In Praise of Hierarchy

futurelab default header

by: John Caddell

After last week’s look at "The Future of Management," it seemed appropriate to look back at one of my favorite HBR articles, "In Praise of Hierarchy," by the late Dr. Elliott Jaques (link – $$). In this article, published in 1990, Dr. Jaques asserts that "35 years of research have convinced me that managerial hierarchy is the most efficient, the hardiest, and in fact the most natural structure ever devised for large organizations."

Despite appearances, "In Praise of Hierarchy" is not diametrically opposed to "The Future of Management." Dr. Jaques is open and candid about hierarchical organizations’ many failures. He blames poor implementation of hierarchy and states for years we’ve been attacking the wrong problem:

…One of the most widespread illusions in business [is] that a company’s managerial leadership can be significantly improved solely by doing psychotherapeutic work on the personalities and attitudes of its managers…. The problem is that our managerial hierarchies are so badly designed as to defeat the best efforts even of psychologically insightful individuals.

Dr. Jaques’ goes on to state that responsibility is best determined by looking at the time span of an employees’ longest task, program or project, and that "the boundaries between successive managerial layers [should] occur at specific time-span increments."

Meaning that an engineer who’s working on a module that is due in two months has a time span of two months. The project manager who is in charge of the new version that is six months out has a six-month span. The product manager who is planning the two year roadmap has a two-year span. And so on.

Dr. Jaques solution for improving hierarchy is simply this:

  1. assure that managers have "just enough authority to ensure that their subordinates can do the work assigned to them"
  2. create a quantum time-span difference between managerial levels (Dr. Jaques suggests examples of three months, 1 year, 2 years, 5 years, 10 years and 20 years)

The above always felt right to me when I was a manager. When conflicts arose it was typically because my manager and I (or I and my subordinate) were managing roughly the same time span. The layers

One would imagine that in the seventeen years since this article was published, there would have been progress. Instead, I think things are worse. I’ve read about CEOs who want to sit up high and assess at some times, and dig into the details at other times. Or we get so enamored with terms like "self-organizing systems" that we want to throw the baby out with the bathwater.

Instead of fixing our hierarchies, we’re patching them or trying to discard them. Maybe we should take a new look at Dr. Jaques’ old idea.

Original Post: