Management Control Systems
as a Package and its role of
enabling Organizational
Ambidexterity in SMEs
An exploratory study
MASTER THESIS WITHIN: Business Administration NUMBER OF CREDITS: 30 ECTS
PROGRAMME OF STUDY: Civilekonom AUTHORS: Hanna Åberg & Richard Lind TUTOR: Timur Uman
Acknowledgments
This thesis marks the end of a journey after almost four years of business administration stud-ies at Jönköping University. An interesting journey may oftentimes be defined by the people that you meet along the way, which is why we would like to express our gratitude towards
certain individuals that have been part of this final project.
First of all, we would like to thank our tutor Timur Uman, who has been infallible in his abil-ity to provide us with guidance during times when it was needed the most. Timur has, with his extensive knowledge in accounting and management control, been a major asset to this thesis.
We would also like to thank External Relations Coordinator Emil Karlsson, who introduced us to the companies that were part of this study. With that said, we would also like to thank the people that participated in our interviews, who were generous and kind enough to share their valuable insights with us. It was nothing but a pleasure to interact with these people and
learn more about their organizations.
Master Thesis in Business Administration
Title: Management Control Systems as a Package and its role of enabling Organizational Am-bidexterity in SMEs
Authors: Hanna Åberg & Richard Lind Tutor: Timur Uman
Date: 2020-05-18
Key terms: Management Control Systems, Managemetn Control Systems as a Package, Levers of Control, Organizational Ambidexterity, Exploration, Exploitation, SMEs
Abstract
Background – A firm that achieves organizational ambidexterity is managing the balancing act of refining existing capabilities while also discovering new opportunities in the marketplace. It has shown that the achievement of organizational ambidexterity will lead to better perfor-mance but also long-term survival. The challenge that comes with achieving organizational ambidexterity is present within all firms regardless of size. Yet, extant literature has focused on the research of larger firms. The characteristics may differ significantly between a larger enter-prise and a small-to-medium-sized enterenter-prise. For instance, the amount of available resources appears to differ relative to firm size. As a result, it is reasonable to assume that the way organ-izational ambidexterity is achieved may also differ depending on the size of the firm.
It is argued that management control systems play an important role with respect to organiza-tional ambidexterity. Further, these control systems may be viewed as a package of controls which through occurring dynamic tensions, provide the complexity needed for a firm to be ambidextrous. Yet, the role of these interrelated control systems and how they enable ambidex-terity is understudied among SMEs.
Purpose – This thesis aims to explore management control systems as a package and its ena-bling role in achieving organizational ambidexterity within SMEs.
Method – The empirical data was collected by interviewing members of the top management team within three different SMEs. The empirics were gathered through semi-structured inter-views and then analyzed in relation to reviewed theory within the package approach of man-agement control systems, by conducting a thematic analysis.
Findings – The findings from this thesis suggest that SMEs need to find ways to allocate their finite resources in order to meet the current strategic challenge at hand. SMEs have a firm spe-cific focus on either refinement of existing capabilities or discovery of new opportunities de-pending on the current state of the marketplace. While both focuses may be present simultane-ously, the SMEs still need to prioritize one side over the other due to resource limitations. Hence, the findings suggest a hybrid between contextual- and sequential ambidexterity as the most appropriate narrative for the studied SMEs. Management control systems as a package fits into this narrative as tensions occur between the systems which triggers ambidextrous activities. A well-integrated package approach work as a navigation system and may thus facilitate the achievement of current strategic challenges. The finalized framework in this thesis comple-ments existing theory by suggesting two different scenarios occurring within SMEs. In order for SMEs to achieve organizational ambidexterity, a shift between these two scenarios occurs in the long run.
Table of Contents
1
Introduction ... 3
1.1 Background ... 3
1.2 Problematization ... 4
1.3 Objective and Research Question ... 7
1.4 Delimitations ... 7 1.5 Keywords ... 8
2
Literature Review ... 9
2.1 Organizational Ambidexterity ... 9 2.1.1 Exploration ... 10 2.1.2 Exploitation ... 102.1.3 Balancing Exploration and Exploitation ... 11
2.1.4 Structural Ambidexterity ... 12
2.1.5 Contextual Ambidexterity ... 13
2.1.6 Sequential Ambidexterity ... 13
2.2 Organizational Ambidexterity in SMEs ... 14
2.3 Management Control Systems ... 15
2.4 Management Control Systems as a Package ... 16
2.4.1 Reasons to study MCSPs ... 16
2.5 MCSPs in SMEs ... 17
2.6 Conceptual frameworks for studying MCSPs ... 18
2.6.1 Malmi & Brown (2008) ... 18
2.6.2 Merchant & Van der Stede (2012) ... 21
2.6.3 Simons (1995) ... 23
2.7 Motivation for Chosen Package Approach (Simons 1995) ... 25
2.8 Tensions and interrelations between the four systems ... 27
2.8.1 Belief Systems and Interactive Systems ... 27
2.8.2 Interactive Systems and Diagnostic Systems ... 27
2.8.3 Boundary Systems and Diagnostic Systems ... 28
2.8.4 Boundary Systems and Belief Systems ... 28
2.8.5 Belief Systems and Diagnostic Systems ... 28
2.8.6 Interactive Systems and Boundary Systems ... 29
2.9 MCSPs and Organizational Ambidexterity in SMEs ... 29
2.9.1 Belief Systems and Organizational Ambidexterity ... 31
2.9.2 Interactive Systems and Organizational Ambidexterity ... 31
2.9.3 Boundary Systems and Organizational Ambidexterity ... 32
2.9.4 Diagnostic Systems and Organizational Ambidexterity ... 33
2.10 Tentative Model ... 33
3
Methodology ... 35
3.1 Research Perspective ... 35 3.2 Research Purpose ... 36 3.3 Research Approach ... 36 3.4 Method ... 38 3.4.1 Choice of Methodology ... 38 3.4.2 Theoretical Approach ... 38 3.4.3 Source Criticism ... 393.4.4 Data collection ... 39 3.4.5 Interview Guide ... 40 3.4.6 Selection of sample ... 41 3.4.7 Semi-structured interviews ... 42 3.4.8 Procedure ... 43 3.4.9 Participants ... 44 3.4.10 Data Analysis ... 45 3.5 Data Quality ... 45 3.5.1 Credibility ... 45 3.5.2 Transferability ... 46 3.5.3 Dependability ... 46 3.5.4 Confirmability ... 47 3.6 Ethical Considerations ... 47
4
Empirical Findings ... 49
4.1 Theme statements ... 49 4.1.1 Resource Allocation ... 49 4.1.2 Belief Systems ... 49 4.1.3 Interactive Systems ... 49 4.1.4 Boundary Systems ... 50 4.1.5 Diagnostic Systems ... 50 4.2 Resource Allocation ... 50 4.3 Belief Systems ... 51 4.4 Interactive Systems ... 52 4.5 Boundary Systems ... 54 4.6 Diagnostic Systems ... 565
Analysis ... 59
5.1 Control Systems Identified within the SMEs from the Empirics ... 62
5.2 Scenario 1: Exploration as primary focus ... 66
5.3 Scenario 2: Exploitation as primary focus ... 68
6
Discussion ... 70
6.1 Implications ... 70
6.2 Contribution to Literature ... 71
6.3 Limitations ... 72
6.4 Suggestions for Future Research ... 73
7
Conclusion ... 75
8
References ... 77
9
Appendix ... 85
1 Introduction
In the first chapter, a background of the research field is presented, also the need for empirical studies of MCSP in SMEs is examined. To create a better understanding of the resrarch, a description of the research problem and research purpose is presented. Lastly, the delimitations of this study is described.
1.1 Background
Have you ever heard of the Roman God Janus who was gifted with two sets of eyes? Janus, also known as the God of beginnings, was able to view the past while concurrently looking into the future. Today, Janus’s skills would have been a desirable asset for enterprises that constantly tries to look backward, attending to the exploitation of their existing product portfolios, while simultaneously gazing forward, exploring what new innovations that may define the future. Do not be too hard on yourself if you were not familiar with the Roman God Janus. It is perhaps equally improbable that a person born in the late 1990s have heard of the American photo-graphic equipment company Kodak. Kodak used to be a giant within the photophoto-graphic industry before the turn of the millennium. Many people would later be surprised when Kodak ended up filing for bankruptcy protection in January year 2012. Kodak fell victim for what most compa-nies, especially the ones in the digital sector, are struggling with. Namely, to constantly keep reinventing themselves in a fast-moving market. One could argue that Kodak failed to make a transition into the digitalization because they failed with the balancing act of exploration and exploitation.
There is no one-size-fits-all with regards to the balance of exploratory and exploitative actions within a firm. For example, an automobile workshop that provides the service of car repairs do not necessarily have to keep their main focus on what March (1991) calls notions of exploration, namely “search, variation, experimentation, and discovery” (March, 1991, p. 102). Instead, it would make more sense to prioritize exploitative activities such as “refinement, efficiency, se-lection, and implementation” (March, 1991, p. 102). Further, when instead examining a new tech start-up which by nature is subject to a higher risk of failure, one could see how their prioritizations with respect to exploration and exploitation would differ significantly from the automobile workshop.
Small and medium-sized enterprises (SMEs), just like larger firms, need to jointly engage in exploration and exploitation. However, one aspect that differentiates SMEs from larger firms
is the relatively lower amount of slack resources within SMEs, meaning that they simply cannot afford to have business units that focus solely on either exploration or exploitation (Lubatkin, Simsek, Ling, & Veiga, 2006). Yet, it is important for SMEs to achieve organizational ambi-dexterity, also known as the balance of exploration and exploitation, as it has shown to generate higher performance and long-term survival (Mihalache, Jansen, Van Den Bosch, & Volberda, 2014). It is argued that achieving organizational ambidexterity requires an element of complex-ity which creates a tension. For instance, it could be the importance of the top management team leadership (Mihalache et al., 2014), or the impact from the CEO (Cao, Simsek, & Zhang, 2010), while others highlight the importance of management control systems (Bedford, 2015; Haustein, Luther, & Schuster, 2014; McCarthy & Gordon, 2011). With that said, scholars seem to have different views on what actually creates the complexity to achieve organizational am-bidexterity and thus, it remains a debated topic within the research field.
1.2 Problematization
The measurement of business performance has engaged both managers and management ac-counting academics for years. However, it seems that the main focus among scholars has been on researching financial performance. The amount of available resources affects the firms’ per-formance, and the ability to balance the allocation of available resources into exploratory and exploitative activities is known as organizational ambidexterity (March, 1991). The amount of resources available may differ significantly between larger firms and SMEs (Lubatkin et al., 2006), which intuitively would suggest different approaches with respect to organizational strategy, depending on the size of the firm. However, almost all research on ambidexterity thus far is designed for larger firms (Soto-Acosta, Popa, & Martinez-Conesa, 2018).
SMEs are not only different from larger firms when it comes to resource allocation, but may also be relatively less bureaucratic and structured (Forbes & Milliken, 1999), and tend to have fewer and less formal control systems (Busenitz & Barney, 1997). Further, it is argued that SMEs, compared with larger firms, deal with more challenges when it comes to managing trade-offs related to exploratory and exploitative innovations (Andriopoulos & Lewis, 2009). As a result, “generalizing current prescriptions for ambidexterity into innovation strategies for these firms [SMEs] might prove incorrect, inappropriate or dangerous” (Chang & Hughes, 2012, p. 2). With that said, one could argue that we do not know how the characteristics of SMEs affect
the dynamics behind organizational ambidexterity, and how the balancing act of exploration and exploitation may differ in SMEs from larger firms.
Even though control systems in SMEs may be less formal in contrast with larger firms (Busenitz & Barney, 1997), they do interrelate with the ability to achieve organizational ambidexterity, which also is true for larger firms (Bedford, 2015; McCarthy & Gordon, 2011). Management control systems used to be viewed by scholars as an inhibiting factor to the innovative efforts of a company (Amabile, 1988; Damapour, 1991). This association of innovative capacity sup-pressed exploration and instead incited exploitation leading to a less ambidextrous organization. However, during recent years that notion seems to have shifted as it is argued that management control systems may instead function as a triggering act to achieve organizational ambidexterity (Bedford, 2015; Haustein et al., 2014; McCarthy & Gordon, 2011). Management Control Sys-tems (MCSs), defined as “the devices or sysSys-tems managers use to ensure that the behaviors and decisions of their employees are consistent with the organization’s objectives and goals” (Merchant, 2007, p. 4), is considered a well-studied topic within the research field of accounting (Malmi, 2013). Although, thus far the main focus within that research has been on studying MCSs in isolation, rather than viewing MCSs integrated in what Simons (1995) call a package approach.
The idea of viewing MCSs as different control systems interacting within a package was first presented by Otley (1980). Even though, the package approach has been mentioned in a range of studies within management accounting research since, it remains an understudied topic (Malmi & Brown, 2008). Further, Malmi and Brown (2008) emphasize the importance of stud-ying MCSs as a package since “…MCS do not operate in isolation”, and thus it could help us to learn more about “…how to design MCS in order to produce desired outcomes” (Malmi & Brown, 2008, pp. 287, 288). Sandelin (2008) together with Kennedy and Widener (2008), share that view in their findings and further expands on the significance of looking at MCSs interde-pendently. It is argued that an examination of a single control system without taking the man-agement control system package (MCSP) into account, carries the risk of leading to incorrect findings within that particular system (Kennedy & Widener, 2008; Malmi & Brown, 2008; Sandelin, 2008). Scholars have come up with different theoretical frameworks with regards to MCSP (Malmi & Brown, 2008; Merchant, 2007; Simons, 1995). Previous studies have shown that MCSs interplay within an MCSP (Kennedy & Widener, 2008; Malmi & Brown, 2008;
Sandelin, 2008). However, there is a lack of studies addressing the issue of how the interrelation between MCSs occurs within an MCSP. According to Malmi and Brown (2008), the concept of management control systems as a package have barely been studied empirically even though it has been around for many years. It is argued that “With more refined conceptual and analyt-ical approaches, case studies could be conducted in a range of theoretanalyt-ically different contexts” (Malmi & Brown, 2008, p. 289), which then could increase the understandings of MCSPs in a practical context (Malmi & Brown, 2008).
According to Gschwantner and Hiebl (2016), a package approach of MCSs may be required to achieve and manage organizational ambidexterity in a successful manner. In Simons (1995), four different levers were presented constituting a package: Belief systems, Interactive systems, Boundary systems and Diagnostic systems. When it comes to interactive controls, it shows that it can support the balancing counteraction of exploration and exploitation (McCarthy & Gor-don, 2011). Interactive controls may enable exploration by facilitating “open channels of com-munication and free flow of information”, which provides employees with opportunities to ex-plore new ideas proactively (Ylinen & Gullkvist, 2014, p. 94). At the same time, McCarthy and Gordon (2011) argue that exploitation may be achieved through the alignment of an organiza-tion’s objectives and values with the employees’ actions. Mundy (2010) suggests that counter-vailing controls such as interactive systems, the fostering interaction between employees, and diagnostic systems, the measurement of goal fulfillment, can in combination trigger organiza-tional ambidexterity. Further, boundary systems being used to assess employees’ performance in reward and compensation controls, may show to be detrimental for the organization’s ambi-dexterity as employees might prioritize measures that favors their own performance evaluation rather than the firm’s objectives (Grafton, Lillis, & Widener, 2010). While tight boundary sys-tems may provide the stable foundation and exploitative direction needed in new product de-velopment, combined with loose belief systems, it may improve exploration and hence the achievement of organizational ambidexterity (Ylinen & Gullkvist, 2014).
Organizational ambidexterity was studied in Mihalache et al. (2014) where smaller firms showed to be dependent on neutralizing tensions between exploration and exploitation for long-term survival and superior firm performance. By looking at a firm’s organizational ambidex-terity and how management control systems as a package are enabling exploration and exploi-tation, one would further study the functionalities of MCSPs within firms. Additionally, the
focus of research would move beyond the traditional view of examining financial performances and look more into non-conventional and unexplored territory. Examining MCSPs within smaller firms have proven to be of high relevance since the control systems tend to have more of an informal characteristic (Medcof & Song, 2013; Sandelin, 2008), and thus apply well in an SME-context. Moreover, Gschwantner and Hiebl (2016) state that “little research on how management control systems support and influence organizational ambidexterity has thus far been conducted” (Gschwantner & Hiebl, 2016, p. 372). Further, Gschwantner and Hiebl (2016) is calling for “…more research investigating under which conditions and circumstances man-agement control systems or combinations of such systems can foster organizational ambidex-terity” (Gschwantner & Hiebl, 2016, p. 373). By studying the tensional circumstances between MCSPs and organizational ambidexterity in an unexplored business environment like SMEs, this study is seeking to explore these ideas further. Based on these insights this study aims at exploring how management control systems as a package, within small and medium-sized en-terprises, may deflect firms ability to balance exploration and exploitation as strategic out-comes.
1.3 Objective and Research Question
The purpose of this study is to explore the role of MCSP in enabling organizational ambidex-terity within SMEs.
Research Question: What is the role of MCSP in enabling organizational ambidexterity in SMEs?
1.4 Delimitations
This thesis has been delimited to focus on an SME-context. The primary focus on smaller firms narrows down the scope of the study and aims at exploring the context of SMEs, which is an understudied area within the existing literature. The companies that are represented in the em-pirical findings are all located in Sweden within the region of Småland. This thesis has also focused on one specific strategic approach, namely organizational ambidexterity. The focus on one specific strategic approach narrows down the scope even further. Given the relatively short time frame of this thesis, it was not viewed as a feasible alternative to conduct a longitudinal study, even though it could have generated interesting obeservations.
1.5 Keywords
2 Literature Review
This section presents the existing literature within the research field. All research is examined separately. First, organizational ambidexterity is described followed by an examination of different conceptual frameworks to explain MCS and MCSP. The chapter concludes by combining the different themes by presenting a tentative model.
2.1 Organizational Ambidexterity
The term organizational ambidexterity can be traced back all the way to the mid-1970s where Duncan (1976) was introducing the concept for the first time. However, it was the landmark article published by March (1991) on exploration and exploitation that was going to spark real interest on the topic among scholars (Raisch & Birkinshaw, 2008). Reason being that up until this point scholars viewed the simultaneous achievement of exploration and exploitation to be practically impossible (Hannan & Freeman, 1977; McGill, Slocum, & Lei, 1992; Miller & Friesen, 1986). Conversely, March (1991) argued that the attainment of both exploration and exploitation was instead a balancing act which, if executed successfully, would reward firms with both long-term success and survival. Organizational ambidexterity may be defined as the simultaneous pursuit of exploratory and exploitative activities within an organization (C. A. O'Reilly, III & Tushman, 2013).
An ambidextrous organization manages to adhere to the alignment of current demands while adapting to changes that occur in the business environment of which the organization will re-main part of in the future (Gibson & Birkinshaw, 2004). The two most common terms that are used to describe the orientations of organizational ambidexterity is exploration and exploitation, and Levinthal and March (1993) present a definition of what an ambidextrous organization is about:
“The basic problem confronting an organization is to engage in sufficient exploitation to sure its current viability and, at the same time, to devote enough energy to exploration to
en-sure its future viability.” (Levinthal & March, 1993, p. 105).
So, is it impossible to achieve exploration and exploitation at the same time, and if not, when is an organization reaching sufficient exploitation while also devoting enough energy to explo-ration? An inherent issue that might need to be addressed to this discussion is that the existing
example, level of innovativeness within an organization. A sufficient level of radical innova-tiveness within company X might be an insufficient level of radical innovainnova-tiveness for company Y. In Simons (2010), he expands on what research to a large extent have omitted, namely the consideration of organizational context, and whether the context and structure of an organiza-tion affect the ability to balance exploraorganiza-tion and exploitaorganiza-tion. We will now go through the dif-ferent aspects that have been, and yet remains, part of the discussion on the topic of organiza-tional ambidexterity. Moreover, before looking at how to balance the activities of exploration and exploitation one may need to further look into how the literature defines these two activi-ties.
2.1.1 Exploration
In March (1991) exploration is linked with the following terms “search, variation, risk taking, experimentation, play, flexibility, discovery, innovation” (March, 1991, p. 71). Further, Andri-oupolos and Lewis (2009) suggest that exploration “entails the development of new knowledge, experimenting to foster the variation and novelty needed for more radical innovation” (Andri-opoulos & Lewis, 2009, p. 696). In He and Wong (2004), exploratory innovation is viewed as “technological innovation aimed at entering new product-market domains” (He & Wong, 2004, p. 483). In an SME-context, exploration requires the development and experimentation of new skills. These initiatives may be triggered by identifying new trends or opportunities in either new or existing market places (Lubatkin et al., 2006).
2.1.2 Exploitation
March (1991) associates exploitation with terms such as “refinement, choice, production, effi-ciency, selection, implementation, execution” (March, 1991, p. 71). Andrioupolos and Lewis (2009) argue that “exploitation hones and extends current knowledge, seeking greater effi-ciency and improvements to enable incremental innovation” (Andriopoulos & Lewis, 2009, p. 696). Moreover, He and Wong (2004) define exploitative innovation as “technological innova-tion activities aimed at improving existing product-market domains” (He & Wong, 2004, p. 483). In an SME-context, exploitation is about the refinement of existing competencies to in-crease the level of customer satisfaction and meet the current market conditions (Lubatkin et al., 2006).
2.1.3 Balancing Exploration and Exploitation
As mentioned earlier, older research thought of it as impossible to pursue exploration and ex-ploitation simultaneously (Hannan & Freeman, 1977; McGill, Slocum, & Lei, 1992; Miller & Friesen, 1986). However, since March (1991) claimed that understanding the relationship be-tween the two concepts was needed in order to gain long-term success and survival, other schol-ars have henceforth arrived at similar conclusions (He & Wong, 2004; Lubatkin et al., 2006; Tushman & O'Reilly, 1996). Nevertheless, researchers have brought up the issue that a trade-off may occur when an organization need to adhere to the allocation of scarce resources to balance between the exploratory and exploitative activities (Andriopoulos & Lewis, 2009; March, 1991; Simons, 2010). Raisch and Birkinshaw (2008) bring up another important aspect, namely that exploratory and exploitative activities may “require fundamentally different organ-izational structures, strategies, and contexts” (Raisch & Birkinshaw, 2008, p. 376). Further, He and Wong (2004) highlight circumstances such as “organic structures, loosely coupled systems, path breaking, improvisation, autonomy and chaos, and emerging markets and technologies” which are associated with exploration, while exploitation is related to “mechanistic structures, tightly coupled systems, path dependence, routinization, control and bureaucracy, and stable markets and technologies” (He & Wong, 2004, p. 481). Given the differences between circum-stances that surround exploration and exploitation, it becomes clear why Raisch and Birkinshaw (2008) suggest that structures, strategies and contexts may be organized differently depending on whether the organization is engaged in exploratory, exploitative, or perhaps both activities simultaneously. In the next sections, a further examination of three different approaches to-wards ambidexterity will be presented, which have been researched extensively over the two past decades within the research field.
One may ask why it is important to assure that exploration and exploitation occur simultane-ously. Especially since when an organization adheres only to exploitation it may actually expe-rience a short-term increase in performance. However, in a long-term perspective, it may be detrimental as the organization may struggle with keeping up with the environmental changes in the marketplace. In contrast, an organization that only pursuits exploration may respond well to changes in the marketplace but will instead struggle with returns on the capital invested (Raisch & Birkinshaw, 2008; Turner, Swart, & Maylor, 2013). Further, organizations have a tendency to focus more on exploitation as it brings certainty, efficiency and short-term benefits while exploration tends to cause uncertainty, inefficiency and costs (Hill & Birkinshaw, 2014;
C. A. O'Reilly, III & Tushman, 2013). According to Gschwantner and Hiebl (2016) a healthy balance of exploration and exploitation is crucial and, if managed properly, will generate suc-cess and survival in the long-term. When resources are scarce, finding a healthy balance be-comes even more important as organizations will have to make trade-off decisions (Cao, Gedajlovic, & Zhang, 2009). As a result, the merging of the learning modes of exploration and exploitation may be achieved through different structures of ambidexterity, namely structural-, contextual- and sequential ambidexterity which will now be examined further.
2.1.4 Structural Ambidexterity
The ideas of structural ambidexterity go back, once again, to Duncan (1976) who stressed the idea of organizations managing trade-offs between exploration and exploitation by implement-ing dual structures. Duncan (1976) did not explicitly term this phenomenon as structural ambi-dexterity but he was going to lay the foundation for Tushman and O’Reilly (1996), who was the first to conceptualize it. Structural ambidexterity may be defined as “the separation of ex-ploratory and exploitative activities through a dual structure in the organization” (Gschwantner & Hiebl, 2016, p. 374). For example, in practice one may often see the R&D and business development units being responsible for the exploratory activities, while the core business units deal with exploitative activities (Birkinshaw & Gibson, 2004). This separation of activities is oftentimes seen as necessary given the comparative disparity between them. As a result, it pro-vides employees who work within these separate business units with clearly articulated objec-tives. At the same time, an inherent risk occurs of business units being isolated due to the dual structuring (Birkinshaw & Gibson, 2004). However, there are ways to mitigate that risk by making sure that exchanges of information on exploration and exploitation occur continuously between the units (McCarthy & Gordon, 2011).
One may ask to what extent different business units should be integrated, which not very sur-prisingly is a debated topic within the research field. Once again, the unique context of an or-ganization needs to be part of the discussion when in this case looking at levels of integration between units. But also as Raisch and Birkinshaw (2008) points out, the importance of coordi-nation among top executives together with a strong culture. Tushman and O’Reilly (1996) ex-pand on this by arguing that organizational ambidexterity cannot be achieved by only adhering to structural ambidexterity alone. Organizations need to create tensions between the exploratory and exploitative business units through leadership processes and control mechanisms (C.
O'Reilly & Tushman, 2004). Emphasis may be put on, for example, cultural controls and reward and compensations controls to align the exploratory and exploitative activities with an organi-zation’s values and objectives (Gschwantner & Hiebl, 2016; McCarthy & Gordon, 2011). It seems like communication plays an important part when applying a structural approach, as the organization would try to avoid unit isolation between exploratory- and exploitative activities to increase organizational ambidexterity (McCarthy & Gordon, 2011).
2.1.5 Contextual Ambidexterity
Contextual ambidexterity was first brought up in Gibson and Birkinshaw (2004) where they define it as "…the behavioural capacity to simultaneously demonstrate alignment and adapta-bility across an entire business unit” (Gibson & Birkinshaw, 2004, p. 209). This approach pro-vides the employees with a higher degree of autonomy as they oftentimes will make trade-off decisions between exploratory and exploitative activities without supervision from managers (Gibson & Birkinshaw, 2004). Accordingly, the business units tend to be more integrated from the beginning which would decrease the likelihood of isolation. However, the relatively high level of employee autonomy comes at a cost, namely that the organization has to implement suitable control systems that will assure alignment of the organization’s values and objectives throughout both exploratory and exploitative activities (McCarthy & Gordon, 2011). Contex-tual ambidexterity should not be viewed as a substitute for structural ambidexterity, rather Raisch, Birkinshaw, Probst and Tushman (2009) consider it as necessary to combine the two in order to achieve sustained organizational ambidexterity. This challenges the organizations since the two structures may require different control systems (Simons, 2010). With that said, organ-izations may be able to cope with these tensions by implementing appropriate control systems to achieve organizational ambidexterity.
2.1.6 Sequential Ambidexterity
The antecedent of the third approach of ambidexterity can also be found in Duncan (1976) where he suggests that in order to accommodate the conflicting activities of exploration and exploitation, organizations need to shift their structure continuously making sure that it is aligned with the organization's strategy over time (C. A. O'Reilly, III & Tushman, 2013). Thus, sequential ambidexterity may occur as “firms can realign their structures to reflect changed environmental conditions or strategies” (C. A. O'Reilly, III & Tushman, 2013, p. 327). Boum-garden, Nickerson and Zenger (2012) support this idea by reasoning how it may be relatively
informal organization and culture. Moreover, it is viewed that sequential ambidexterity can suit smaller firms with scarce resources that may not be able to apply structural ambidexterity. Alt-hough, a sequential change of structure may cause disruption and should be managed with cau-tion. Another critique that sequential ambidexterity has received is the long time spans between exploratory and exploitative activities (C. A. O'Reilly, III & Tushman, 2013). Hence, one may ask how suitable this approach is for organizations operating in an unpredictable and fast-mov-ing marketplace.
2.2 Organizational Ambidexterity in SMEs
Mihalache et al (2014) emphasize that, for smaller firms to survive in the long run, they need to achieve organizational ambidexterity. This may be a challenge for all firms regardless of size. However, it has shown that smaller firms tend to be more resource-constrained than larger firms which makes it even harder for them to balance exploration and exploitation (Lubatkin et al., 2006). According to O’Reilly, Harreld and Tushman (2009), causes of death in smaller firms may be numerous but oftentimes they relate to the general lack of resources. Moreover, smaller firms may also be subject to structural issues that are related to the resource-constrained reality that they often face (C. A. O'Reilly et al., 2009). Smaller firms simply cannot afford the luxury of an organizational structure where multiple business units coexist and are focusing on either exploratory or exploitative actions (Lubatkin et al., 2006). One may ask what smaller firms can do to mitigate the resource-related disadvantages that they are facing when trying to achieve organizational ambidexterity. It seems, at least according to some scholars, that a rather dystopian setting is outlined for smaller firms and their chances of overcoming challenges re-lated to ambidexterity. Lubatkin et al (2006) disagree as they describe it as illusive beliefs with regards to the claimed difficulties of achieving organizational ambidexterity. Instead, Lubatkin et al (2006) argue that the top management team (TMT) in general, and the CEO in particular, may be the decisive factors to determine whether a smaller firm manages to be ambidextrous. SMEs may have a general disadvantage when it comes to the amount of resources available, as it has shown to be an important factor for firm survival. At the same time, top managers in SMEs oftentimes monitor both strategic and operational positions. Hence, smaller firms may be able to leverage their relative size disadvantage, by benefiting from closely interrelated con-trol systems with less resistance and grasp the benefits of stimulated organizational ambidex-terity (Gschwantner & Hiebl, 2016; Lubatkin et al., 2006).
2.3 Management Control Systems
The aim and purpose of management control systems (MCSs) are to provide the needed and important information for managers to perform their work and to lead the organization in the right direction. MCSs are also intended to help managers to maintain viable patterns of behavior (Otley, 1999). According to the contingency theory, there is no comprehensive system of man-agement control in every organization, and therefore the appropriate control system will be chosen depending on the different core objectives, situations and goals existing within the or-ganization (Otley, 1999). Management control system is a concept that has been defined in several ways. Some of them overlapping each other, while others are different. Some of the definitions are broader, while others are narrower (Abernethy & Chua, 1996; Chenhall, 2003; Malmi & Brown, 2008; Merchant, 2007; Simons, 1995). Because of the several, but also dif-ferent definitions, a major challenge of how to define MCSs has occurred among researchers within the management control field (Fisher, 1998).
Management Control Systems was first defined in Robert Anthony’s work in 1965 (Otley, 1999). Anthony defined management control system as “the process by which managers assure that resources are obtained and used effectively and efficiently in the accomplishment of the organization’s objectives” (Anthony, 1965, p. 17). From this, several conceptualizations of MCSs have been created. Hopwood (1976) and Ouchi (1978), two of the first to deliberate socialization processes as a type of control. These controls were titled social control and clan control. These controls were informal and are evolving without encouragement from the organ-ization’s management. Furthermore, Hopwood (1976) explains an additional control, namely self-control. This control needs to be internalized by the organization’s personnel and create the effect of the administrative controls within the organization (Haustein et al., 2014). Adler and Borys (1996) created a formalization of control. This conceptualization included coercive control and enabling control, where coercive controls “are designed to force reluctant compli-ance and to extract recalcitrant effort” (Adler & Borys, 1996, p. 69). Enabling controls creating “users with visibility into the processes they regulate by explicating its key components by codifying best-practice routines” (Adler & Borys, 1996, p. 71). Another framework is Simons’ (1995) lever of control. This framework is built on four controls and is designed to focus on the use of MCSs rather than their design.
In 2008, Malmi and Brown evolved the “package of controls” framework. This framework rec-ognizes that organizations usually tend to use complicated control systems. Malmi and Brown’s
(2008) framework consists of five different types of controls. Ferreira developed, in 2009, Ot-ley’s (1999) framework of performance measurement systems (PMSs). This new framework consists of twelve dimensions, with the main focus on formal controls. Ferreira and Otley’s framework does also characterize Simons’ (1995) levers of control (Haustein et al, 2014). Lastly, in 2012, Merchant and Van der Stede created an object of control. Their framework consists of four types of controls. Merchant and Van der Stede´s control system reflects Hopwood’s (1976) framework, but also Ouchi’s (1979) (Merchant, 2017).
2.4 Management Control Systems as a Package
In this paragraph, a discussion on why management control systems should be investigated as a package is presented. This section also provides information about different frameworks, but also a motivation for the framework that will be used in this study.
2.4.1 Reasons to study MCSPs
The thought and idea of management control systems (MCS) operating as a package has existed since 1980 when Otley (1980) first introduced it. Otley explained this phenomenon as “It is often impossible to separate the effect of an AIS from other controls; they act as a package and must be assessed jointly” (Otley, 1980, p. 422).
Looking at and studying MCS as a package is important due to several reasons. The first reason is, according to Malmi and Brown (2008) that “MCS do not operate in isolation” (Malmi & Brown, 2008, p. 287). This definition is the underlying reason for the existence of the packaging phenomenon. While prior studies have focused on MCSs as single themes or practices, without a connection between them, they are substantially operating within a context where they are invariably being included in a broader control system. Management control systems should, therefore, be studied together as a package (Chenhall, 2003). This was also argued by Flam-holtz, Das, and Tsui (1985), who called for research to view and study accounting from a broader perspective since they identified several control systems that create an organizational control system (Abernethy & Chua, 1996).
Prior researchers have, in previous studies of innovation in practice, focused on activity-based costing/management (ABC/M), the balanced scorecard (BSC), value-based management (VBM), rolling forecasting, and target costing. During these studies, the purpose was to explain their advancement, approval and adoption. However, studying these systems individually may affect the conclusions that can be drawn if the adoption and approval of a new MCS element
are related to the function of the existing MCS package (Malmi & Brown, 2008). This consti-tutes another important aspect of why it is relevant to study MCSs as a package.
The third reason why study MCSs as a package is important is that many prior studies of MCS are investigating what MCS design is needed to generate the anticipated result. A lot of already existing literature is about accounting-based controls, related to formal systems, and it is, there-fore, an understudied area of other types of controls, e.g. administrative and cultural controls. There is also a limited understanding of if, but also how, they complement or substitute each other in different contexts. By study MCS as a package, a better understanding of how organi-zations can choose and range their MCSs, to meet the organization's goals and core objectives, but also the desired outcome is presented (Malmi & Brown, 2008).
Kennedy and Widener (2008) conducted a study where they could present findings on a more complex control system. They found “evidence that the control components are coupled in an interlocking system through multiple unidirectional and bidirectional relations” (Kennedy & Widener, 2008, p. 320). Another study, conducted by Sandelin (2008), presented similar find-ings where he argued that individual control systems should be studied together in a package to find accurate and trustworthy findings (Sandelin, 2008). Control systems do act as a package when they are designed to achieve similar ends (Abernethy & Chua, 1996). Some controls act as substitutes for each other, while others are contradicting (Merchant, 1985). Studies based on single instruments or control mechanisms have contributed to unclear and deceptive findings (Fisher, 1998). Several authors, therefore, recommend the application of a more comprehensive and integrative approach (Ferreira & Otley, 2009; Merchant, 2007).
To achieve different purposes and goals various types of controls can be used (W. Ouchi, 1977; W. G. Ouchi & Maguire, 1975). Simultaneous use of the different control systems to achieve the different goals makes it very hard, if not impossible, to isolate the effect of any specific means of control (Otley, 1999). As mentioned before, MCSs do not operate in isolation (Malmi & Brown, 2008), which makes it of high importance to study them as a package.
2.5 MCSPs in SMEs
During the last decades, studies of management control have increased, but the majority of them are conducted on larger firms. Hence, there is a lack of evidence on how management controls are operating in small to medium-sized enterprises (SMEs) (Pešalj, Pavlov, & Micheli, 2018).
This section will, therefore, present information on common control systems, but also how they are operating in SMEs (Pešalj et al, 2018). Because of a changing economic environment, or-ganizations need to be flexible to meet the changes and new demands (Mundy, 2010). Henri (2006a) describes the competing demands as the need to find a balance between “tension, con-trol and flexibility” (Henri, 2006a, p. 77). Since MCSs are formal systems, used to maintain and organize organizational activities, companies now need to include a flexible culture, infor-mal controls, to find the balance between stability and flexibility (Heinicke, Guenther, & Widener, 2016). Formal and informal controls can either work as complements or substitutes (Tiwana, 2010) and these controls can work as tools for companies to establish strategies and promote positive behavior, but also to increase the performance in both short term and long term (Pešalj et al, 2018).
As previously mentioned, a majority of prior empirical studies have been conducted on larger firms (Pešalj et al, 2018). This is problematic since SMEs display specific characteristics that differentiate them from larger firms. SMEs tend to be more flexible and thus more capable of innovation than larger firms, but are bound by relatively stricter human resources and financial boundaries (Pešalj et al, 2018). SMEs do also rely more on informal control systems but this may be necessary to develop and maintain employee behavior in line with organizational goals and values, but also to understand and improve performance (Pešalj et al, 2018), as is created by formal control systems in larger firms (Kaplan, 2008). With that said, this could indicate that SMEs need both formal and informal control systems to create a balance between stability and flexibility.
2.6 Conceptual frameworks for studying MCSPs
When studying MCSs as a package, there are several frameworks available for this purpose. There are three wide frameworks existing and these are developed by Malmi and Brown (2008), Simons (1995) and Merchant and Van der Stede (2012) (Sandelin, 2008).
2.6.1 Malmi & Brown (2008)
Malmi and Brown’s (2008) framework is built and based on a work of Brown’s (2005) and is provided to study MCSs as a package. Brown first developed the typology by analyzing and synthesizing nearly four decades of MCS research. Malmi and Brown (2008) explain that their typology and framework is a broad approach to make empirical studies of the phenomenon. The framework aims to aid and encourage discussion and research within this subject, instead
of providing a final result and explanation (Malmi & Brown, 2008). Malmi and Brown’s (2008) framework consists of five dimensions, as will be explained in detail in the next section: (i) cultural controls, (ii) planning controls, (iii) cybernetic controls, (iv) administrative controls and (v) reward and compensation controls (Malmi & Brown, 2008).
Cultural controls
Culture is a control system when it is used to regulate behavior. Malmi and Brown (2008) argue for three aspects when talking about cultural controls. These are (i) value-based controls, (ii) symbol-based controls and (iii) clan controls. Malmi and Brown (2008) argue that values im-pact behavior on three levels. The first level is that personnel is enlisted if they carry values that are in line with the organization. The second level is when employees are socialized and there-fore have their values changed to be harmonious with the organization (Alvesson & Kärreman, 2004). The third is when values are communicated to the employees, and the individuals are acting by them, even if they do not comply with them (Malmi & Brown, 2008). Symbol-based controls are related to visible expressions, created by the company with achievement to create a convinced culture. Malmi and Brown (2008) explain clan controls as in each organization there exists subcultures or clans. Within these subcultures or clans, there are individuals who have shared values, set of skills and beliefs, because they have been exposed to a socialization process (Malmi & Brown, 2008).
Planning controls
Planning controls are important for organizations in several ways by serving many purposes. Flamholtz (1985) explains planning as “… an ex ante form of control” (Flamholtz, 1985, p.39) based on three arguments. First, planning direct effort and behavior through setting goals of the organization’s functional areas. Second, it presents a clear level of effort and expected behavior by organizational members, by providing standards that should be followed to achieve the or-ganizational goals. Lastly, planning can empower coordination by coordinate objectives across the functional areas, and thereby may control the activities of groups and individuals so their actions are in line with the organization’s core values (Malmi & Brown, 2008). Malmi and Brown (2008) explain that planning is important and has a major role in directing employee behavior.
Cybernetic controls
Cybernetic controls are defined as “a process in which a feedback loop is represented by using standards of performance, measuring system performance, comparing that performance to standards, feeding back information about unwanted variances in the systems, and modifying the system’s comportment” (Green & Welsh, 1988, p. 289). Cybernetic systems can either be information systems or control systems depending on how they are used by the managers. Cy-bernetic controls contain four systems. These are (i) budgets, (ii) financial instruments, (iii) non-financial measures and (iv) hybrids as contains both financial instruments and non-finan-cial measures. Budgets do, even if it has received a lot of criticism, play an important role in contemporary organizations (Ekholm & Wallin, 2000). Budgets are used in several ways but when it is used as MCS, budgets have its focus on “planning acceptable levels of behavior and evaluating performance against those plans” (Malmi & Brown, 2008, p. 293).
Reward and compensation controls
The main purpose of reward and compensation controls is to motivate individuals and groups, but also to increase their performance by having congruence between their goals and their ac-tivities (S. E. Bonner & Sprinkle, 2002). Bonner and Sprinkle (2002) present findings that in-centives do increase employees’ performance by focusing on employees’ efforts on the task. This linkage, between effort and the task, can have significance on performance in three ways. These three are effort direction (the assignment the employees are focusing on), effort duration (how long time employees allot themselves to the assignment) and effort intensity (the amount of attention the employees allot to the assignment) (Malmi & Brown, 2008).
Administrative controls
Administrative controls aim to direct employee behavior. Malmi and Brown (2008) argue that it can be done by (i) organizational design and structure, (ii) governance structures within the firm and (iii) procedures and policies (Abernethy & Chua, 1996; Alvesson & Kärreman, 2004; Emmanuel, Otley, & Merchant, 1990; Macintosh & Daft, 1987; D. T. Otley & Berry, 1980; Simons, 1987). Organizational design and structure is a form of control, intending to organize employees, both individually and in groups (Abernethy & Chua, 1996; Alvesson & Kärreman, 2004; Emmanuel et al., 1990), but also to reduce the changeable behavior and, in turn, increase its predictability (Flamholtz, 1985). Governance structure within an enterprise includes the
formal lines of authority and accountability (Abernethy and Chua, 1996), comprehends the or-ganization’s board structure and its composition. Policies and procedures, on the other hand, are the bureaucratic way of specifying the processes and behavior within a company. Policies and procedures include standard operating procedures and practices (Macintosh & Daft, 1987), but also rules and policies (Simons, 1987) and can be related to what Merchant and Van der Stede (2017) explains.
2.6.2 Merchant & Van der Stede (2012)
The use of management control systems is important and the need exists because of three main reasons: lack of direction, motivation problems, and personal limitations. To solve these prob-lems, Merchant and Van der Stede have developed the objective of control framework, and it is dividing the control practices into four groups. These four groups are result controls, action controls, personnel controls and cultural controls (Merchant and Van der Stede, 2017). The following sections will present these groups of control in more detail.
Results controls
Results controls are an indirect form of control since the focus is not explicitly on employees’ actions or decisions. Results controls do provide several pre-emptive benefits and are used at many levels within companies. By having well-defined and structured results, employees are conscious of expectations that encourage them to work in line with the organization’s desired goals. Another benefit with results controls is that the controls can be an effective tool when the desired behavior is unclear. This can yield adequate control simultaneously as the employ-ees are given higher levels of autonomy. The autonomy increase because the personnel are in charge but also responsible for the produced outcome, not the actual action. Results controls do not regulate employees’ desired actions and decisions, but to motivate them to take actions and decisions that will generate satisfying results in line with the desired goals of the organization (Merchant and Van der Stede, 2017).
Action controls
Opposed to results controls, action controls are a direct form of management control since they aim to assure employees perform and take actions in line with the organization’s desired direc-tion. The effectiveness and usefulness of action controls depend on the managers, within the organization, and their knowledge. To implement action controls, managers need to know what
the (un)desirable action is but also having the ability to assure it to (not) occur. It is therefore important for managers to have the knowledge of which action that fits the organization, and its desired objectives, in the best way (Merchant and Van der Stede, 2017). Merchant and Van der Stede also argue that action controls can be broken down into four types of control: (i) behavior constraints, (ii) preaction reviews, (iii) action accountability and (iv) redundancy.
Behavioral constraints can be both physical or administrative and make it almost impossible for employees to take actions or do things they are not supposed to do. Action accountability is referred to holding employees responsible for the actions they take and redundancy can, ac-cording to Merchant and Van der Stede (2017), be seen as an action control because of its increased probability to complete a task reliably. This is done by assigning more people but also equipment than needed (Merchant and Van der Stede, 2017).
Personnel controls
Personnel controls aim to motivate employees to believe in, but also control, themselves on natural tendencies. Merchant and Van der Stede (2017) argue that personnel controls serve three purposes. First, personnel controls work as a tool to clarify expectations by creating an under-standing among employees regarding what the organization wants. Secondly, the controls sometimes work as a device to make sure employees can perform their tasks and assignments in a good way. This is done by offering the right capabilities and resources needed for an achievement. Lastly, some personnel controls act as a fuel to increase the employees’ willing-ness to be committed to self-monitoring, which is an inborn force related to the willingwilling-ness to always perform in a satisfactory and committed way (Merchant and Van der Stede, 2017).
Cultural controls
Some similarities can be seen between cultural- and personnel controls since cultural controls aim to encourage mutual monitoring. Cultures within an organization are built on shared tradi-tions, norms beliefs and values, but also ways of behaving. Finally, cultural controls do create energetic peer pressure on individuals who deviate, and thus are not following, the organiza-tion’s core values and norms (Merchant and Van der Stede, 2017). Merchant and Van der Stede (2017) argue that cultural controls often remain the same, even while goals are changing due to business conditions and the environment they are operating in. Hence, it may be difficult for managers to change a culture within an organization, but it can be done by using five different
methods. These are (1) codes of conduct, (2) group rewards, (3) intra-organizational transfers, (4) physical and social arrangements and (5) tone at the top (Merchant and Van der Stede, 2017).
Codes of conduct are formal, written documents providing overall information about the organ-izational values, but also commitments to stakeholders. Codes of conduct aim to help employ-ees understand recognize the expected behavior, without a specified rule. Group rewards aim to increase communication and expectation, but also the mutual monitoring within the group (Merchant and Van der Stede, 2017). The third way, intra-organizational transfers (also known as employee rotation) seeks to transfer culture within the organization through the socialization of employees (Merchant and Van der Stede, 2017). Physical arrangements as office plans, ar-chitecture and interior décor, but also social arrangements as dress codes, institutional habits, and behaviors can also be a tool to create and shape organizational culture. Lastly, Merchant and Van der Stede (2017) argue that organizational culture can be shaped by setting the proper tone of the top. Therefore, managers act as role models and are an important factor when cre-ating a specific culture within an organization (Merchant and Van der Stede, 2017).
2.6.3 Simons (1995)
Simons (1995) framework is frequently used in management accounting literature and consists of four control systems. These are (i) beliefs, (ii) boundary, (iii) diagnostic and (iv) interactive systems. An effective control environment is accomplished by combining all levers of control since “The power of these levers in implementing strategy does not lie in how each is used alone, but rather in how they complement each other when used together. The interplay of pos-itive and negative forces creates a dynamic tension...” (Simons, 2014, p. 301). The four control systems do create opposing forces, where belief systems and interactive control systems create positive and inspirational forces. The remaining levers, boundary systems, and diagnostic con-trol systems create constraints and ensure compliance with orders. These two contradicting forces are used by managers to create a dynamic tension that allows the effective control of strategy. In the following section, a more detailed explanation of the levers will be provided.
Belief and boundary systems
Simons (1995) define belief systems as “the explicit set of organizational definitions that senior managers communicate formally and reinforce systematically to provide basic values, purpose,
inspire but also guide organizational search and discovery. Belief systems seek to convey in-formation about how the organizations create value, what the desired level of performance is but also information on how individuals are expected to create internal and external relation-ships. The communication of beliefs is often conducted through a mission or vision statement and credos (Simons, 1995; Widener, 2007). These mission statements, credos and statements of purpose can be treated as part of a system when they are (1) formal, (2) information-based and (3) used by managers to maintain or alter patterns in organizational activities.
Boundary systems, the second lever of control, “delineates the acceptable domain of strategic activity for organizational participants” (Simons, 1995, p. 39). Boundary systems establish lim-its to strategically undesired actions, based on business risks. The boundaries are often commu-nicated through codes of business conduct (Simons, 1995: Widener, 2007). By implementing codes of conduct, managers try to ensure that employees, and other subordinates, are not jeop-ardizing the well-being of an organization by wasting or ineffectively using resources through projects, but also actions, that are not in line with the organization’s core strategy (Simons, 1995).
Belief systems, but also boundary systems, aim to motivate employees’ opportunity seeking. The difference between the two systems is that belief systems do it positively, through inspira-tion, whilst boundary systems do it more negatively, by demarcating the opportunity domain. By working together, these two levers do create tension between commitment and adaptation. It is also worth mentioning that these two levers are very important in organizations. When opportunities and pressures for performance increases, a structured belief system, and enforce-able boundary system is the key to success. These two elements give the managers the ability to concentrate on positioning their firm to meet the competitive challenges of the marketplace (Simons, 1995).
Diagnostic and interactive systems
Diagnostic systems are “designed to ensure predictable goal achievement” (Simons, 1995, p. 59) and the implementation of intended strategies. Diagnostic systems give the managers the ability to monitor, but also reward, the achievement of pre-established goals through the review of critical performance variables. These systems are formal information systems and are also used by managers to correct deviations from present standards of performance. For this to be
possible, a reinforced analysis must be correct since these analyses are tools to create an effec-tive allocation of resources, definition of goals, motivate, but may also serve as an evaluaeffec-tive analysis. This creates more time for managers to focus on other important aspects of the organ-ization (Simons, 1995).
Interactive systems, contrary to diagnostic systems, stimulate search and learning, allowing new strategies to emerge as participants throughout the organization respond to perceived opportu-nities and threats. Interactive control systems are formal information and are used by managers to active search. Contrary to diagnostic control systems, managers usually pay more frequent and regular attention to interactive control systems, where they also are more personally in-volved. Interactive control systems do force dialogue within the organization but do also pro-vide a framework for debate and motivate information gathering outside of routine channels. With this, interactive control systems can increase organizational learning and hopefully gen-erate new improved strategies (Simons, 1995).
2.7 Motivation for Chosen Package Approach (Simons 1995)
According to Malmi (2013), there is a need to compare and contrast different frameworks to find a common agreement on what management control systems as a package means. In this section, we will, therefore, present a short comparison of the three presented frameworks to motivate for the one best suited for our purpose.
The presented frameworks (Simons, 1995; Merchant and Van der Stede, 2012 and Malmi and Brown 2008) differ from each other. For instance, one difference is the way managers are using them. Simons’ (1995) framework is focusing on the use of management control systems rather than the design of them. Further, Simons’ (1995) framework does not concentrate on senior management (Haustein et al, 2014). Simons argues that diagnostic and interactive systems are related to the way they are used by managers. In other words, Simons explains that control systems either can be used in a diagnostic or interactive way (Simons, 1995). Malmi and Brown (2008) and Merchant and Van der Stede´s (2017) frameworks do have another focus. These two frameworks are focusing on what controls that are available for senior managers for the man-agers to guide employees, where the guidelines aim to control the behavior and actions so they are in line with the organization’s goals (Malmi and Brown, 2008; Merchant and Van der Stede, 2017). Even if Simons’ (1995) framework has received criticism of being too narrow (Bisbe,
Batista-Foguet, & Chenhall, 2007), but also have vague and ambiguous definitions (Tessier & Otley, 2012), the levers of control framework has been used in prior studies within SMEs.
In Pesálj, Pavlov and Micheli’s (2018) study of the use of management control and performance measurement systems in SMEs relies on Simons’ (1995) levers of control framework given that the framework makes it possible to explore different control systems jointly, as well as to ex-amine the approach balance in a deeper context (Pesalj et al, 2018). Companies, due to the changing environment they are operating within, need to create a more flexible culture. A flex-ible culture relies on formal, but also informal control systems (Mundy, 2010). A framework consisting of both formal and informal control systems is Simons’ (1995) levers of control where organizational performance is managed through four systems. Simons’ (1995) also ar-gues that levers of control framework help managers to manage the changing organizational demands (Simons, 1995).
A second study, conducted by Arumugham, Maelah and Aman (2018) used Simons’ (1995) framework to identify usable control to meet the identified challenges in SMEs (Arumugham, Maelah and Aman, 2018). Another argument for using the levers of control framework is pre-sented by Simons himself. He argues that the framework consists of different controls and to find out if one control is effective, to be a flexible culture, one need to take the other levers into consideration as well (Simons, 1995). A third argument using levers of control framework as a tool in our study is because of its explicit concern with the dual use of management control systems in order to facilitate creativity, but at the same time provide boundaries for the employ-ees’ behavior (Mundy, 2010).
Lastly, Simons’ (1995) framework levers of control has been used in Massaro, Moro, Aschauer and Flink’s (2019) study of trust, control and knowledge transfer in small business networks since it has been argued that a correct use of the framework can promote an organization’s performance because of the creation of tensions between innovation and goal achievement. By looking at prior use of Simon´s (1995) levers of control framework, one could argue that Si-mon’s (1995) framework is the most suitable to explore how management control systems as a package enables organizational ambidexterity in SMEs and is, therefore, the used framework in this paper.
2.8 Tensions and interrelations between the four systems
The control systems generate a “dynamic tension between opportunistic innovation and pre-dictable goal achievement that is essential for positive growth” (Simons, 1995, p. 153) and these tensions are a central element in Simons’ (1995) framework (Mundy, 2010). The dynamic tensions do arise from the interrelations between the levers of controls that facilitate the devel-opment of innovativeness, organizational learning, a market orientation that together increase the organizational performance (Mundy, 2010). Tensions can be found between each of the control systems and may arise due to a situation with many opportunities, but only a limited amount of attention (Simons, 1995). These tensions will be explained and exemplified in the following paragraphs.
It is of high importance to study the tensions between the control systems because the increased use of one lever of control enhances the benefits attained from increasing the use of the other levers (Widener, 2007) as it can create unintended consequences (J. M. Bonner, Ruekert, & Walker, 2002; Dent, 1991). In the following paragraphs, 2.8.1 to 2.8.6, the interrelations be-tween each system are described and explained. Later, in paragraph 2.9, a description of each systems interrelation with ambidexterity is described.
2.8.1 Belief Systems and Interactive Systems
Belief systems aim to motivate and inspire employees, but also to secure commitment towards the organizational goals (Simons, 1995). The purpose of interactive systems is to inform, but also to challenge, the underlying assumptions and action plans that initiates an organization’s activities (Abernethy & Lillis, 1995; Speklé, 2001). A tension between these two systems could, for instance, arise during meetings. During such meetings, employees’ individual views are communicated and exchanged. Through this dialogue, the organization’s core values act as a source of inspiration to search but also to find new opportunities (Pešalj et al, 2018). These two systems are both established as positive systems, and together they ease innovation but also increasing employee motivation and commitment to the organization and its core values (Bisbe & Otley, 2004; Tuomela, 2005).
2.8.2 Interactive Systems and Diagnostic Systems
Interactive systems enlighten the employees about the action plans and the assumptions that initiate a firm’s actions and activities (Abernethy and Lillis, 1995, Speklé, 2001). Diagnostic
