Knowledge-intensive work is quite different from the physical manufacturing work that birthed the corporate hierarchies widely prevalent today. Consider the incompetent Pointy-Haired Boss in the Dilbert comic strip: He cannot lead through wisdom or better information, and he is unable to control his subordinates because their work depends on effort that he cannot observe or comprehend.
Add the usual challenges of transmission losses – such as control and information – in multi-layered hierarchies, and one might justifiably ask if hierarchies are an anachronism and simply have no place in today’s knowledge-intensive economy. Some academics have even argued that hierarchies may be effective instruments of execution but are ultimately ineffective at adaptation in the face of complexity and uncertainty.
In line with this thinking, many organisations – from tech companies to professional services firms – have experimented with agile or flat designs that dismantle traditional forms of hierarchy in order to harness the distributed knowledge of specialised individuals.
But are there certain situations in which the presence of a hierarchical structure can be useful to cope with complexity and uncertainty? In other words, can hierarchy and authority figures create agility even if the boss is no wiser than the subordinates, nor able to directly control or monitor their behaviour?
In our research, we used a computational model to build “digital twins” – simulations of teams that possessed either hierarchical or flat structures. We investigated the function of authority in these teams and the structural role that bosses play. Our findings reveal that organisations that need to achieve agility – in the sense of rapid results – in environments that require coordinated action among team members may benefit from a hierarchical structure of influence in which a boss influences subordinates more than the other way round.
Balancing variety-seeking and convergence
We discovered that flat organisations may explore too many options and take too long to make a decision, while hierarchical organisations may attain convergence more rapidly but can also miss out on identifying the very best alternatives. Therefore, hierarchies can be most useful when the benefits of making a good (if not the best) coordinated decision quickly outweigh the potential downsides of not finding the absolute best solution.
Crucially, this benefit of hierarchical influence is a pure structural effect and does not depend on the characteristics of each roles’ occupants within the structure. The “agents” who occupy the apex role in our computer-simulated hierarchies were no wiser than their subordinates and were incapable of perfectly controlling their actions.
When organisational adaptation requires all team members to contribute to the search for valuable interdependent actions, bosses provide stability, while subordinates produce the variation needed for search. Put simply, Dilbert’s boss can be useful simply by virtue of exercising some authority on his team.
These results offer a perspective on why hierarchical structures – in the form of multiple layers of asymmetric influence – may persist even when the leader has no distinguishing knowledge, foresight or authority. Even within teams performing highly creative work, a hierarchical structure can be more useful than arms-length or peer-to-peer interactions if the team operates in task environments in which both search and coordination are important.
Why hierarchies persist
In the urge to begin flattening organisations and democratising hierarchies, there are some natural breaks. Pushing things to the limit by creating completely flat systems, where everyone exerts equal influence on each other, is unlikely to be the best approach to adaptation. These structures may not do well when the need for innovation also entails the need to coordinate.
While organisations striving to implement flat structures can remove hierarchical layers, it may not be the best course of action for them to fully dispense with some degree of asymmetry in terms of influence. Consider online communities that were initially set up as flat systems, but soon discovered they required some form of asymmetric influence with certain members of the community taking on leadership roles in order to function.
The design challenge for managers attempting to create alternatives to the traditional command-and-control hierarchy may be to find ways to preserve the benefits of asymmetric influence. However, they need to do this without creating disparities in power that can quash the diverse set of ideas of organisational members or lose the valuable sense of fairness and participation among employees.
Rachel Eva Lim
About the author(s)
is an Associate Professor of Organisation and Management at Emory University’s Goizueta Business School.
Daniel A. Levinthal
is the Reginald H. Jones Professor of Corporate Strategy and a Professor of Management at The Wharton School.
About the research
“The Dual Challenge of Search and Coordination for Organizational Adaptation: How Structures of Influence Matter” is published in Organization Science.