My fair employee | Brunswick Group

My fair employee

It may sound like self-help, but it’s actually sound science: leaders’ beliefs shape how their teams perform

Having vowed to ignore women for the rest of his life, the mythical – and presumably lonely – sculptor Pygmalion decided to carve a female companion out of ivory. He fell so deeply in love with his creation that he talked to it, showered it with gifts – some impressive (gold); others less so (smooth pebbles) – and even laid the sculpture down on his couch at night to sleep. The normally vengeful Roman gods took pity on the delusional artist and granted his wish: with a kiss, the statue came to life.

In a landmark article for the Harvard Business Review, J. Sterling Livingston saw the mythical sculptor as an apt metaphor for business leaders. “More often than one realizes, the manager is Pygmalion,” Livingston wrote. Citing a wealth of research and case studies – and this was in 1969 – Livingston argued that leaders’ beliefs about their teams change how they interact with its members. This, in turn, changes how those team members act, causing them to live up or down to their leader’s beliefs.

A leader’s expectations – not demands or performance targets, but truly held beliefs – can become self-fulfilling prophecies that can produce changes as seemingly impossible as turning ivory into flesh.

Since Livingston’s article, the Pygmalion-esque power of our expectations has been studied in healthcare, the military and education. It’s such a well-established psychological phenomenon that there’s even a name for it: the Pygmalion effect. (The reverse, where low expectations cause lower performance, is called the Golem effect).

Though the research goes back 50 years, it feels especially relevant and necessary today. Gallup’s most recent global survey found that worldwide, only 15 percent of employees feel engaged. The main culprits, at least in the US, according to Gallup, are managers – who, Gallup estimates, account for “70 percent of variance in employee engagement.” The survey didn’t find reason for optimism: only one out of five employees globally feel like they’re “managed in a motivating way.”

Which isn’t to suggest that harnessing the Pygmalion effect will solve the employee engagement crisis overnight. The takeaway is much subtler: what you think about your employees shapes how you act toward them, which affects how they perform, which in turn affects what you believe about them. A virtuous cycle.

The same is obviously true in families, in friendships, and with ourselves. It bears thinking about. Changes in beliefs take time to become embedded in our thinking and even longer to show up in the responses of others. But it’s a step worth taking, given the huge potential rewards.

It was a study in the mid-1960s, led by Robert Rosenthal, a Harvard professor, and Lenore Jacobson, an elementary school principal in San Francisco, that first drew attention to the phenomenon (the Pygmalion effect is sometimes called the Rosenthal effect). The pair administered a test to students at the start of a school year to measure students’ IQs.

When Rosenthal and Jacobson shared the test results with teachers, they told teachers the students had taken the “Harvard Test of Inflected Acquisition,” which identified the students mostly likely to “bloom.”

The Harvard Test was, of course, a sham, and no more accurate than a coin flip at predicting a student’s potential. Rosenthal and Jacobson had chosen the “bloomers” at random – but the teachers didn’t know that.

At the end of the year, Rosenthal and Jacobson found a marked difference in IQ scores between “bloomers” and “non-bloomers” – it seemed the fake Harvard Test had created very real distinctions in students’ performances.

Rosenthal, who sat in on classes throughout the year, concluded that the gulf had been caused by teachers’ beliefs. Since the teachers had expected students to perform differently, they had treated them differently. Rosenthal categorized the four most important differences.

1 Climate Through both verbal and non-verbal communication, teachers tended to create a warmer, friendlier environment for “bloomers.”

2 Input factor Since teachers believed “bloomers” were capable of handling more information, they gave them more information to handle.

3 Response opportunity Teachers gave “bloomers” more time to respond, called on them more often, and let them speak longer, believing these students would know or eventually arrive at the right answer.

4 Feedback “Bloomers” were praised more. And, more importantly, “bloomers” were given differentiated feedback; instead of a simple “wrong” or “good job,” teachers told “bloomers” what specifically they could do better or what they did right – crucial to learning and improving.


More than 50 years of research suggests that our beliefs about others can act as a kind of self-fulfilling prophecy – for better, or worse.

The study has its detractors. IQ is, at best, an incomplete measure of intelligence and academic achievement. And the results were more pronounced in younger children than in older ones, where expectations didn’t always influence outcomes so drastically. But the bigger question is: why did they influence them at all?

Rosenthal explained in a 2015 interview: “The bottom line is that if we expect certain behaviors from people, we treat them differently – and that treatment is likely to affect their behavior.”

Dov Eden, a professor at Tel Aviv University, followed a similar approach in one of his most famous experiments. Using results from a fake test, Eden told commanders in the Israeli army that certain soldiers under their command were especially gifted. After only seven weeks of training, Eden found a 15 percent difference in performance between soldiers who had been assigned the highest potential versus those who had been given the lowest. His takeaway: “Leaders get the performance they expect.”

So do online daters, apparently. In 2014, Christian Rudder, co-founder of the online dating site OkCupid, wrote about a few experiments the website had run on its users. One of which, again, involved deliberate deception: when the site’s algorithm had determined two people weren’t meant for each other, OkCupid occasionally told them they were. And the data showed, in Rudder’s words, “when we tell people they are a good match, they act as if they are. Even when they should be wrong for each other.”

Our beliefs shape relationships and outcomes in the office no less powerfully than they do online. Livingston’s HBR article, “Pygmalion in Management,” focused on the power of leaders’ expectations on a team’s performance. The piece told the story of organizational experiments overseen by Alfred Oberlander, a district manager for the insurance company MetLife, who made three simultaneous changes: (1) he grouped his best sales agents together, (2) put them under the leadership of his best assistant manager, and (3) set the group ambitious sales targets. Oberlander also grouped his average-performing agents into a team, and did the same with his low-performing agents.

The results? The “super staff” performed superbly, and helped improve overall performance by 40 percent. The lowest group’s sales declined, living down to the expectations set for them.

Unexpectedly, the average group delivered better-than-average performance – in fact, their sales figures grew the most sharply. Oberlander’s explanation was that the average team had a better-than-average leader, one who believed they were capable of delivering results on par with the “super staff.”

In other words, their leader expected them to perform at the highest level (and told them as much), and they did.

A manager in a separate division at MetLife tried to replicate Oberlander’s results. He divided agents into three groups and set them more ambitious goals. But from the outset this manager admitted that he believed his employees, even his highest performers, were “either average or incompetent.” While the manager followed the same steps, he didn’t get the same results: sales stagnated.

This is one of the most important wrinkles – and limitations – of the Pygmalion effect: it can’t be faked. Leaders have to genuinely believe that their team is gifted or capable of hitting an ambitious target. Pretending just doesn’t cut it.

Why? Because belief shines through in millions of nonverbal, often subconscious, cues. Even if we’re trying to believe, even if we want to believe, people can tell that we don’t. The consensus, established by two 1967 studies, is that body language, facial expressions, and tone of voice account for 93 percent of the information we communicate – only 7 percent comes from the words we actually speak. “Indeed, managers often communicate most when they believe they are communicating least,” Livingston wrote.

The Pygmalion effect is real, though some leaders might argue it’s not grounded in business reality. They’ve been working with their teams long enough to know who’s gifted and who’s not.

But the message of the research is as compelling as it is inconvenient: performance today reflects not just the talent of employees, but the underlying beliefs of the people who lead them. Continue with the same beliefs and you can expect the same results. Change the beliefs – truly, fundamentally change them – and better performance is possible.

15% of employees worldwide are engaged

Ellen Langer, a Harvard psychologist, writes in Counter Clockwise of an experiment where people asked to complete 100 jumping jacks reported feeling tired after finishing around 70; another group, which had been asked to do 200 jumping jacks, reported feeling tired at the 140 mark. A number of factors are at play in this discrepancy, but one is simply that people did more because they were asked to do more.

And in what’s known as “Sweeney’s Miracle,” Dr James Sweeney of Tulane University trained George Johnson, a janitor at the university whose IQ was assessed as below average, to use sophisticated computing technology. This janitor, under Sweeney’s tutelage, became head of Tulane’s Biomedical Computer Center – and would go on to train others to do the same.

Few will be shocked to read that people are often capable of more than we think they are – and more than they believe they are. Though we’ve heard this before, it doesn’t mean we’re actually practicing it.

Beliefs are quickly formed and slow to change. In Thinking, Fast and Slow, the Nobel Prize-winning Daniel Kahneman writes of a study that presented people with the text below:

Alan: intelligent—industrious—impulsive—critical—stubborn—envious

Ben: envious—stubborn—critical—impulsive— industrious—intelligent

Then they were asked what they thought of Alan and Ben. Overwhelmingly, Alan was liked more than Ben – even though the two had exactly the same traits. People couldn’t even get to the end of a sentence before forming judgments and beliefs about Alan and Ben; nor did they have the patience to step back and reconsider their beliefs when asked about them. Malcolm Gladwell’s best-selling book Blink, which highlights the prevalence and power of “two-second judgments,” suggests we do this far more often than we think.

Another set of beliefs leaders should examine: the ones they hold about themselves. Plenty of us hold flawed beliefs about our abilities – a landmark study found that 93 percent of US drivers rated themselves as “above average. ” Leaders are no exception. Many think they’re better at communicating expectations than they actually are: “Managers are more effective at communicating low expectations to their subordinates than in communicating high expectations to them,” Livingston wrote, “even though most managers believe exactly the opposite.” Given the power that belief has on performance of individuals and teams, this is a serious problem, but also a wonderful opportunity for individual employees and managers.


93% of US drivers rated themselves as "above average"

The Pygmalion effect can sound so simple as to be glib. Belief and positive thinking sound nice, but they can’t change hard skills or a host
of other factors that determine how a team performs. And what of employees themselves? Don’t they bear some responsibility for how
they perform?

These points make two mistakes: first, they mischaracterize the Pygmalion effect, and second, they ignore the overwhelming body of research behind it. Both are easier to do than take responsibility for the beliefs we hold about others and ourselves.

Dov Eden, who led the research on the Israeli army, admits: “It sounds so simple; it seems too good to be true.” And yet, time and time again, a change in leaders’ beliefs produces measurable changes in performance – in Eden’s study, a 15 percent gap between the “gifted” and the “non-gifted”; in Oberlander’s study at MetLife, a 40 percent jump in sales. And there are hundreds of other studies like them.

And if it’s not glib, perhaps it’s just obvious. Who in their right mind would suggest that believing your team is mediocre will lead to better outcomes? But Gallup’s data suggests that many leaders hold precisely these kinds of beliefs about their colleagues.

The myth of Pygmalion ends abruptly – we know he married the statue-turned-woman, Galatea, and that they had a child. But whether he continued to give Galatea smooth pebbles, or if she continued to sleep on the couch, remain a mystery.

The psychological effect that bears his name is much less ambiguous. Decades of research show that when people change what they believe about their colleagues and what they expect from them, measurable, if not miraculous, improvement almost inevitably follows.

And perhaps most challenging of all, the only price tag for this wonderful change: a willingness to have faith in the people we lead.


Giovanna Falbo, a Partner, specializes in labor relations and employee engagement. Edward Stephens is Deputy Editor of the Brunswick Review. Both are based in Brunswick’s New York office.

Download (174 KB)