The necessary evil of hierarchy

Harold J. Leavitt was a professor at Stanford University until his death in 2007. He is regarded as one of the early pioneers in the field of Organizational Behavior.

He spent five decades doing research on communication, decision making, and small groups, and teaching managers about leadership and organizational development. He wrote the textbook “Managerial Psychology” in 1958, which is still used, 54 years after it was first published.

Leavitt was a self-confessed “humanizer”, he was one of the “people people”, he said, being concerned about the necessity of creating more employee engagement by delegating responsibility and fostering widespread participation in decision making.

He contrasted his group of people with the “systemizers”, people who are primarily concerned about increasing organizational effectiveness using tools based on analysis, logic, consistency, and control.

Toward his end of his career, however, he became much more of a systemizer himself. He concluded that we should acknowledge the existence of authority and the inevitability of hierarchy in large organizations.

His starting point was the claim that the traditional hierarchical form of organization is dead and that it will be supplanted by flatter and radically decentralized organizations.

As he points out, this claim is far from new: It was made by Mary Parker Follett (a British leadership expert) in the 1920’s and has been repeated ever since. Below I have included some quotes from more recent writings by management gurus Michael Hammer, Tom Peters, Jim Collins, and Gary Hamel.

If we were to believe these humanizing management gurus, you would conclude that every modern organization consists of (or soon will consist of) self-managing teams, with no formal management hierarchy to direct them. As Gary Hamel says, you can fire all the managers!

Leavitt agreed that many changes have occurred during the last few decades. Employees are more highly qualified and need less direct supervision. Organizations are (somewhat) flatter. More work is outsourced to external partners. Managers spend more of their time coordinating work horizontally across formal units. Yet he concluded that the despite these changes, the hierarchical organization is not dead – it is not even in ill health. Here’s a quote from one of his articles:

Even today, just about every large organization remains hierarchical. The organizations of the knowledge economy—whether loosely coupled, networked, or federalized—seem to be no more than modifications of the same basic design (…) the basic blueprint is unchanged. Subordinates continue to report to superiors, much as they historically have done, at GE and IBM. Department heads report to division managers, who report to group VPs and so on.

Leavitt agreed that hierarchies suffer from many dysfunctions, and he said that he didn’t particularly like them (who does?). They seem to conflict with our egalitarian values. As Gary Hamel complains, you give power to a “caste” of managers who have authority by means of their position instead of their competence or contribution.

But Leavitt concluded that there was no other viable system for accomplishing complex tasks. He had not observed even a single case where a non-hierarchical form had been able to match this capability.

It’s a bit like Winston Churchill who remarked that democracy is the worst form of government, except all those other forms that have been tried.

If we accept this reasoning, what does it imply, or what should it imply? It doesn’t mean that we should accept authoritarian management styles. It doesn’t mean that we shouldn’t try to remove the dysfunctions of large organizations. It doesn’t mean, either, that we shouldn’t experiment with new organizational forms.

It may imply, however, that we should read what management gurus write with a bit of skeptical realism. That we should ask for documentation when somebody proclaims that there is a new and revolutionary organizational form. That we should accept the need for a system of authority and accountability in an organization that intends to operate efficiently.

Accepting the inevitability of hierarchy actually represents an opportunity for improvement. Rather than pretending that the formal structure does not exist, it means that we should focus on whether we have a well designed hierarchy, whether we have an organization that develops its people, gets the job done – and delivers what key stakeholders expect. That is what the effective hierarchies have done in the past – and what they will continue to do in the future.