Full Terms & Conditions of access and use can be found at
European Accounting Review
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rear20
Cascading Control Changes, Incoherence, and
Dialogue: Insights from a Longitudinal Case Study
Thomas Toldbod & Berend van der Kolk
To cite this article: Thomas Toldbod & Berend van der Kolk (2022) Cascading Control Changes,
Incoherence, and Dialogue: Insights from a Longitudinal Case Study, European Accounting
Review, 31:2, 377-407, DOI: 10.1080/09638180.2020.1813185
To link to this article: https://doi.org/10.1080/09638180.2020.1813185
Published online: 22 Sep 2020.
Submit your article to this journal
Article views: 399
View related articles
View Crossmark data
Citing articles: 1 View citing articles
European Accounting Review, 2022
Vol. 31, No. 2, 377–407, https://doi.org/10.1080/09638180.2020.1813185
Cascading Control Changes, Incoherence,
and Dialogue: Insights from a Longitudinal
THOMAS TOLDBOD ∗and BEREND VAN DER KOLK ∗∗
∗Aalborg University Business School, Aalborg, Denmark; ∗∗Department of Accounting and Management Control, IE
Business School, IE University, Madrid, Spain
(Received: March 2019; accepted: August 2020)
Abstract Environmental shifts regularly urge organizations to adapt, which may entail management
control (MC) changes. Changes to an MC element such as a performance measurement system, however,
may in turn create incoherence with other, non-changed elements, generating a need for more changes, and
thus trigger a cascade effect. To date, however, we know little about how this sequential process unfolds
and what managers can do to deal with incoherence. This paper contributes by enhancing the understanding
of sequential changes, drawing on the organizational ecology literature, and we empirically inform our
research with a ﬁve-year longitudinal case study. Our data illustrates in detail how initial MC changes,
intended to cope with an environmental shift, trigger a cascade effect. This sequential process results in an
extensive change period, during which various incoherent MC elements coexist. Our study acknowledges
that incoherence among MC elements can decrease control effectiveness by creating intra-organizational
frictions, yet we highlight the role of managers in mitigating such negative effects. Speciﬁcally, we show
how managers can alleviate the unfavorable effects of incoherence by changing their use of performance
measures in order to better facilitate organizational dialogue, learning and problem solving.
Keywords: Change; Management Control; Organizational Ecology; Performance Measurement
Organizations are regularly confronted with environmental changes, for instance related to
demand, innovation and regulation. To thrive, they continuously need to adapt to the new cir-
cumstances and consequently they rarely reach a steady state, as they keep evolving from one
state to another (Harrington et al., 2011). Changes in operational and strategic objectives likely
involve or affect the management control (MC) elements used, because they also have to ﬁt the
new state in order to be effective (Marginson, 2002). Therefore, the conﬁguration of MC ele-
ments used by the organization is frequently subject to change. The MC change literature has
argued that change is only successful if an MC element ﬁts the existing structure of the organiza-
tion (Burns & Vaivio, 2001; Scapens, 1994), if the change process is skillfully managed (Jansen,
*Correspondence Address: Thomas Toldbod, Aalborg University Business School, Aalborg, Denmark. Email:
Paper accepted by Teemu Malmi
© 2020 European Accounting Association
378 T. Toldbod and B. Van der kolk
2011;Malmi,1997), and if the change process in general overcomes a set of barriers embed-
ded in the actual change process (Kasurinen, 2002). Research on MC change often focuses on
the change process of the design1of one MC element, such as the performance measurement
system (Busco et al., 2007; Jansen, 2011; Kasurinen, 2002), the cost accounting system (Ashraf
& Uddin, 2015; Granlund, 2001;Malmi,1997) or a budget (Collier, 2001). Recent research,
however, points to the interplay among MC elements, highlighting that changes to the design of
one MC element may inﬂuence the effectiveness and presence of other MC elements and may
subsequently require MC redesign (Friis et al., 2015; Henttu-aho & Järvinen, 2013; Østergren
& Stensaker, 2011). Despite this, whether, when and how this interplay among MC elements
takes place is an under researched issue in the literature. Scholars have therefore called for more
holistic approaches that take multiple MC elements (as a system or package) and changes into
account, preferably at different moments in time (Grabner & Moers, 2013;Malmi,2013;Malmi
&Brown,2008; Sandelin, 2008; van der Kolk, 2019). Our study addresses such calls by adopting
a process perspective (cf. Langley et al., 2013) that allows us to examine cascading changes of
MC elements in a longitudinal case study. We theoretically underpin our study with the organiza-
tional ecology literature and we particularly focus on the interplay between MC elements in the
years following a signiﬁcant change to a core MC element. In addition, we show how managers
can cope with these changes in effective ways.
Although coherence in its purest form is not likely to be found in empirical settings, various
MC studies implicitly or explicitly suggest that organizations should strive for coherence of MC
elements in order to be effective (cf. Bedford et al., 2016; Friis et al., 2015; Fullerton et al., 2013;
Kristensen & Israelsen, 2014; Widener, 2007). The MC change literature adds that if a change has
revolutionary characteristics, that change is probably not coherent with the existing MC practices
and structure of the organization (Scapens, 1994). This poses a challenge for organizations that
are frequently confronted with change and still want to exercise effective control. To address this
conundrum, we mobilize the organizational ecology literature (cf. Hannan & Freeman, 1984;
Hannan et al., 2003a), which acknowledges that engagement with revolutionary types of change
complicates the aspiration for coherence.
The organizational ecology literature poses that change often starts with an external and/or
internal crisis, which triggers an initial organizational change as a ﬁrst response to deal with the
new situation. Subsequently, other changes may be required to align with the changed objectives,
leaving a period during which organizations employ incoherent MC elements. Organizational
ecology states that incoherence may be more present in situations of high cultural asperity (i.e.
strong differences between the existing and the new MC elements) and structural opacity (i.e.
the difﬁculty to predict a change’s consequences), which in turn require purposeful managerial
interventions. What such managerial interventions mean, in terms of the use of MC elements,
and whether and how such interventions ‘ﬂow back’ to affect the design of MC elements, has,
however, received scant attention in prior MC literature (Burns & Vaivio, 2001; Hall, 2016).
This paper addresses these issues and studies how changes related to MC design can cause inco-
herence, and how managers can use this situation to create organizational dialogue through a
different use of MC elements. We empirically inform our analysis with a ﬁve-year longitudinal
case study of a multinational company. The case reveals how the formalization and centralization
of a planning process (via two new MC elements: an integrated planning system and a supply
1In this study, we distinguish between design and use of MC elements, in line with prior MC literature (Merchant &
Van der Stede, 2007; Naranjo-Gil & Hartmann, 2007; Tessier & Otley, 2012). In general, design refers to the content
(e.g. what is measured) and the speciﬁc qualities of the MC element (e.g. what is the target, how many performance
indicators are included) as they are developed. Use refers to how the MC element is actually used by managers (e.g. how
do managers use the MC element to exercise control).
Cascading Control Changes, Incoherence, and Dialogue 379
chain council) created incoherence and triggered further changes to the existing MC elements.
In particular, we discuss the incoherence of the new MC elements with the extant performance
measurement system. We show how managers ﬁrst tried to mitigate the unfavorable effects of the
incoherence by changing their use of the performance measurement system, and later (formally)
changed the design of that system.
This paper makes two contributions to the literature. First, this paper complements the MC
literature (cf. Friis et al., 2015; Henttu-aho & Järvinen, 2013; Østergren & Stensaker, 2011)by
showing when and whether a speciﬁc MC change will spread and cascade to other MC elements.
In particular, organizational ecology’s (cf. Hannan & Freeman, 1984; Hannan et al., 2003a) rele-
vant analytical terms, such as cultural asperity and structural opacity, enhance our understanding
of interactions among MC elements. Second, we highlight how incoherent MC elements can
yield positive outcomes through constructive managerial actions, which mitigate the negative
effects of incoherence on control effectiveness. MC change does not just require skillful man-
agement of the speciﬁc change, but also managerial interventions outside the intended change
locus. Our study contributes to the MC change literature by showing in a longitudinal case study
how existing MC elements can adjust to new MC elements, extending previous research that
focused on how new MC elements should ﬁt existing ones (Burns & Vaivio, 2001; Scapens,
The following section explores coherence among MC elements and introduces relevant
notions from the MC change literature and the organizational ecology literature. Section three
discusses the research methods and section four presents the case of cascading changes, incoher-
ence and managerial interventions at MULTICORP. In section ﬁve, we reﬂect on the meaning
and signiﬁcance of our ﬁndings and situate our contributions in the extant literature. Section six
includes conclusions, limitations of our study and avenues for further research.
2. Literature Review
2.1. Management Control and Coherence
MC is about motivating employees to behave in line with organizational objectives and is, there-
fore, considered to be a crucial function in any organization (Merchant, 1982; Merchant & Van
der Stede, 2007; Simons, 1995). In order to obtain goal congruency throughout the organization,
managers can design and use a broad range of MC elements, such as values, work rules, and per-
formance measures (Malmi & Brown, 2008). Prior research stated that MC elements should not
be studied ‘in isolation,’ but rather as a package or system, because MC elements may inﬂuence
each other’s effectiveness (Flamholtz et al., 1985; Malmi et al., 2020; Malmi & Brown, 2008;
Merchant & Van der Stede, 2007; Otley, 1980; Simons, 1995,2005; van der Kolk, 2019). The
simultaneous application of various MC elements implies that the control effectiveness of a spe-
ciﬁc MC element can (partly) be determined by another MC elements. Henri (2006), for instance,
showed that interactive and diagnostic uses of MC elements can complement one another to
achieve higher effectiveness. Widener (2007) found that belief systems complement most other
types of MC elements. Because of the potential interplay between MC elements (they might
strengthen, but can also hinder, each other’s effectiveness), the coherence among MC elements
seems important for the way they function (Flamholtz, 1983). Coherence refers to the extent to
which the MC elements form a uniﬁed whole that is designed to achieve similar ends, which
requires ‘alignment and coordination’ (Ferreira & Otley, 2009, p. 275). Prior research stated that
MC systems are coherent, while this is not necessarily the case for MC packages, which may
consist of elements that were ‘designed and implemented by different people, in different parts
380 T. Toldbod and B. Van der kolk
of an organization, at different times’ (Ferreira & Otley, 2009, p. 276; and see also Otley, 1980;
Malmi & Brown, 2008). Coherence among MC elements implies that they compatibly work
together and motivate employees to act in line with the organizational goals.
The control literature has pointed out that different MC elements can be incoherent, i.e. aimed
at achieving different, or even opposing, goals, which may give rise to tensions (Curtis &
Sweeney, 2017; Mundy, 2010; Simons, 1995; van der Kolk et al., 2020). Some tensions may
be persistent, needed or even appreciated in organizations, for instance to facilitate learning and
change, or to deal with conﬂicting stakeholder interests. Van der Kolk et al. (2020, p. 12) suggest
that in such cases managers can play an active and important role in balancing such tensions, for
instance by ‘sometimes emphasizing and sometimes downplaying one of the MC elements when
that ﬁtted the departmental needs.’ We acknowledge that, in this context, perfect coherence or
perfect incoherence are theoretical constructs that are not likely to be found in their pure form
in empirical settings. At the same time, we consider the notions of coherence and incoherence to
be useful here, since control combinations can be closer to coherence or closer to incoherence.
More coherent MC elements, argues the extant MC literature, are helpful in setting out clear
directions for employees. Thus, when one MC element is changed, other elements might also
have to change in order to reduce incoherence among the MC elements.
Change processes have been under the attention of accounting scholars for a long time. Pre-
vious research examined, for instance, the process of replacing an existing cost system with an
activity-based costing system (Malmi, 1997) and the implementation of control elements related
to total quality management (Sharma et al., 2010). Various calls have been made to examine the
changes to MC elements in relation to other MC elements that operate in the organizational set-
ting (Grabner & Moers, 2013;Malmi,2013; Malmi & Brown, 2008; van der Kolk, 2019), while
traditionally, the focus was often on one newly implemented or changing control element. A
more inclusive and holistic approach to MC change receives increasing attention in the literature
and a key challenge for such studies is to theorize empirical ﬁndings and further advance the MC
literature to support better organizational performance (cf. Malmi & Brown, 2008, p. 288). In
the following section, we introduce the organizational ecology literature and explore how it can
theoretically inform the study of the interplay of MC elements.
2.2. Organizational Ecology
The organizational ecology literature is concerned with change. It argues that an organization’s
reluctance to change, or its failure to adapt more rapidly to change than its competitors, can
lead to organizational mortality (Hannan et al., 2003a). An underlying assumption of the ecology
literature is that organizations may become victims of a ‘Darwinistic’ selection process – survival
of the ﬁttest – if they do not respond rapidly to changes. In our study, we are not so much
interested in this organizational mortality aspect, but more in the question what effective MC
change in organizations constitutes, which is also analyzed in detail by this stream of literature.
The organizational ecology literature uses the term architectural element, which is deﬁned as ‘a
set of values on the relevant organizational features (e.g. forms of authority, patterns of control
relations, accounting principles, compensation policies)’ (Hannan et al., 2003a, p. 466). This
deﬁnition closely resembles that of MC elements, described as ‘systems, rules, practices, values
and other activities management put in place in order to direct employee behavior’ (Malmi &
Brown, 2008, p. 290). It seems that the organizational ecology literature and the MC literature
both have an interest in governance and control mechanisms and assume that these are central to
the well-functioning of organizations. As such, the ecology literature has much to offer to the MC
literature that is focused on issues that arise following changes in the design of MC elements. We
will draw on three key notions from the organizational ecology literature: the cascade effect of
Cascading Control Changes, Incoherence, and Dialogue 381
change, cultural asperity and structural opacity. Below, we explore how they can inform ongoing
discussions in the MC literature on interrelations among MC elements.
The organizational ecology literature acknowledges that changes in organizations often come
as a response to internal or external threats or opportunities (Hannan et al., 2003a; Hannan &
Freeman, 1984). Initially, organizations may be reluctant to change, which is coined ‘inertia’
in this literature. An initial response to an internal or external threat or opportunity, however,
reduces inertia and fosters the possibility of more changes. This idea that one change may cause a
chain of new changes (Amburgey et al., 1993) is also called the cascading effect of organizational
change (Hannan et al., 2003a). This means that organizational change seldom appears as an
isolated event, which resonates with the claim by Malmi and Brown (2008, p. 287) that MC
elements should not be studied in isolation.
Not all changes to architectural elements cause a cascading effect. The cascading effect is more
likely to materialize if the initial change relates to a core architectural element than a peripheral
element, and if the element that is changed is strongly connected with other elements (cf. Han-
nan & Freeman, 1984). In addition, a cascading effect is also more likely to occur if the changes
challenge or contradict previous ways of doing things. The organizational ecology literature
uses the term cultural asperity for changes that entail signiﬁcant differences and challenge the
current practices, instead of complementing them.2The reasoning behind this is that an existing
(micro)culture puts limitations on the extent to which a new MC element can be easily embedded.
Hence, cultural asperity emanates if new MC elements break with the status quo, for instance,
when they clash with extant norms or do not align with the organizational identity. A strong
departmental or organizational culture can thus create inertia, which renders changes to the MC
elements more difﬁcult and potentially problematic. To illustrate this with an MC-related exam-
ple, it has been argued that introducing private-sector MC elements such as ﬁnancial incentives
into a public-sector culture can create problems (cf. Frey et al., 2013), because such a practice
signiﬁcantly deviates from the existing norms and values in that (micro)culture. This would be a
situation of high cultural asperity. Contrastingly, in a strong performance culture, the introduction
of ﬁnancial incentives will probably complement the existing culture. This would be a situation
of low cultural asperity (Hannan et al., 2003b). High cultural asperity between an old and a new
element enhances the likelihood that such a change will trigger other changes according to the
organizational ecology literature. High cultural asperity enhances the incoherence between ele-
ments, which may necessitate changes to other elements to reduce the incoherence again. Such
subsequent changes to existing architectural elements do not happen overnight as organizations
need time to adjust, develop and implement elements. Organizations probably ﬁrst evaluate the
(potential) beneﬁts of new architectural elements before changing other architectural elements
in such a way that they are coherent with the new elements (Denis et al., 2001; Hannan et al.,
2003a). Consequently, when new elements are implemented, they probably coexist, at least for
some time, with old elements.
The coexistence of ‘new’ and ‘old’ MC elements can create incoherence, because the objec-
tive of the new MC elements may differ from the objective of the old (Hannan & Freeman,
1984). Incoherence based on diverging or opposing objectives can be difﬁcult to foresee and will
2The notion of cultural asperity partly resembles the idea of ﬁt or misﬁt from contingency theory. Contingency theory’s
notion of ﬁt or misﬁt is often used when examining a relationship between an (internal) MC element and an external
contingency variable at one moment in time (i.e., a snapshot), for example environmental uncertainty. Furthermore,
a contingency approach greatly beneﬁts from particular theories that can provide predictions and/or explanations of a
speciﬁc phenomenon (Chenhall, 2003; Gerdin & Greve, 2004; Otley, 2016). The organizational ecology literature allows
us to understand a purely intra-organizational phenomenon from a dynamic perspective, namely how a change to a
speciﬁc MC element aligns with existing, stable MC elements. This enables a more dynamic and local view compared
to how ﬁt and misﬁt have been operationalized in the contingency literature (cf. Chenhall, 2003).
382 T. Toldbod and B. Van der kolk
probably ﬁrst be noted ex-post, i.e. after the initial change. The organizational ecology literature
introduces the notion of structural opacity, which relates to the lack of ex-ante knowledge about
the possible interplay among new and old elements. To clarify what this means, consider a situ-
ation in which an initial change A triggers or requires another change B. Hannan et al. (2003b)
use the notion of structural opacity to describe situations in which: (1) B cannot be foreseen;
(2) the characteristics of B cannot be deﬁned at the time of the initial change (A); or (3) actors
cannot undertake all these adjustments (A +B) in parallel. A high degree of structural opacity
thus limits the predictability of future events, which means that the organization cannot know
aprioriwhich adjustments will be needed in order to mitigate or reduce incoherence. There-
fore, changes often occur as a sequential search-and-adjustment process instead of as a parallel
process (Hannan et al., 2003b). In the context of MC, this notion may help to explain why in
situations of structural opacity changes to one MC element may trigger sequential changes to
other MC elements, that may span a long period of time. Furthermore, if the initial change is the
response to a crisis, the newly implemented or changed elements are designed only to respond
to that crisis within the speciﬁc organizational entity that is affected. A manager is given a man-
date to make speciﬁc changes to address the crisis, but an assessment of the ﬁt of these changes
with other architectural elements is often not part of the initial change process. If this is the case,
inter-architectural connections are disregarded and, as a consequence, the old and new archi-
tectural elements coexist, causing incoherence (Barnett & Carroll, 1995; Hannan et al., 2003a).
This resonates with observations from the control literature that MC elements should be studied
in the context of the ‘package’ or ‘system’ in which they operate (cf. Malmi & Brown, 2008),
and that managers should have ‘a good understanding of the MC package conﬁguration within
their organization [ ...] before deciding to add new MC elements.’ (van der Kolk, 2019, p. 519)
To summarize, the organizational ecology literature can theoretically inform our investigation
of control changes and (in)coherence, yet it received very limited attention in the MC literature.
Speciﬁcally, we identiﬁed three notions from the organizational ecology literature that are help-
ful to examine MC change, namely the cascading effect (i.e. the idea that one change may cause
a chain of new changes), cultural asperity (i.e. signiﬁcant differences that challenge and/or con-
tradict the current practices) and structural opacity (i.e. the lack of ex ante knowledge about the
possible interplay between ‘old’ and ‘new’ MC elements). In the following section, we discuss
our research methods.
3. Research Methods
We adopt a longitudinal case study approach for two reasons. First, Hannan et al. (2003a) argue
that studies on change in organizations should not be limited to a single change at one moment
in time but rather should examine change in the context of other, related changes. Change is
aprocess that takes time, and ignoring the temporal dimensions can result in an incomplete
or distorted understanding of organizational processes (cf. Langley et al., 2013). Therefore, we
situate the changes of interest within the context of related changes using a longitudinal per-
spective, which allows us to pay particular attention to the dimension of time, addressing prior
calls (Baines & Langﬁeld-Smith, 2003; Beaubien, 2012; Granlund, 2001; Jansen, 2011). Sec-
ond, we conduct a qualitative case study to examine ‘how’ and ‘why’ questions related to MC
change. This approach seems most suitable to scrutinize complex relations among MC elements
‘in order to guarantee data quality’ (Malmi & Brown, 2008, p. 298). Furthermore, a longitudi-
nal case study is ‘the most consistent way of analyzing developments’ (Flick, 2009, p. 138) and
therefore forms a powerful method for examining MC change processes over time.
Cascading Control Changes, Incoherence, and Dialogue 383
3.1. Case Selection
This study was conducted in a global Manufacturing Company (MULTICORP), because we
wanted to study a large organization that was embarking on an MC change. Through conversa-
tions with researchers who were familiar with MULTICORP, we learned that MULTICORP met
these requirements. Furthermore, we wanted to study an organization that strongly relied on MC
elements to direct employee behavior in order for the changes and effects to be more pronounced
and observable. MULTICORP operates in multiple countries and prior research has stated that
this enhances an organization’s reliance on MC elements (Moores & Yuen, 2001), because of the
increased operational and geographical complexity related to operating internationally (Chenhall,
2006; Galbraith, 1973).
To give further depth to our analysis and to focus our empirical work, we studied the part of
the company that is responsible for operations (manufacturing). The focus on a speciﬁc part
of the organization is also in line with our theoretical foundation; changes to MC elements
yield the most signiﬁcant effect within the part of the organization where the change is situated
(Hannan & Freeman, 1984; Hannan et al., 2003a). Our level of analysis is the senior management
level, speciﬁcally the interaction between top management and plant managers. The interviewed
members of the organization (see Appendix A) were all working in this part of the company and
the documents we used and analyzed were either about the company as a whole or about the
operations part of the organization.
In the 15 years before this study, the company experienced exponential growth and more than
tripled its revenue. This growth was rooted in a global expansion; MULTICORP historically
manufactured and sold its products within Europe, but from the beginning of this millennium
onwards, the company started to manufacture and sell products globally. Simultaneously, the
company expanded its product portfolio, which allowed it to sell to a much broader array of
segments. The expansion came from the internal development of new technology and various
investments, for which operations had to be built up from the ground. Furthermore, MULTI-
CORP took part in the market consolidation that characterized the early 2000s, which allowed
them to acquire competitors within their historical core market segments. This helped MUL-
TICORP to acquire new technologies that enabled further expansion. At the time of our study,
MULTICORP operated more than 60 manufacturing units in 18 different countries across all
continents and employed between 18,000 and 20,000 employees worldwide. MULTICORP’s
turnover amounted up to approximately three billion euros, and individual plants were highly
differentiated and took on diverse roles. For instance, some units were highly nested in their
own geographical market and supplied that market with their ﬁnished products. Other organiza-
tional units were worldwide suppliers of a speciﬁc sub-assembly, implying that the speciﬁc unit
was only produced at that speciﬁc plant and would be shipped to other plants that also needed
it in their processes. This variety of activities and attention on plants can be related to MUL-
TICORP’s expansion period and the fact that many plants were acquired by MULTICORP at a
given moment. Consequently, the company evolved into a complex organization with many local
variations but also with global interdependencies.
3.2. Data and Data Analysis
As outlined by Langley et al. (2013) and Kouamé and Langley (2018), process studies follow an
inductive methodology where the connection between data and theory unfolds during the pro-
cess. In some of the ﬁrst interviews in 2012 and 2013, we learned that the initial MC changes had
effects on the coherence among the MC elements and that some changes triggered or demanded
changes to other MC elements. To theorize these observations, we went back to the theory to
384 T. Toldbod and B. Van der kolk
search for terminology and theoretical models that could help us to make sense of the messy
process data (Langley, 1999). We found that the ecology literature could theoretically inform our
observations. Consequently, our theoretical foundation and research question evolved throughout
the research process (cf. Langley et al., 2013).
Initial contact with MULTICORP was established via researchers who had carried out research
projects within the company before. The data collection for our longitudinal research took place
between 2012 and 2017. This allowed us to study the organization before the initial MC change
and during the period in which the organization experienced incoherence. Furthermore, we were
able to discuss the MC changes and the managerial interventions with key informants directly
after these events occurred, which increases the reliability of their recollection of these events
and, hence, the reliability of our data. We partly rely on retrospective insights from our intervie-
wees, which may potentially lead to biases and imperfect reﬂections on the process. Therefore,
we followed rigorous methods to mitigate these issues and protect data quality (Flick, 2009).
Interview data is inadequate as a sole data source (Kouamé & Langley, 2018;Yin,2009), which
is why we also examined other data sources (see Table 1). For instance, to increase the reliability
of the collected evidence, we requested interviewees to back up their stories with company doc-
uments. Furthermore, we triangulated the interview data by asking similar questions to different
interviewees that had detailed knowledge and ﬁrst-hand experience about the processes we were
interested in, such as interviewees who were involved in developing the new MC elements or
experimented with the new MC elements in MULTICORP’s trial setting (see section 4.3). This
allowed us to search for consensus between different interviewees on important issues and events
presented in this study. We use quotes from these interviews throughout the forthcoming analysis
sections. Furthermore, following Pratt (2008,2009), we present additional quotes in Appendix
B to demonstrate the consensus-making process.
The primary data source in this study is a series of semi-structured interviews with rele-
vant actors in MULTICORP, complemented by observations of organizational events and desk
research. The themes that were discussed in the interviews were deﬁned beforehand, but within
these themes the respondents had the possibility to elaborate. Interviewees were selected based
on their knowledge of MC elements under investigation and can be divided into two groups.
First, we interviewed those involved in the manager-subordinate relationships, i.e. members of
the top management team of Operations and their direct subordinates, titled general managers,
and their direct subordinates titled plant managers. A plant manager is the manager of a single
plant, whereas a general manager is in charge of group of plants within a deﬁned geographical
area. Second, we interviewed people who were seen as MC specialists in MULTICORP, such as
the Director of Supply Chain Excellence and a Demand planner (see last column, Appendix A).
Table 1. Case data.
Interviews 33 interviews with 26 organizational members (approximately 37 h of interview data)
Observations Observed four performance review meetings
Archival data Performance Measurement template, including KPI deﬁnitions and calculations
(Performance Management – KPI deﬁnitions)
Four performance measurement reports from four different manufacturing units
The action planning module in SAP and four-unit speciﬁc action plans
The “sales and operational planning” template that outlines this process (see section
MULTICORP’s online process management system
Cascading Control Changes, Incoherence, and Dialogue 385
The ﬁrst group was selected because they were able to provide insights into how various con-
trols functioned in practice between various hierarchical levels in the ﬁrm. The second group of
interviewees has profound knowledge about speciﬁc MC elements, for example the performance
measurement system. Combined, this provided us with a representative sample, enabling us to
gain control-related insights from both subordinates and managers (Tessier & Otley, 2012), and
from different organizational entities within operations. Interviews were conducted in Danish
and were recorded to enable an in-depth analysis of the interview responses. In total, 33 inter-
views were conducted with 26 unique organizational members for a total of approximately 37 h.
All interviewees were of Danish nationality and most interviewees worked in Denmark.3
NVivo was used to organize the qualitative data. We coded interviews into nodes equivalent
to the theoretical underpinning of this paper, such as incoherence and cultural asperity. Quotes
were assigned to the cultural asperity node, when interviewees used (Danish words equivalent to)
‘massive,’ ‘huge’ or ‘signiﬁcant’ when discussing changes or differences among MC elements
or practices. In addition, quotes were assigned to the incoherent node when interviewees used
words equivalent to ‘problem,’ ‘conﬂict’ or ‘dispute,’ when referring to the relationship between
new and old MC elements.
Due to the conﬁdentially agreement with the case company, some of the documents were made
available with conﬁdentiality restrictions and cannot be quoted here, but they helped neverthe-
less to interpret and validate information we obtained elsewhere. One of the authors passively
observed four performance meetings, i.e. without intervening with ongoing processes. These
observations allowed witnessing the real use of the performance measurement system. During
these meetings, ﬁeld notes were made. We used pattern matching and cross-referencing of state-
ments and explanations from various data sources in order to get a better understanding of the
data. Also, informal observations and talks during, for instance, coffee breaks and lunches served
as important sources to better understand the change processes at MULTICORP.
Combined, these data sources allowed data triangulation (Eisenhardt, 1989), which enhanced
the authenticity and credibility of the ﬁndings (Baxter & Chua, 1998; Golden-Biddle & Locke,
1993). We stopped collecting data when we reached adequate theoretical saturation (Eisenhardt,
1989). Table 1provides an overview of our case data.
The data collection and analysis described above allowed us to scrutinize the sequential change
process at MULTICORP (see Figure 1), which we divide into the following sections: MULTI-
CORP and its MC package before the control crisis (section 4.1), the control crisis, initial MC
changes and incoherence (section 4.2), phase one of the cascade process: changing the use of the
performance measurement system (section 4.3), and phase two of the cascade process: adjusting
the design of the performance measurement system (section 4.4).
4. The Case of MULTICORP
4.1. Before the Control Crisis: Autonomy and Decentralization
MULTICORP’s focus from the early 1990s onwards was on growth. To achieve its growth objec-
tives, the organization’s strategy was to enhance the entrepreneurial spirit of general managers
and plant managers by ‘handing them the keys’ to their speciﬁc unit. With this, they meant
that local managers could operate their unit autonomously (either a plant or a group of plants,
depending on the managerial level) and use (local) information to let it grow and further develop
3Only two interviewees (ID 4 and 9) had their main responsibility outside Denmark – in Australia and Germany. Although
it is too strong to say that we “control” for culture in our study, we believe that potential cultural differences did not play
a signiﬁcant role in the phenomena studied in this paper.
386 T. Toldbod and B. Van der kolk
Figure 1. A timeline highlighting the key changes at MULTICORP. Performance measurement system is abbreviated
here (PM system).
it. This self-governing culture and the autonomy of individual plant managers was one of the
most important aspects of MULTICORP’s MC package before the great ﬁnancial crisis surfaced
in 2008. The sales and operational planning manager (ID 2) explained the situation:
It had a lot of advantages that for many years [in the 1990s and early 2000s], in [a speciﬁc country], we handed
over the keys to that country’s general manager, so only he decided everything that should happen in that country.
No one else would be interfering with what was happening in that country. This local performance accountability
caused a great spirit of entrepreneurship and ownership.
In line with the philosophy of ‘handing over the keys,’ the general managers and plant man-
agers were responsible for developing their own control practices. Consequently, each unit
developed and maintained their own MC elements (related to, for instance, operational planning
and organizational structure), which resulted in a myriad of MC practices that varied locally. To
exemplify this, a company-wide forecasting process existed in which the sales function estimated
expected demand levels, but local managers did not integrate these forecasts into their own plan-
ning process. As an explanation for not using the company-wide planning data, local managers
indicated that they ‘knew better’4what was required to be a successful, growth-oriented unit.
Such local interventions would affect the performance of the individual unit, for which these
managers would eventually be held accountable. In general, the MC practices at MULTICORP
were very fragmented, diverse and differentiated across local MULTICORP units.
At this time, however, one element within MULTICORP’s MC package was strongly cen-
tralized and formalized: the performance measurement system. The performance measurement
system included 12 (ﬁnancial and non-ﬁnancial) operations-related performance measures,
addressing topics such as safety, quality, delivery, costs and staff. The performance measurement
system monitored performance levels of the globally dispersed individual units and individual
general managers or plant managers were accountable for these performance measures. In line
4According to Supply Chain Conductor 1 (ID 21) who had multiple years of experience as a plant manager.
Cascading Control Changes, Incoherence, and Dialogue 387
Table 2. Example of one of the 12 standardized performance measures.
Performance measure First time through percentage
Purpose: Provide an overview of the stability of the process quality on production
Approach: Measuring scrap and rework
Scope: All manufacturing plants in operation
Impact: Primary: delivery. Secondary: productivity, cost, and capacity
Deﬁnition: % of produced parts with no defects on ﬁrst pass compared to total amount
of parts produced
Aggregation: Number of plants not on target divided by number of plants on target
Formula: 100% * (Amount of produced parts – [scrap +rework]) / (total amount of
with the philosophy of ‘handing over the keys,’ local managers were held accountable for the
performance of their own unit, and were given high levels of decision authority to achieve their
performance goals. For example, several interviewees (e.g. ID 2, 3 and 21) mentioned that local
managers took great pride in documenting ‘superior performance’ of their units, which was seen
as a possibility to help advance careers within the ﬁrm. All performance measures were stan-
dardized and well deﬁned, while targets could vary among units within MULTICORP. Table 2
is an example of a deﬁnition of one the measures, the ‘ﬁrst time through percentage’, taken
from the internal document Performance Management – KPI deﬁnitions. The formula used to
calculate the performance indicator is made very explicit (i.e. the amount of produced parts –
[scrap +rework], divided by the total amount of parts produced). The other information that is
given per performance indicator explains, amongst other things, its purpose and to which part of
MULTICORP it applies (Scope). Based on this information, this measure was used to estimate
one dimension of performance for each individual manufacturing unit.
4.2. The Control Crisis, Initial MC Changes and Incoherence
For many years, MULTICORP’s MC philosophy was characterized by the self-governance of
decentralized units. However, as MULTICORP went through a period of rapid growth by acquir-
ing several other companies, its internal structure became more complex. The great ﬁnancial
crisis that surfaced in 2008 exposed a problem with this self-governing philosophy and the
increased complexity. The sales and operational planning manager (ID 2) explained:
The basis was the [great ﬁnancial] crisis. Suddenly, we stopped billing customers, but it took almost six months
before we stopped [scaled down] production. That’s what happens in a company when there is a lack of coordination
between sales and production departments. Therefore, we wanted to create a new [MC] element, which we called
Demand and Supply Integration. This is basically about having one set of numbers and a transparent process from
sales planning to operational planning [ ...] We didn’t have a forecasting process that worked well, so we had to
establish that ﬁrst.
This quote points out that the repercussions of the great ﬁnancial crisis in 2008 made this lack
of internal integration visible. As argued by Galbraith (1973), increased complexity requires
more integrated mechanisms, yet this need became more salient to managers after MULTICORP
demonstrated a lack of response to the sudden drop in demand, caused by the ﬁnancial crisis. In
effect, the great ﬁnancial crisis functioned as an eye-opener that led MULTICORP’s top man-
agement to search for ways to avoid similar problems in the future. MULTICORP had outgrown
their old approaches to MC and needed a more structured MC approach. During an early inter-
view in 2012, the sales and operational planning manager (ID 2) commented on the development
of new MC elements in the period after the great ﬁnancial crisis:
388 T. Toldbod and B. Van der kolk
We have developed these [MC] elements [to address the mismatch between internal supply and demand] over, more
or less, the last three years. We have applied them for about one year in one of our supply chains, and in 2013–2014
it will be implemented in all MULTICORP’s supply chains [...] What we did was to gather some of our highly
skilled logistical engineers and others alike in a project room [ ...] It was fun to go down to the project room
[located in the building’s basement] and sense the activity and be part of the discussions, [for instance] about how
to set a reorder point optimally, [how to calculate] stock levels and so on?
As captured in the quote, the MC changes were developed ‘in isolation’ (literally, as the project
room was located in the basement) by supply chain specialists to respond to the problems caused
by the lack of coordination. The scope of the project was to solve the speciﬁc problem at hand, i.e.
the lack of coordination between internal supply and demand. Consequently, the discussions in
the room were oriented toward this (narrow) area of aligning supply and demand better, without
considering existing MC elements, such as the performance measurement system. It was not the
great ﬁnancial crisis per se that triggered the need for more integration, but it rendered the need
salient. This process resonates well with the organizational ecology literature in that a signiﬁcant
crisis can catalyze changes to MC elements as a ﬁrst response to deal with the crisis (Hannan &
Freeman, 1984; Hannan et al., 2003a). In the case of MULTICORP, it was a sequence of crises
that initiated the MC change; an external (the great ﬁnancial) crisis exposed the internal control
issues and triggered the control crisis (see Figure 1).
In 2011, MULTICORP pre-tested the new MC elements in one of their supply chains and,
in 2013, MULTICORP fully implemented the new MC system to solve the lack of coordina-
tion and titled this change program ‘Integrated Demand and Supply.’ The MC system consisted
of two related MC elements: (1) a new, centralized sales and operational planning department
would issue forecasts for the whole company; and (2) ‘supply chain councils’ would be orga-
nized frequently to discuss and ﬁne-tune MULTICORP’s planning for the next period. We will
describe both initiatives in more detail below. These descriptions form the basis for our analysis
and discussion of why and how the MC changes enhanced incoherence among MC elements and
triggered new changes to MC elements.
4.2.1. Centralized sales and operational planning
The main goal of centralized sales and operational planning is to obtain ‘one set of numbers’5
derived from both qualitative and quantitative forecasting methods. Subsequently, this one set of
numbers should be used for action planning in each of the supply chains, which then ﬂows down
to the individual plants.
As part of this new MC element, on the seventh workday of every month, demand planners
released forecasts for each of MULTICORPs supply chains. These forecasts are called the inter-
nal delivery plans. These internal delivery plans deﬁne the output requirements for every week in
the upcoming month based on the projected sales. Master planners have to ‘translate’ the internal
delivery plan to numbers that can be used to schedule the production in the local plants. When
the master planner has ﬁnished this job, the weekly output requirements are handed to a produc-
tion planner, who translates it to the daily production schedules and individual production orders.
Just as the master planner decomposes the output requirements from the internal delivery plan,
the production planner decomposes the weekly output requirements from the master planner and
sets up production orders to execute the level output production plan. The production orders they
send to the production lines must correspond to the weekly output requirements that the mas-
ter planner established. The plan that the master planner and the production planner produce is
labeled the ‘supply network plan’, which is communicated to the supply chain council.
5Multiple respondents referenced this vision, for instance sales and operational planning manager (ID 2), Master Planner
2 (ID 14) and Supply Chain Conductor 1 (ID 21).
Cascading Control Changes, Incoherence, and Dialogue 389
This description outlines one of the essential new practices: a new way of planning, which
implies a novel way of formulating short-term plans – very centralized and structured. This
means that autonomy and self-regulation, which were central to the MC philosophy before
the control crisis manifested (see section 4.1), were strongly suppressed. In the words of our
interviewees: ‘the keys’ were essentially taken away from the local plant managers and general
4.2.2. The supply chain council
Another structural change regarded the organization of supply chain councils. A supply chain
council is a cross-organizational council that groups the supply chain’s main stakeholders – either
general managers or plant managers from various business units – and is chaired by the supply
chain conductor,6i.e. the executive manager of the speciﬁc supply chain.
After the master planner and production planner have drafted the supply network plan, the
planning manager presents the plan to the supply chain council at their monthly meeting. Before
the supply chain council meeting takes place, the individual supply plans from each individual
plant in the supply chain are aggregated to a supply network plan for the full supply chain, which
is approved at the supply chain council meeting. Either the supply network plan is accepted and
approved with only small changes, or a major constraint somewhere in the supply chain urges
the council to alter the plan. A major constraint occurs, for instance, when a speciﬁc plant cannot
operate at normal capacity because of scheduled maintenance, local holidays or repetitive quality
issues. This means that in such a situation, a plant cannot deliver the amounts required by the
‘internal delivery plan’. In this case, it is the responsibility of the council to solve the constraint,
while the responsibility for the supply chain’s compliance with the overall ‘internal delivery
plan’ lies with the supply chain conductor, who also has the ﬁnal say in the council meeting.
Decisions made by the supply council can have a big impact, such as the reallocation of output
requirements to other plants in the supply chain that hold similar manufacturing capabilities
and have excess capacity. Ultimately, the supply chain conductor is responsible for, and deﬁnes,
short-term output requirements for the plants in the supply chain. A supply chain conductor (ID
27) summarized his role as follows:
To improve lead time to customers, while reducing inventory and cost. That is what I am supposed to do. That’s my
most important task [ ...] and it’s not possible to give 30 units in a [supply] chain that task individually.
The supply chain council can be seen as another manifestation of a shift away from the decen-
tralized MC philosophy in which general managers and plant managers held ‘the keys’ and had
full decision autonomy. The objective for the newly implemented MC elements now seemed to
be to optimize the full supply chain, rather than to maximize the returns from individual plants
To break down the functionally fragmented culture and align internal operations better, MUL-
TICORP’s top management established the supply chain councils. The coordination of a complex
supply chain with between 10 and 30 distinct units is, they argued, too complex for local man-
agers to grasp. So, top management saw no other option than to ‘force them [the individual plant
managers and general managers] to talk to each other [in the supply chain council], otherwise it
wouldn’t happen’ (sales and operational planning manager, ID 2). Consequently, they needed a
person to take on that role and hold total responsibility and decision rights for the supply chain.
Regarding the joint changes that the newly implemented centralized sales and operational
planning and the supply chain council caused, a Supply Chain Conductor (ID 21) elaborated:
6Initially there was only one supply chain conductor, but later this number increased to four for MULTICORP.
390 T. Toldbod and B. Van der kolk
A plant manager was accustomed to make his own decisions. He was in charge of his own planning and many other
things. Now he doesn’t have that anymore, I mean, he can’t even decide what he wants to produce anymore [ ...]
There’s a lot of power that has been moved away from units that once held that power. Instead, [the plant manager]7
now must become extremely effective in executing a plan [made by others].
According to the Supply Chain Conductor, the new MC elements formed a clear break with the
company’s historical functional and self-governing way of organizing control, since the control
focus shifted from the plants to the broader value chain. As part of this process, local manufac-
turing units also changed from what resembled a proﬁt center to a cost center structure, which
severely affected their autonomy on variables that in turn affect their unit’s ﬁnancial perfor-
mance. For instance, transfer prices were not up to local negotiations anymore, but deﬁned by
the corporate ﬁnance department and as described above, output levels were deﬁned via these
new structures. This reduced their autonomy over variables that affected their performance. The
fact that they now had to ‘execute [action] plans’ made by others decreased their autonomy sig-
niﬁcantly. For instance, they had no (or very little) inﬂuence anymore on the operational rate,
i.e. the level of utilized production capacity compared to total available production capacity.8
The fact that they are handed an output requirement means that they cannot affect the utilization
of production capacity anymore. In effect, local units lost large parts of their sovereignty and
autonomy to inﬂuence their own performance outcome.
Through the theoretical lens of organizational ecology, this exempliﬁes the notion of cultural
asperity, as planning controls are a core architectural element in operations (Daft & Macintosh,
1984). These new MC elements do not complement the entrepreneurial spirit that MULTICORP
historically was built on where local units had high degrees of autonomy to manage their own
unit. On the contrary, they clashed with the existing entrepreneurial culture at MULTICORP.
To illustrate this, one interviewee (ID 6) used the term ‘disempowering of local managers’ to
describe the effects of these new MC elements. In other interviews, adjectives such as ‘massive’
and ‘huge’ were used to describe how these new MC elements deviated from status quo, signaling
cultural asperity (Hannan et al., 2003b).
MULTICORP’s top management wanted to dismantle the historical focus on individual units
and shift the focus to the well-functioning of the total supply chain. This change in focus led to
the development of the new MC elements that signiﬁcantly deviated from the previous approach
to MC. This high degree of cultural asperity raises, according to the literature, the risk that
new MC elements are incoherent with old MC elements, which may urge an organization to
make further changes to MC elements, hence triggering a cascade effect (Hannan et al., 2003a,
2003b). In the next section, we address how the two new MC elements were incoherent with
MULTICORP’s performance measurement system.
4.2.3. Incoherence and cascading changes
MULTICORP implemented the internal delivery plan and the supply chain council in one of
their supply chains in 2011. During the introduction of these new MC elements, the performance
measurement system (that was used to assess the performance of MULTICORP’s units) remained
unchanged. The operational excellence manager (ID 1) indicated that ‘their [i.e. the general and
plant managers] KPIs [key performance indicators] are still the same.’ Therefore, the period after
implementing the new MC elements left this speciﬁc supply chain in a situation characterized by
the coexistence of ‘old’ and ‘new’ MC elements, each focused on a different objective. Already
in 2012, the sales and operational planning manager (ID 2) acknowledged that the coexistence
of these ‘old’ and ‘new’ MC elements caused problems:
7To respect the conﬁdentiality agreement with the interviewees, our interviewees are anonymized.
8This speciﬁc example is based on the internal document Performance Management – KPI deﬁnitions.
Cascading Control Changes, Incoherence, and Dialogue 391
Responsibility for operating performance lies only with the individual plant or group of plants [ ...] Each unit is
measured on its own productivity and results, but no units are measured on the contribution to the [performance
of the] supply chain [ ... Consequently,] you can’t get anyone to give anything away [for instance, output levels],
because that affects their performance [measures]. In addition, you cannot take anything away from them without
them complaining [ ...] It inspires suboptimization. That’s the nature of this [ ...] It’s a challenge in the current
structure [with the performance measurement system still in place] to get people to think “horizontally.” We are
right in the middle of that issue. This performance [measurement] system has been there for historical reasons, but
it absolutely doesn’t ﬁt anymore.
The performance measurement system was designed for evaluating the performance of isolated
units and was directed at the company’s old control objective and functional focus. This MC
element still motivates individual managers to optimize their local performance levels by hold-
ing them accountable for unit performance instead of giving them a ‘supply chain perspective’.
According to the sales and operational planning manager, one of the effects of this ‘silo think-
ing’ is that it inspires ‘suboptimization’. Consequently, it is ‘a challenge [...] to get people to
think horizontally’ because the performance measurement system does not communicate com-
mon goals across the value chain. In effect, it gives no impetus for cross-unit collaboration.
A plant manager (ID 7) further elaborated on the effects of coexistence of old and new MC
The decisions made in the supply chain council can conﬂict with my own supervisor [a general manager] with
respect to the performance measures, because we might make some decisions that are good for the value chain as a
whole but not so good for our production plant, and that gives rise to conﬂicts.
According to this interviewee, the coexistence of old and new MC elements fueled a con-
ﬂict, as the different MC elements simultaneously encouraged contradictory objectives. The
performance measurement system did not ﬁt the horizontal objective stimulated by the new MC
elements, in other words, the old and new MC elements were incoherent. Some interviewees
mentioned the placement of inventory as an illustration of a conﬂict related to the two contra-
dictory objectives. One supply chain found it optimal that a speciﬁc plant carried an additional
inventory with a net worth of approximately 15 million euros. However, this would lower the
inventory turnover in this speciﬁc plant, thus hampering the performance levels captured by the
performance measurement system. The plant manager therefore found himself in a conﬂict of
interest with the general manager, while the plant manager remained uncertain about the proper
course of action because different MC elements dictated different actions, aimed at different
objectives. MULTICORP’s employees acknowledged the incoherence of the MC elements and
the conﬂicts that resulted from the contradictory objectives, a plant manager (ID 8) explained:
[The new MC elements] can create conﬂicts. For instance, our CFO [Chief Financial Ofﬁcer] and ﬁnance department
recently indicated that we’re having too much working capital. They may have a point. [They say] ‘You must be able
to free up some capital’, yet in the council we optimize while focusing on minimum delivery time, availability and
a lot of other things, and only after that we look at our capacity and investments. That is a conﬂict [ ...]So,Ioften
talk with our COO [Chief Operations Ofﬁcer] and supply chain director and ask them if they can please agree on the
[prioritization of the] tasks, [but] there’s this form of pressure [The pressure from the vertical focused performance
measurement system and the horizontal focus of the new MC elements ...] I can’t navigate freely when I feel this
pressure from the supply chain and from my local boss, while our performance measures dictate local parameters
that [for example] imply having a lower working capital [in my unit], but that does not make sense if you look at
the [supply] chain.
The new MC elements have a clear focus on the full supply chain, while the performance
measurement system still focuses on the individual units, prompting ‘tasks’ intended to optimize
performance for the individual, local units. When these different MC elements are juxtaposed, a
conﬂict emerges. The different MC elements inspire achieving different performance goals and
are thus incoherent (Abernethy & Chua, 1996).
MULTICORP was debating and considering the need for more drastic adjustments to the
performance measurement system, as expressed by the operational excellence manager (ID 1):
392 T. Toldbod and B. Van der kolk
We need to get all these performance measures integrated [with the new MC elements].
A plant manager (ID 7) also elaborated on the need for changes:
We’re about to take some of the performance measures, for instance delivery performance, away from general
managers. We’re going to make it a matter of the [supply chain] council.
Senior managers at MULTICORP thus recognized that the new MC elements affected
the functioning of the performance measurement system, and vice versa, and that the arrival
of the new MC elements triggered a need for changes to other, existing MC elements. However,
the performance measurement system had not yet changed, while the new MC elements were
already in place. When asked why the performance measurement system had not changed yet,
the sales and operational planning manager (ID 2) explained:
Changing takes time. We are slowly convincing our managers of the usefulness of these [new MC elements] [...]
It took us almost a year before we could show better results, which probably also is the time it takes from imple-
mentation to the moment you can see the results. [ ...] We believe in it [the new MC elements] because they have
improved performance, but they have not been operational for, let’s say, three years, so we can’t be certain on
whether the results are sustainable. It’s a long process [and] we have these control challenges we have to live with
almost every day [... .] There is a need to rethink that setup [so] we discuss performance measurement: ‘Do we
use the correct measures?’ and ‘How should we structure them?’ [But] we have not yet ﬁgured out how to measure
the performance of units across [the supply chain].
This quote points to some central notions of organizational ecology. As captured in the ﬁrst
part of the quote, managers at MULTICORP did not know whether the new MC elements would
actually solve this control crisis. Therefore, they wanted to document the advantages of the new
MC elements before implementing further changes to other MC elements. This trial period is
central in organizational ecology literature, because organizations need time to develop and
implement new elements, during which they document the beneﬁts before making additional
changes to realign old elements with new ones (Denis et al., 2001; Hannan et al., 2003a). The
last part of the quote resembles what is coined structural opacity in the organizational ecology lit-
erature (Hannan et al., 2003b). As outlined by an interviewee (ID 2), managers at MULTICORP
acknowledged that, indeed, the implementation of the new MC elements had some unintended
consequences and that changes to the performance measurement system were also necessary.
Yet, as indicated in the above quote, they did not know exactly ‘how to structure’ a performance
measurement system that would be coherent with these new MC elements. Therefore, MUL-
TICORP could not specify upfront which performance measurement design changes would be
needed to reduce incoherence between the old performance measurement system and the new
MC elements (i.e. structural opacity). In 2012, the corporate manufacturing and supply chain
manager (ID 3) also reﬂected on the need for design changes to the performance measurement
system and why MULTICORP had not changed it yet:
We’ve deduced that the performance measurement system should be better integrated than currently is the case,
because the performance measurement system should align with all our new systems [ ...] We cannot have a
situation where they [the MC elements] are not in harmony. But, there are already so many new things going on that
we had to give up! Or to put it differently, we just wait a little longer [with adjusting the performance measurement
system to the new situation].
This quote indicates that there were ‘already so many new things going on’ that they decided
to delay the adjustment of the performance measurement system. MULTICORP did not have
the resources to develop and implement these changes parallel with the development and
implementation of the new MC elements. This resonates with one of the aspects of structural
opacity, namely that adjustments cannot take place in parallel. Furthermore, another intervie-
wee (ID 2) argued that MULTICORP seemed not able to deﬁne what the design changes should
entail, resembling another aspect of structural opacity from the organizational ecology literature
(Hannan et al., 2003b).
Cascading Control Changes, Incoherence, and Dialogue 393
The combined notions of cultural asperity and structural opacity allow us to understand two
issues. First, they explain in which situations the initial changes may be a catalyst triggering
further changes. Speciﬁcally, the high level of asperity means that existing MC elements also
have to change, because they severely deviate from the new MC elements. Second, these notions
help to explain why the need for further changes only follows subsequently and not in parallel.
The structural opacity at the time of the initial change gave that they could not deﬁne the required
changes at that time and did not have the resources to take on these additional changes at the same
time of implementing the new MC elements. Accordingly, when the new MC elements were fully
implemented in 2013, the following period in MULTICORP was characterized by incoherence
among MC elements. This led to conﬂicting objectives (Hannan et al., 2003a) and increased
uncertainty about the proper cause of action related to different tasks demanded by the incoherent
MC elements (Abernethy & Chua, 1996; Barnett & Carroll, 1995). Managers, seemingly aware
of the conﬂicting objectives and the risk of organizational impasse, tried to mitigate the negative
effects of incoherence, which we will provide further details on in the following section.
4.3. Phase One of the Cascade Process: Changing the Use of the Performance Measurement
The previous section highlights the incoherence between the old performance measurement sys-
tem and the new, supply chain-focused MCs. Because MULTICORP’s top management did not
want to make changes to the performance measurement system (yet), local managers started to
explore ways to deal with the incoherence. They often indicated that, if there were any conﬂicts
related to MC, they would discuss it with those involved. For example, the general manager (ID
6) indicated that ‘if there’s anything related to performance management, we talk about it’ and
What it basically all boils down to, is that we must manage differently than before. That is what it’s basically all
about at the end of the day, and that requires a lot of managerial resources [...] Today managers are much more
“out there”, in touch with their subordinates, which was not the case before. Now they have a more interactive and
coaching management style. It’s more a “dialogue” than a “command;” so, you’ve got to change as a manager [ ...]
You have to talk with each other, not talk to each other.
A plant manager (ID 7) acknowledged that this was indeed the case and highlighted how the new
style of management was more supportive:
This means that the individual managers are much closer to the subordinates [ ...] They offer much more sup-
port and attention when there are conﬂicts, and they’re present to help solve conﬂicts that hamper production and
continuous improvements. That’s the difference.
These quotes suggest that a dialogue-oriented style of managing was gaining momentum at
MULTICORP, as managers are more often ‘out there’ among employees and peers ‘talk[ing]
with each other’. Yet the same plant manager also acknowledged that, in the end, he would
still be held accountable for the performance measures – from the ‘old’ unchanged performance
measurement system – because it was still in place:
Sometimes there are embedded conﬂicts. Consequently, we have a dialogue about it and debate how it all ties
together, so that he is informed about it. Nevertheless, he will hold me accountable for this unit’s numbers [in the
performance measurement system], and that’s how it is.
Depending on the managerial level, the output from the performance measurement system is
discussed in the monthly or weekly performance review meetings. Before the new supply-chain
oriented MC elements, the structure of performance review meetings followed a ﬁxed order.
First, the performance measures were presented in detail, while actual performance would be
compared to the standards that were set in the beginning of the period, applying basic variance
394 T. Toldbod and B. Van der kolk
analysis methods. Second, attention would be given to potential ways to improve the perfor-
mance. The ﬁrst part of the meeting was a plenary session, chaired by the plant manager or
general manager and the second part of the meeting was in smaller subgroups. The goal of
the second part of the meeting was to come up with solutions for how to deal with speciﬁc
problems relevant for that (sub)group, so the group composition depended on the problem at
hand. For instance, two plant managers experienced quality issues and formed a group together
with a quality engineer to discuss how they could deal with these quality issues. Another group
reviewed problems related to delivery performance and potential solutions were discussed within
that group. In meetings that took place after the new coordination-related MC elements were
implemented, the focus was shifted. The time spent on the second part of the meeting had sig-
niﬁcantly increased at the expense of the time spent on the ﬁrst part. A general manager (ID 17),
for instance, explained the change in the structure of the performance review meetings:
It’s uninteresting to go through the numbers [as we did before]; we already know them now, before we arrive at the
meeting. What is of interest, however, is that we have a discussion [...] To have a good meeting means to have
a good dialogue. This is where we meet and where we have a dialogue across functions and factories and discuss
issues with respect to, for instance, productivity. Unit A [for instance] has this issue, so now the “ﬁre” is there. We
[Unit B] also have had the same issue in our unit, so we should try to get together [and help one another ...]Imight
have some useful knowledge [for them]. It is this sort of collaboration that is of interest, not the numbers.
This quote highlights that the ﬁrst part of the meeting had become ‘uninteresting’. Instead,
the increased room for discussion enabled opportunities for communication and discussion, as
managers from similar levels now worked to solve complex problems and conﬂicts, sometimes
even before they materialized and escalated. In retrospect, a number of interviewees conﬁrmed
that this was different before, when the emphasis of the meeting was more on the ﬁrst part,
discussing performance levels and variances. A general manager (ID 5) further explained:
The essence is, we measure performance, but that is not what we are really interested in [ ...] So we don’t look at
performance levels per se. [Instead] they provide, I don’t want to say an excuse, but an opportunity to discuss issues
[...] It [the performance review meeting] is a place where you share problems. [For instance], “I have had a similar
problem and I solved it is this way, and I have an employee that knows about this. I will ask him to call you.” [...]
No one is being yelled at if they don’t achieve their [performance] objective, but they will be yelled at if they don’t
want to engage in the dialogue [ ...] That’s the whole point [ ...] It’s the behavior at the [performance] meetings
that is important. [ ...] You have to be very consistent in the way you manage [the meeting]. Actually, it has to be a
kind of business philosophy [to have a dialogue] some things need to fall in place and then it immediately provided
results [ ...] I cannot imagine going back.
Therefore, managers did not participate in performance review meetings to actually review per-
formance. Instead, we observed that the performance review meetings turned into a time and
space to discuss and solve problems. In some cases, managers did not even discuss the perfor-
mance measures and variance analyses that used to be discussed in the ﬁrst part, but instead
started with the ‘second part,’ i.e. organizing small work groups to focus on solving particular
tasks. Consequently, the performance measurement system became an important starting point
for dialogue, while the performance measures themselves were still the same. The so-called per-
formance review meetings were reinvented as a space where individuals could share experiences
and gain new insights. As a result, the performance review meetings functioned as a forum for
cross-unit interaction and knowledge sharing.
Our case study shows how incoherence among MC elements inspired MULTICORP’s man-
agers to spot new opportunities and start using the performance review meetings in a novel way
that made more sense to them, given the new supply chain focus. Managers effectively created a
safe space in which open communication was encouraged, allowing organizational members to
engage in discussions and ﬁnd solutions.
As this style became more common, positive results of the improved cross-unit interaction
could be observed. As an illustration, a general manager (ID 5) stated in 2015:
Cascading Control Changes, Incoherence, and Dialogue 395
What happened – and I still ﬁnd this very exceptional because production managers are intelligent people who want
to deliver results – is that they really started to embrace it [the dialogue with other organizational members], when
they realized that this gave them the opportunity to deliver the performance they were never able to deliver before.
To illustrate how the changing use of the performance measurement information affected per-
formance, a plant manager (ID 6) provided us with an example of the operationalization of new
equipment in 2015. While before it would take at least two years to make new equipment oper-
ational, the knowledge he gained from the performance review meetings enabled him to achieve
very high productivity and an uptime of 80% within four months after he started to use the
new equipment. The plant manager (ID 6) put the results into context and made the following
comment when explaining the success:
That was a new internal ‘world record’ in MULTICORP [...] What made the difference is this different manage-
rial approach. We now follow measures and directly involve employees in continuously improving and analyzing
performance; that’s the cornerstone.
The new style of the performance review meetings was highly appreciated by the employees
and managers; interviewees particularly liked the new emphasis on dialogue and learning and
the structure of problem-oriented discussions in smaller, knowledgeable teams. In other words,
the interactions and the knowledge sharing in the meetings were perceived to be an important
element in achieving the ‘new internal world record’.
This section shows that the original changes indeed cascaded. First, as a result of the introduc-
tion of the new MC elements, managers changed their use of the (old) performance measurement
system (but not the design itself), which was already outside the original locus of the initial
changes, to respond to the initial control crisis. The design features, such as the ‘scope’ dimen-
sion of the performance measurement system, remained unchanged in this phase of the cascade
4.4. Phase Two of the Cascade Process: Adjusting the Performance Measurement System
Although managers were able to deal with the incoherence among the MC elements by adjusting
their use of the performance measurement system, various interviewees also acknowledged that
the system itself (eventually) had to change. In 2015, MULTICORP started to develop a new
performance measurement system that would be more in line with the new supply chain objec-
tive. In 2017, a supply chain conductor (ID 21) commented on the new performance measures,
highlighting the shift from the individual units to the supply chain as a whole:
I now have some measures that I am responsible for. Now, strictly speaking, we do not really look at the individual
unit’s performance, but only how they affect the supply chain [...] So, now I run the performance review meeting
every week. This means that I follow up on the measures and performance.
Also, in 2017 the director of supply chain excellence (ID 24) seemed content about the
changes and particularly highlighted the commitment to supply-chain-related goals that the new
performance measures fueled:
[The supply chain conductors] are now responsible for these measures and for supply chain performance. [ ...]This
has changed [ ...] The performance measures are now measuring the [performance of the] supply chains [ ...]The
reason for this shift is that we wanted more commitment to these [new MC elements] and to emphasize that all units
are part of a supply chain.
Contrary to the old performance measurement system, which focused on local unit performance
(see ‘scope’ dimension in Table 2), the new system focuses on supply chain performance. Besides
this change, the performance measures for which the supply chain conductor was responsible
also included a few adapted measures from the old performance measurement system. The new
396 T. Toldbod and B. Van der kolk
performance measurement template covered ﬁnancial measures (revenue and cost measures),
purchasing (supplier delivery time and quality), distribution (stock ratios and delivery perfor-
mance), customer service (lead time to customer) and internal quality measures. As expressed
by the director of supply chain excellence in the quote above, the new supply chain focus was
an attempt by MULTICORP’s top management to generate further commitment to the supply
chain objective. This made the performance measurement system more coherent with the new
MC elements (i.e. the centralized sales and operational planning and supply chain councils).
Summarizing, the cascade effect in the ﬁrst phase entailed a change in the use of the perfor-
mance measurement system, while in the second phase the change was manifested by intended
changes to the design of the performance measurement system. The two phases thus included
different processes, which became more visible by distinguishing between the use and design
aspects of the performance measurement system.
We began this paper by posing the question how sequential MC changes affect incoherence
among MC elements, and how managers can deal with this. We make two contributions, and
below we discuss how these contributions can be situated in the relevant literatures.
First, by mobilizing the notions of cascade effect (including the idea of sequential change and
‘old’ and ‘new’ elements), cultural asperity and structural opacity from organizational ecology
(cf. Hannan & Freeman, 1984; Hannan et al., 2003a) we extend the analytical lenses available
to the MC change literature. Our analysis shows how a control crisis triggers cascading MC
changes, which subsequently create incoherence among MC elements. This ﬁnding speaks to
MC research interested in the interactions among MC elements within a package or system (cf.
Malmi & Brown, 2008; Grabner & Moers, 2013; Friis et al., 2015; Henttu-aho & Järvinen, 2013;
Østergren & Stensaker, 2011; van der Kolk, 2019).
As demonstrated in the case study, the notions of cultural asperity and structural opacity help
to explain whether and when MC design changes will likely spread outside the locus of the
intended change – cascading to other MC loci. In our empirical material, we observed that high
levels of cultural asperity, i.e. strong differences between ‘old’ and ‘new’ MC elements, created
incoherence between MC elements. This triggered a cascade effect, where the original design
changes required a design reconﬁguration of other MC elements in order to reinstate coherence.
Cultural asperity facilitates further theorizing on the question when new or changed MC ele-
ments cause incoherence with ‘old’ MC elements, and may form a starting point for a cascade
of changes. In a situation of cultural asperity, an MC change is not likely to be an isolated event,
but will likely become a catalyst for further MC change. Cultural asperity thus informs when
and why some ‘new’ MC elements affect ‘old’ MC elements. Hence, to understand change well,
studying one change to a speciﬁc MC element is only the starting point, and observations about
the magnitude of MC change (e.g. radical vs. incremental, see Burns, 2000; Scapens, 1994)
should not only focus on the initial MC change, but also take cultural asperity and possible sub-
sequent changes into account. Thus, asperity allows us to understand when the initial change
likely has effects outside the intended change locus on other elements in the MC package or sys-
tem (cf. Malmi & Brown, 2008; van der Kolk, 2019). In other words, what initially may seem to
be an incremental MC change, might cascade to a more radical change, when the initial change
is characterized by high cultural asperity.
While asperity helps us understand when and why a MC change cascades, it provides little
insights into how this process unfolds. However, the concept of structural opacity (Hannan et al.,
2003b) can also help to contextualize our ﬁndings. In our case, interplay between the ‘new’ and
‘old’ MC elements was not part of original change mandate. Furthermore, at the time of the initial
Cascading Control Changes, Incoherence, and Dialogue 397
changes, MULTICORP did not know how to redesign the performance measurement system in
such a way that it aligned with the new supply chain focus. Lastly, the cultural asperity between
‘new’ and ‘old’ MC elements means that it brings multiple unknowns, both regarding how to
make the ‘new’ MC elements work and which other changes are needed to ‘old’ MC elements.
The search and adjustment process required to align the ‘old’ performance system would demand
signiﬁcant resources, yet those resources were already engaged with implementing the ‘new’ MC
elements. This exempliﬁes organizational ecology’s concept of structural opacity, where lack of
aprioriknowledge and scarce resources generate sequential (rather than a parallel) change.
This causes a period in which incoherent ‘old’ and ‘new’ MC elements co-exist. The notion of
structural opacity helps to understand how and why MC changes cascade. That is, if the initial
change is structurally opaque, the cascade will likely be sequential and protracted. Although our
ﬁndings do not include a situation that is not structurally opaque, the organizational ecology
literature would argue that in such situations the cascade process could be parallel and relatively
swift, because organizational members would know which further changes would be needed and
would have the organizational resources to initiate these changes (Hannan et al., 2003b,2003a).
Together, the notions of cultural asperity and structural opacity enhance our understanding
of cascading MC changes and when, why and how incoherence between MC elements is more
likely to emerge and exist for longer periods of time. When a ‘new’ MC element breaks the
status quo of ‘old’ MC element (i.e. cultural asperity) it creates incoherence between ‘new’ and
‘old’ MC elements. This initiates a cascade of subsequent MC change(s) to realign ‘old’ and
‘new’ MC elements. Our longitudinal case study does not allow any generalizations about a
speciﬁc timeframe for MC changes and cascade effects. Conceptually, however, the presented
theorized narrative points to the potential of cultural asperity and structural opacity to impact the
severity of the changes needed and the length of the period, although it would also be too sim-
plistic to suggest any linear relationships on the basis of our study alone. Some may even argue
that – from an economics perspective – very high levels of cultural asperity can be dealt with
rather sooner than later, as high levels of incoherence may jeopardize organizational effective-
ness and likely receive immediate attention (Bedford et al., 2016; Friis et al., 2015; Kristensen &
Israelsen, 2014). However, based on the organizational ecology literature and our case insights,
we argue the contrary. When the initial change signiﬁcantly breaks status quo (i.e. cultural asper-
ity), it can make the need for subsequent changes more difﬁcult to predict (i.e. structural opacity).
Hannan et al. (2003b) call this the fog of change. While the resulting search-and-adjustment pro-
cess requires more organizational resources to analyze and deﬁne further changes, organizational
resources are scarce and may already be reserved for implementing the initial change. Thus, the
change becomes sequential and protracted, and incoherent ‘old’ and ‘new’ MC elements co-exist
for a longer period of time, in our case multiple years. The combined insights from these con-
cepts contribute to theory building on the interplay among MC elements (cf. Friis et al., 2015;
Henttu-aho & Järvinen, 2013; Østergren & Stensaker, 2011; Malmi & Brown, 2008).
The idea of a cascade effect challenges current conceptualizations of MC change as a process
within the locus of a speciﬁc MC element (e.g. Busco et al., 2007; Kasurinen, 2002;Malmi,
1997). Our longitudinal study incorporates the element of time and focuses more on the change
process, hence, complementing earlier research that predominantly used a variance perspective
(cf. Baines & Langﬁeld-Smith, 2003; Beaubien, 2012; Granlund, 2001; Jansen, 2011). In line
with prior research that argued that MC elements should not be studied in isolation (Malmi &
Brown, 2008), we argue that changes to MC elements should also not be studied in isolation–
Our analysis shows how existing structures can adapt to ‘new’ MC elements in two different
phases. The ﬁrst phase entailed the response by managers, related to the use of the MC elements
outside the intended change locus, and in the second phase – in our case years after the initial
MC change – the design features of the existing MC elements changed. This ﬁnding nuances the
398 T. Toldbod and B. Van der kolk
MC literature that proposes that a signiﬁcant change to a speciﬁc MC element should align with
the existing structures (Burns, 2000; Burns & Vaivio, 2001; Scapens, 1994), by showing how the
existing MC elements can also adapt to a newly implemented MC element. What might not seem
as a ‘radical’ change at ﬁrst can cascade to other MC loci and cause multiple other MC changes,
rendering the total change process more radical than originally intended. Together, the notions
from the ecology literature theoretically inform a more holistic understanding of MC change in
the context of an MC package or system (cf. Malmi & Brown, 2008; Grabner & Moers, 2013;
van der Kolk, 2019), and help to understand why incoherence may exist for extended periods of
The second contribution relates to how managerial actions can ﬂow back to the design of MC
elements and how this affects the functionality of incoherent MC elements (cf. Hall, 2016). Pre-
vious research on coherence maintained that managers should strive for coherence among MC
elements in order to achieve higher levels of control effectiveness (Abernethy & Chua, 1996;
Bedford et al., 2016; Friis et al., 2015; Fullerton et al., 2013; Kristensen & Israelsen, 2014;
Scapens, 1994; Widener, 2007). We nuance this position by highlighting that there may also be
positive (side) effects of incoherence. Our data suggest that if managers are indeed aware of the
existence of incoherence, this can motivate them to take action. More speciﬁcally, in the case
of MULTICORP, it inspired managers to use the current performance measurement system dif-
ferently – the ﬁrst phase of the cascade effect. In essence, managers changed their use from a
basic ‘variance analysis’ tool into a platform for learning and sharing knowledge about poten-
tial solutions, more in line with MULTICORP’s new supply chain objectives (cf. Melnyk et al.,
2014; Pinheiro De Lima et al., 2013). Thus, the initial incoherence triggered positive outcomes
through managers’ choices, such as generating more organizational dialogue about alignment
issues within the supply chain. This ﬁnding resembles research that studied how the imperfec-
tions of a performance measurement system ‘helped to provide a fertile arena for productive
dialogue and discussion’ (Chenhall et al., 2013, p. 282). Our ﬁndings can also initiate a more
in-depth discussion on the concept of coherence, and whether striving for coherent MC elements
is indeed necessary. Prior literature has already pointed to tensions among MC elements that
need managerial attention (Simons, 1995; Mundy, 2010; Curtis & Sweeney, 2017; van der Kolk
et al., 2020), and in line with such studies we ﬁnd that incoherence in itself does not per se
produce organizational impasse, i.e. when handled well by managers. Thus, discussions related
to (in)coherence must also grasp how speciﬁc MC elements are used and not just whether MC
elements are designed (Abernethy & Chua, 1996; Ferreira & Otley, 2009). Speciﬁcally, we show
that incoherence can create more dialogue, which allows organization members to ‘sort things
out’ and get to the root(s) of problems, hence stimulating innovative problem solving. The use of
the performance measurement system changed into serving as a learning platform, and to a lesser
extent as a device for feedback and monitoring. For such situations to materialize it is essential
that managers encourage open communication and that members of the organization can engage
constructively in such discussions. A sole focus on the negative implications of incoherence
would have obscured its potential to trigger positive (side) effects.
The observation that skillful management is required also speaks to the MC change literature.
While the MC change literature acknowledges that MC change requires skillful management
within that locus (Malmi, 1997), we add that it also requires skillful management outside this
locus. Speciﬁcally, managerial actions outside the original change locus are needed to support
the original MC change and ensure implementation of the ‘new’ MC elements. Although we
cannot derive this directly from our study, it seems that if managers would not have smoothened
the change process by adjusting their use of the ‘old’ MC elements (i.e. the performance system),
tensions and conﬂicts could have increased as this would have made the incoherence more clear.
The resulting organizational dialogue facilitated higher levels of performance at MULTICORP.
Cascading Control Changes, Incoherence, and Dialogue 399
This ﬁnding extends the literature by highlighting how incoherence, triggered by cascading MC
changes, does not per se have to result in organizational impasse through ineffective and dys-
functional control. On the contrary, the incoherent MC elements can trigger managerial actions
that can eventually result in positive organizational outcomes.
This paper examines how MC change can create incoherence, which may subsequently trig-
ger other MC changes, and how managers can use this situation to facilitate dialogues among
organization members. We introduced notions from organizational ecology – cascading change,
cultural asperity and structural opacity – that allow a better understanding of the relationship
between MC and incoherence. The case of MULTICORP demonstrates that it can take years
before the effects of an initial change become visible and result in new changes. During this
time span, new and old MC elements coexisted – even though they were incoherent. Yet, the
incoherent MC elements did not result in an organizational impasse, but instead, managers saw a
possibility to change their use of the performance measurement system, in an effort to facilitate a
dialogue among organizational members. Our paper extends the MC change literature and helps
to explain why not only the study of MC elements, but also the study of MC change, beneﬁts
from a more holistic approach that takes the wider organizational context into account (cf. Malmi
Although our longitudinal case study provided in-depth insights into the ways in which
accounting and control elements operate in their contexts (Hopwood, 1983), the ﬁndings and
conclusions of our study should be seen in the light of the limitations of the used research method.
One limitation regards the study’s timeframe; although the organization was studied for a long
time, we also refer to events such as the great ﬁnancial crises that happened before the start of
this study. Therefore, we also had to rely on retrospective interviews and company documents
regarding (the effects of) these events before we entered the case organization. As suggested by
Golden-Biddle and Locke (1993), people may imperfectly recall their actions or what exactly led
them from one action to the next. To address this potential retrospective bias, we followed estab-
lished and robust methods to triangulate the obtained data with other sources, such as company
documents, notes from meetings and corroboration of certain aspects with other interviewees.
We studied one part of one organization for ﬁve years, which allowed us to examine managerial
responses in detail and identify mechanisms that managers used to mitigate incoherence among
MC elements. It may, however, be the case that some of (the effects of) the observed mecha-
nisms are contingent on organizational factors, which is why we provided a detailed description
of MULTICORP’s characteristics and the relevant organizational events.
Future research may further investigate the role of managers in the MC change process and
examine and compare the cascade effect across different settings (e.g. different departments,
organizations, or sectors). Such a study would enable a more robust identiﬁcation of the factors
(e.g. a manager’s educational or cultural background or an organization’s ﬁnancial history or
prospects) that affect the choice to use speciﬁc MC elements differently. Furthermore, our case
illustrated how a change process unfolds when cultural asperity between MC elements is high
and where structural opacity limits the a priori knowledge about changes. Future research could
further explore how, for instance, change processes take shape if cultural asperity is high, while
structural opacity is relatively low. Under such circumstances, managers know which changes are
required to deal with the incoherence. It would particularly be interesting to learn about tradeoffs
that managers face in such situations, addressing questions such as: ‘Would managers make all
the MC changes simultaneously?’ and ‘Is (part of) the reconﬁguration process still sequential?’.
400 T. Toldbod and B. Van der kolk
Although there are reasons to assume that the answer to the latter question would be positive – for
instance, because of the high level of stress associated with impactful organizational changes that
are made at the same time (Dahl, 2011) – more in-depth research in this area has the potential
to enhance our understanding of the effects of managerial (micro) interventions and improve
MC-related decisions in the future.
The authors would like to thank Salvador Carmona, Matthew Hall, Robin Roslender, Thomas Borup Kristensen and
Henrik Nielsen, for their constructive comments on earlier versions of this paper. We would also like to thank Teemu
Malmi and the two anonymous reviewers for their helpful feedback. Earlier versions of this paper were presented at the
9th EIASM Performance Measurement conference in Nice, the 12th Manufacturing Accounting Research Conference
in Copenhagen and at a research seminar at Aalborg University. The authors thank the participants for their valuable
Thomas Toldbod http://orcid.org/0000-0001-8800-242X
Berend Van Der Kolk http://orcid.org/0000-0003-2011-4585
Abernethy, M. a., & Chua, W. F. (1996). A ﬁeld study of control system “redesign”: The impact
of Institutional processes on strategic choice. Contemporary Accounting Research,13(2), 569–606.
https://d oi.org /10.1111/j.1911-3846.1996.tb00515.x
Amburgey, T. L., Kelly, D., & Barnett, W. P. (1993). Resetting the clock: The dynamics of organizational change and
failure. Administrative Science Quarterly,38(1), 51–73. https://doi.org/10.2307/2393254
Ashraf, J., & Uddin, S. (2015). Military, “managers” and hegemonies of management accounting controls: A critical
realist interpretation. Management Accounting Research,29, 13–26. https://doi.org/10.1016/j.mar.2015.07.002
Baines, A., & Langﬁeld-Smith, K. (2003). Antecedents to management accounting change: A structural equation
approach. Accounting, Organizations and Society,28(7–8), 675–698. https://doi.org/10.1016/S0361-3682(02)
Barnett, W. P., & Carroll, G. R. (1995). Modeling internal organizational change. Annual Review of Sociology,21(1),
Baxter, J. a., & Chua, W. F. (1998). Doing ﬁeld research: Practice and meta-theory in counterpoint. Journal of
Management Accounting Research,10, 69–87.
Beaubien, L. (2012). Technology, change, and management control: A temporal perspective. Accounting, Auditing and
Accountability Journal,26(1), 48–74. https://doi.org/10.1108/09513571311285612
Bedford, D. S., Malmi, T., & Sandelin, M. (2016). Management control effectiveness and strategy: An empirical analysis
of packages and systems. Accounting, Organizations and Society,51, 12–28. https://doi.org/10.1016/j.aos.2016.
Burns, J. (2000). The dynamics of accounting change inter-play between new practices, routines, institutions, power
and politics. Accounting, Auditing & Accountability Journal,13(5), 566–596. https://doi.org/10.1108/09513570010
Burns, J., & Vaivio, J. (2001). Management accounting change. Management Accounting Research,12(4), 389–402.
Busco, C., Quattrone, P., & Riccaboni, A. (2007). Management accounting. Issues in interpreting its nature and change.
Management Accounting Research,18(2), 125–149. https://doi.org/10.1016/j.mar.2007.04.003
Chenhall, R. (2003). Management control systems design within its organizational context: Findings from contingency-
based research and directions for the future. Accounting, Organizations and Society,28(2–3), 127–168.
Chenhall, R. H. (2006). Theorizing contingencies in management control systems research. Handbooks of Management
Accounting Research,1, 163–205. https://doi.org/10.1016/S1751-3243(06)01006-6
Chenhall, R. H., Hall, M., & Smith, D. (2013). Performance measurement, modes of evaluation and
the development of compromising accounts. Accounting, Organizations and Society,38(4), 268–287.
Collier, P. M. (2001). The power of accounting: A ﬁeld study of local ﬁnancial management in a police force.
Management Accounting Research,12(4), 465–486. https://doi.org/10.1006/mare.2001.0157
Cascading Control Changes, Incoherence, and Dialogue 401
Curtis, E., & Sweeney, B. (2017). Managing different types of innovation: Mutually reinforcing management con-
trol systems and the generation of dynamic tension. Accounting and Business Research,47(3), 313–343.
Daft, R. L., & Macintosh, N. B. (1984). The nature and use of formal control systems for management control and
strategy implementation. Journal of Management,10(1), 43–66. https://doi.org/10.1177/014920638401000105
Dahl, M. S. (2011). Organizational change and employee stress. Management Science,57(2), 240–256.
Denis, J.-L., Lamothe, L., & Langley, A. (2001). The dynamics of collective leadership and strategic change in pluralistic
organizations. The Academy of Management Journal,44(4), 809–837. https://doi.org/10.5465/3069417
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review,14(4), 532–550.
Ferreira, A., & Otley, D. (2009). The design and use of performance management systems: An extended framework for
analysis. Management Accounting Research,20(4), 263–282. https://doi.org/10.1016/j.mar.2009.07.003
Flamholtz, E. G. (1983). Accounting, budgeting and control systems in their organizational context: Theoretical and
empirical perspectives. Accounting, Organizations and Society,8(2/3), 153–169. https://doi.org/10.1016/0361-36
Flamholtz, E. G., Das, T. K., & Angeles, L. (1985). Toward an integrative framework of organizational control.
Accounting, Organizations and Society,10(1), 35–50. https://doi.org/10.1016/0361-3682(85)90030-3
Flick, U. (2009). An introduction to qualitative research ( (4th ed.). Sage Publications.
Frey, B. S., Homberg, F., & Osterloh, M. (2013). Organizational control systems and pay-for-performance in the public
service. Organization Studies,34(7), 949–972. https://doi.org/10.1177/0170840613483655
Friis, I., Hansen, A., & Vámosi, T. (2015). On the effectiveness of incentive pay: Exploring complementarities and
substitution between management control system elements in a manufacturing ﬁrm. European Accounting Review,
24(2), 241–276. https://doi.org/10.1080/09638180.2014.976055
Fullerton, R. R., Kennedy, F. A., & Widener, S. K. (2013). Management accounting and control prac-
tices in a lean manufacturing environment. Accounting, Organizations and Society,38(1), 50–71.
Galbraith, J. (1973). Designing complex organizations. The European Institute for advanced studies in management.
Gerdin, J., & Greve, J. (2004). Forms of contingency ﬁt in management accounting research - A critical review.
Accounting, Organizations and Society,29(3-4), 303–326. https://doi.org/10.1016/S0361-3682(02)00096-X
Golden-Biddle, K., & Locke, K. (1993). Appealing work: An investigation of how ethnographic texts convince.
Organization Science,4(4), 595–616. https://doi.org/10.1287/orsc.4.4.595
Grabner, I., & Moers, F. (2013). Management control as a system or a package? Conceptual and empirical issues.
Accounting, Organizations and Society,38(6–7), 407–419. https://doi.org/10.1016/j.aos.2013.09.002
Granlund, M. (2001). Towards explaining stability in and around management accounting systems. Management
Accounting Research,12(2), 141–166. https://doi.org/10.1006/mare.2000.0151
Hall, M. (2016). Realising the richness of psychology theory in contingency-based management accounting research.
Management Accounting Research,31, 63–74. https://doi.org/10.1016/j.mar.2015.11.002
Hannan, M. T., & Freeman, J. (1984). Structural inertia and organizational change. American Sociological Review,49(2),
Hannan, M. T., Pólos, L., & Carroll, G. R. (2003a). Cascading organizational change. Organization Science,14(5),
Hannan, M. T., Pólos, L., & Carroll, G. R. (2003b). The fog of change: Opacity and asperity in organizations.
Administrative Science Quarterly,48(3), 399–432. https://doi.org/10.2307/3556679
Harrington, L. H., Boyson, S., & Corsi, T. M. (2011). X-SCM: The new science of X-treme supply chain management.
Henri, J. F. (2006). Management control systems and strategy: A resource-based perspective. Accounting, Organizations
and Society,31(6), 529–558. https://doi.org/10.1016/j.aos.2005.07.001
Henttu-aho, T., & Järvinen, J. (2013). A ﬁeld study of the emerging practice of beyond budgeting
in industrial companies: An institutional perspective. European Accounting Review,22(4), 765–785.
Hopwood, A. G. (1983). On trying to study accounting in the contexts in which it operates. Accounting, Organizations
and Society,8(2–3), 287–305. https://doi.org/10.1016/0361-3682(83)90035-1
Jansen, E. P. (2011). The effect of leadership style on the information receivers’ reaction to management accounting
change. Management Accounting Research,22(2), 105–124. https://doi.org/10.1016/j.mar.2010.10.001
Kasurinen, T. (2002). Exploring management accounting change: The case of balanced scorecard implementation.
Management Accounting Research,13(3), 323–343. https://doi.org/10.1006/mare.2002.0191
402 T. Toldbod and B. Van der kolk
Kouamé, S., & Langley, A. (2018). Relating microprocesses to macro-outcomes in qualitative strategy process and
practice research. Strategic Management Journal,39(3), 559–581. https://doi.org/10.1002/smj.2726
Kristensen, T. B., & Israelsen, P. (2014). Performance effects of multiple control forms in a lean organiza-
tion: A quantitative case study in a systems ﬁt approach. Management Accounting Research,25(1), 45–62.
Langley, A. (1999). Strategies for theorizing from process data. The Academy of Management Review,24(4), 691–710.
Langley, A., Smallman, C., Tsoukas, H., & Van de Ven, A. H. (2013). Process studies of change in organiza-
tion and management: Unveiling temporality, activity and ﬂow. Academy of Management Journal,56(1), 1–13.
Malmi, T. (1997). Towards explaining activity-based costing failure: Accounting and control in a decentralized
organization. Management Accounting Research,8(4), 459–480. https://doi.org/10.1006/mare.1997.0057
Malmi, T. (2013). Management control as a package — The need for international research. Journal of Management
Control,23(4), 229–231. https://doi.org/10.1007/s00187-013-0169-z
Malmi, T., Bedford, D. S., Brühl, R., Dergård, J., Hoozée, S., Janschek, O., Willert, J., Ax, C., Bednarek, P., Gosselin,
M., Hanzlick, M., Israelsen, P., Johanson, D., Johanson, T., Madsen, D. Ø., Rohde, C., Sandelin, M., Strömsten,
T., & Toldbod, T. (2020). Culture and management control interdependence: An analysis of control choices that
complement the delegation of authority in Western cultural regions. Accounting, Organizations and Society,In
Malmi, T., & Brown, D. a. (2008). Management control systems as a package-opportunities, challenges and research
directions. Management Accounting Research,19(4), 287–300. https://doi.org/10.1016/j.mar.2008.09.003
Marginson, D. E. W. (2002). Management control systems and their effects on strategy formation at middle-
management levels: Evidence from a UK organisation. Strategic Management Journal,23(11), 1019–1031.
Melnyk, S. A., Bititci, U., Platts, K., Tobias, J., & Andersen, B. (2014). Is performance measurement and management
ﬁt for the future? Management Accounting Research,25(2), 173–186. https://doi.org/10.1016/j.mar.2013.07.007
Merchant, K. a. (1982). The control function of management. Sloan Management Review,23(4), 43–55.
Merchant, K. a., & Van der Stede, W. a. (2007). Management control systems performance measurement,evaluation and
incentives (2nd ed.). Prentice Hall.
Moores, K., & Yuen, S. (2001). Management accounting systems and organizational conﬁguration: A life-cycle perspec-
tive. Accounting, Organizations and Society,26(4–5), 351–389. https://doi.org/10.1016/S0361-3682(00)00040-4
Mundy, J. (2010). Creating dynamic tensions through a balanced use of management control systems. Accounting,
Organizations and Society,35(5), 499–523. https://doi.org/10.1016/j.aos.2009.10.005
Naranjo-Gil, D., & Hartmann, F. (2007). Management accounting systems, top management team heterogeneity and
strategic change. Accounting, Organizations and Society,32(7–8), 735–756. https://doi.org/10.1016/j.aos.2006.
Otley, D. (2016). The contingency theory of management accounting and control: 1980-2014. Management Accounting
Otley, D. (1980). The contingency theory of management accounting: Achievement and prognosis. Accounting,
Organizations and Society,5(4), 413–428. https://doi.org/10.1016/0361-3682(80)90040-9
Østergren, K., & Stensaker, I. (2011). Management control without budgets: A ﬁeld study of ‘beyond budgeting’ in