Can we manage without managers?

The management column (Bartebly) in this week's The Economist makes the case that managerial hierarchies are valuable & inevitable, and flat organizations represent a "false utopia:"

Why companies need middle managers
Organisations embrace flat hierarchies at their peril

The article is paywalled, but the basic logic is as follows:

  1. Flatter hierarchies will create a "leadership vacuum [that] risks being filled by petty tyranny."
  2. Models like Holacracy may wok for small operations, but "clearly won't do for a Unilever or Goldman Sachs."
  3. If you remove layers, you'll end up ovewhelming people at the top: " Not every Tesla factory hand is going to seek their annual appraisal from Mr Musk."
  4. Supervising people is a core leadership function in the organization, so multiple of layers "means those in charge of managing lots of people have had experience managing fewer people before."

This is an evergreen argument that surfaces periodically (you can find versions of it on the pages of Fortune or Harvard Business Review in the 1960s).  There are several  questionable claims and assumptions embedded in it, but let me start with the first, which is a classic: that flatter hierarchies lead to chaos and dysfunction, "Lord of the Flies"-style.

If you replace managerial hierarchies (which work-sort of) with nothing other than hope or vague operating principles, you're going to get a chaotic, wasteful, and arbitrary environment. The 1970 piece on the "Tyranny of Struturelessness" by Jo Freeman is often cited by people who make this point (and so did Max Weber, sixty years prior).  Fair enough... but this is a bit of a strawman (there are better options).

The Tyranny of Stuctureless
The Tyranny of Stuctureless by Jo Freeman

Second, not all self-management models that are popular these days may easily scale to support a 50,000+ organization.  The Bartelby column takes aim at Holacracy--I'm no Holacracy expert so I don't have a strong point of view here, other than to observe that the largest-scale example of it I'm aware of was Zappos, with a few thousand employees. But again, Holacracy is by no means the only way to run a large organization without layers upon layers of  managers (there are better options, still).

Third, firms like Haier, Nucor, Handelsbanken and Buurtzorg employ tens of thousands of employees while operating with a small managerial & administrative footprint-- they have a couple of layers of management between the CEO and the front line, plus tiny support functions relative to competitors (see Humanocracy for more details). Despite a super-lean management structure, they have industry-leading performace on productivity, efficiency, customer satisfaction, and innovation.  How do they do that?  The key is to separate the "what" from the "how."

These vanguard organizations have figured out how to get the control and coordination that managerial hierarchies deliver, without paying the "management tax" (e.g., the overhead, inertia, insularity, and politicking).  For instance, they rely on accountability to customers (internal or external) and peers, as well as on transparency and team-based incentives, to ensure there's no slacking or infighting. They don't get rid of management as a set of activities (e.g., planning, allocating, reviewing)--but they syndicate it to the broader organization.  For instance at Nucor, the US steelmaker, safety audits are peformed by operating teams visiting sister plants, as opposed via safety inspectors sent by HQ (turns out, those who work in a similar setting are best positioned to identify issues or blindspots).  The work of managing doesn't go away, obviously.  It just happens with far fewer managers and administrators.

Fourth, the notion that leadership is primarily about managing people is the relic of a bygone era--one in which most workers where unskilled, information was hard to obtain and share, and administrative competence was in short supply.  In the post bureaucratic vanguard organizations cited earlier, leadership is defined as one's ability to multiply the capacity of others to achieve shared goals, irrespective of position or rank.  

Just because the ladder has fewer rungs doesn't mean leadership opportunities are scarce-quite the opposite.  By giving people the ability to gain influence (and compensation) based on accomplishment as opposed to advancement,  an organization ends up with more, not fewer leaders.  And these leaders don't have to devote their talents and energy to politicking or sabotaging each other in zero-sum promotion battles. It's not often mentioned, but in a managerial hierarchy-with influence largely defined by position-there's only so much power to go around.  This doesn't elicit the best workplace behaviors, as you might have noticed.


The experience of the post-bureaucratic vanguard shows that, with some ambition and imagination, it's possible to run large, complex organizations with new and improved models.  However useful the manager-centric template described by business historian Alfred Chandler was in the 19th/20th centuries, we need something better suited to the challenges of the 21st.

Skepticism about the applicability of managerially lean models to large organizations reminds me of the pre-18th century critiques of democracy.  Most political philosophers couldn't imagine that form of governance "scaling" beyond a city, like Rousseau's Geneva.  And if we keep on assuming that the bureaucratic model is a lamentable yet inevitable feature of large organizations, then surely we won't make much progress.

There's plenty of evidence to suggest another viable path.  We just need to muster some courage and conviction to take it.