You
no doubt have heard about Zappo’s, the on-line shoe retailer that offers $2000
to new employees who decide after their initial training that they no longer
wish to continue with the company. A couple of years ago, Zappo’s founder, Tony
Hsieh, decided to introduce a management concept called holacracy (a variation
of the self-managed team concept) and offered all employees the choice of
staying with the company and embracing this new concept or taking a generous
severance package. Eventually, by the time the deadline had passed, 260 of the
company’s roughly 1500 employees (about 18%) decided to take the offer – which
was on average about 5.5 months’ worth of salary.
Zappo’s
specific implementation of holacracy was developed by Brian Robertson, a former
programmer, and is perhaps a more structured and extreme variation of the
concept of self-managed teams. Zappo’s
is not the first company to experiment with some form of self-managed teams;
Thorell (2013) claims to have identified 18 “bossless” companies in the U.S.
(not including Zappo’s), with 6 of them operating internationally.
Another
company, W. L. Gore, has become famous not only for its revolutionary fabric
but also for its commitment to self-managed teams. Here is an excerpt from its
web site:
“How we work at
Gore sets up apart. Since Bill Gore founded the company in 1958, Gore has been
a team-based, flat lattice organization that fosters personal initiative. There
are no traditional organizational charts, no chains of command, nor
predetermined channels of communication.
“Instead, we
communicate directly with each other and are accountable to fellow members of
our multi-disciplined team. We encourage hands-on innovation, involving those
closest to a project in decision making. Teams organize around opportunities
and leaders emerge. This unique kind of corporate structure has proven to be a
significant contributor to associate satisfaction and retention.”
The
distinguishing feature of such self-managed teams is the absence of any formal
hierarchy. This means there are no formal titles or promotions, and no one has
a “boss.” Team members negotiate their responsibilities with each other,
employees rank each other to help with pay decisions, and team members can hire
and fire. These applications of
self-managed teams are consistent with Carson et al.’s (2007) definition: those
in which team members have greater responsibility for setting their own goals,
monitoring their own progress, and making their own decisions than do team
members in manager-led teams.
Note
that self-managed teams are not, strictly speaking, leaderless. Rather, leadership
is “shared” or distributed. In their meta-analysis on the impact of “shared”
leadership on team effectiveness, Wang et al. (2014) conclude that the effects
of this type of leadership is stronger for attitudinal outcomes and group
processes than on team performance.
Actually,
the majority of work organizations not only in the U.S. but globally are based
on hierarchical principles. Some have tried to experiment with flat
organizations and have not had much success. Even Google learned this the hard
way, when its founders Larry Page and Sergey Brin flattened the organization
early on but pulled the plug a few months later when they found that this was
just not working.
According
to Galinsky and Schweitzer (2015), hierarchy is the most dominant form of
social organization across all species: “Hierarchy helps people know who does
what, when, and how. These rules promote efficient interactions by setting
clear expectations for the behaviors of people of different ranks. Essentially,
hierarchy facilitates social interactions by simplifying them.” (p. 67)
Kenney
and Anderson (2012) have summarized the literature on status hierarchies in
groups, and they point to two theories of hierarchy: the functionalist and the
dominant theories, and propose a third one, what they call the micropolitics
theory. In brief, the functionalist theory says that hierarchies form to help
groups make decisions more efficiently. They give some individuals who are seen
to be the most competent more control over decisions. According to the
dominance theory, hierarchies form as a result of competition and the
assertiveness of some members over others. While the research shows
overwhelming evidence for the functionalist theory over the dominance theory,
the authors propose a blending of the two: that hierarchies form as a result of
who can help the group most and also through individuals’ desire for status.
Regardless of the evidence for each of these theories, the reality is that most
social and work groups establish hierarchies early on, and they are relatively
difficult to eliminate.
In
my view, four factors need to be taken into account for any organization
interested in moving towards self-managed teams: cultural preferences,
individual differences, organizational context, and the nature of the work or
task. First, cultural preferences. As Hofstede and others have demonstrated
over the years, cultures that are high in uncertainty avoidance and high in
power distance in particular have a strong preference for hierarchy. Several
years ago, I was coaching a Taiwanese executive who was the general manager of
a global company’s subsidiary. His company had just introduced the matrix form
of organization, and Ron (not his real name) commented to me: “This is difficult.
Most Asians want to know ‘who is my boss?’”
In
a recent study, Herbert et al. (2014) found that perceptions of shared
leadership varied depending on the collectivistic orientation of the
participants in their study (over 350 members of various virtual project
teams). And in a survey of over 200 Mexican executives, Nicholls et al. (1999) also
found numerous challenges in the implementation of self-managed teams in the
Mexican work setting. As demonstrated by Hofstede and others, Mexican cultural
values are closely related to collectivism, high power distance, and high
uncertainty avoidance. While a collectivistic orientation might help provide a
positive reaction to self-managed teams, the last two certainly do not, as their
interview findings suggest. Anicich et al. (2014) have been studying the
effects of hierarchy both structurally and culturally. In their studies of
expert mountain climbers from 56 countries on over 5,000 expeditions, they
found that expeditions from more hierarchical countries had more climbers
reaching the summit, but also more climbers dying along the way. These findings
illustrate both the benefits and drawbacks of hierarchy, which can create
efficiencies but, by suppressing participation and voice, can be dysfunctional.
Second,
individual differences. People will vary in their need for structure and their
own motivational level, as well as in their testosterone levels. I would
predict that individuals who tend to have a strong desire to be open to
experience (one of the personality characteristics of the Big Five theory of
personality) are more likely to be comfortable with such self-managed teams. Furthermore,
many studies have shown that high-testosterone individuals are dominant and
like being in high-power positions. By implication, such individuals like to be
in charge and prefer hierarchy. In fact, as Galinksy and Schweitzer (2015) and
others have written, two baboons that have high testosterone levels become very
competitive and when testosterone levels vary, the baboon with the lower
testosterone will walk away. Similar findings have been found in research on
other primates and animals. Perhaps this is why co-CEOs are rare and when they
do happen, they do not last long. It is rare for two alpha dogs to cooperate
over a long period of time! As an aside, Galinksy and Schweitzer (2015) suggest
a simple test to determine a rough estimate of your testosterone level: compare
your ring finger with your index finger. The longer the ring finger is relative
to the index finger, the more you were exposed to greater levels of
testosterone in the mother’s womb – and this exposure is an indicator of your
testosterone.
Third,
the organizational context. Moving from an authoritarian corporate culture to
one that is self-managing is not an overnight transition, as anyone knows who
has worked in companies trying to create large-scale change. In fact, there is
something ironic in a leader who dictates that henceforth his or her
company/division/group will implement self-managing teams. In addition, because
there will inevitably be a period of confusion and uncertainty when this
concept is implemented, it is generally not a good idea to introduce it when an
organization is going through a number of changes at the same time (e.g.,
introducing a new business model while restructuring and downsizing).
One
of the successful examples of a self-managed team is the Orpheus Chamber Orchestra,
an ensemble of musicians that has operated without a conductor for many years.
The orchestra has been written about and studied over the years (e.g.,
Vredenburgh and He, 2003) and it has certainly thrived. Yet if you examine the
context, several features are notable. First, it has only about 25 musicians,
since it is a chamber orchestra, not a full orchestra. Second, it has
established a culture of collaboration which is congruent with the self-managed
team approach. Third, members have self-selected into the orchestra so that its
musicians are motivated by self-leadership and reciprocal influence. The
orchestra is, however, NOT leaderless; rather, leadership roles rotate. Here is
one of their principles and a brief explanation: “Working in shared leadership. Every musician has the opportunity to play leadership
roles such as leading rehearsals or directing the performance of a new musical
composition. The decision rests on the group; the leader is chosen according to
expertise, strengths and interests (Luc, 2011).”
Fourth,
the nature of the work or task. Self-managed teams seem to work best when members
are working on tasks that are complex and require high levels of collaboration
and information sharing. Ronay et al. (2012) performed a series of experiments
in which they assigned participants to different groups working on tasks that
were either high in procedural interdependence or low in procedural
interdependence. Participants’ power levels were “primed” so that there were
three different types of groups: groups of 3 high-power participants, groups of
3 low-power participants, and groups of 1 high-power, 1 low-power and 1
baseline participant. The mixed-power groups were significantly more productive
than the other two groups but only for the tasks high in procedural
interdependence. In a second experiment, they created different types of groups
(groups of high-testosterone participants, groups of low-testosterone
participants, and a mix of high-, low- and average-testosterone participants) and
had them working on the interdependent tasks. The mixed groups were
significantly more productive than the other two groups, with the high-testosterone
groups experiencing much more intragroup conflict than the other two types of
groups. The researchers conclude that “… the functional benefits of hierarchy
are most pronounced under conditions of high procedural interdependence …” and
“…intragroup conflict mediated the performance decrements for the
high-testosterone groups, but not the low-testosterone groups.” (p. 675)
The
intent of organizations promoting self-managing teams is certainly admirable.
Too many organizations have created cultures where people are afraid to speak
up, and where ideas (especially from the rank-and-file) are not listened to.
The effects of too much hierarchy, combined with the presence of
high-testosterone individuals, many of whom are also narcissistic, can be damaging
for teams and organizations that not only need to be high performing, but also need
to be innovative and resilient. Giving employees more of a voice and empowering
them is in principle a great idea. And moving away from the command-and-control
model to a more collaborative model will resonate with many employees today.
For
organizations interested in making the move towards self-managing teams and
less hierarchy, here are three pieces of advice. First, assess your
organization’s readiness for this change (more on this in a subsequent blog).
Second, do some experimenting and piloting, while making sure you involve
organizational members and get their feedback. Third, allow for transitions,
keeping in mind that the choice is not necessarily a binary one between an
authoritarian, hierarchical model and a self-managed model – but may well lie
in variations on this continuum.
Anderson,
A., and J. Kennedy. (2012). Micropolitics: A New Model of Status Hierarchies in
Teams, in Margaret A. Neale, Elizabeth
A. Mannix (eds.) Looking Back, Moving Forward: A
Review of Group and Team-Based Research (Research on Managing Groups and Teams,
Volume 15.) Emerald Group
Publishing Limited, 49 – 80.
Anicich,
E.M., Swaab, R.I., & Galinsky, A.D. (2014) Hierarchical cultural values
predict success and mortality in high-stakes teams. Proceedings of the
National Academy of Sciences. doi: 10.1073/pnas.1408800112
Carson, J. et al. (2007). Shared Leadership in Teams: An Investigation
of Antecedent Conditions and Performance. Academy
of Management Journal, 50 (50: 1217-1234.
Galinsky, A. and Schweitzer, M. (2015). Friend & Foe. New York:
Crown Business.
Garvin, D. (2013). How Google Sold Its Engineers on Management. Harvard
Business Review.
Luc, E. (2011). The 8 Leadership Principles of Orpheus, the
Conductor-less Chamber Orchestra.
http://blogue.edithluc.com/the-8-leadership-principles-of-orpheus-the-conductor-less-chamber-orchestra/?lang=en
Nicholls, C. et al. (1999). Taking Self-Managed Teams to Mexico. Academy of Management Executive, 12 (2):
15-25.
Ronay, R. et al. (2012). The Path to Glory Is Paved with Hierarchy: When
Hierarchical Differentiation Increases Group Effectiveness. Psychological Science, 23 (6): 669-677.
Thorell, L. (2013). How Many Bossless Companies Exist Today? http://www.innovatini.com/how-many-bossless-companies-are-there/
Vredenburgh, D., and I. He. (2003). Leadership Lessons from a Conductorless
Orchestra. Business Horizons,
September-October, 19-24.
Wang, D. et al. (2014). A Meta-Analysis of Shared Leadership and Team
Effectiveness. Journal of Applied
Psychology, 99 (2): 181-198.
