Abstract
A vital function of organizational structure is the allocation of human resources to tasks. Based on an empirical study of five organizations in the state sector in Norway, we identify and analyze a structure of allocation that potentially constitutes a theoretical category between a unitary, hierarchical structure and a resource/task matrix. The core trait of this structural category is the partial overlap of task and resource management among middle managers. Such partial overlap provides the basis for analyzing other structural dimensions: -the principle of grouping of resource units, -mechanisms of resource allocation from resource units to task units, -task management, and -challenges for middle managers. The main advantages are a higher degree of resource flexibility, the weakening of rigid “employee ownership” at the unit level, and decisions and solutions based on cross-disciplinary and cross-unit competencies. A key challenge is to address the new ambiguities and lack of role clarity that such a structure implies.
Plain language summary
In unitary organizations, each employee has one manager and role clarity and predictability prevail. In a matrix organization, an employee has two managers: A resource manager and a task manager. Here we describe a middle ground between the two well known structures.The core trait is the partial overlap of task and resource management among middle managers. Thus for some employees, the middle manager is responsible for resource management, while the actual work is done under the management of other manager. Among other employees, the middle manager is responsible for task management, while a different manager is responsible for resource management. The main advantages of such a structure are a higher degree of resource flexibility, the weakening of rigid “employee ownership” at the unit level, and decisions and solutions based on cross-disciplinary and cross-unit competencies. A key challenge is to address the new ambiguities and lack of role clarity that such a structure implies.
Keywords
Introduction
The allocation of human resources to tasks is a vital function of the organizational structure. When organizations strive to become flexible and adapt to turbulent environments, they may modify their structure, de-bureaucratize, and deviate from a unitary hierarchy (Adler, 2001; Fjeldstad et al., 2012; Lee & Edmondson, 2017; Martela, 2019). While research has analyzed several issues and dimensions of such organizational adaptations, the question of how tasks and resources are matched has not received much attention. For example, in their review of post-bureaucratic organizations, Lee and Edmondson (2017) cite environmental turbulence and a higher incidence of knowledge work as important drivers of flexibility and innovation, but they mention neither the significance of resource allocation, nor the mechanisms and challenges of matching resources to tasks. In those cases when resource (or task) allocation is included in the analysis of organizational structure, research presumes (a) that it mainly takes place inside a unitary hierarchy (Puranam, 2018), (b) that it is captured by the ideal types of bureaucracy, adhocracy or self-managed organizations( Martela, 2019), or the HR Quadriad or similar project matrix (Bredin & Søderlund, 2011b; Keegan et al., 2017; Søderlund & Tell, 2012), or that (c) such decisions can be assessed in terms of degrees of decentralization (Lee & Edmondson, 2017).
Thus, there is a research gap as regards structures of resource allocation outside the generic forms of unitary hierarchy, matrix and self-allocation. Here, we identify and analyze a particular structure for resource allocation, that occupy a middle ground between a unitary hierarchy and a matrix of resource and task units. The core trait of this structure is the partial overlap of resource and task management for middle managers. The research questions are: What is the rationale for such partial overlap, what structural dimensions are related to this trait and how are the challenges of these dimensions addressed?
The empirical study is based on data from five organizations in the Norwegian state administration that have moved away from a unitary hierarchy. The main contribution of the study is the identification and analyzes of structures of resource allocation that previously have only received scant attention in research. In addition to partial overlap of task and resource management we analyze the principle of grouping of resource units, mechanisms of resource allocation from resource units to task units, task management and challenges for middle managers. We argue that the structure identified in the analysis constitute a potential theoretical category, and suggest several avenues for further research. A secondary contribution is to application of concepts and findings from the domain of Human Resource Management (in project-based organizations) to enrich the generic understanding of organizational structures.
In the next section, we describe the conceptual basis of the paper. We then discuss the data and methods that is the basis of the empirical analysis. Next, we describe and analyze the new structures for resource allocation empirically, continue with a discussion of the findings and provide a short conclusion.
Theoretical Basis and Previous Research
Task/Resource Allocation
Here, we aim to provide insights to better understand the match between an organization’s set of tasks on the one hand and its human resources on the other hand. This matching process is referred to as task allocation in the organizational structure literature and as resource allocation in the human resource management (HRM) and project management literature. Puranam (2018, p. 10) defines task allocation as “the problem of mapping tasks obtained through task division to individual agents and groups of agents.” In the literature on HRM in project-based organizations, the concept of resource allocation (or “flows”) refers to job rotation and assignment to and release from project teams (Bredin & Søderlund, 2011b; Keegan et al., 2017). Thus, while the two concepts have their origin in different disciplinary domains, they capture the same organizational phenomenon; how human resources with competencies and labor power are allocated to tasks (time-limited projects or activities of an ongoing, recurrent nature) that are to be performed.
Three Models of Task Allocation
Inspired by Martela (2019) we distinguish between three basic models of resource allocation; the unitary hierarchy, the resource/task matrix and self-assignment. The latter model comprises contributions that maintains that self-selection of a task is a feature of many new forms of organizing (Puranam, 2018; Raveendran et al., 2021), and an actor-oriented model of organizational structure based on actors who have the abilities and values to self-organize inside and across organizational boundaries (Fjeldstad et al., 2012). Since the empirical study here captures structure along the hierarchy—matrix continuum, we will not go into further details on models of self-assignment.
Unitary Hierarchy (“R Equals T”)
In a traditional line organization, the one-employee-one-boss principle applies. A unit such as a department establishes a system of common supervision among positions and units, requires that common resources are shared, creates common measures of performance, and encourages mutual adjustment (Mintzberg, 1979). The department is part of a task hierarchy, in which there is a successive decomposition of tasks into more specialized ones that on the lowest level can be performed by an individual or a team (Kolbjørnsrud, 2018; Martela, 2019). According to Martela (2019, p. 9): “Having established the necessary broader tasks for the accomplishment of organizational goals, the top management allocates these to the next level of management, which in turn makes the allocation decisions in regard to their subordinates. Thus, each manager is responsible for making the task allocation for the layer of employees directly below themselves.” The implication is that an operative organizational unit in this structure shares both a set of tasks and the human resources required to perform these tasks. Thus, as shown in Figure 1 the resource hierarchy matches the task hierarchy and the resource and task dimension for the frontline manager overlap.
Unitary hierarchy (R equals T).
Two Hierarchies: The Resource/Task Matrix (R Differs From T)
There are many possible principles of grouping in organizations, related to outputs (products, market segments, projects), inputs (functions, specialized competencies, work processes) and geography (Mintzberg, 1979; Worren, 2012). An organization often faces dilemmas in terms of which principle of grouping to apply. However, rather than choosing one principle, the organization may apply a combination of these (Mintzberg, 1979). In a multidimensional organization, several principles of grouping are built into the structure at the same level in the hierarchy (Worren, 2012). Considerable variation exists among such multidimensional organizations, in terms of which grouping principles the units are based on, the nature of the interdependencies between the units, as well as the balance of power between these units (Worren, 2012, Mydland & Nesheim, 2017).
To understand resource allocation, a specific multidimensional (matrix) constellation is particularly relevant. The secondment matrix (Knight, 1976) is based on grouping according to input (functions) and output (projects) units. Here, a project manager oversees the project. The project participants are assigned from several functional units and work on a full-time basis, and the functional managers are not involved in the project work. This type of structure was “rediscovered” and analyzed in greater depth by researchers working at the intersection of HRM and project management (Nesheim, 2021; Bredin & Söderlund, 2011a; Keegan et al., 2017). While project participants belong to a line unit, they also report to one or more project managers in relatively unstable relationships: “People are temporarily grouped together around specific tasks to be solved, after which the team disbands and may or may not collaborate again in different compositions later” (Bakker et al., 2012, p. 384). Thus, in project-based organizations, each project initiates processes of assignment and allocation to the project, project participation and engagement, and project dispersion. An implication is that the criteria for and the mechanisms of the allocation of human resources to a project become core issues in these organizations (cf Keegan et al., 2017; Nesheim 2018).
Bredin and Söderlund (2011a) identified the HR quadriad, a model that includes the main human resources tasks and key actors in inter-functional project organizations. Building on previous studies, they differentiated between two separate roles, the line (resource) manager and the project manager. This is illustrated in Figure 2 below. The former is the manager of a resource unit, and his or her role tends to focus on human resources issues such as the assignment of personnel, competence development, appraisals, and career counseling. The project manager is responsible for project (task) execution and is dependent on the line manager for personnel resources. He or she may also be a vital source of input to the line manager’s appraisal of employees.
Resource/task matrix (“R separate from T”).
A partly similar matrix is described in the literature on agile organization. According to the new “orthodoxy” of agile management (Nesheim, 2023; De Smet, 2018), cross-disciplinary, stabile product teams (rather than temporary project) typically constitute the operational units (the task dimension) of the organization, while the resource dimension may be handled by a separate resource manager.
Thus, in a resource/task matrix the principle of unitary hierarchical authority and responsibility is violated (Nesheim, 2021; Keegan & Den Hartog, 2013. Instead, there are two co-existing hierarchies (of resources and tasks), where the task and resource units are not hierarchically ordered. Employees report simultaneously to a resource manager |and (one or more) task managers in relatively stable (De Smet, 2018) or unstable relationships (cf. Keegan & Den Hartog, 2013). The resource/task matrix begs the question of the mechanisms of allocation or “matching” of resources to tasks. In one study, Engwall and Jerbrant (2003) analyze how the scarcity of resources has resulted in a resource allocation syndrome, enhanced by dependencies and competition between projects, reallocation of resources, and short-term managerial decision-making. Sankaran et al. (2019) evaluate how project managers influence the assignment of team members and identify five different lateral strategies through which influence is exerted. Nesheim (2021) 7 found that networking and relationship building were a key factor for resource managers in their endeavors of resource allocation to projects.
Between the Unitary Hierarchy and the Resource/Task Matrix
Unitary, hierarchical organizations where resource and task management overlap have well-known benefits such as clear roles, a low level of ambiguity, and a high degree of internal predictability (cf. Lee & Edmondson, 2017; Mintzberg, 1979; Worren, 2012). However, changes away from such a structure of allocation are common, and driving forces include environmental turbulence and unpredictability (Lee & Edmondson, 2017), a higher incidence of projects at the core of the organization (Bredin & Søderlund, 2011b), and the potential gains from inter-departmental problem solving and task execution (Lee & Edmondson, 2017). The resource/task matrix is an alternative to a unitary hierarchy, but also have its own potential contradictions and weaknesses. For example, an implication of the resource/task matrix is that complementary manager roles are required along both dimensions, thus increasing the total number of managers (Ford & Randolph, 1992). In addition, this matrix creates a great deal of dependencies between units, which must be handled in direct dialog between managers and through formal arenas of coordination (Sy et al., 2005; Fjeldstad et al., 2012).
Thus, in terms of resource allocation, there are pros and cons of both a unitary hierarchy and a resource/task matrix. Another alternative, not captured systematically in research, is to depart from a unitary hierarchy in a selective manner, without adopting a matrix structure with separate roles for resource and task management. For example, many organizations combine recurrent (or syclical or seasonal) task and project-based tasks in the core activities (Nesheim 2020, Nesheim & Schou, 2024). As the level of complexity increases due to variation in types of tasks, a pure resource/task matrix may not be appropriate. Momeni and Martinsuo (2018) studied organizations where the carrying out of projects and non-projects is based on allocation of personnel from the same resource units. The two case firms were system suppliers that deliver systems as projects to industrial customers, and complemented their systems with non-project service activities. They found that the customers’ service recurrent requirements were a source of uncertainty for resource allocation in addition to changes and delays in projects, and that firms responded with bottom-up or hybrid rather than hierarchical resource allocation processes.
In the product management literature, studies have shown that a firm’s operational and innovative activities often are organized differently. The main operational and exploitative activities (including incremental innovation efforts) may be managed through a unitary hierarchy, while developmental, exploratory explorative activities are often built around teams that span functional boundaries (Hobday, 2000; Liker, 2004). For example, Artto, Kulvik, Poskela & Turkulainen (2011) found that innovation groups used part-time representatives from various “line” organization functions when selecting, processing, and advancing companywide innovation projects.
Thus, the literature has described a number of cases that indicate a middle ground between a unitary structure and a resource/task matrix. Here, a middle manager partly serves as a combined resource and task manager (similar to a unitary hierarchy), and partly as a resource manager that allocates human resources to development teams or projects (similar to a resource/task matrix). While the existence of such structures has been documented, their conceptual dimensions, specific of challenges of resource allocation, and how these are handled have received scant attention in the research literature.
The Empirical Context of the Study
This article is based on data from three separate studies of a total of five organizations in the state sector in Norway. Here, the state administration has two levels: the ministries and the state agencies. The ministries are led by a minister who is a member of the government. The ministries assist the minister in running the various sectors of government administration. In Norway, most employees in state administration have open-ended employment contracts and are not replaced when a new government is appointed. The state agencies implement government policies in the various areas, and open-ended employment contracts prevail.
Data and Methods
Five Case Studies
We identify and explore structures in five organizations that have departed from a unitary hierarchy and selectively implemented elements from a resource/task matrix, using qualitative data. Qualitative methods can be used in range of ways to study and to theorize organizational phenomena. Among the different strategies and templates of qualitative research (Langley & Abdallah, 2011), we emphasize the nomothetic approach described by Eisenhardt (1989). Such an approach is recommended when little is known about a phenomenon or where current perspectives are conflicting or confusing, and where case study evidence can therefore be seen to contribute novel insight (Langley & Abdallah, 2011). Here, we explore allocation challenges when resource and task management partly overlap, and the organization combines elements from a unitary hierarchy and a resource/task matrix. Such structures have been identified in the literature (see above), but have not been analyzed in depth. The five case organizations have all departed from a unitary hierarchy but differ in the specifics of the structure chosen after the change.
The analysis comprises a total of five cases and is based three separate studies on de-hierarchization in the state sector in Norway:
Study A: two government ministries (A1 and A2), data collection conducted in 2014 (Mydland & Nesheim, 2017).
Study B: a state agency (B1), data collection conducted in 2018 (Nesheim, 2018; Nesheim, 2020).
Study C: two state agencies (C1 and C2), data collection conducted in 2020 (Brattli et al., 2021).
An overview of these studies (cases, data and sub-themes covered) is provided in Table 1 below. Here, we (a) build on the content of these studies, that is, the arguments, descriptions and analysis reported in previous publications and (b) supplement with new analysis of the data collected for the three studies. Simply put, previous publications from three studies comprised similar, but not identical themes. Consequently, there were “gaps” in the analysis of the various sub-themes. In order to cover all the relevant sub-themes (for the present paper) in an appropriate manner for each of the five cases, we analyzed the primary data (collected in 2014, 2018, and 2018 respectively) in October 2020.
Threes Studies of De-Bureaucratization.
Choice of Sub-themes and Narrative Structure in the Empirical Analysis
This choice was informed by the specific research questions and sub-themes analyzed in the four publications described in Table 1, as well as the author’s evolving understanding of the issues in question. The following narrative structure was chosen: We start with the background of the structural changes (5.1), before we describe the core characteristics of the new structures (5.2). Then, four elements of the structure are analyzed in further detail (5.3–5.6.).
Analyzes of Each Sub-theme
At a more detailed level, under each subtheme we present content from the previous publications, supplemented with new analysis of the original data. These data are partly “objective” such as the descriptions of formal roles and organizational units, and partly interpretative and based on the respondents’ perceptions of issues and challenges. Regarding the latter, we applied the principles of analysis at several levels of abstraction, in accordance with Gioia et al. (2013) For each interview, all relevant utterances have been coded (first order). In the next step we have assigned each code from the first order to a subtheme (second order), and then subthemes were assigned to themes (third order). This process has been conducted in an iterative manner, enabling us to back-and forth both between orders, and between respondents and conceptual categories. In addition, we also report analysis from the survey in study B.
Empirical Analysis
Background
The five organizations in question have gone through similar changes, in departing from a unitary hierarch. The changes were based on perceptions of weaknesses of the traditional hierarchical structure. In the two government ministries (A1 and A2), attention was on the horizontal division of labor and especially the relationship between the organizational units (sectors) at the lowest managerial level. The structure was perceived as rigid, with clear boundaries between the units, and as ill-suited for the solution of cross- disciplinary tasks and tasks that involved expertise across sectors. Middle managers were reluctant to allocate resources to other units, and a culture of resource ownership prevailed among these managers. An implication was that it took too long to adapt to new tasks and changing political priorities. Such tasks included time-limited tasks (projects) or activities of a more long-term nature. In addition, from an employee perspective, the structure was not perceived to be well-suited for development and empowerment (Mydland & Nesheim, 2017; Difi, 2012).
Study B analyzed a state agency that changed its structure around 2000. After the change, it has maintained a similar structure for two decades, with only minor adjustments. Before 2000, the agency had a functional-based line organization with three managerial levels between the director and ordinary employees. At the time, the organization was perceived as having too many managers and too many hierarchical levels and as being rigid in the allocation of personnel to tasks. In addition, it did not enable employees to develop their competencies and fulfill their potential. Decision processes and adaptations to new issues and priorities were perceived as too slow (AFF, 2000; Nesheim, 2018).
Similar motives were observed in C1 and C2 (Brattli et al., 2021). Here, many employees found that they were “locked in” their organizational unit, which often presented limited opportunities for development. Their perception was that they needed to apply for positions in other units to further their careers. They also perceived the allocation of resources between units as rigid, and middle managers typically “owned” the personnel in their own units. A culture of “organizational silos” weakened the potential for flexibility across internal units.
The Key Dimension: Partly Overlapping Resource and Task Management
Previously, the five organizations had a traditional hierarchical structure, where the manager at the lowest level was responsible for a set of tasks to be performed by the employees of the unit. Thus, the structure was consistent with the unitary, “R equals T” model for resource allocation. After the change, the resource and task hierarchy no longer coincided; rather, the elements related to the resource and task dimensions became clearly differentiated. One possible option was to create separate roles for each dimension and establish a resource/task matrix. None of the organizations chose this structure. Instead, in the five cases we find that a middle manager holds the roles of both resource manager and task manager. However, the two roles do not involve the same employees. In her capacity as manager of a resource unit, the person is responsible for the development and appraisals of several employees, while in the role of task manager, the incumbent coordinates several tasks that involve many employees in its execution. In all five organizations, the task dimension included both time-limited tasks (projects) and recurrent, ongoing tasks. The employees assigned to the resource dimension only partly overlap with the employees assigned to the teams of the task dimension.
Figure 3 illustrates the core trait of the five organizations and how they differ from a unitary hierarchy and the resource/task matrix described in the theoretical part of the article. The circle on the left-hand side refers to resource management, while the circle to the right-hand side refers to task management. “R and T” depicts a condition of overlap between the two dimensions, while “R only” and “T only” refers to a situation where only one dimension applies.
The manager role at the lowest hierarchical level: partial overlap of resource and task ownership.
While the structures in the five case organizations were similar in terms of resource and task management, they were the outcomes of different processes. In A1 and A2 the boundaries between the units at the lowest hierarchical level were made more flexible and permeable. In B1, C1, and C2 the new structures were to a larger extent the outcomes of a redesign of the organization structure. In the further elaboration of this structure, we analyze four specific elements; the basis of membership for the resource unit, how resources are allocated to tasks, task management in teams, as well as challenges of middle managers in the new structure.
Resource Unit: Basis of Membership
In each of the five organizations, all employees belong to a resource unit and are allocated from this unit to various tasks in the organization. The basis of a resource unit, that is, the criteria for membership, differs among the organizations. In A1 and A2, the employees belong to the same units as before the change in structure. Here, the membership basis of the units did not change. Rather, it was emphasized that the boundaries between existing units became more permeable, allowing more flexible allocations and exchanges of personnel across units. In B1, several options have been applied during the past 20 years. Today, a resource unit typically consists of employees working on similar tasks and constituting more than one discipline.
C1 and C2 (two of four departments) chose a more radical approach in assigning employees to resource units. Before the change in 2017, each unit in these organizations was based on a discipline or employees with similar competencies. After the change, employees were assigned randomly to the resource units. The rationale was to break up strong networks and dissolve rigid structures. By “loosening” the relationships connected with the resource units, the intention was to improve cooperation between units and promote the flexible allocation of employees to tasks.
After the change was implemented, many employees expressed concerns about the loss of social affiliation and disciplinary identity. Two illustrative statements are:
There is a lack of social affiliation. We were randomized to units where no one works on the same issues that I do. If I am not present at work, no-one will notice. (Employee, C2). The challenge is that people lost their sense of belonging and their disciplinary identity…. Previously, the disciplinary and resource manager was the same person … and the former role was very strong…. Now everything is up in the air and distributed. (Middle manager, C2).
Thus, the bases of membership of resource units and the degree of change differ markedly between the cases. In A1 and A2, there was no significant change in membership. In contrast, the randomization observed in C1 and C2 was the outcome of a radical change, that involved breaking up established networks in the organization. In B1, there has been minor adjustment to the resource units in two last decades.
Mechanisms for Resource Allocation
In a unitary hierarchy, allocation of personnel resources is a matter of intra-departmental task assignment. When resource and task management no longer overlap, it begs the question of how resource allocation takes place across departments and internal units. We found that in A1 and A2, such allocation was handled partly on a case-by-case basis by managers and partly through formal application processes for positions in projects in the organization. The following statements illustrate these internal differences:
For some temporary tasks in the ministry, we get formal applications and appoint a project manager. But this is not necessarily the way it is done. There are not clear procedures (for resource allocation) …typically we discuss this on an ad-hoc basis (Middle manager, A1). The choice of project manager differs according to the whether the project is initiated…from the political level to the internal unit…. You will find all variants (Middle manager, A2).
The allocation processes involved managers at various levels in the ministries. Further, it was emphasized that the leader group at the department level should function as a team, by taking on a common strategic responsibility and working more closely together. A culture encompassing such a broader mindset across mangers was perceived to increase resource flexibility.
In B1, flexible staffing—“the right person for the right task unit at the right time”—was strongly emphasized. It is the organization as a whole (rather than a specific unit) that should “own” the human resources. Most positions in task units are announced internally, and employees apply for these. The decisions on the allocation of personnel to work units are taken in a collective arena of 12 middle managers.
Employees often have preferences for certain work groups and may influence assignments. They are expected to rotate between work groups and gain experiences from different types of work. However, in the survey we found that 53% worked in long-term work groups only. In addition, 58% reported that they have a strong influence on their participation in long-term work groups, and 75% noted that they prefer to continue working in the same long-term work groups in the next 2 years. A high score for a preference for further participation in long-term work groups, combined with a perception of high influence on the allocation to such groups, indicates that the level of rotation is lower than intended in the organization.
C1 and C2 have modeled the decision-making on resource allocation on the B1 example. C1 include the whole organization in the allocation arena, while C2 has one allocation meeting arena for each department. In both C1 and C2, a key motive for a more flexible allocation was to reduce the impact of strong personal networks and unit “ownership” of personnel. However, respondents in the five organizations noted that personal networks and relationships still influence allocation decisions. For example, one respondent noted:
Some of the middle managers still have a certain amount of ownership to some of the employees and make sure they are allocated to their own projects…. [While] there are strong connections between a middle manager and an employ, this is also what the employee wants (Middle manager, C2).
There is a large variation of resource and competencies, and this tend to complicate the decision process:
The whole organization is involved in the meeting, and it is not easy to present a case to the others, because the employees work in different areas. To prioritize between two different areas and decide which is most important is not easy (Middle manager, C1).
The “politics” of resource allocation is another matter. Participants in the allocation arena represent different internal interests. Since participants differ in competence and experience, there may be unbalances in the decisions taken. As expressed by one respondent:
It is impossible for all members (of the meeting) to be fully updated in all areas. It may well be that the person who “shout the loudest” or are able to present the most persuasive arguments win through (Middle manager, C1).
Thus, when an organization occupies a middle ground between the unitary organization and resource/task matrix, novel mechanisms of resource allocation arise. The cases reveal a combination of both formal and informal mechanisms, as well as a number of dilemmas in such decisions.
Task Management: Coordination Rather Than Control
As described previously, the incumbent of the resource manager role is also a task manager, but the two roles only partly overlap in terms of the employees involved. A1 and A2 underwent few formal changes in the practice of task management after the structural change. However, it was emphasized that the middle managers should work closely and have a “departmental” rather than a “narrow sector” mindset. There were also examples of selective delegation of task management roles, such as the delegating the role of project manager to employees. A dilemma between control and autonomy among middle managers was observed. On the one hand a flexible and more team-based organization enabled delegation. On the other hand, the political organization in Norway (where authority ultimately rests with the minister) and the middle managers perception of a threat to their position and status limited this effect. An illustrative statement is:
The work is delegated to the employees, but there is no delegation of decision authority. This is a major difference between a political system and other systems, that the minister has all authority (Middle manager, A1).
In contrast B1, C1, and C2 made more drastic changes to the organizational structure. B1 redesigned the structure, taking the tasks at the operational level as a starting point. First, the core of the organization was perceived to consist of products (outputs) to be delivered to its prime “customer,” a government ministry. Between 30 and 40 products were identified and described at the time. It was stated explicitly that these were either long-term tasks (e.g., related to a specific geographic area or a group of firms in the relevant industry) or short-term tasks that required problem solving inside a pre-determined time horizon.
Secondly, each product (or set of products) was allocated to a work group or team. Contrary to the typical hierarchical model of state agencies and ministries, these operative task units now became the main building blocks of the organization. The teams should largely be cross-disciplinary, self-organized, and coordinated by one person serving as team coordinator. Typically, an employee participated in several teams. While minor adjustments to the team-based organization have been made in the past 20 years, the principles remain the same.
Such an operative, product mindset differs from both a top-bottom and an employee-centered approach to the structuring of an organization. At the same time, even the relatively “flat” structure of B1 requires certain hierarchical elements to ensure coordination. One such element is the coordination of teams that work in a similar area. and the need to handle task boundaries and interdependencies between teams. Today, such task management is the responsibility of a middle manager, who also serves in the resource manager role. The intention is that coordination of teams should not require detailed control of and involvement with each team, to preserve the self-managed nature of these work units.
C1 and C2 have modeled task management on the structure applied by B1. After the change, C1 and C2 established a similar team-based structure, with a middle manager that coordinates several teams inside an area, in addition to serving in the resource manager role. In B1, C1 and C2, the task manager has some discretion in how the role is executed, based on the type of task, the actual challenges in a specific team, and his or her experience. Many respondents noted that there is a balancing act between autonomy and control in the team-based organization. For example, the survey in B found that 18% of employees perceived that the middle manager intervened too strongly in the operation of the teams.
Challenges for Middle Managers
As described above the core characteristic of the structure is the partial overlap between the resource and task dimensions of middle managers. In four of the five cases (exception: B1) the challenges of such overlap were emphasized by middle managers. One recurrent issue in the interviews was the perceived lack of role clarity and high degree of ambiguity experienced. Two illustrative statements are:
I find the lines of decisions to be very unclear. They may be unambiguous in the eyes of top management but are not perceived in such a manner at the bottom of the organization. You wonder: Who is really responsible? (Middle manager, C1). When you are involved in different tasks, you may be pulled in many different directions, in projects and other tasks. This is a source of frustration for many employees. (Middle manager, A1)
When a manager relates to an employer in an “R only” capacity, the manager often lacks the information necessary to perform her role in a satisfactory manner. One middle manager said:
If you are their resource manager (only), you are not able to observe how they perform the job and assess the quality of their work. You lack relevant knowledge of your employees, and therefore the quality of the appraisal interview suffers. (Middle manager, A2).
In a “T only” constellation, informants emphasize a lack of formal authority over employees. An illustrative statement from a project manager is as follows:
I perceive that the role of project manager requires a certain “fingerspitsgefuhl” as regards the execution of leadership. The challenge is that you do not own the resources in question. So, people must deliver, right, while they have other deliveries and tasks at the same time. It is not always easy to gain progress when you are not the resource manager, it challenges you more (Resource manager, A2).
In contrast, the middle managers in B1 did not perceive these relationships as especially challenging. Here, the change from a unitary hierarchy had been an institutionalized aspect of the organization for a long time before data collection. Therefore, we expect that roles and practices had been developed and internalized, enabling middle managers to cope with these challenges.
Discussion
Comparisons: Overview of Findings
Table 2 provides an overview of the empirical findings, that enables us to compare the five cases. Table 2 shows that while the background for the changes and the core trait of the structure (partly overlapping task and resource management) were similar, we observed variety in the basis of membership of resource unit, mechanisms of resource allocation, task management as well as the challenges related to middle management in “R only” and “T only” situations.
Overview of Empirical Findings.
Contribution of the Study: Toward a New Structural Category
While the partial overlap of resource and task management is implied by earlier research (Artto et al., 2011; Momeni & Martinsuo, 2018), it has not been analyzed in detail in relation to resource allocation. Through the application of the concepts of (a) resource vs. task management and (b) project vs. recurrent tasks we have described and analyzed how five organizations have selectively departed from the overlapping resource and task management, that is a vital aspect of a unitary hierarchy. The core trait of the new structures is the partial overlap of the two dimensions. If we go beyond the findings of this particular empirical study, a key question is whether this “middle ground” represents a stabile structure of allocation that should be theorized, rather than simply empirical observations at a point in time that deviates from the well-known structures of unitary organization and the resource/task matrix. Here, we suggest five arguments that supports the first position, enabling us to develop a theoretical category.
1. A unitary organization is appropriate when task dependencies mostly are handled inside an organizational unit (cf. Thompson, 1967; Worren, 2012). When task dependencies cross organizational unis (as described in the cases above), organizational alternatives will be considered. Thus, if task dependencies cross organizational units, both a full resource task matrix and more selective departures from the hierarchy is a possibility. Which of these alternatives that are appropriate, are bases on four factors (2–5 below).
2. The resource task matrix is a two-dimensional structure. It is only appropriate when there are several task units (such as projects or product teams) and these task units require contributions from several disciplines or special competencies (cf. Bredin & Søderlund, 2011b). If the work in many of the task units are based on or two disciplines, there is a much weaker rationale for the full resource/task matrix, and the “partial overlap of T and R” is an alternative.
3. The literature has described organizations where core, operative activities (in line units) are combined with change and innovation projects with selective participation of employees in line units (f.ex. Artto et al., 2011). Here, a selective departure from a pure unitary organization rather than a resource/task matrix is appropriate.
4. Even if there are several disciplines that contributes to most task units, the resource task matrix may not be chosen. In a unitary organization, middle managers are required for their competencies and experiences related to the actual work, for coordinating and delegating work; in addition to the resource dimension. A change toward the resource task matrix would require they take on a pure resource manager role, with a perceived loss of influence, status and changes in work tasks (Nesheim, 2021). Thus, such a change may be perceived as not being favorable, and it may also be difficult to recruit managers for such positions.
5. Another consideration is that a resource task matrix requires a double set of specialized middle manager roles. Thus, the implementation of a full matrix implies larger costs as well as the complex administrative mechanisms often associated with this structure (cf. Bredin & Søderlund, 2011b; Knight, 1976).
Thus, when condition 1 (implying a move away from the unitary organization) is combined with conditions 2, 3, 4, and/or 5 (a resource/task matrix is not appropriate), there are reasons to expect that the organization structures of allocation occupy a middle ground between a unitary organization and a resource task/matrix, as shown in Figure 4 below. These arguments are suggested as boundary conditions for the theoretical category. We believe that the partial overlap between the resource and task dimension is a stabile characteristic of the organization, and should be investigated as such, rather that an empirical “outlier” or a temporary pattern. The study is the first that has identified and analyzed such a structure in depth. From the empirical analysis, we have described and analyzed four different aspects of the structure; emphasizing what the challenges are and how they may be handled.
Structures for resource allocation in organizations.
Avenues for Further Research
Based on the study, there are a number of opportunities for further research. A key issue for research is the further clarification of the structures for allocation identified here, such as the number and content of the various aspects of and challenges of the structure “type.” Research may also analyze the five boundary conditions suggested in above in further depth. Another issue is the significance of the various motives and drivers of de-hierarchization. For example, if there is an external shock (not observed in the cases here), or management perceives large weaknesses of the existing structure (such as in B1, C1, and C2), this may drive more radical change, compared to other motives.
While the middle management roles in the five cases are similar regarding the contrasts to the “R equals T” role in a unitary hierarchy, future research may also build on the variation documented here. What is the actual degree of overlap between the resource and task dimension, and how are the challenges perceived by middle managers related to different degrees of overlap? How do middle managers handle the “R only” and “T only” challenges? How do mindsets and practices related to “R” and “T” management evolve and (may) become institutionalized over time? Other avenues are to look out for other hybrid constellations of “R” and “T” than the one identified here, and the possible co-existence of different “R and T” constellations in the same organization (see also Nesheim and Schou, 2024).
As regards the resource dimension separately, we have shown that resource units may be based on different membership criteria. Especially, we have documented randomized membership of such units, that differ from membership based on discipline, specialized competencies and similar work area or tasks. This particular change was regarded as radical and, to a large extent, perceived in a negative manner by the employees involved. Further research could investigate (in a larger sample) the association between a strong, rigid network of human resources before the change and the choice of membership basis of resource units after the change. In addition, one could analyze whether the lack of social affiliation and knowledge sharing documented here associated with randomized resource units are sought to be compensated (or ignored) over time.
Regarding the task dimension, B1, C1, and C2 have all designed a team-based organization. Both long-term tasks and projects are allocated to cross-disciplinary teams with a high degree of autonomy. The teams are intended to be self-managed and coordinated by a team coordinator, which is not a formal managerial role. The middle manager, in her capacity of task manager, oversees several teams. Further research may describe and analyze dilemmas of control and autonomy in such structures, as well as the challenges of balancing long-term and project teams.
In the process of de-hierarchization, the norm of allocating resources in “a flexible manner that reflects strategic priorities” is emphasized in all five organizations. We have shown that different mechanisms may be applied to realize such intentions. In three organizations, allocations decisions were formalized and outcomes of collective arenas, while they tended to be more informal and ad-hoc in A1 and A2. Future research could describe the various mechanism of allocation when organizations deviate from a unitary hierarchy in a more detailed manner. In the analysis of such structures of allocation, three aspects may be taken into consideration. First, in the absence of a unitary, unambiguous hierarchy, the total time and administrative costs of allocation should be compared to the gains from a more flexible organization. Second, allocation involves several units and organizational interests, and power based on superior information, network connections, and negotiation skills may have a larger influence on outcomes compared to organization-wide, strategic priorities. Third, because allocation decisions are important for competence development and careers of employees, considerations of organizational justice, legitimacy, and transparence are also relevant. Finally, how do employees influence such allocation decisions, and what are the (potential) drivers toward self-assignment?
Conclusion
Organizational structures have an impact on the allocation of human resources, but research has not adequately analyzed how structures that don’t conform with a unitary hierarchy or the resource/task matrix influence resource allocation. This article has intended to address this research gap, and contributes to the literature in several ways. We identify structures of allocation that occupies a middle ground between a unitary hierarchy and such a resource/task matrix. We analyze characteristics and challenges of a such structures of resource allocation, and argue that they constitute a potential theoretical category, and not only empirical patterns observed at a point in time. The conceptual distinctions of resource versus task management and time-limited versus recurrent tasks have made it possible for us to identify this category, as well as related issues that have received scant attention in the literature, such as challenges when resources and tasks “belong” to different units, the criteria for membership in resource units, and the mechanisms for resource allocation. A specific contribution is the application of concepts and findings from the domain of Human Resource Management (in project-based organizations) to enrich the understanding of organizational structure.
Inside its boundary conditions, advantages of the type of structure in question are a high degree of resource flexibility, the weakening of rigid “employee ownership” at the unit level, and decisions and solutions based on cross-disciplinary and cross-unit competencies. To realize these intentions, middle managers have a key role. They should to able to overcome the challenges of lack of role clarity and ambiguousness in relations to employes and other middle managers. When the roles of resource manager and task manager partly overlaps, it is important to engage in dialog to handle different and more complex relations to employees, as well as issues of resource allocation in arenas with other middle managers.
While the “deviant” structures of allocation have been described earlier in relation to innovation (Artto et al., 2011) and the combination of production and service activities (Momeni & Martinsuo, 2018) in market-oriented firm, our analysis is based on a study of the state sector in Norway. Here, there is a relative high degree of environmental stability and low task uncertainty. We believe that the relative lack of research attention to allocation in the (more uncertain and dynamic) private sector reflects the path dependencies of existing perspectives and research themes, rather than a low significance of resource allocation in organizations as such. Thus, future research might investigate structures for allocation in several organizational contexts. Among firms in the private sector, a higher level of adaptation flexibility is often expected, given the high degree of competition and the emphasis on innovation related to organizational ambidexterity, business models, products, and processes. Therefore, one should expect to find an abundance of structures, mechanisms, and challenges of allocation for research in this context.
