Managers’ work is largely a process of working across boundaries. In carrying out strategy, and in communicating workers’ needs and know-how, managers must work across hierarchical boundaries in which they represent and translate the experiences of different actors, including themselves. Increasingly, studies of middle management have highlighted mangers’ position as one of in-betweenness, and their skills as those of deftly navigating the ambiguities and misunderstandings, as well as the political differences, between various groups.

Increasingly, the idea of working across boundaries has drawn scholarly attention, particularly when these boundaries are in flux or are contested. In such situations, boundaries cannot be taken for granted and themselves become acted upon as part of the job role. What has been termed “boundary work”, however, has classically been studied in terms of how boundaries are reinforced or subverted to protect groups or allow their collaboration. Our recent paper in Journal of Management Studies attempts to nuance understandings by asking the question “how do middle managers employ boundary work so as to create forms of agency in their everyday practice?” Our core observation from this study is that middle managers’ boundary work does not simply reinforce or undermine boundaries, but selectively configures boundaries in certain ways to make them more or less visible, and more or less porous.

To better understand managerial boundary work, we ethnographically examined a Brazilian audit firm, using observations and interviews to understand the experience of boundary work and managers motives for carrying it out. We observed how boundary making was a highly political activity, constituting an important source of managerial agency as middle managers were able to filter and translate top-management ideas and to establish spaces from within which action was possible. They could close off or open such spaces in way that allowed them to consolidate their positions and to influence the organization “from the middle”. They could selectively promote social interactions and identifications between different units or ranks, or conversely, to distance actors from each other. Specifically, we identified four kinds of managerial boundary work:

Barricade Boundary Work – Barricades are boundaries that are easy to see and difficult to cross. Middle managers erected barricades to highlight and legitimize their own positions, making themselves essential gatekeepers for organizational practices. Positioning themselves as the spokespersons for management to their employees, and the employees’ spokespersons to management, they emphasized autonomy and authority.  Adopting unique practices for their own units, they rendered these indecipherable to others, creating parallel action spheres where they could exert unique control.

Façade Boundary Work – Façades give the impression of solid boundaries but are in practice porous and insubstantial. Middle managers engaged in façade boundary work when they desired to symbolically mark differences between groups that in practice required ongoing collaboration and exchange. The façade allowed them to regulate exchange at key moments and maintain a balance between autonomy and relationality. In façade boundary work, managers were able to de facto exert agency without challenging or calling into question established organizational hierarchies, a strategy that allowed them flexibility in practice.

Taboo Boundary Work – Taboos boundaries are those that must not be spoken, although they remain operative. Taboo boundary work occurred largely among peer managers, where cultivating a spirit of collegiality and cooperation required overlooking established power differences or diverse interests. Whereas façades involve publicizing differences while engaging in connections, taboos maintained surreptitious differences while maintaining a patina of togetherness. Much of what was “taboo” involved competition over scarce resources or status among managers, where making such competition visible would undermine the ability to draw on each other for essential resources and skills.

Phantom Boundary Work – The most elusive of boundaries, phantoms have an ethereal existence and can be passed through but haunt everyday work in a “spectral” way. In distinction from a situation of boundarilessness, managers would selectively ignore or bypass formal boundaries so as to establish connections, where “doing it despite” was felt as a part of the relationship building. In this way, phantom boundaries involved a presence-in-absence, where the knowledge of boundary crossing was part of the interactional functioning of the crossing.

Overall, complementing existing literature on boundary work, managers did not simply create or destroy boundaries, but worked on the composition of boundaries themselves, making them visible or invisible and modifying how much they could be crossed and by whom.  This paints a picture of middle managers’ work as foundation to the constitution of the organization itself, because the “walling in and walling out” practices of the managers were an ongoing kind of organizing work that is distinct from and less visible than top management strategy making. It implies that the everyday life of managers is key to understanding how organizations work, and how strategy will be operationalized in practice.

Our paper suggests that a future research agenda should focus on how everyday practices of boundary work shape the spaces of action within organization, and thus carries strong political implications.  We also note, however, that these practices were experienced as an unrecognized form of work by managers and were often subjectively exhausting. In effect, by changing the boundaries that define themselves and others, middle managers effectively shift their own identities in their day-to-day lives, and these practices can lead to confusion and personal burden as they seek to understand who they are and how they relate to others in their activities. Boundary work makes managers just as managers make boundaries, and the organization that results is not always predictable in advance. Our study thus suggests that we should pay more attention not only to how managers exist within organizations, nor to how they themselves make organizations, but about how they make themselves through their action on those organizations.