Are Diversity Initiatives Worthless?

Recently, Jack Welch, former CEO of GE, made some controversial comments at a Women in the Economy conference sponsored by the Wall Street Journal.  What did he say that upset at least some of the female attendees? First, he said that working hard and showing how your skills can benefit the company are the keys to getting ahead.  In effect, he said, “over deliver … performance is it.” Unlike what some may believe, this is not what upset them; who could argue against this, in the first place?

He then criticized mentorship programs and other diversity initiatives for women, referring to them as “victims’ units.”  He even mentioned some female executives who approached him while he was at GE, telling him that they refused to participate in these kinds of programs.  By inference, Welch would probably also argue against any kind of diversity initiatives for African-Americans or minorities.

We all know the numbers.  Of the Fortune 500 companies, only 3% have a female CEO today.  A survey of 60 major companies by McKinsey shows women occupying 53% of entry-level positions, 40% of manager positions, and only 19% of C-suite jobs.

In my experience, Welch represents the mindset of a generation of white male executives (mostly in their sixties), some of who still believe that there is true meritocracy in corporations, that there are no barriers to anyone getting ahead other than your own internal ambitions, and that regardless of the culture or work environment, those who are successful find ways to make it to the top.  In this Darwinian world, there is no need to do anything special or different for diverse groups. You just have to figure it all out, since “the cream rises to the top.” For these executives (I know; I have worked with quite a few of them in my career), diversity initiatives, affinity groups, and support networks for women (and by extension, African-Americans) are unnecessary and even unfair.  And some women and African-Americans agree with them! Taken to an extreme, what Welch implies is that managers should have no responsibility in developing others. Just leave them alone and let them figure it out for themselves.

Contrary to what Welch implies, there continue to be cultural, systemic, and organizational barriers to success in today’s work place.  The evidence is overwhelming, and I don’t need to rehash this in this column. Here are a couple of points I would like to offer based on what we know from the science and practice of Industrial-Organizational Psychology.  

First, we know from research and from schema theory that we have filters and expectations about individuals that tend to bias our perception of them.  And one of these pervasive biases is a “similar-to-me” bias. We tend to like those who are like us, and tend to react favorably to those with whom we perceive to have similarities.  No question that this has been a barrier to females getting ahead. Fortunately, through diversity programs and the track record of many outstanding women in the work force, I believe that individuals in corporations today are more “enlightened” than they have been in the past.  But the biases still exist. In Europe during the eighties, many orchestras changed their practice from having judges watch and evaluate potential orchestra members audition in front of them to having them audition “blind.” That is to say, the applicants performed behind a curtain so that the judges could not tell whether the applicants were male or female.  This simple practice led to a dramatic increase in the proportion of female orchestra members.

Second, while very few if any corporate executives would argue against evaluating people other than for their performance (as Welch suggests), how that performance is viewed can be subject to bias.  Here, attribution theory can shed much light.  Attribution theory states that we as managers not only evaluate performance, but also try to determine the causes of that performance.  Is the reason for their performance based on ability, effort, luck, or some other factor? A manager’s evaluation of the potential of an individual may depend therefore not just on his or her performance but also ona the manager’s answer to this question of what caused the performance.

Welch implies that it is all about performance.  But wait. Isn’t this the same Jack Welch who in GE introduced the famous 2 X 2 matrix where managers were evaluated not just on their performance (on the one axis), but also on their values (the other axis), and that a manager who performs well but who does not have the right values should be “terminated?”

Unfortunately, our biases creep into our evaluation of the causes of performance.  There is a lot of evidence, for example, that male managers tend to attribute the performance of their female subordinates more to luck than to ability or effort.

So what are the implications to individuals and to corporations of the Welch assertions?  First, for individuals, there is no question that your performance is your “foot in the door,” your ticket for punching your way to the dance.  This will mean making some personal sacrifices and trade-offs, and working some long hours to build a successful track record if your ambition is to be a successful executive.  But I don’t believe that this means rejecting whatever support and help you can take advantage of, whether within your company or outside the company. For example, many of us need to build our networks (as Reid Hoffman calls it, your personal board of directors) and if your company offers programs to help you with this, there is no reason not to take advantage of them.  Believe me, the Welches of the world (white males in their sixties and seventies), when they were rising stars, had their own network and support system. It may not have been formalized, but they still took advantage of them. And many of these groups excluded women, whether intentional or not.

For managers, this means that your responsibility as a manager includes developing and coaching others.  Catalyst just published some recent research demonstrating that a majority of high potentials received developmental support and are in turn developing others in their organizations.  This “culture of talent development” is critical for companies today, and yet Welch, of all people, would seem to suggest it is not necessary, or even desirable.

For corporations, continuing to provide mentoring programs, affinity groups, and similar initiatives – and more broadly implementing diversity initiatives – will provide them with a competitive advantage.  After all, the business case for diversity in attracting, developing and retaining talent is well-established, notwithstanding the opinion of Mr. Welch.

Global Mindset Part II

An example:  you are an expatriate manager of a multinational company in a Middle Eastern country  and you have just found out that no women are allowed to even apply for certain jobs in your department.  You say to yourself, “I just don’t get it.”  Another example:  an executive who works with Korean nationals once expressed his frustration to me that Koreans will never tell you what they really think.  “Why can’t they just be candid like Americans?”

I could give many more examples to illustrate reactions to differences in cross-cultural management practices that suggest a gap in global mindset, especially in one aspect:  that of developing empathy, which suggests an ability (and willingness) to understand another person’s or group’s perspective.  Actually, lots of research suggests that this skill differentiates effective negotiators from average ones.  For managers working cross-culturally, I believe that this “perspective-taking” skill is critical.  As two researchers from the University of Chicago (Epley and Caruso, 2008) have stated, “… the ability to accurately adopt someone’s perspective is better than chance but less than perfect.”  They point to three barriers, which I will paraphrase here. 


The first barrier is “activating” or switching on in our minds a willingness to do this.  As managers and leaders of global teams, this is sometimes difficult to do when there are so many mental balls that we are juggling.  And if we have not even made the effort to learn about other cultures, or to recognize that our way is not the only way, switching mentally to consider practices from another person’s perspective will be tough.  Our default mode is our own perspective, our own way of viewing things.    

The second barrier is our natural tendency is to react to things from our own perspective.  In one experiment which they cite, participants were asked to send either sincere or sarcastic messages to another participant, either over the telephone or via e-mail.  They were asked to predict, for each of 10 sincere and 10 sarcastic messages, whether the recipient would interpret the message correctly or incorrectly.  Recipients were not significantly better than chance at distinguishing between sarcasm and sincerity over e-mail, but not surprisingly, were significantly more accurate over the telephone.  But the senders did not think there would be any difference in the recipients’ accuracy when communicating over e-mail or the telephone.  “The senders’ intentions to communicate sarcasm or sincerity were so clear that it rendered them unable to appreciate … that the perception of the person on the other end of the computer monitor would be very different from the person on the other end of the telephone.” 

From my experience, I can recall many times when executives say they don’t understand why their messages are not being understood, or are being misinterpreted by employees.  If the executive working with Korean nationals has asked them for their opinions and they don’t give him any, it must be because they prefer not being candid!  The perspective that in some cultures, authority is so respected that voicing an opinion is tantamount to challenging the boss, is not something that would occur right away to this executive.           

Third, if we do recognize that we need to understand another person’s perspective, our ability to do this may depend on whether we believe that person is similar to us or not.  In either case, this may lead to problems.  Let’s say that you are a manager for a global company working with a group of Japanese employees in the Tokyo subsidiary.  You could make the assumption that because these employees belong to the same company as you they should react similarly to you.  Or you could make the assumption that because these employees are Japanese, they will react based on your “stored knowledge” of what Japanese are like – which may or may not be accurate.  Each of these assumptions will not necessarily reflect the Japanese employees’ perspectives.

I was recently in Singapore to teach a class in Global Leadership to a group of intelligent and experienced Asian executives, most of whom have regional roles working in global companies.  One of their challenges is in managing within a matrix environment and convincing senior management that certain global policies and strategies might have to be adapted for different markets.  In discussing their situation, we had a productive dialogue in looking at the situation from the senior managers’ perspective – what could be going on in their minds, what might be driving their behavior?              

Although empathy and perspective-taking are sometimes difficult, developing this skill can be learned through practice and mindfulness.  I have three simple suggestions.  One, get to know the other person or group better, as well as their cultures.  By doing this, you will minimize your tendency to stereotype.  Second, learn to describe first before judging.  We have a quick tendency to evaluate based on first impressions.  But in cross-cultural situations, what you see is often not what you get, because our observations are filtered through our own cultural frame of reference.  And third, try to reflect on what is going on and what might be causing the behavior.    

So for the expatriate manager and the executive working with Korean nationals, learning about the local cultures might give them insight into why these practices exist.  It does not mean accepting these practices, but it may mean developing alternative approaches.  The executive working with Korean nationals, recognizing that he is an authority figure, might put more effort in asking specific questions rather than asking them generally for their opinion.  Ultimately the benefit of developing empathy and of having a global mindset will help you become a more effective global leader.      


Epley, N. and Caruso, E.  (2008).  Perspective taking:  misstepping into others’ shoes.  In K. D. Markman et al. (Eds.), Handbook of Imagination and Mental Simulation.  New York:  Psychology Press.

Global Mindset Part I

Larry Parker (not his real name) was a marketing executive for the Asia Pacific division of a multi-national company.  He would hold regular teleconferences with his marketing directors in Asia and, according to him, he found it difficult to make much progress with them.  Asking for my advice, he commented, “Why is it that when I tell them that they need to meet a certain deliverable by a certain time, they all say they will do it, and yet nothing happens by the deadline.  I can never tell if they have agreed to do something or not.  Why can’t they just be straight with me?”

Does Larry have what many management experts are calling “global mindset?”  What is global mindset, anyway?  How do we know when someone has it?  Professors Anil Gupta and Vijay Govindarajan define global mindset as “combin(ing) openness to and awareness of diversity across cultures and markets with a propensity and ability to synthesize across this diversity.”  And The Thunderbird School of Management says that global mindset is made up of your:  intellectual capital (e.g., your global business savvy, your cosmopolitan outlook), psychological capital (e.g., your passion for diversity, your quest for adventure), and social capital (e.g., your intercultural empathy, your diplomacy).    

These are certainly reasonable.  As implied, global mindset is a mental attitude, an inclination.  It is not a behavior, but it should predict behavior.  In my own experience and interviews with executives and students, I would say there are four components which can be easily remembered with the acronym FACE:  Flexibility, Acceptance/openness, Curiosity, and cross-cultural Empathy.

I asked my students near the beginning of my course in Cross-Cultural Management to describe what global mindset means to them.  Here is a sampling of what they wrote:   

“Global mindset means that you are aware of your environment, of others and the impact of ideas and events in your business, strategy or position.”

“Taking a more macro look at things … understanding that things won’t work the same all over the world, and taking that into account.”

“Having an understanding that countries have different cultures, and going into each country, one must always be aware and sensitive to that country’s cultural ways.”

 “Someone who understands or has an open mind to understand different cultures and how these affect the outcomes of decisions.”

“Putting yourself in the other culture’s shoes.”

“Listening and resisting reflexive judgments.”

“Your way is not always the right way.”

“Understanding that different countries/cultures have different ways of doing things.  They value certain things differently.  A global mindset has to take all of that into consideration and be open-minded and willing to compromise.” 

When I asked Larry (a mid-westerner who had only begun to travel to Asia) what he thought was going on, he said that it was either because Asians don’t have the same sense of urgency as Westerners, and/or that they are not as candid.  Six months later, Larry requested a transfer from his position and eventually moved to a staff job in headquarters.

We can reasonably assume that Larry did not have a global mindset and was perhaps a poor fit in his global role.  He showed little curiosity for the geography he was managing, was not willing to explore other ways to accomplish his objectives, and could not imagine viewing things from his direct reports’ perspective.      

Developing a global mindset, on the other hand, is not easy.  Traveling to other countries, or reading about different cultures, may help, but is not sufficient.  And of the four components, developing inter-cultural empathy is probably the most difficult.  In a subsequent article, I will explain why, and also some of the ways you can develop global mindset.

The Leadership of Steve Jobs

By now, many of us have read, watched, and listened to many accounts of Steve Jobs’ many contributions can achievements.  There is a passion from consumers about Apple and Steve Jobs that is rare in the corporate world.  Not long ago, I walked past an Apple store in Soho and saw hundreds of Post-It notes and flowers from so many thanking Steve Jobs.  As his biographer Walter Isaacson and others have pointed out, however, Steve Jobs was far from perfect.  I’d like to comment in particular on his leadership and management style.  It is well-known that Steve Jobs could be arrogant, dictatorial, and mean-spirited.  Yet he was a great leader.  So does this invalidate the claims of some management writers and thought leaders today that effective business leaders today need to be nice, kind, humble (Level 5 leadership), and practice “servant leadership?”  Does this mean that executive leaders should now not worry about being ruthless, imperial and aloof?

Not at all.  I think this apparent contradiction can be explained by two sets of factors.  One, we have to recognize that leadership style is situational.  A style that might work under some circumstances might not work in others.  Of course this concept has been around for years, but I am still surprised at the claims being made about “universal” leadership characteristics and behavior.  Those of you who have worked overseas and led cross-functional global teams will surely recognize that your leadership needs to be adapted to specific cultures.  I believe that Mr. Jobs’ leadership style (not to mention his genius in design) was a key ingredient in Apple’s success; had he used a different style, he might not have achieved the same spectacular results at Apple.

Two, despite the observations of some about Mr. Jobs’ arrogant style, I believe that he had at least three qualities that great executive leaders have:  a clear vision, a passion for the company and its people, and an ability to inspire trust.  This is what I would consider his leadership character.  In fact, Mr. Jobs not only had a vision, he made sure that everyone in the company bought into that vision, and this created a “higher purpose” for the company that really excited Apple employees.  Of course, his passion for the company and its products is legendary.  And employees trusted Mr. Jobs – not because he founded the company but because he showed time and again his competence in many areas, especially product design and marketing.  And because employees saw – through his behavior – that Mr. Jobs was not driven by his own ego or by some self-interested needs (like the outrageous pay packages of some executives), they trusted him.  So if Mr. Jobs was at times arrogant, even nasty, employees viewed these behaviors in the context of these underlying qualities.

I think the lessons for executives today are clear.  Leadership style is situational – your behavior can and should vary depending on circumstances.  What is important to consider is the character of your leadership.  Do you have a clear vision for your team or your company?  Do your team members believe in that vision, and are they excited enough to become part of the journey towards achieving that vision?  And do they trust you to do what is ultimately best for the company, the stakeholders, the customers, and employees – not what’s best for you?                      


The Limitations of Self-Managed Teams

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.

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?

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.

Lessons for Mr. Buffett

Recently, I spoke to my class in Organizational Behavior about what I would advise Mr. Buffett after the recent incident with Mr. Sokol.  Here is a summary of what I said.

Number one, be aware of your cognitive biases.  You like people who are like you.  So do most people.  But don’t let that color your judgment.  You liked Sokol because he came from the same school as you, has had a good track record and is results-oriented like you.  You have many things in common with him.   But you still need to judge him based on his performance AND his values.  In OB, these are called “similar-to-me” and “knowledge-of-predictor” biases.

Number two, make sure you get multiple sources of feedback – preferably using a 360-degree  process.  Recognize that you are the boss, and information to you gets very filtered – sometimes you are the last to know.  Jack Welch recognized this early on, and this is part of the reason why he created the Crotonville Management facility.  By the way, if you reach out and ask people two levels down what they think of their boss, chances are you won’t get the straight scoop – at least not right away.  In many cases, after a manager is fired, people will come to you and say, thank you for firing him – he was a terrible leader; what took you so long?  So having a process like 360-degree feedback will help you get a more rounded view of your team.  

Number three, adapt your management style, and specifically your decision-making style, to the situation.  You like to delegate.  It seems that you spend most of your time in your office, reading or talking on the phone.  You don’t generally manage by walking around.  You trust your direct reports.  You like to empower them.  This suits your personality and is understandable, given the diversified nature of your company.   That’s fine in most circumstances.  But when there are yellow or red flags, or when the situation changes, you have to be willing to change your style also.  Sometimes as a boss, you have to dig deep and you need to ask a lot of questions.  This is not the opposite of delegation nor does it suggest that you no longer trust your people.    This is to make sure you have a good understanding so you can best support your team.  In my experience, your team will not view this as “interfering” but as a positive perception that you are interested in their business and willing to help.  It will also help you connect the dots among your various businesses and see where you as CEO can truly add value.

Boss, Manager or Leader?

After reading two excellent books with the word “boss” in their titles (Robert Sutton’s “Good Boss, Bad Boss,” and Linda Hill and Kent Lineback’s “Being the Boss”), I became intrigued with the connotations of this term, and how a boss differs from being a manager and a leader.

Sutton, Hill and Lineback don’t really make a big deal over these distinctions; it seems that for them, the three terms are synonymous.  I’ve observed that we use the term “boss” more frequently and more informally, both in the context of work and outside of work.  My children used to wonder who the real boss in the family was, and some of my male friends would sometimes defer making decisions by letting others know about their spouse that “she’s the boss.”  In my experience working internationally, I find that the word “boss” is commonly understood and used in many countries.  Sometimes the English word “boss” is used; at other times, its local equivalent is.  In Japan, for example, where titles are important, there are many different gradations for the title of boss, for example, honbuchofuku-honbuchobuchojichokacho.

Most good managers at work will avoid throwing the weight of their authority around by telling their employees to do something because “I’m the boss.”  Many employees, however, will comply with their manager’s request (although they may not say it out loud) because “he (or she) is the boss.”  In fact, the dictionary definition of a boss is “a person who exercises control or authority.”

You won’t find many organizations where the word “boss” is in a job title, but you will find that a great majority of organizations uses the terms “manager” or “leader” in their job titles.  While managers tend to avoid referring to themselves as the boss, they are not reluctant to describe their job as managing or leading a group, department, or business unit.

In my opinion, Kotter’s article on managers versus leaders did no favors for managers.  When he wrote that article in 1990, he claimed that “most U.S. organizations today are over-managed and underled.”  Although he stated in the article that both managers and leaders are needed, the implication is that in a world of constant change and complexity, it is more important to be a leader than a manager.  As Sutton has pointed out, however, the distinction may be accurate but dangerous.  Why so?  Let me illustrate (with details disguised).

A number of years ago, I was coaching a marketing executive who I shall call Julia.  She had just been promoted and assigned to another country from her native Australia, where she had been the marketing head for one of her company’s product lines.  In this country, she was going to be the Chief Marketing Officer for the subsidiary.  The subsidiary had also just hired a new CEO, Ron, a native of the country who had been educated in the U.S. and Europe, and had actually come from a competitor’s European operations.  Ron was charismatic and energetic, and he went about exciting the subsidiary with his grand vision and plans for turning the subsidiary around.

Julia was excited too.  She had met Ron soon after she arrived in the country’s capital, and was impressed by his passion and zeal.  However, she did have a bit of a concern about him.  Looking at his background and experience, Ron had an MBA and had been in sales for most of his career.  Now he was being asked to run a subsidiary that had a strong R&D function but had grown somewhat “bloated” over the years.  The subsidiary had not been turning out enough innovative products, and was not profitable enough, according to internal company and external industry benchmarks.  Did Ron know enough about the technical aspects of the business and about its operations to manage the entire subsidiary enterprise?

A few months after Julia arrived, the subsidiary began its profit planning for the following year, and she spent a couple of weeks with her team, individually and as a group, to get a detailed understanding of expenses and sources of revenue.  She grilled them on each line item, and made sure she understood exactly where the money was going, and how it was being spent.  She remembered her old boss in Australia, who would spend several hours with her on her budget every year that she learned to come in thoroughly prepared to respond to questions he might have about any line item on the budget.

With this experience, Julia prepared for her first meeting with Ron.  She had sent him her profit planning figures a few days before, and came into the meeting with back-up notes and documents, ready to answer any question he might throw at her.  To her surprise, he did not have any questions.  He had not even bothered to open the e-mail she had sent him with the profit plan figures she had attached.  He glanced at the numbers she showed him, nodded, and then told her he would get back to her if his CFO had any comments or requests for more information.  He then started to talk to her about his vision for what Marketing could do to help launch some new products the following year.

Julia was a bit stunned, but she went along and brainstormed some new ideas with her boss.  Ron lasted a couple of years with the subsidiary before he left.  While he injected a breath of fresh air into the subsidiary, his lack of attention to detail and the operational aspects of running a business did not help.  He was a leader, not a manager.

In my experience, the best bosses today both lead and manage.  They are able to wear both hats, and know when to “zoom in” and when to “zoom out.”  Take Alan Mulally, who was until recently CEO of Ford Motor Company, and who engineered a very successful turnaround of the company.  He set a clear direction, aligned his team and Ford employees towards a common purpose, and inspired people.  Yet, from all reports, he also was very conscious of Ford’s challenges in returning to profitability, and spent considerable amounts of time managing the bottom line and diving deep into the operational aspects of the business.

My advice for today’s bosses?  First, you can’t be a good leader without also being a good manager.  Get to know your functional area, and what your team is doing.  Ask questions and get into the details.  Second, as a leader, one of your first orders of business is to create a compelling purpose and direction for your team.  Don’t do it in a vacuum, or on a mountaintop where you come down to make your pronouncements.  Involve your team, find out what might excite them, and connect your team’s purpose with the larger goal of the company.  Third, as a boss, use your authority to set a direction; recognize and reward those who perform and who show the right values; and take action on those who don’t.

Then there is the leader as coach.  That’s a subject for another post!


Hill, L. and Lineback, K.  (2011).  Being the Boss. Boston:  Harvard Business Review Press.

Kotter, J.  (1990).  What Leaders Really Do.  Harvard Business Review.

Sutton, R.  (2010).  Good Boss, Bad Boss.  New York:  Business Plus.

New Year’s Resolutions, Managers and Nudge Strategy

About this time of the year, popular magazines are filled with articles about helping you make sure that your New Year’s resolutions stick this time. In their best-selling book Nudge, Professors Richard Thaler and Cass Sunstein show that an effective way to change people’s behavior is to “nudge” them, rather than say, demand big changes in attitudes or behavior. These nudges can be passive (e.g., placing your work-out clothes and bag right by your desk or bedside) or can be self-imposed (e.g., using a smaller-sized plate when having a meal). The effectiveness of this approach has been demonstrated time and again, even in the UK, where in 2010 then prime minister David Cameron created a Nudge Unit. One of its successes was dramatically increasing on-time tax payments by simply reminding taxpayers that many British citizens pay their taxes on time (an example of a nudge using social norms).

This is their description of what nudging is: “A nudge … is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Putting the fruit at eye level counts as a nudge. Banning junk food does not.” (p. 6)

As summarized by Ann Cuddy in her book Presence, the underlying reasons for why nudging is effective are the following. First, nudges are small and require minimal psychological and physical commitment. Rather than promising yourself that you will never again be late for meetings, for example, you might make a resolution to not be late for the next meeting you will have with your boss. Second, nudges operate via psychological shortcuts; for example, using normative influence (showing what other people would do in a situation) rather than informational influence (giving all kinds of reasons why you should do something). Third, our attitudes follow from our behavior rather than vice-versa.

One of the underlying concepts behind nudging is that of choice architecture. Choice architecture refers to the design of an environment that can influence the choices that people make without necessarily intruding on their freedom of choice. A nudge is “any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives.” A nudge is not an order, nor is it forced compliance; it’s an attempt to make a better option more visible.

In their book, Thaler and Sunstein describe different types of nudges; many of these apply more to public policies than to organizational policies. However, as Thaler and Sunstein mention, employers are themselves choice architects. In implementing their policies, organizations do in fact nudge employees one way or the other towards certain choices, for example, when they opt in or out of certain benefits. I have not read many examples of how managers can apply nudging strategy in the work place, although managers do in fact nudge, whether they are conscious of this or not. My focus here is not so much on organizational or employer policies, but on what individual managers can do to become more effective choice architects.

In doing this, I am using as a framework Peter Drucker’s (1973) five critical responsibilities of a manager: setting objectives, organizing (e.g., analyzing activities, structuring, selecting people for jobs), motivating and communicating, measuring, and developing people. Sunstein (2014) has outlined ten types of nudges, which I list along with an example or two of what managers can do to utilize each of these types of nudges.

  1. Default rules. The common example here is when employees are automatically enrolled in retirement plans so that they don’t even have to choose actively. Per Sunstein, “…in many contexts, default rules are indispensable, because it is too burdensome and time-consuming to require people to choose.” In the work place, managers who want to set up regular interactions with their team can create schedules on their employees’ calendars so that weekly team meetings are blocked off.
  1. Simplification. Consultant William Scheimann has reported that in his surveys, only 14% of employees have a good understanding of their company’s strategy and direction. A recent HBR article found percentages in the same ballpark as Scheimann’s. There can of course be many reasons for this, but certainly one of them is that many strategies are complicated, and have not been simplified enough for employees. Managers can make these strategies understandable to employees by simplifying and explaining how the elements of the strategy align with team members’ own objectives. Jack Welch, former CEO of General Electric, once said: “The more simply your idea is defined, the better it is. You communicate, you communicate, and then you communicate some more. Consistency, simplicity, and repetition is what it’s all about.”
  1. Uses of social norms. Informing people what most others do in similar situations has a big influence on behavior, and is an effective nudge. For example, when working for a large financial services company, my team and I collaborated with several consultants and academics to identify the most effective management practices in this firm. We interviewed 60 of the managers nominated by their superiors as among the best managers in the company, and 60 so-called average managers (no one admitted to having poor managers in their divisions). We selected those practices that best differentiated the outstanding from the average managers, and used this to build a leadership development program for the company, including 360-degree surveys so managers could compare their results with the best. Google recently did something similar when they came up with eight critical management behaviors of their best managers (Garvin, 2013).
  1. Increases in ease and convenience. A nudge that makes it easy for people to choose is effective, other things being equal. To facilitate communication, for example, managers can make sure that employees’ work spaces are contiguous (much as what Apple is doing for its new corporate offices and Google did when it created “bullpens” or open spaces so that workers could interact more frequently with top managers). For virtual teams, managers can make sure that employees have access to social technology tools that make it easy for team members to communicate with one another.
  1. Disclosure. Explaining the hidden costs of some behaviors (e.g., the full cost of certain credit cards) is also an effective nudge. Leaders can encourage managers who are hiring potential employees on the importance of considering diverse candidates, and point to the benefits of diversity as well as the disadvantages (and legal implications) of not hiring diverse candidates.
  1. Warnings, graphic or otherwise. For Sunstein, such nudges should be used especially when serious risks are involved. Managers can explain clearly to employees the dangers of accepting bribes especially when doing business in countries that score relatively high in the Corruption Index, and provide examples of businesspersons from other companies who have been fired or worse, jailed, for these crimes.
  1. Precommitment strategies. This is a type of nudge where, if people precommit to engaging in certain actions, they are more likely to follow through. During project review meetings, for example managers can review action items with their team and ask specific team members to explain what they will do next about action items on project tasks for which they are responsible.
  1. Reminders. These are simple nudges to remind people to perform certain actions (e.g., the e-mail alerts we receive letting us know when to pay our credit card bills). Managers can set alerts to schedule specific times when they can check in with specific employees, either personally or through e-mail, and inquire about certain follow-up items.
  1. Eliciting implementation intentions. In this nudge, one asks a question about a future conduct to draw out their intention (e.g., do you plan to fill out the employee survey?). With an employee who might be hesitant to collaborate with others, a manager may have a conversation with an employee about finding out when and how the employee might reach out to colleagues.
  1. Informing people of the nature and consequences of their past choices. Through data about past behavior (e.g., people’s expenditures on car insurance), people can be nudged into either continuing or increasing their commitment to that behavior. Harkin et al. (2016) conducted a meta-analysis of the relationship between monitoring goal progress and goal attainment and found overwhelming evidence for such a relationship. In fact, they report that “… progress monitoring had larger effects on goal attainment when the outcomes were reported or made public, and when the information was physically recorded.” For example, during their weekly updates with direct reports, managers could review progress on specific task milestones and reinforce how making progress in these milestones can make a difference not only to the team but also to the department.

While use of these different types of nudges might not seem very unusual or unique, I have found this checklist helpful in reminding managers of the different types of nudges at their disposal, especially keeping in mind Drucker’s five managerial responsibilities. As managers think through the appropriate nudges for their direct reports, they also need to consider the following four questions. Ly et al. (2013) refer to these as bottlenecks; I’ve adapted their four questions, which I think are an excellent starting point for managers as they consider the most effective nudges to use for their direct reports:

  1. Are your direct reports aware of what they need to do but are unable to perform the behavior, or does the desired behavior need to be activated? This is the classic “skill” versus “will” question that managers inevitably need to answer.
  2. Are they motivated enough to impose a nudge on themselves?
  3. How much cognitive or information overload is there? A nudge that relies on providing more information may not work at certain times (e.g., when there are a significant number of change initiatives or there are pressing deadlines that need to be met urgently).
  4. Are there competing actions or is inertia involved? If the former, then the manager might want to focus on discouraging those other actions first.

A key takeaway here is that changing behavior and influencing others do not have to require herculean efforts; through effectively using nudges, managers (as well as individuals) can make strides in achieving their goals – and our New Year’s resolutions. Furthermore, since managers are already choice architects, they should be aware of and make use of these different types of nudges to motivate their employees and build a high-performance team.


Cuddy, A. (2015). Presence: Bringing Your Boldest Self to Your Biggest Challenges. New York: Little, Brown and Company.

Drucker, P. (1974). Management: Tasks, Responsibilities, Practices. New York: Harper & Row.

Garvin, D. (2013). How Google Sold Its Engineers on Management. Harvard Business Review, December.

Harkin, B. et al. (2016). Does Monitoring Goal Progress Promote Goal Attainment? A Meta-Analysis of the Experimental Evidence. Psychological Bulletin, 142 (2), 198-229.

Ly, K. et al. (2013). A Practitioner’s Guide to Nudging. Rotman School of Management: Research Report Series.

Sunstein, C. (2014). Nudging: A Very Short Guide. Journal of Consumer Policy, 37.

Thaler, R. and Sunstein, C. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press.

Networked and Mentored

Frank, a management consultant with his own independent business, believes in networking especially as a source of potential clients. His week is filled typically with lunches with potential and former clients, and evening meetings with various business groups, e.g., other consultants, business networking events, etc. Fortunately, Frank is very extroverted, and he loves schmoozing and meeting new people. He claims to have gotten some new business out of these networking opportunities but I suspect that he would go to these events regardless.

Harriet works for a global consumer products company that had piloted a mentoring program for high-potentials in the company. She was selected to participate in the program, and after an internal matching process, was assigned a formal mentor – a senior executive from another division who had also been selected to participate but as a mentor. They met twice a month; while Harriet found their meetings helpful, she did not feel that she was getting much out of these meetings. After a year, the pilot program was disbanded, and Harriet’s meetings with the senior executive stopped.

Two important pieces of advice for workers of all ages today are to network and to have a mentor. In fact, Reid Hoffman, the founder of LinkedIn and the author of The Start-Up of You, advocates that hiring managers should give preference to candidates who are connected because workers today need to have “network intelligence.” Both networking and having a mentor are important to one’s career, but is one more important than the other, and what do these mean in practice? And what is the evidence other than anecdotal that these actually work?

For networking, I will use the definition proposed by Gibson et al. (2014): networking is a form of goal-oriented behavior, both inside and outside of an organization, focused on creating, cultivating, and utilizing interpersonal relationships. (p. 150) Reid Hoffman emphasizes the relationship-building aspects more so than the traditional notion of networking – a way to meet as many people as possible in order to see what they can do for you. There is evidence from research that networking works, but the types of networks and the measures of success vary significantly across studies. I like the concept of reputation capital, which Meister and Willyerd (2010) define as the “sum total of your personal brand, your expertise, and the breadth, depth, and quality of your social networks.” (p. 214) They predict that companies will increasingly hire and promote individuals based in large part on their reputation capital.

The traditional concept of mentoring is that of a one-on-one relationship between a more experienced organizational member and a less experienced employee. There is plenty of research suggesting that mentoring has many benefits to the mentee (or protégé) and to the organization. In one study of U.S. Army officers, Payne and Huffman (2005) showed that officers who had mentors tended to show greater “affective commitment” (i.e., emotional attachment) and “continuance commitment” (i.e., awareness of costs and benefits) to the organization than those who did not. Mentoring was also negatively associated with turnover. DeLong et al. (2008) have argued persuasively that professional service firms in particular would benefit greatly from mentoring their young professionals. Unfortunately, competitive pressures have eroded the traditional practice of mentoring in many firms. As a result, retaining talent has become a key issue. Allen et al. (2004) did a meta-analysis of hundreds of studies of mentoring and found that mentoring does have a positive impact, but the impact (especially on objective career outcomes such as compensation and promotions) was relatively small compared to the impact on such psychosocial outcomes like job and career satisfaction. It is important to note that other studies have shown that mentoring is especially beneficial among traditionally discriminated-against groups, such as African-Americans, Hispanics and Asian American men and women.

For women in particular, however, the Catalyst organization found that their mentors tended to play more of an advisory role, while male employees’ mentors tended to play more of a sponsorship and advocacy role for them.

Perhaps because of this, research and practice on mentoring over the past several years has advocated both formal and informal mentoring, as well as the use of peer and multiple mentors.

As Kram (1985) has indicated, mentoring has a dual purpose: helping you to enhance your career (e.g., through exposure and visibility, coaching) and helping you to enhance your professional effectiveness (e.g., counseling, friendship). She refers to these as the career function and the psychosocial function. An ideal mentoring relationship is when both are present. According to the research, the most important variable to predicting the outcomes of mentoring is the quality of the mentoring relationship.

Both networking and mentoring have benefited from recent technologies, with the increase in the use of social media such as LinkedIn, Twitter, and Facebook. For example, a recent trend in mentoring is that of online or e-mentoring. I have not had direct experience with these sites, but the idea is that you can find an anonymous mentor on-line and get advice from anywhere from six months to a year. Some NGOs like the Global Action Networks have mechanisms to match mentors and mentees globally and provide support for e-mentoring.

Networking and mentoring can be especially challenging for the global manager. Carraher et al. (2008) have pointed to the challenges expatriate managers have in being mentored. In my experience, I have seen expatriate managers struggle to find mentors in their host country for various reasons. First, host country executives may not be fully aware of the concept of mentoring and what its benefits are, both to them and to the expatriate manager. Second, they may also be somewhat resentful of the expatriate manager, or at least view them with suspicion, and therefore establishing a trusting relationship becomes quite difficult. Third, expatriate managers themselves may not believe that they would benefit from mentoring, especially if they adopt an ethnocentric attitude.

Another challenge especially for the practice of mentoring is that European and Asian mentoring practices may vary significantly from American practices. In many Asian cultures, for example, mentoring takes place, but it is heavily influenced by Confucian values, so that the mentorship tends to be more familial but also hierarchical. Goto (1999) has found that “Culturally, Asian mentor-like relationships differ from their Western counterparts in that they are much more formally hierarchical, and they blur the distinction between family and social ties.” (p. 53) In Sweden, mentoring follows a process called “Handleduing” which is very similar to coaching and self-directed learning.

A third challenge is the perception of cost-benefit ratios. Research has shown that those who have not engaged in mentoring tend to overestimate the costs of this relationship and underestimate the benefits (which can be intangible as well as tangible).

For those interested in building your network and having mentors, here are a few recommendations. (Also note that Reid Hoffman has some excellent advice in his chapter on “It Takes a Network” from the book I referenced earlier). First, take small steps and start with your current connections. For those of you who are not as extroverted as Frank and who are loathe to simply schmooze and introduce yourselves to total strangers, consider your current contacts from LinkedIn. Re-establish relationships you may have with some colleagues you have not been in touch with. Perhaps there are some colleagues who have risen to senior positions who might be good potential mentors. Send them a short message updating them on your professional doings; you will most likely hear back from only some of them, but you can then follow up with those who replied. Second, take the initiative and get out of your comfort zone. If you are a global manager, take the time to learn more about the cultures of the colleagues with whom you are interacting. This is especially the case when a mentor you might consider approaching is of a different race, gender, or nationality than you. In fact, the potential mentor himself (or herself) may feel a bit of discomfort. You may need to be sensitive to this and seek to find common ground especially with these individuals. Sarah was a Human Resources manager who worked for an organization and had met Vic, VP of Finance, casually at the cafeteria. She knew that the best time to catch Vic was after six when he was in his office wrapping up his work for the day. She decided to stop in one day to ask him a work-related question. She ended up spending a half hour with him, and this started an informal mentoring relationship which lasted for over two years, until she left the company for a promotion elsewhere. Remember that quality is better than quantity. Make sure that those who end up as your mentors are willing to take a more active role in your development.

Third, be clear on your own goals and time frame, and break these down into bite-size pieces. What do you hope to achieve with your networking? And what are some specific goals you can set for yourself at your next networking opportunity? For example, tell yourself that in your next networking meet-up, you will meet at least two individuals you have not known before and get their business cards or contact information. Fourth, remember that networking and having a mentor is a two-way street. Determine what you have to offer and consider how you can help.

Fifth, be open to chance encounters and mentoring moments and take advantage of them. A few years ago, I scheduled a meeting with a client in Boston. We were going to meet in the office of a Harvard Business School professor with whom I was partnering. The client called to say he was running late. The professor, who I will call Sam, and I had an extra hour on our hands. Since he had a fairly light day, I decided to take advantage of the time and spent the next hour engaging in an amazing conversation with Sam. Sam was in turn gracious and tremendously helpful, and to this day, consider this one-hour session one of the most memorable highlights of my professional career.


Allen, T. et al. (2004). Career Benefits Associated with Mentoring for Protégés: A Meta-Analysis. Journal of Applied Psychology, 89 (1): 127-236.

Carraher, S. et al. (2008). Mentoring Across Global Boundaries: An Empirical Examination of Home- and Host-Country Mentors on Expatriate Career Outcomes. Journal of International Business Studies, 39: 131-1326.

DeLong, T., Gabarro, J. and Lees, R. (2008). Why Mentoring Matters in a Hypercompetitive World. Harvard Business Review, January.

Gibson, C. et al. (2014). Understanding the Role of Networking in Organizations. Career Development International, 19 (2), 146-161.

Goto, S. (1999). Asian Americans and Developmental Relationships. In A. Murrell et al. (Eds.), Mentoring Dilemmas: Developmental Relationships Within Multicultural Organizations. Mahwah, NJ: Lawrence Erlbaum.

Kram, K. (1985). Mentoring at Work: Developmental Relationships in Organizational Life. Glenview: IL: Scott, Foresman and Company.

Hoffman, R. and Casnocha, B. (2012). The Start-Up of You. New York: Crown Business.

Marcus, B. (2014). Advice from Top Women Leaders about Finding a Mentor.

Meister, J. and Willyerd, K. (2010). The 2020 Workplace: How Innovative Companies Attract, Develop, and Keep Tomorrow’s Employees Today. New York: Harper Collins.

Payne, S. and Huffman, A. (2005). A Longitudinal Examination of the Influence of Mentoring on Organizational Commitment and Turnover. Academy of Management Journal, 48 (1): 158-169.

Leadership in the Age of Robots

In March of this year, a Google computer defeated the world’s reigning Go champion in four out of five matches. Several years earlier, IBM’s Watson computer defeated two of Jeopardy’s greatest champions and IBM’s Deep Blue competed successfully against the former world chess champion Garry Kasparov. We are all aware of the use of robots in hospitals and combat zones, and the evolution of autonomous cars.

The military uses robots for several purposes, including the recovery of improvised explosive devices. Hospitals are also increasingly using robots for guiding patients and delivering drugs. More recently, as reported in the Economist (2016), researchers have created “robodocs,” robot surgeons that successfully stitched up the intestines of piglets with a minimum of human supervision. Under a surgeon’s supervision, the Smart Tissues Autonomous Robot (STAR) was able to sew piglets’ guts together after the doctors had severed the piglets’ intestines. In fact, STAR was able to carry out about 60% of the procedure without human intervention and its stitches were more evenly spaced and the sutured guts less leaky than what surgeons would have done.

There is no question that we are in the midst of a new age of artificial intelligence and the use of robots. Bryjolffon and McAffee (2014) have argued that we are now in the second machine age. The first started with James Watt’s steam engine, which kicked off the Industrial Revolution. The second machine age started with using computers and digital tools. Carlopio (1988) has described the different phases of new technology a bit differently. In the 15th and 16th centuries, printing, silk-throwing machinery, the screw press and the windmill were seen as labor-saving devices. Traditional trades were not affected but were actually enhanced. Then the Industrial Revolution began, and technology became labor-enslaving, with work processes becoming more standardized and specialized. In the early twentieth century, scientific management became popular, with the assembly line perhaps its most widely used application. Now we are in the third phase of technology which he has described as labor-replacing, where computers and robots will replace jobs.

There has been a lot written on the impact of automation on workers and jobs, and I am not an expert in this area. However, the consensus seems to be that computers are getting smarter. Many experts predict that computers will displace jobs, not just at low-end but also at the high-end. Acemoglu and Autor (2010) suggest that work can be divided into a two-by-two matrix: cognitive versus manual and routine versus nonroutine. Demand for routine tasks to be performed by humans has been falling due to automation, whether these are routine cognitive tasks (e.g., bank tellers, mail clerks) or routine manual tasks (e.g., machine operators, cement masons, dressmakers). But nonroutine cognitive and manual work has been growing, and there has been much debate on the extent to which computers and robots will be able to perform these tasks and replace humans.

In the meantime, we are seeing more and more robots working alongside humans. In 2012, Amazon bought Kiva Systems, a company that makes robots. These robots are used in Amazon’s warehouses to “pick” and bring goods from storage shelves to employees. What do we know about interactions between humans and robots in the work setting? Actually, research on human-robot interaction has been going on for a while. For example, Hinds et al. (2004) wrote about the rise of “professional service” robots (as distinguished from industrial robots) that share the workplace and help people perform their tasks, e.g., supplying troops with ammunition in the battlefield, delivering medications from pharmacies to nursing stations in hospitals. In the future, these robots will be more highly interactive with people. The researchers did a study to determine the effects of the robot’s appearance and its relative status on how people work with robots. They created a lab experiment that required subjects to interact with robots to accomplish some tasks. They manipulated three levels of appearance (human as the baseline; human-like where the robot had a face, torso, arms and legs, and wore an outfit; and machine-like, where the robot covering was metallic and angular). Then they manipulated status by telling the subjects that their robot partner was their supervisor, their peer, or their subordinate. They found that subjects interacting with a more machine-like robot had an increased sense of personal responsibility they felt for the task, with subjects feeling most responsible when interacting with a machine-like subordinate. Furthermore, subjects felt less responsible when collaborating with a robot supervisor as compared with a robot peer or subordinate.

As another example, Kim et al. (2014) reviewed the literature on social distance and developed some hypotheses on people’s reactions to interactions with robots. They had participants play a card-matching game on a computer with Wakamaru, a robot developed by Mitsubishi Heavy Industries, Ltd. As the researchers described it, here is what the robot did:

“In its interactions with the participants, the robot used three key behaviors: gaze, speech, and navigation. The robot’s three-degrees-of-freedom head allowed it to direct its gaze toward the participant and other targets in the environment. The robot communicated with the participants using synthesized natural language and moved toward and away from them at different points during the interaction.” (p. 786)

The participants played the game with the robot, which made suggestions on moves. The researchers manipulated power distance (with the robot either as supervisor or subordinate) and proxemic distance (how close or distant the robot was to the participant).  What they found was that participants who interacted with the supervisor robot at close distances performed better and reported a more positive experience and stronger rapport with the robot than those who interacted with the supervisor robot at far distances. They also found that participants who interacted with a close subordinate robot reported a more positive user experience and more rapport than those who interacted with a distant robot.

What these and other studies show is that there are interaction dynamics between humans and robots that we need to take into account when designing the work of the future. Now if robots can replace humans in many tasks, can they also replace humans as leaders? Here, there is very little research on this topic. I did find an article by Samani et al. (2012) who used the term robotics leadership to describe the work that robots can perform in stock brokering (robots handling stock trades) and avionics (robots replacing human pilots in airplanes). However, I don’t consider these as leader behaviors as much as expertise that robots can provide. There is also some emerging research on leader-like behaviors of robots among themselves; for example, using simulation techniques among foraging robots, Pugliese et al. (2015) found that the most skilled robots became “leaders” and that robot groups with leaders were more effective than robot groups without leaders.

Part of being a leader is influencing others to perform certain actions. We have seen evidence from research by Milgram (1963) and others on the effect that those perceived to be in positions of authority can have on getting others to comply and follow orders. Parasuraman et al. (1997) wrote about this in the context of automation, when they summarized the research on the dangers of automation. A robot leader who gives instructions (especially if the robot has a deep male voice and perhaps is made up to look dominant) might have others following orders unquestioningly. In fact, in an interesting experiment where participants worked with robots on a set of tasks, Gombolay et al. (2015) found that participants preferred robots who made decisions about how the tasks were to be allocated rather than the participants having to make the decisions themselves. In another experiment, Robinette et al. (2015) found that participants followed the robot in an artificially created emergency situation (where they had to be evacuated from a room) even when they had seen the robot make navigation mistakes earlier and continued to make mistakes in directing them to a wrong exit!

We do know quite a bit about effective leadership, and what people look for in leaders. Kousez and Posner (2007) have been conducting surveys over the past 30 years on qualities most admired in leaders, and they have found that four qualities are what most of the people they surveyed around the world want in a leader: honest, forward-looking, inspiring, and competent.

In his now classic article, Kotter (2001) stated that leaders do things differently than managers: leaders set a direction, align people, and motivate them. More recently, Google found in its own research (Garvin, 2013) that outstanding managers (versus average managers): coach well, empower their teams and do not micro-manage, express interest in employees’ success and well-being, are productive and results-oriented, are good communicators and listen to their teams, help their employees with career development, have a clear vision and strategy for the team, and have key technical skills so they can help advise the team.

Can robots be as good as humans in performing these behaviors? Potentially, yes. I can envision robot leaders being programmed, for example, to express interest in employees’ success, to listen and provide a clear vision for the team. I can also envision robot leaders being programmed to be honest, forward-looking, inspiring, and competent.

However, effective leadership is not only about performing these behaviors. As Kousez and Posner (2007) argue, leadership is about establishing a relationship between leader and follower. While certain robot leadership behaviors might lead to compliance among followers, organizations also want to create high-performing cultures characterized by motivation and commitment. Can robots get us to trust them so that they inspire us to do our best and engage us? Can a robot ever cause our brains to release oxytocin, which is a chemical that helps promote many kinds of social behavior (Stix, 2014)? Based on the evidence, it seems that our brains are wired differently when it comes to reacting to those who inspire, engage, and motivate us versus those who simply get us to comply.

As Colvin (2015) has pointed out, we are asking the wrong question if we are trying to figure out only what computers cannot do that humans can: “Rather than ask what computers can’t do, it’s much more useful to ask what people are compelled to do – those things that a million years of evolution cause us to value and seek from other humans, maybe for a good reason, maybe for no good reason, but it’s the way we are.” (p. 53)

In my view, we will continue to want human leaders in the work setting especially in three (and very “human”) areas where leadership is critical: making decisions on business and people issues (specifically around strategic decisions, and on who to select and promote), communicating those decisions and related issues, and inspiring and motivating. The value-add of human leaders is evident in these areas, where commitment rather than compliance is of critical importance to building a high-performance organization.

For each of the above, note the following continuing patterns:

  • Despite years of evidence that statistical methods of selecting job candidates are superior to human methods, almost every firm that I know still wants to see a candidate face-to-face before hiring him or her, especially for higher-end work and/or professional positions.
  • When managers congratulate someone for a job well done, or let them know that they have been promoted, they much prefer to do this face-to-face rather than sending them an e-mail or handing them a piece of paper with the news. The same is true for communicating bad news, such as when someone has to be let go. Evidence suggests that people also would much prefer to hear this information face-to-face from their leaders.
  • When a team needs to be inspired and motivated, the most effective managers engage in face-to-face meetings and “high-touch” actions to lift spirits up and boost morale.

Brynjolfsson, E. and McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W. W. Norton.

Carlopio, James. (1998). Implementation: Making Workplace Innovation and Technical Change Happen. Synergy Books International.

Colvin, G. (2015). Humans Are Underrated: What High Achievers Know That Brilliant Machines Never Will. New York: Penguin.

The Economist (2016). Who Wields the Knife? May 17, p. 74.

Garvin, D. (2013). How Google Sold Its Engineers on Management. Harvard Business Review.

Gombolay, Matthew C., et al. (2015). Decision-making authority, team efficiency and human worker satisfaction in mixed human–robot teams. Autonomous Robots 39 (3): 293-312.

Hinds, P., Roberts, R. and Jones, H. (2004). Whose Job Is It Anyway? A Study of Human-Robot Interaction in a Collaborative Task. Human-Computer Interaction, 19: 151-181.

Kim, Y. and Mutlu, B. (2014). How Social Distance Shapes Human-Robot Interaction. International Journal of Human-Computer Studies, 72: 783-795.

Kotter, J. (2001). What Leaders Really Do. Harvard Business Review.

Kousez, J. and Posner, B. (2007). The Leadership Challenge (Fourth Edition). New York: Wiley.

Milgram, S. (1963). Behavioral study of obedience. The Journal of Abnormal and Social Psychology67 (4), 371-378.

Parasuruman, R. and Riley, V. (1997). Humans and Automation: Use, Misuse, Disuse, and Abuse. Human Factors, 39 (2): 230-253.

Pugliese, F. et al. (2015). Emergence of Leaders in a Group of Autonomous Robots. PLoS ONE 10(9): e0137234. doi:10.1371/journal.pone.0137234.

Robinette, P et al. (2015). Overtrust of Robots in Emergency Evacuation Scenarios.

Samani, H. et al. (2012). Towards Robotic Leadership. In SS. Par et al., (Eds.), ACHRS Part II. Hedelberg: Springer, pp. 158-165.

Stix, G. (2014). Fact or Fiction: Oxytocin Is the Love Hormone. Scientific American, September 8: