SELF-MANAGING ORGANIZATIONS AND GOAL CONGRUENCE:
An Exploratory Study of Management Control Systems in a Dutch Telecom Holacracy
Marleen Kappert
S2283867 – 11 970 words (excluding references & appendix) June 21, 2018
Supervisor: dr. R.A. Minnaar Co-assessor: dr. C.A.P. Heijes
Master thesis Accountancy University of Groningen Faculty of Economics and Business Accountancy & Controlling Department
Nettelbosje 2, Postbus 800, 9700 AV Groningen, The Netherlands +31 (0) 633291193
SELF-MANAGING ORGANIZATIONS AND GOAL CONGRUENCE: AN EXPLORATORY STUDY OF MANAGEMENT CONTROL SYSTEMS IN A DUTCH TELECOM HOLACRACY
Abstract
Studying control in self-managing organizations becomes increasingly relevant to the existing management (control) literature and society, due to the fact that the number of self-managing organizations is rapidly rising. Given the common belief that hierarchical organizations are natural and enduring, understanding how radical self-managing organizations could function effectively is of practical and scholarly concern. This study aims to address this concern by the means of a qualitative case study conducted at an organization that operates through the most implemented type of self-managing, holacracy. The findings provide relevant insights into the way a self-managing organization (SMO) could distribute authority and control throughout the organization. The SMO studied appears to rely mostly on the cultural, administrative and non-financial cybernetic management control systems (MCS) to ensure employees’ behavior is aligned with organizational objectives. These MCS seem to facilitate that the assigned decentralized authority related to ‘typical’ managerial tasks was perceived and carried out by employees.
Keywords: self-managing organizations, goal congruence and management control systems
1. Introduction
Organizations in a wide range of industries consider less-hierarchical organizations. Although the concept dates back to mid-twentieth century, nowadays experiments with less-hierarchical organizations are booming. This popularity could be driven by managerial hierarchies’ inability to cope with the fast-changing environment (Bernstein, Bunch, Canner and Lee, 2016).
Less-hierarchical organizations embrace and facilitate employees’ needs to respond more rapidly to emerging opportunities than the current managerial controls and hierarchical reporting allow them to (Lee & Edmondson, 2017; Starkey, Barnatt & Tempest, 2000). For instance, by implementing self-managing teams, organizations flatten hierarchies and cut hierarchical reporting structures. Working with self-managing teams enables them to increase the speed of response to the fast-changing environment (Burns and Stalker, 1961; Barker, 1993). In situations where work cannot be fully specified in self-managing teams, there is still a manager responsible for control and accountability (Ouchi & Maguire, 1975; Williamson, 1981). When control issues arise less-hierarchical organizations have the tendency to aim for control by centralizing and concentrate authority (Edmondson & Smith, 2006; Pfeffer & Leblebici, 1973; Staw, Sandelands, & Dutton, 1981).
Next to the less-hierarchical organizations, non-hierarchical organizations are rising, also called self-managing organizations. These organizations have radically decentralized authority throughout the organization. A self-managing organization (SMO) has eliminated the whole hierarchical reporting
structure between manager and subordinate, at least for work execution. These SMOs lack managers to ensure control and accountability, yet these organizations still have essential ‘managerial’ work, such as monitoring, that has to be carried out (Lee & Edmondson, 2017). Whether and how these SMOs pursue control and accountability remains unclear (Foss & Dobrajska, 2015; Okhuysen & Bechky, 2009), because what is currently known about internal relations in less-hierarchical organizations is mainly derived from studies that still rely largely on managerial authority, such as examinations in self-managing teams (Kellogg, Orlikowski & Yates, 2006; Burns & Stalker, 1961; Turco, 2016) Therefore, it remains unclear how SMOs align their employees’ behavior with organizational objectives (Foss & Dobrajska, 2015; Okhuysen & Bechky, 2009).
Given the common belief that hierarchical organizations are natural and enduring, understanding how radical self-managing organizations could function effectively is of practical and scholarly concern (e.g. Gruenfeld & Tiedens, 2010; Lee & Edmondson, 2017). This study aims to explore the way a SMO could operate effectively, by gaining a better understanding of which controls a SMO could implement to align employees’ behavior. This will facilitate our understanding of how radically decentralized organizations can flourish in dynamic internal and external environments (e.g. Foss & Dobrajska, 2015; Okhuysen & Bechky, 2009). The main research question of this study is:
How do SMOs ensure their employees’ behavior is aligned with organizational objectives? The number of SMOs is rapidly rising. Studying this type of organization becomes increasingly relevant to the management (control) literature and society (e.g. Laloux, 2014; Lee & Edmondson, 2017). Yet, apart from Kornberger, Pflueger and Mouritsen (2017), who studied the shift from hierarchies to less-hierarchical structures, little research is conducted to review control in non-hierarchical organizations, such as self-managing organizations This study aims to extend the literature on management control by exploring the way authority and control could be distributed in SMOs to effectively pursue organizational objectives.
Since different types of SMOs exist, different types of SMOs might need different control mechanisms to align employees’ behavior (Lee & Edmondson, 2017). According to Bernstein et al. (2016), a better understanding of the way the most implemented form of SMO, holacracy, operate effectively will contribute to our understanding of how most SMO could achieve organizational objectives. Therefore, this study contributes to the existing literature by providing insight into the distribution of decentralized authority for managerial tasks in the Dutch SMO Voys. Voys operates through a holacracy design and provides telecom services through the cloud. Furthermore, this study will contribute to society and the existing literature by providing insight into the way a SMO could deal with control issues, coordination and alignment.
Additionally, this study examines whether formal and executed decentralized authority could diverge, which has not been studied yet (Dobrajska, Billinger & Karim, 2015). Overall, the insights obtained contribute to our understanding of how employees’ behavior could be aligned with organizational objectives.
The remainder of this paper is structured as follows. Section two presents the literature review. Section three includes the research methodology. Section four presents the case and section five includes the case analysis. The discussion explains the findings in relation to the existing literature in section six. In the final section, the research question sis discussed along with limitations of a single case study and suggestions for further research.
2. Theory
In order to understand how SMOs ensure employees’ behavior is aligned with organizational objectives, this theory section is divided into the following sections: characteristics of SMOs, the most common type of SMO, drivers of employees’ behavior, management control systems and subsequent empirical questions.
2.1 Characteristics of self-managing organizations
Insight into the organizational forms and heterogeneities of SMOs, could improve the understanding over which tasks employees could have decentralized authority and therefore, freedom to decide to work towards organizational objectives or not.
Lee & Edmondson (2017) define SMO as: “radically decentralize authority in a formal and systematic way throughout the organization.” To function as a SMO, Lee & Edmondson (2017) state three conditions. First, a formal system should prescribe how authority is decentralized in the organization. Second, employees, alone or together within teams, should at least have the managerial authority “to manage” how they execute their work. Third, the hierarchical reporting relationship between manager and subordinate should be fully eliminated for the tasks for which authority is decentralized, also called self-managing tasks (Lee & Edmondson, 2017).
Only with these three radical conditions, employees, alone or together within teams, have at least the authority “to self-manage” how they execute their work without the corresponding obligation to report to a hierarchical manager (Lee & Edmondson, 2017). Without this elimination, the manager could take back the delegated power to solve conflicts, disagreements and crisis (Foss, 2003).
Heterogeneity between SMOs. SMOs could differ in the way they organize self-managing. SMOs could differ in organizational design, organizational structure, work context, and nature of work. These characteristics are likely to influence the level of coordination, the way authority and controls are formally implemented, and the way work is organized. Therefore, different types of SMOs may need different control mechanisms to ensure they achieve organizational objectives Over time, controls could differ per SMO as well, due to the changing environments, the complexity of work and the size and growth rate (Lee & Edmondson, 2017).
The type of SMO could influence the degree to which alignment of employees’ behavior is results from impersonal sources such as organizational governance model, rules and formal roles versus from
personal sources such as status, popularity, and social influence and pressure. Coordination could be informally distributed by teams or, for instance, formally distributed by implemented governance processes. As well the degree to which behavior is supposed to be aligned with organizational objectives by fixed-order processes, rules and procedures, or by people (Simons, 1987; Lee & Edmondson, 2017).
In SMOs, ‘managerial’ work could be organized through a role-based structure or through a team-based structure (Lee & Edmondson, 2017). This typical ‘managerial’ work is mostly divided over different employees, split up into different ‘managerial’ tasks. The tasks, for which authority are decentralized, could be formally distributed in a democratic way, limited by clear boundaries, or bounded by the temporary existence of a task (Lee & Edmondson, 2017). For which tasks, beside work execution, authority is decentralized is likely to depend on the nature of the organizational work context and organizational design. As shown in figure 1, researchers defined a couple of typical managerial tasks and arranged them from most likely to be decentralized in a SMO to at least: work execution, managing and monitoring work execution, organization and work design, work and resource allocation, personnel and performance management and at last firm strategy (Lee & Edmondson, 2017; Hackman, 1986; Puranam, Alexy & Reitzig, 2014).
2.2 The most common type of SMO, holacracy.
Worldwide the most adopted type SMO is holacracy (Bernstein et al., 2016). According to the conditions of Lee & Edmondson (2017), the holacracy governance model classifies as a SMO.
The holacracy perspective is that the whole organization works towards one goal, the organization’s purpose. To pursue this goal, the work is divided into roles. A bunch of roles works together to pursue a fragment of the organizational goal, indirect all roles should contribute to the organizational goal. An employee can apply for roles, these roles include tasks and responsibilities. A role can be returned to the circle at any time, a meeting is not required unless otherwise agreed. Some roles are fulfilled by one-person others by more (Lee & Edmondson, 2017; Robertson, 2015)
While fulfilling a role, the daily activities could change with the emerging changes in roles and circles due to tension. Tension is defined as the gap between the current situation and the desired outcome an employee envisions (Robertson, 2015). These small changes are implemented to diminish tension that employees experience from the role they participate in. These small changes result in continuously dynamic mini reorganizations.
Rules. Manager responsibilities should be delegated to roles instead of individuals and should continuously shift with the creation, changes and elimination of roles. The objectives will be realized through teams, the circles (Robertson, 2015). Instead of fulfilling one position, employees can apply for one or more roles in different teams, also called circles (Lee & Edmondson, 2017).
Typical managerial tasks such as creating a strategy, allocating resources, assigning employees to roles, evaluating alignment with goals, representing the department are distributed over different roles and persons (Lee & Edmondson, 2017). Furthermore, there are rules related to embodying each of these
‘managerial’ roles. One role holds more managerial tasks at once, the Lead Link. This role may not be combined with some other roles that hold managerial tasks (Robertson, 2015).
The Lead Link controls any defined responsibilities within the circle, except for the ones that have been delegated to roles. The Lead Link is responsible for strategy creation within the circle, allocating resources and assigning employees to roles. The Lead link role is responsible for connecting their super-circle to the sub-super-circles. The Lead Link has some characteristics in common with a traditional manager, however, this role is bounded by the circle’s governance guidelines (e.g. Lee & Edmondson, 2017).
Furthermore, some roles are fulfilled by election and employees are elected for a certain period. These are bounded by the temporary existence of the task, mainly selected for six months. However, the Lead Link role is for an indefinite time. The person for this role is selected by the super-circle above the circle. Additionally, other not typical managerial roles could exist in the circle (Robertson, 2015).
Decentralized tasks. The “holacracy constitution” prescribes fixed guidelines, rules and processes that facilitate organizational changes. This document includes all the rules regarding the distribution of authority throughout the organization. This governance structure provides a framework that formally enables decentralized authority for the tasks work execution and organizational and work design (Lee & Edmondson, 2017; Robertson, 2015). When holacracy is carried out properly, the authority for managing and monitoring work execution and work and resource allocation should be partly decentralized. The firm strategy is set by the broadest super-circle, and therefore is not decentralized according to the “Holacracy Constitution” (Lee & Edmondson, 2017; Robertson, 2015).
It provides an overview which roles, and indirect the employees’ embodying those roles, are responsible for which accountabilities. Theoretically, these formally defined accountabilities attached to roles seem to facilitate high formal coordination through impersonal and bureaucratic sources of control. The guidance is provided by imposed rules and fixed-order processes addressed to roles, instead of employees following orders imposed by managers. This way the governance structure enables monitoring of employees’ behavior in their roles. Theoretically, the fixed holacracy structure aims to facilitate behavior towards the organizational purpose, therefore could be seen as a form of control (Lee & Edmondson, 2017, Robertson, 2015).
The “Holacracy Constitution” does not include information about the evaluation process of functioning and assessment or hiring and firing. The Lead Link may take a role from a colleague when the role is not fitting the person, because the person is not functioning towards the purpose of the role (Robertson, 2015). All the rules are describing roles, what happens if a person loses all of his roles is not determined in the holacracy guidelines.
An example of a SMO operating through the holacracy governance model is Zappos. The tasks, for which, authority is decentralized at Zappos are shown in figure 1 (Figure created by Lee & Edmondson, 2017).
Figure 1. Decentralized tasks at Zappos versus managerial hierarchy (Lee & Edmondson, 2017).
2.3 Drivers of employees’ behavior
This section aims to understand what behavior employees could carry out in the domains they have authority over, and what drives employees’ behavior towards organizational objectives. Gaining an understanding of the drivers of employees’ behavior could help understand which mechanisms could be implemented to pursue goal congruence in SMOs.
In SMOs, employees have at least the decentralized authority to decide how to organize work execution. However, to achieve organizational objectives, employees need to carry out behavior aligned with these objectives. Researchers have different ideas about how to ensure employees’ behavior is aligned with operational or formal organizational objectives. Some believe self-managing teams can only be productive if they have an effective control system that controls and integrates employees’ actions into organizational behavior (Tannenbaum, 1968). Others suggests, organizations could emphasize empowering mechanisms, provided that they trust their employees enough to grant them a certain amount of authority (Jensen & Meckling, 1994; Davis, Schoorman & Donaldson, 1997).
Considered that humans are driven by different motives, drivers for employees’ behavior could also vary among employees. It seems unlikely that controls would be redundant because it seems unlikely that employees would always achieve maximal utility through executing pro-organizational behavior (e.g Jensen & Meckling, 1976; Donaldson & Davis, 1989, 1991; Merchant & Van Der Stede, 2007). In order to know which control mechanism could be implemented to align employees’ behavior, it is sufficient to know what could drive employees’ behavior.
Drivers of employees’ behavior. Mainly there are two renowned theories that explain what drives employees’ behavior and how an organization could achieve goal congruence. These two perspectives are the Agency theory and the Stewardship theory (Jensen & Meckling, 1976; Donaldson & Davis, 1989, 1991). Both theories are based on the existence of information asymmetry between the owner and the manager, created by the separation of ownership and control. The owner and manager pursue utility maximization, however the owner does not have full oversight over the company, because he
delegates his work to the manager (Jensen & Meckling, 1976). However, the theories divert on the view towards employees’ behavior. They differ in explaining what drives employees’ behavior. Therefore, they differ in the way to deal with the information asymmetry between owner and manager (Van Slyke, 2006; Jensen & Meckling, 1976; Davis et al., 1997).
The agency perspective is based on the assumption that managers’ interests may diverge from organizational interests. Due to the information asymmetry, there is always the possibility that the managers take the given opportunity to pursue their individual utility at the expense of the owner’s utility. The owner invests in (coercive and compliance-based) controlling mechanisms, reporting mechanisms and implements financial incentives and sanctions to minimize the abuse of control (Jensen & Meckling, 1976; Demsetz & Lehn, 1985; Walsh & Seward, 1990; Van Slyke, 2006).
Controls are imperfect; therefore, some opportunities will always remain. Several studies admit the agency is based on a limited and simplified view and may not hold for every manager (e.g. Doucouliagos, 1994; Jensen and Meckling, 1994). Still, this perspective could provide useful insights into employees’ behavior (Davis et al., 1997).
The stewardship perspective could be seen as a complementary perspective to the agency (Donaldson & Davis, 1994; Davis et al., 1997). In contrast with the agency perspective, the stewardship perspective is based on the situations, in which, managers pursue organizational objectives in order to meet personal objectives. This theory also applies to managers that receive a higher utility to pursue organizational objectives when these are not aligned with their self-interest (Donaldson & Davis, 1989, 1991). Furthermore, this theory assumes that intrinsically motivated managers, managers that highly identify with organizations and managers who pursue self-actualization and growth, are more likely to pursue organizational objectives (Davis, et al., 1997). Managers fitting these descriptions seem to be trustworthy and pro-organizational. Also, organizations that have a collectivist culture, low power distance and focus on involvement and trust perceive to have more stewardship relations in their company. The stewards’ performance depends on the space a steward is given to operate effectively. Stewards should be facilitated and empowered with autonomy to maximize the owner’s utility (Davis et al., 1997). Monitoring, incentive schemes and other agent costs can be reduced because these employees act like steward towards organizational objectives. For maximization of the owner and manager’s utility, these mechanisms should be decreased, because they could be perceived as a lack of trust and decrease a stewards’ motivation (Argyris, 1964; Davis, et al., 1997). Van Slyke (2006) points out that the stewardship perspective is more focused on alignment in the long term. The principal focusses on developing trustworthy relations with the steward, where frequent contact, feedback and periodic reporting are implemented to achieve alignment.
Attaining goal congruence. Van Slyke (2006; p.158) emphasizes that both theories on drivers of employees’ behavior point out the following mechanisms to achieve goal alignment between owner and manager: “monitoring, trust, reputation, incentives and sanctions.” It is plausible that a mix of
monitoring, trust, reputation, incentives and sanctions mechanisms are implemented to achieve goal alignment in SMOs.
In SMOs the monitor and the monitored are not hierarchy distinguished (Conger & Kanungo, 1988; Thomas & Velthouse, 1990), and employees are at least trusted with the autonomy to execute decentralized authority within the provided organizational guidelines and boundaries. The assigned decentralized authority could be perceived as a sign of trust instead of monitoring (Van Slyke, 2006). According to Davis et al. (1997) organizations with signs of trust and a low power distance are more likely to have more stewards. SMOs could implement decentralized authority as an ‘empowering’ mechanism within clearly defined boundaries as a token of trust to create more committed employees that pursue their own goals through achievement of organizational goals, also named stewards. Stewards could feel empowered and trustworthy when decentralized authority is delegated to them, what is associated with positive effects on the employees’ satisfaction and organizational performance (e.g. Verme, 2009, Turco, 2016, Conger & Kanungo, 1988).
Decentralized authority and autonomy over certain tasks are within by provided guidelines and boundaries, they seem to hold conditions that delimit employees’ behavior to a certain level. Therefore, these empowering mechanisms, share characteristics with control mechanisms that implemented to direct and control the behavior of agents in a particular way.
The boundaries and guidelines of decentralized authority might be implemented to control employees to behave towards organizational objectives without demotivating stewards. Supported by the trend that management techniques shift to more indirect techniques, as stated by Clegg (2012; p. 76): “managing in and through vision, mission, culture and values, leading to a lot less imperative instruction and command and a great deal of more dialogue and discussion.” Therefore, employees might not recognize the boundaries as a control, because they are subtly integrated in the way of working, or because the employees strong identify with the guidelines, team and values (Barker, 1993; Barker and Cheney, 1994).
Independent of the reason why the decentralized authority might be implemented, it might function as a mechanism to facilitate alignment of employees’ behavior with organizational objectives (Van Slyke, 2006). The decentralized authority could function as control because an organization wants to provide guidance and guidelines to help employees to pursue organizational objectives and avoid non-alignment. When non-alignment occurs, it does not mean this is derived from self-interest, it might be a lack of knowledge. To be sure most employees’ behavior is aligned with organizational objectives, SMOs might be better off to implement additionally control mechanisms (Merchant and Van Der Stede, 2007, Appelbaum & Batt, 1993; Heckscher & Donnellon, 1994). Decentralized authority might function as a control, without demotivating the pro-organizational stewards (e.g. Jensen & Meckling, 1994; Davis et al., 1997).
2.4 Management Control systems
This section is focused on which management control mechanisms or systems exist within SMOs and which SMOs could implement to align employees’ behavior.
Unintentional management controls. There could be mechanisms that unintentionally functioning as management controls. These could be the efforts organizations carry out to empower employees such as delegating decentralized authority. These efforts could create more pro-organizational employees, by improving the work experience and indirectly increasing the employees’ work motivation and organizational commitment, sense of control, and job satisfaction (e.g. Cohen & Ledford, 1994; Hackman, 1980). Langfred (2004) disputes whether decentralized authority works similar or as effective compared to managerial authority.
The perceived experience of empowerment beyond the increased share in decision-making, is also influenced by the work design, nature of reward system, supervisory style and organizational factor such as culture and transparency (Conger & Kanungo 1988; Hackman, 1980; Spreitzer, 1996). Employees might perceive all these organizational constructs and empowerment efforts as signs of trust and are more willing to operate towards organizational objectives. Employees can feel empowered or trusted when they perceive to have decentralized authority, although, this does not mean that those employees carry out their decentralized authority (Conger & Kanungo, 1988; Thomas & Velthouse, 1990). Therefore, more research is needed to examine whether these factors might function as controls (Foss and Dobrajska, 2015).
In SMOs, all employees have a certain level of managerial authority. Difficulties to perceive and execute the formal assigned decentralized authority could arise, partly due to the fact the monitor and the monitored are not hierarchy distinguished, concerns for feelings of co-workers and fear for unpleasant consequences (e.g. Conger & Kanungo, 1988; Thomas & Velthouse, 1990; Clegg, 2012). There are factors that might decrease the perception of decision empowerment about certain ‘managerial’ tasks (Foss and Dobrajska, 2015).
Research about the ‘managerial’ task monitoring is ambiguous. Some research found that a high degree of trust in SMOs could eventually lead to the induced unwillingness to monitoring others. Monitoring could be perceived as a violation of trust itself (e.g. Langfred, 2004; Lewicki & Bunker, 1996). On the contrary, other research on SMOs highlights the possibility of peer control in SMOs (Barker, 1993; Hinds & Bailey, 2003; Sole & Edmondson, 2002). Barker (1993) found a case at which an informal socially rules system emerged. The rational rules and peer pressure combined create a tight control system. The strong controls caged the employees, this kind of peer control and monitoring by peers could evolve to function ultimately as coercive as managerial control (Barker, 1993; Barker and Cheney, 1994).
In some organizational models, employees are willing to submit to value consensus of colleagues and informal socially emerged rules system because they perceive these actors as authorized, as legitimate sources of control (Whitley, 1977). According to Marcson (1960, p.130), in these situations
colleagues could form an informal network, using their authority to constrain the ‘inappropriate’ behavior of an employee and to secures a certain degree of uniform behavior (Mills, 1983; Trist, 1981). This way alignment could be enforced (Gruenfeld & Tiedens, 2010; Barker 1993). Peer control might enable control, and therefore alignment, however, it could decrease the motivation of stewards (Davis, et al., 1997). Whether the empowering efforts and organizational factors, such as delegating decentralized authority to employees, outweigh the possible peer controls of informal hierarchies and high trust work environments remains unsolved.
The impact of organizational factors, and degree of trust and informal hierarchies on the extent to which decentralized authority is perceived and executed could vary among employees (Foss and Dobrajska, 2015). A work context without managers, might need a more robust and qualitatively different set of control mechanisms for work coordination (e.g. Gruenfeld & Tiedens, 2010).
Some emphasize the need of an effective controls system to control and integrate self-interested employees’ behavior towards organizational objectives (Tannenbaum, 1968), where others argue that controls should not demotivate pro-organizational employees (Jensen & Meckling, 1994; Davis et al., 1997). Various researchers affirm hierarchal organizations should not solely rely on managerial authority for effective control and goal congruence, because they question the effectiveness of managerial authority and the control managers have (e.g. Appelbaum & Batt, 1993; Heckscher & Donnellon, 1994; Piore & Sabel, 1984; Thompson, 1967). Organizations should consider other management controls, besides managerial authority, that enable for instance coordination across boundaries, such as protocols and practices of display and representation (e.g. Kellogg et al., 2006; Okhuysen & Bechky, 2009).
Management control systems in general. There are many forms of management controls, that organizations implement to align employees’ behavior. When these controls form a complete system they can be named management control systems (Malmi & Brown, 2008).
Management control systems are defined in many divergent ways, which makes it challenging to define the concept management control systems (Anthony, 1965; Ouchi, 1979; Otley and Berry, 1980; Flamholtz, Das & Tsui, 1985; Green and Welsh, 1988; Emmanuel, Otley and Merchant, 1990; Simons, 1995; Abernethy and Chua, 1996; Langfield-Smith, 1997; Fisher, 1998; Chenhall, 2003; Alvesson and Karreman, 2004; Merchant and Van Der Stede, 2007; Malmi & Brown, 2008). Divergent MCS definitions have complicated the interpretation of research results (Fisher, 1998; Malmi & Brown, 2008). The hierarchical reporting relationship seems to be an important part of most management control systems definitions (Ouchi, 1979; Flamholtz et al., 1985, Simons, 1995; Abernethy & Chua 1996; Merchant & Van Der Stede, 2007). For instance, Simons (1995) described MCS as a way for senior management to successfully pursue organizational objectives.
There are MCS definitions that seem to focus more on other elements than the managerial relationship. From a sociological perspective, management control systems (MCS) could be described as a way to empower employees in order to achieve their own goals (Chenhall, 2003). Merchant and
Otley (2007) state MCS only exist to achieve profitable results craved by stakeholders, by supporting an organization with changes in the organizational environment.
This paper will employ the MCS definition of Malmi and Brown (2008; p. 290): “all systems, rules, practices, values and other activities management put in place in order to direct employee’ behavior should be called management controls.” Their definition focuses on the mechanisms that are implemented to provide direction. Furthermore, they argue that there are different MCS systems in an organization and they do not have to work as a united system.
Management control systems package. Malmi and Brown’s (2008) perspective is that an organization has multiple MCS and all together they form an MCS package. Malmi and Brown (2008) analyzed forty years of MCS research and developed a conceptual typology of an MCS package, as shown in figure 1. There are different types of MCS a SMO could implement, there is no universal optimal mix. The choice may depend on the organizational design and structure, the work context and the nature of work. This package provides an approach to study MCS empirically.
Figure 2: MCS Package (Malmi & Brown, 2008).
This package includes cultural, planning, cybernetic, reward and compensation and administrative controls systems. The cultural controls are all the controls settled to influence employees’ values, social norms and beliefs such as symbols and value-based controls (e.g. Birnberg & Snodgrass, 1988; Simons, 1995, Ouchi, 1979; Schein, 1992).
Planning controls include goal setting and directing employees’ behavior towards those goals. An expected level of effort and behavior is attached to these goals.
Cybernetic controls include, for instance, budgets and, non- and financial measures (e.g. Bunce, Fraser & Woodcock, 1995; Hansen, Otley & Van Der Stede, 2003, Ittner & Larcker, 1998). The cybernetic controls are evaluated on the existence of quantified activities and targets to enable measuring, analysis of the difference between outcome and target and the ability to modify employees’ behavior in the activities carried out.
Reward and compensation controls aim to motivate employees towards organizational objectives and increase the duration and intensity of their efforts towards targets through rewards (Flamholtz et al., 1985, Bonner & Sprinkle, 2002).
At last, the administrative controls aim to direct employees towards pro-organizational behavior. The organizational and governance structure aims to organize employees and attach accountabilities to employees, what enables monitoring of their behavior (Otley & Berry, 1980, Emmanuel et al., 1990, Abernethy & Chua, 1996, Alvesson & Karreman, 2004). Policies and procedures are implemented to specify the way tasks and behavior has to be carried out (Macintosh & Daft, 1987, Simons, 1987). In SMOs
Specific control issues for SMOs. Research on how SMOs deal with control issues, such as organizational crisis and conflicts is scarce (Dobrajska, et al., 2015; Lee & Edmondson, 2017, Okhuysen & Bechky, 2009). When non-alignment occurs in hierarchical organizations, monitoring and compensation mechanisms are commonly used to gain more alignment with organizational objectives (Davis, et al., 1997). Several studies in less-hierarchical organizations suggest control issues have the tendency to lead to centralization and authority concentration (Edmondson & Smith, 2006; Pfeffer & Leblebici, 1973; Staw et al., 1981).
To be classified as a radical less-hierarchical organization, a SMO, employees do not have to report to a manager and this should theoretically stand in case a conflict or a crisis occurs (Lee & Edmondson, 2017). However, how decentralized authority is formally distributed, perceived and carried out, is likely to influence the actions that will happen or not happen when a conflict of a crisis emerges (Foss & Dobrajska, 2015). Langfred (2007) found self-managing teams acting dysfunctional when dealing with a conflict, they avoided collaboration by restructuring their team. How emergent coordination is accomplished without managerial authority remains unclear. Guessing that policies and other practices that enable coordination in SMOs (Okhuysen & Bechky, 2009) might include prescriptions about conflicts and crises.
2.5 Subsequent theoretical and empirical questions
The existence and effectiveness of MCS could differ between SMOs, due to many factors, such as, the culture, governance structure and the drivers of employees’ behavior. The tasks, for which authority is decentralized, might influence the way control is pursued in SMOs. According to the agency and stewardship perspectives, employees’ behavior could be aligned by implementing a mix of monitoring, trust, incentives and sanctions (Van Slyke, 2006). According to the stewardship theory, it is likely that pro-organizational employees’ motivation will be negatively influenced when they perceive monitoring or incentive-driven control systems (Van Slyke, 2006; Davis et al., 1997). The organizational efforts that employees perceive as empowering, such as the formally assigned decentralized authority, might positively influence the alignment towards organizational objectives (e.g. Cohen & Ledford, 1994; Conger & Kanungo 1988; Hackman, 1980; Spreitzer, 1996).
Furthermore, the governance model provides a holacracy with fixed processes, policies and procedures what might function as an effective management control system (Malmi & Brown, 2008).
Subsequently, empirical questions emerge. For which tasks is authority decentralized at the case studied? How effective are the implemented management control systems? Lastly, does the assigned, perceived and carried out decentralize authority diverge among employees?
By examining which factors and which MCS are present at the case, this study increases the understanding of the way the most implemented type of SMO could operate effectively.
3. Research Methodology
Qualitative research and inductive reasoning are chosen to gain a better understanding of the experience of control mechanisms and decentralized authority in a specific context, a SMO (Lee & Edmondson, 2017). A single embedded case study in a SMO increases the robustness, considers the specific context and facilitates the exploration of the complex experience of decentralized authority in practice (Baxter & Jack, 2008; Merchant & Van Der Stede, 2007).
This case study combined a document analysis, work context analysis and semi-structured interviews. The combination of three different data gathering techniques, also called triangulation, were carried out to increase the accuracy, validity and robustness of the results. Although the case study was carried out by as subjective outsider, the triangulation reduced the research biased, by proving information from different perspectives (Denzin, 1978).
The document analysis and work context analysis aimed to minimalize projections of case specific results on other holacracy driven organizations and SMO in general. This is important, because case study interpretations are bounded to their context (Lee & Lings, 2008).
Interviews are chosen because they enhance the divergent experiences employees could have about decentralized authority in a rich and comprehensive way (Sandberg, 2000). A better understanding of the way an employee perceives their organization and their position in it, improves the understanding of the way their behavior is carried out (Lee & Lings, 2008).
The interviews are examined from a phenomenological interpretive perspective. An interpretation of a human is interrelated to their perceived position in world, where past experiences influences future experiences (Berger & Luckmann, 1991; Schutz, 1945). This approach assumes the existence of more realities, employees’ experiences and their reality are contemplated as subjective (Lee & Lings, 2008; Sandberg, 2005). This study aimed to provide insight into the layer underneath the employees’ behavior, the intersubjective reality.
It should be clear that the study results are likely to be limited to subjective personal perspectives of the specific organizational context. Even though interpretations are bounded to the specific context, the relevance of the results is not necessarily completely bounded to the specific context (Lee & Lings, 2008).
3.1 Case selection
Increasing the relevance of the result, this study chose a case operating through the most implemented SMO type, holacracy, to understand how a SMO can operates effectively (Bernstein et al., 2016). The relevance it not necessary completely bounded to the specific case context (e.g Lee & Lings, 2008). Furthermore, the case’s classification as a SMO, ensured that a ‘real’ SMO was studied. The selection process for the interviewees was based on a minimal employment of one year, their involvement in the customer support process and their willingness to cooperate. In the end, nine employees were interviewed, they carried out divergent roles in different levels of the organization, three employees had roles at the organizational circle, Voys United circle and smaller circles, two employees had roles at the Voys United and smaller circles and the residual had mainly roles at the customer support circle, described as a smaller circle of the case organization. Mainly roles were interviewed that held some kind of ‘managerial tasks, such as Lead Links. Furthermore, an organizational holacracy expert, also behavioral coach, was interviewed. After seven interviews, a clear understanding of the concept was accomplished, and no new information emerges. At this point data saturation was already achieved (Glaser & Strauss, 2017). Data saturation is the point where content validity is reached, and the sample is adequate (Guest, Bunce, & Johnson, 2006). The last two interviews were already scheduled and therefore are taken into account as well to create more robustness in the findings.
3.2 Data generation
The document analysis contains a review of the formal control mechanisms, such as the implemented procedures, rules and policies. These were reviewed to explore which formal MCS are implemented to align employees’ behavior towards organizational objectives. The work context analysis focused on the work environment, work routine, and the decentralized ‘managerial’ tasks.
In consultation with a role that guides internships and with the conjunction of nine employees, data was generated through semi-structured interviews. Within two sequent weeks, all the interviews were conducted at the office of Voys. They lasted between 35 and 70 minutes per person and took place in the second quarter of 2018.
To minimalize the bias, the ambiance of an informal coffee date was sought for every interview. The interviewer was fully informed about the roles the interviewee employed, read out loud the protocol, and carried out a similar appearance for every interview. The interviewer strived for an equal approach and paid attention to the way questions were empathized and how intonation was used in all interviews taken.
All the interviews were recorded and were anonymously converted into written records. The first half of the questions were focused on the employees’ perspective on the whole organizational design, formal and informal MCS, conflicts, evaluation of functioning and assessment in order to obtain an
understanding how employees perceive to be aligned to organizational objectives and for which tasks decentralized authority was perceived and carried out. The second part was focused on the past personal experiences of perceived and carried out decentralized authority in their role. The interview protocol is shown in appendix A.
3.3 Data analysis
The information conducted from the document analysis and the work context analysis were processed in the case introduction and provided input for the interviews and the analysis.
The interview data was processed by using open, axial and selective coding. This method facilitated the account of emerged topics and it enhanced the structuring and the interpretation of the interview transcripts. In Appendix B the resulted coding scheme is set out, including coding examples to sort statements into the related topics within the category (Saldaña, 2015). Finally, the information was evaluated to assess the possible pattern in the experiences to answer the research question and provide additional insights into a SMO.
4. Case introduction
The telecom company involved in this study is Voys, a Dutch telecom SMO operating through a holacracy design. Since 2006 Voys provides business telecom services through the cloud and operates in a dynamic competitive market. Most parts of the organization, especially the earnings model, are focused on flexibility. The organization is influenced by the market demand and occupancy rate and the work agreements move along with this occupancy rate. The customer can construct and change their modules themselves and Voys strives for self-reliant customers. Customers are able to transfer and to cancel on a daily basis, the notice period is one day (Voys, 2018).
Since 2012, this company defines itself as a self-managing organization. Over the years, the organization Voys has been operating in different industries. Therefore, the company is separated into different entities. All the companies are gathered in the organization Bedstone. Voys could be seen as one of the core brands, or domains, of Bedstone. Voys United is the entity that should cherish the Voys brand and their associated purpose is: “Bringing game-changing communication and inspiration” (Voys, 2018).
4.1 Voys governance and organizational structure
Since March 2015, Voys operates through a highly structured governance model, holacracy. Therefore, the organization pursues one overall organizational purpose. The purpose of the whole organization, Bedstone, is “bringing equality by providing everybody with access to free communication.” The associated strategy is “promoting equivalence by providing one billion people with free communication.” The super-circle includes among other roles, Spindle, VOIPgrid, Voys and
overhead roles. Voys contributes to this organizational goal by delivering telecom services through the cloud and pursuing as many happy customers as possible. To pursue this purpose, the role Voys is divided into smaller roles that together form the Voys circle. The Voys circle is part of the Voys role and the Voys role is a part of Bedstone. A smaller role of the Voys circle is “Klant Geluk”, a customer support role. This role’s purpose is “A culture of independence & happiness for customers & colleagues.” The customer support role is divided in different roles and forms a circle as well, the customer support circle. The main role that is embodied by every Customer support circle role is the customer support role, with the purpose to “make customers happy with amazing support” (Voys, 2018).
Holacracy aims to facilitate the flexibility for the work Voys executes. Any moment, roles and circles could be added, changed or eliminated within Voys. Apart from changes in ‘small’ roles, the current holacracy organogram did not radically change the last year. In Figure 3, the current holacracy ‘organogram’ of Voys is shown in figure 3 (add-on & adjusted figure of Lee & Edmondson, 2017).
Figure 3. Overview of Voys (Voys, 2018)
This study is focused on the mutual interaction and alignment between processes related to Customer support. Interviewees belonging to Bedstone, Voys, Customer Happiness and their sub-circle Webcare were interviewed. The following section will give an overview of Voys’ governance and organizational structure.
5. Case Analysis
The analysis is structured along themes that emerged from the literature and the case. The themes discussed are the self-managing tasks, organizational culture and work environment and the observed controls. Per theme, the formal implemented mechanisms and the actual state of affairs are explored to provide a two folded analysis of the case studied. When applicable, experiences with non-alignment and crises are interwoven.
5.1 Self-managing tasks
Over the years, the organization studied has grown, this growth had consequences for the way the SMO was organized. From minimal bureaucratic organizing the organization grew into a highly structured formal governance model, holacracy. Simultaneously the scope and impact of decisions have changed along over the years (Voys, 2018).
Voys operates through a holacracy governance model that prescribes to partly decentralize authority for the tasks managing and monitoring work execution, and work and resource allocation. Furthermore, roles have the joint responsibility to ensure control and accountability over work execution, organizational and work design, and personnel and performance management. Furthermore, Voys partly decentralized the authority for the task strategy creation. The self-managing tasks, tasks for which authority is decentralized at Voys are further explained in Appendix C. An overview of the self-managing tasks is shown in figure 4 (Voys, 2018).
Figure 4. Decentralized tasks at Voys (Lee & Edmondson, 2017).
5.2 Organizational culture and work environment
Voys describes itself, among other things, as open and transparent and they focus on personal contact based on equivalence (Leenders, Ottens, Beltman, & Vletter, 2014). Almost all interviewees brought along their experienced contact in the organization was based on openness, equivalence, personal contact and addressing accountabilities related to actions and organizational goals (Voys, 2018). One interviewee mentioned: “Self-managing facilitates discussions from a place of equivalence (Voys, 2018).” The transparency and openness are shown by working with Google drive for all work-related tasks and the current statistics that are hanging on the wall in the kitchen area (Voys, 2018). An employee addressed she has access to every document in the organization, but she did not care about how other people were evaluated. Furthermore, functioning evaluations were not surprising at all. As someone emphasized: “At Voys, you know where you stand (Voys, 2018).”
Employees express Voys’ mission, integrate Voys’ values in their answers and they indicate the willingness to contribute to the organizational goals. In general, interviewees mentioned they feel responsible for the clients and for Voys by stressing they feel responsible for their accountabilities and they really wanted to help customers as best as possible. Most interviewees point out that colleagues
are supporting and more than helpful. However you have to ask for help yourself. The holacracy coach mentioned that Voys aims to carefully select employees that both fit the job and seem to fit in the organizational culture.
If an organizational crisis would occur, many believed they would fix it together with self-managing in place. Most brought up they would stay loyal to Voys colleagues even if the organization would cease to exist, supported by one employee mentioning Voys is a real family.
In short, the data shows that the interviewees seem to be pro-organizational, highly committed to the organizational objectives. Furthermore, they address they feel the responsibly towards organizational goals (Voys, 2018).
5.3 The observed controls
Organizational and governance structure. Voys implemented a radically decentralized self-managing structure, where employees are trusted with self-self-managing over various ‘managerial’ tasks. The holacracy governance structure formally clarifies which role has which ‘managerial’ tasks and related decentralized authority. The authority that comes along with decentralized ‘managerial’ tasks’ gives employees in roles formally the responsibility and the freedom to ensure organizational accountabilities and purposes are pursued.
The holacracy coach mentioned that for a while it was not clear in which direction the organization was heading, what made it harder to define activities which were aligned with organizational objectives. Employees had different interpretations of the same values. The holacracy coach mentioned:
“There has to be a fence or framework, they need to know these are the boundaries, I can rub against these, however, I know if I stay in between them, I have the freedom to do everything I want and I know which direction is the way that Voys wants to go.. otherwise the freedom is too big and you do not know if going left or going right is beneficial.. You do not know which step is a step in the right direction. Therefore we need to give each other direction where we are heading.. I think that in every domain, every colleague at Voys, need to know his own context so he can undertake activities (Voys, 2018).”
The attached accountabilities and responsibilities in roles, and the changing and flexible work environment requires roles to monitor their tasks regularly to evaluate aligned with the organizational objectives (Voys, 2018). The fixed-order processes, guidelines, policies and procedures, attached to roles enable a role to monitor and evaluate alignment of their accountabilities.
These formally governance mechanisms are to authorize employees to identify and manage tension, also called gaps between the current situation and the desired outcome. This way the holacracy structure and radically decentralized authority for ‘managerial’ tasks formally facilitates alignment of work progress towards organizational objectives. Provided that the organizational objectives are clear and formally recorded, the employees can rely on the prescribed administrative guidelines to evaluate whether actions are contributing towards organizational objectives.
As mentioned in the findings related to managing and monitoring work execution, the data showed that employees perceive and executed some decentralized authority related to tensions because they inserted their grounded point of view about a situation. An example of perceived and executed decentralized authority: “During an internet outage, there were some disputes about something. Somebody wanted to decide what to do, however, it was not his domain. I said great opinion, however your roles are not related, I decide, and I chose something else. If somebody does not have the associated role, and the decisions do not impact their role, the tension is ungrounded (Voys, 2018).”
This incident also indicates that employees only have the managing authority to change tasks and roles that are related to their own. Furthermore, during this particular crisis, Voys still operated through their usual governance structure. Indicating that Employees of Voys can count on their formal self-managing and holacracy system at times, control and alignment is needed.
Flexibility. Within the fixed-order procedures, rules and guidelines, self-managing creates space for every circle and role to decide how they design their work and how to contribute to the organization.
When circles decide which policies they want to maintain, these should be registered in a ‘living’ support mechanism, named ‘Orakel’ (Hendrickx et al., 2016). For instance, a potential project must meet the conditions set for projects in the ‘Orakel’. ‘Orakel’ should explain how circles should operate, how they could assign roles, what boundaries could be used and how the interaction between circles could take place (Hendrickx et al., 2016).
The formalized prescriptions and the actual state of affair about functioning evaluation quite diverges because the information provided in ‘Orakel’ was outdated. Employees know they should update ‘Orakel’ more and they should look more into the registered guidelines of ‘Orakel’. The interviewees mentioned they rather asked colleagues for information, than looking it up in ‘Orakel’. One mentioned: “The information is too much in the minds/heads and too little in ‘Orakel’ (Voys, 2018).” Indicating not all formalized policies and procedures are updated or strictly followed. Assuming that control is not always derived from the holacracy governance structure.
Targets, metrics and progress. Different circles handle targets and metrics differently because their goals are very diverse. For instance, marketing is focused on attracting as much leads as possible, advisory is focused on concluding as many new customers as possible, while customer support or Human resources have more purpose focused on retaining as much happy customers and colleagues. Indicating that Voys steers more explicit on non-financial targets than financial targets.
A tool to keep every employee aware of progress are the televisions with statistics about, for instance, the week’s average client satisfaction rate. Every employee can monitor work execution and progress towards targets by, for example, evaluating metrics and softer key performance indicators. Some employees mention that they actively monitor the statistics (Voys, 2018).
Within Voys, the marketing and advisory circle, steer mainly at metrics. Marketing together with Advisory are mainly responsible for the work of other parts of the organization, when these circles are performing less the whole organization is going to feel it (Voys, 2018). This metric system is clear,
provides progress and has an early alarming functioning when performance is beneath the prognose. This way Voys can act quickly on changes in the market and the way circles are functioning towards goals.
The whole circle is responsible for pursuing their goals and together they evaluate the progression towards goals. When the metrics disappoint, the circles explain themselves at the super-circle or are asked for the reasons why the metrics disappoint. The early warning enables the circle to investigate the variance between the actual and the desired outcome and evaluate the ability to change their course. They are given time to modify their actions and behavior to boost the metrics, however, when it not improves they are expected to dive into the causes and come with a solution plan. This way, monitoring metrics ensure that the organization can quickly react on changes in the market and progress towards organizational objectives is pursued.
The marketing Lead Link addresses that the statistics somewhat disturb her: “Every morning the statistics are red, because at the end of the day you are close to or you have reached your target and then the colors turn yellow and green (Voys, 2018).”
However, she addresses the importance of meeting the metrics for marketing and advisory because these circles are target driven and determine the growth rate of the company.
Rewards and compensation. It depends on the circle, whether they steer more on metrics and whether they set rewards for meeting targets. For example, Customer Happiness circle has defined a social activity as reward for meeting one of their targets. It seems that customer supporters perceive the feeling of proudness when they achieve metrics and are in general really focused on solving the customers’ problems. They do not really expect something in return and feel valued when the customer just addresses their gratitude.
Only the marketing circle has a budget that is a solid percentage of the turnover A good campaign leads to more customers, the turnover grows, and the marketing circle is rewarded with an absolute growth in their budget and vice versa. Furthermore, Voys does not steer on personal bonusses or sanctions for (not) achieving metrics. The functioning evaluation and salary system are based on equivalence and fully decentralized.
Enabling control. The governance model, holacracy, formally specifies responsibilities explicit and therefore, clarifies who is accountable and can be count on for fulfilling a task. When necessary, the interviewees embodying Lead Link roles point out they are ultimately responsible for ensuring monitoring and managing work execution, work and resource allocation and firm strategy are aligned with organizational objectives. They address, they are not responsible for employees to have work. Most employees help the Lead Link with managing and monitoring alignment in roles, work execution and strategy questions.
The supportive and transparent culture with pro-organizational open and supportive employees, could enhance the way control is ensured. Employees mentioned that roles or tasks are divided, based on who is most capable and who really wants to fulfill the role. Usually, employees embody more roles
at the same time and could change the roles they embody at every moment. The overall employment of an employee is not at stake when an employee does not fit a role. Lead Links mention they address tensions and feedback face-to-face when roles are not properly carried out, they rarely take away roles from employees. The employees might evaluate the monitoring by Lead Links as logical because all roles just as much as the Lead Link roles have to pursue their roles’ purpose and accountabilities. They might perceive the monitoring and role take back more as a tension related to role they embody, than feeling personally judged. Supported by the Voys manual, the purpose of feedback is to strengthen competencies and to attribute to the colleague’s self-knowledge, it is not meant for assessment (Hendrickx et al., 2016).
Solving non-alignment issues. Employees have access to the purposes and accountabilities of other roles and they know what to minimally expect from a role. An employee mentions: “When something occurs, then we can we address a related role (Voys, 2018).”
Not all employees were up-to-date about the way formal functioning evaluation actually took place. When asking interviewees about feedback and functioning, the answers were contradicting. One interviewee mentioned there was no formal feedback system at the moment and another mentioned Flyndt was actively used, however the frequency was unclear (Voys, 2018). The Voys manual and all interviewees brought forward that face-to-face feedback was the most important way of evaluating functioning and behavior in roles (Hendrickx et al., 2016, Voys, 2018).
Mainly alignment happens in consultation, when your role is influenced by a change and it creates tension, all interviewees point out they first use one-on-one face-to-face feedback to clear up the tension. When tension is not solved, the tension is brought in at a circle meeting. All addressed the ability and authority to speak up to others and claimed they used their decentralized authority in their roles to solve tensions.
Over the years, most main activities and purposes of circles, such as customer support, are not drastically changed. Questioning, the scope of non-alignment activities. It seems that non-alignment issues are mostly related to projects circles set, these seem sometimes ambiguous or repeated.
At the moment, the strategy alignment about those activities might be in jeopardy. Control about these activities is not ensured by the provided fixed structure, policies and procedures, nor by the Lead Link of the concerning circle or broader circle. The Voys United Lead Link evaluated first the bottom-up strategy session as really circle driven with a lack of alignment with the circle above. She emphasizes that in general most decisions that circles decisions are all good choices when you are zoomed in, however, they are not all aligned with the organizational purpose. Her perspective indicates that the Lead Links responsible for alignment and some failed to ensure alignment of their circle strategy to the broader circle.
These non-alignment issues seem not be driven by self-interest. These seem to be driven by differences in interpretations of organizational values and the inability to recognize value-adding projects. However, roles and Lead Links of the associated circles might think these activities are
aligned. She mentioned that some discussions should not take place because they lack valid tensions. Overall, she concludes: “I think we are good players of holacracy at tactical and operational level, however at strategic level we are in the busy with learning phase (Voys, 2018).”
6. Discussion
The discussion is structured as follows. First, the decentralized tasks and heterogeneity between SMOs are discussed. Second, the assigned, perceived and carried out decentralized authority is discussed. Third, possible MCS to align employees’ behavior and decentralized authority are discussed.
6.1 Tasks for which authority is decentralized
A quick overview of the case decentralized tasks compared to other researched SMOs cases is illustrated in figure 5 (add-on & adjusted figure of Lee & Edmondson, 2017). Zappos represents a ‘typical’ holacracy by only decentralizing authority for tasks that are prescribed by the ‘holacracy constitution’ (Lee & Edmondson, 2017; Robertson, 2015). On the contrary, Valve stresses informal and personal control and ‘managerial tasks’ are bounded by the temporary existence of a task (Foss & Dorbrajska, 2015). At Morningstar managerial work is distributed in a democratic way (Lee & Edmondson, 2017).
Figure 5. Decentralized tasks at Voys versus other organizations (Lee & Edmondson, 2017).
As shown in table 1 of Appendix F, Voys shows similarities with the holacracy governance structure, by fully decentralizing the authority for the tasks work execution and organization and work design. As prescribed by holacracy, the case partly decentralized managing and monitoring work
execution, and work and resource allocation. The case implemented control mechanisms related to these decentralized tasks could provide guidelines for controlling these tasks at other holacracies (Robertson, 2015).
In contrast to employees working under managerial authority, employees in roles formally have the joint responsibility to ensure control and accountability over work execution, functioning and performance evaluations and over organizational and work design, whenever work cannot be fully specified in order to achieve organizational objectives (Ouchi & Maguire, 1975; Williamson, 1981). By employing pro-organizational employees, monitoring tasks and roles might be more likely to be executed in SMOs, because stewards want to pursue and contribute to the organizational objectives (Davis et al., 1997)
According to the ‘holacracy constitution’, firm strategy is set by the broadest super-circle, and therefore should not be decentralized (Lee & Edmondson, 2017; Robertson, 2015). Firm strategy is partly decentralized at the case studied. Therefore, the organization studied diverges on this aspect from the holacracy perspective. While, the ‘holacracy constitution’ lacks guidelines about personnel and performance management, the case studied has fully decentralized personnel and performance management (Robertson, 2017).
Alignment efforts related to these two tasks might be more specific to the case studied than holacracy specific. Therefore, related control mechanisms should be interpreted carefully with the specific case context into account. However, the findings could provide relevant information about the way a SMO could implement controls regarding to these decentralized tasks (Lee & Lings, 2008).
The studied case operates according to highly formal coordination mechanisms. Therefore, the findings of this studied case seem less representable for studies operating through more personal controls as at Valve (Lee & Edmondson, 2017).
6.2 Decentralized authority
The data shows that most employees show behavior in common with stewards (Jensen & Meckling, 1994; Davis et al, 1997). Additionally, there are strong indications that Voys treats their employees more like stewards than as agents.
Voys has implemented a radically decentralized self-managing structure, where employees are trusted with a high level of free space to operate. The authority that comes along with decentralized ‘managerial’ tasks’ gives employees the responsibility and the freedom to ensure their accountabilities and purposes are pursued. In the hiring process, Voys carefully selects employees that both fit the job and fit into the organizational culture. Furthermore, Voys steers on personal contact and feedback. Aligned with the stewardship perspectives, these mechanisms function as controls and these are likely to increase the stewards’ performance and motivation and indirect the firm performance (Argyris, 1964; Davis et al., 1997). Contradicting with common ways to align self-interested employees from the agency perspective, Voys does not steer on financial incentives, nor on reporting obligations, and lacks
