicting logics of efficiency and integrity of care
generate tensions on an everyday basis. The
inclination of staff is to prioritise integrity of care but
this can be problematic due to the pressures of
accountability. Staff responds to the tensions as
“street-level bureaucrats” looking for the best
outcome for patients, which can lead to unintended
consequences (from the point of view of the
performance management system, but not from the
point of view of patients)
§ Over time, tensions may lead to a dysfunctional work
culture focused on ‘making the numbers’ affecting
intrinsic motivations and morale.
Merchant
(1990)
ACC
For profit (Large
Fortune 300,
US); Interviews
and survey
Directive:
§ Performance
measures and targets
(budgets)
§ Manipulation
§ Myopia (short-term orientation)
§ Gaming
§ Discouragement of new ideas
§ Improvements in the short-term
financial results while harming the
company’s long-term interests
§ No reference
§ Increase pressure to meet targets leads to dysfunctions
§ Managers operating in relatively uncertain
environments were significantly more likely to react
to budget pressure by gaming the system
Murphy
(2004)
GM
For profit –
various
industries (Sales,
UK); Survey
Directive:
§ Performance
measurement and
targets
“Problematic behaviours” due to short-
term targets
§ Gaming (extra purchases,
overemphasis on target products,
increased credit risks)
§ Neglect of work aspects that are not
related to short-term targets
§ Reduce helping and sharing
information
§ Exaggerate claims of targeted
offerings
§ Implicit
reference to
negative
consequences
for society due
to unethical
behaviours
§ Increase pressure leads to dysfunctions
§ They are less likely to occur in organizations with
safeguards in place (e.g., corporate reputation, codes,
reinforcing mechanisms encouraging ethical
behaviour) and high commitment/trust environments
§ Some people are more (or less) prone to problematic
behaviours (people with high-status aspiration and
competitiveness may require more controls; highly
educated people, people with high affective
commitment, and those with close relationships to
supervisor may require less control)

56
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Otley
(1978)
ACC
For profit
(Manufacturing
subunits, UK)
Survey and Case
study
Directive:
§ Performance
measurement, targets,
evaluation (budgets)
§ Gaming
§ Misrepresentation
§ No reference
§ Distortion of information can occur even when the
system is technically adequate.
§ The way in which accounting information is used is
conditioned partly on managers’ managerial
philosophy (key shared assumptions) and it may vary
from unit to unit according to the toughness of its
operating environment and its size and profitability.
§ The way in which the system is used by different
manager has a marked effect on the accuracy
of budget estimates.
§ There was collusion between mangers to provide
biased information to higher levels, but to manage
locally recognizing these inaccuracies.
Rankin,
Schwartz &
Young
(2008)
ACC
Management
students;
Experiment
Directive:
§ Performance
measurement, targets,
reporting (budgets)
§ Dishonesty (misrepresentation of
financial information – depending on
the design of control system)
§ No reference
§ Explanations based on agency theory assumptions: “in
the absence of truth-inducing contracts, subordinates
will misrepresent their private information to serve
their own interests, regardless of the effect on the
firm”; however, their level of dishonesty depends on
budget design decisions.
§ One of these decisions refers to authority over
budgets. When subordinates have final authority over
budgets their concerns for fairness and honesty are
higher, when superiors have final authority there is no
significant difference in honesty levels.

57
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
Ridgway
(1956)
GM
Various sectors,
industries (US,
Russia);
Conceptual/
Literature
review
Directive:
§ Performance
measurement and
targets
§ Measure fixation
§ Work distortion (gaming)
§ Neglect of work aspects that can’t be
measured
§ “Storming” at the end of target period
§ Distrust (both on data due to
misrepresentation and on people)
§ Blaming
§ Wasted effort
§ Rejection of complexity that may
affect planned measured results
§ Reduced morale
§ Ratchet effect
§ Adverse influence on overall goal
accomplishment of the organization
No reference
§ Motivational and behavioural aspects affected by the
use of performance measures
Salterio &
Webb
(2006)
ACC
For profit and
non-for-profit
sectors;
Literature
review
Directive:
§ Performance
measurement, targets,
review (budget),
performance-based
rewards (individual
incentive pay)
§ Lying/ dishonesty (misrepresentation
of information)
§ No reference
§ Assumptions about people’s behaviour related to
unintended consequences
Schiff &
Lewin
(1970)
ACC
Various
contexts;
Conceptual
Directive:
§ Performance
measurement and
targets (budgets)
§ Gaming (Slack)
§ No reference
§ Managers satisfaction of personal interests
Schwepker
& Good
(2012)
GM
For profit sector,
(Business-to-
business
salespeople,
US); Survey
Directive:
§ Performance
measures and targets
(focus on perceived
difficulty of target
achievement)
** The research does
not pay attention to
planning, review or
§ Decreased trust in the organization
§ Decreased customer-oriented selling
§ Decreased effectiveness of sales-
oriented behaviour
§ No reference
§ Design issues (increased target difficulty leads to
greater unintended consequences)

58
Authors
Date
Field7
Context &
method
Control mechanisms
investigated
Unintended consequences
Insights that explain how and why unintended
consequences occur
For actors
For others -
external
stakeholders
reward mechanisms but
it assumes they exists.
Segal &
Lehrer
(2012)*
GM
Non-for-profit
(Education,
Canada);
Literature
review and case
study
Directive:
§ Performance
measurement, targets,
review
Enabling:
§ Social controls, peer
monitoring,
partnership culture,
self-monitoring &
reflective thinking,
share leadership
Directive:
§ Mistrust
§ Deviance
§ Alienation
Enabling:
§ Minor corruption
Directive:
§ Low quality
education
§ Serious
corruption
Enabling
§ No reference
§ Assumptions about people’s behaviour (self-fulfilling
prophecy)
§ Increased pressure to perform
§ Organizational complexity
Smith
(1995a)*
PSM
For profit and
non-for-profit
sectors (UK);
Literature
review
Directive:
§ Performance
planning,
performance
measures, targets,
review, performance
related rewards
(monetary and non-
monetary)
§ Tunnel vision
§ Sub-optimization
§ Myopia
§ Measure fixation
§ Misrepresentation
§ Misinterpretation
§ Gaming
§ Ossification
§ Decreased
benefits for the
public/society
(implicit in
text)
§ Lack of congruence between organizational objectives
and measurement system
§ Inability to measure complex phenomena with
precision and accuracy
§ Inability to respond to changing circumstances
Tan & Rae
(2009)
PSM
Various
contexts, but
most research
selected comes
from public
sector; Literature
review
Directive controls:
§ Performance
measurement, targets,
evaluation,
performance-based
rewards (individual
incentive pay)
§ Commodifying services
§ De-professionalization of public
sector workers
§ Social relationships transformation
(emphasis on measured aspects)
§ Values become less important than
measures
§ Time costs
§ Environmental
consequences
due to
measures in
airports
§ Increased pressure
§ High complexity (e.g., airports, health care
organizations) l

59
Table 3. Summary of literature review insights
ECLECTIC
SYSTEMATIC
Literature
Performance management systems
Definition
Set of control mechanisms (planning,
measurement, targets, review, performance
related rewards) used to facilitate the
delivery of organizational goals by
influencing people’s behaviours and
performance
Shared assumptions
§ Interrelated controls
§ In constant evolution
§ Formal and informal
§ Based on predictive mental models
Theories
§ Contingency theory of accounting
(Otley, 1980, 2016)
- ‘Universal approach’ to the design
and use of controls unlikely to be
effective.
- Controls are to be tailored to fit the
organization’s specific circumstances
to provide their intended results.
§ Agency theory (Eisenhardt, 1989;
Jensen and Meckling, 1976)
- Assumptions: agents (employees) =
‘Economic man’ (self-serving or
opportunistic, effort-averse, risk-
averse, extrinsic drive/motivation);
the ultimate objective is
organizational goals maximization
- Problems that need addressing:
interests/goals alignment; uncertainty
(information asymmetry)
- Proposed solution - adoption of
‘directive’ control mechanisms:
monitoring (explicit goals, measures,
targets, evaluation) and incentives
§ Stewardship theory (Davis et al., 1997b;
Hernandez, 2012)
- Assumptions: stewards (employees) =
shared responsibilities and long-term
goals, intrinsic motivation
- Problems that need addressing:
uncertainty (environmental) and
maintenance of alignment (associated
to intrinsic drives/motivation)
- Proposed solution – adoption of
‘enabling’ mechanisms: fostering
high-trust environments through share
leadership, collective responsibility,
intrinsic rewards, continuous
development.
Literature
Unintended consequences
(sociology)
Definition
Effects of purposive action
which are different from those
expected and intended
Shared assumptions
§ There is an intention from
the point of view of the
designer of a performance
management system
§ Unintended consequences
(desirable and undesirable)
are inevitable.
§ Undesirable consequences
can turn to be perverse
(more harm than good)
§ Undesirable consequences
can be minimized, never
eliminated
§ It is important to theorize
undesirable consequences to
minimize them
Theories
§ Merton (1936) explains
unintended consequences as
resulting from the interplay
of action and situation. The
lack of anticipation might be
due five limiting factors:
lack of knowledge, error,
acute concern for short-term
results, values (or ideology),
self-fulfilling prophecy.
These factors are not
mutually exclusive (they can
reinforce each other).
§ Elias (1997, 1998) explains
unintended consequences
Literature
Unintended consequences of
performance management
systems
§ Organisations relying on
directive performance
management systems (i.e.,
agency theory related) in
highly uncertain contexts
are likely to experience
more undesirable
unintended consequences
not just for the actors but
also for their stakeholders
What are the most
salient unintended
undesirable
consequences?
§ Gaming (81% of selected
literature)
§ Information manipulation
(74%)
§ Selective attention (55%)
§ Illusion of control (24%)
§ Alter social relationships
(81%)
Other:
§ Administrative overload
and managerial time costs
(14%); ‘ossification’
(14%); de-
professionalization (10%);
decreased well-being and
morale (10%); stifled
innovation (5%);
unfairness and inequality
(5%).
How and why do
unintended undesirable
consequences occur?
§ High pressure
§ Design and use issues
§ Complexity and
uncertainty
§ Underlying assumptions
(false assumptions about
people, importance of
relationships and context
complexity)

60
Figure 1. Conceptual framework of the unintended consequences of performance
management systems (PMS)
,