“I just love your new haircut.” “Of course it’s no problem to take you to the airport.” “That’s a great sweater…”
Why do we say things we don’t always mean? Does everyone use some form of deception from time to time? According to Operations and Information Management Professor Maurice Schweitzer, not only does everyone use deception, but it’s something we start learning to use as early as age 3.
“We start to test it out as children,” he explains. “A 3-yearold might say something like, ‘No, I didn’t have a cookie.’ testing to see if his use of deception will be rewarded with another cookie. And then you’ll give the child feedback like, ‘I see cookie crumbs on your face,’ and the child will learn to remove physical evidence because that is how he got caught.”
Schweitzer maintains that the “laboratory in which we live gives us great feedback with which to improve our ability to tell lies. We get clear and quick feedback as we learn almost every time whether or not our lie worked.”
On the other hand, he notes that our ability to detect lies is often delayed and imprecise. “When I say that I can’t go out because I have to wash my hair, you just don’t know if that is true. Or if I tell you that I love that sweater, you won’t know. And whenever you visit me, you might put on that sweater. So you get imperfect—or perhaps delayed—feedback. We just don’t learn as well in those types of settings.”
As a result, Schweitzer says that deception represents a tool—that some people are better at using than others—that pervades many of our social exchanges, particularly negotiations. “I became interested in understanding the mechanics of deception and in particular the judgment process for why people are more or less likely to lie. And that led me to think about trust and the conditions under which we are more or less likely to trust others.”
Promises and Lies
How many times have we heard about an executive behaving in an untrustworthy manner or even caught in an outright lie? Can that individual ever regain trust? According to Schweitzer in a new paper, “Promises and Lies: Restoring Violated Trust,” coauthored with his former PhD advisor Professor John Hershey in the Operations and Information Management Department, and Professor Eric Bradlow in the Marketing Department, the answer is yes.
“A lot of people say that talk is cheap, but we’ve found that people really do care about what you say after a rupture of trust,” he notes.
The authors reached their findings through an experimental study in which people played a game where they trusted another individual, then the trust was harmed, and attempts were made to recover that trust. Schweitzer explains that people were given $6 and then given the option of either keeping the money or passing it to an anonymous partner, where it would triple to $18 and then the partner could either split the larger amount of money with them—or keep all of it themselves. This game would be played with the same partner at least seven times in a row.
“In the first round, would you pass the $6 to your partner? You would if you trusted that he would return money to you,” says Schweitzer, noting that in the first round 90% of people passed the $6, as there was no reason not to trust the partner at that point.
However, after the first round, things became more complicated. After people handed over the first $6, their partners (who were really Schweitzer’s research assistants) kept all of the money, giving nothing back. They repeated this untrustworthy behavior in the second round. After this, trust had been severely damaged and most people of course kept the $6 rather than risk losing it again.
So the trust recovery process kicked in on the third round. “After that, I’d have the partner do one of four things: 1) send an apology note; 2) send a promise to return the money; 3) send a note with both an apology and a promise; or 4) send no message. Then, the partner would start behaving in a trustworthy manner. Also, the test subjects would get to see what the partner would have done in each round regardless of how they chose to play. So if the test subject kept the $6, he or she would still find out whether the partner would have split the money or kept it all.”
He continues, “We had a measure of how much the subjects were willing to trust over time. This enabled us to compare the trust recovery rates across different conditions. One key part of this research was our focus on deception. In some conditions, we had our research participants engage in deception as well as behave in an untrustworthy way.”
The findings were surprising: trust that is harmed by untrustworthy behavior can be restored to initial levels. However, trust that is harmed by untrustworthy behavior and deception never fully recovers, as the deception causes enduring harm to the trust.
He maintains that this is an important lesson for business leaders. “If a CEO behaves in a way that is untrustworthy or even just perceived as untrustworthy then it’s important to know that communication can help to restore trust. Maintaining credibility in what you say is the key. For example, a CEO can make a promise to change a course of action and this by itself can be very effective in speeding trust recovery—as long as he hasn’t been caught lying in the past. That is, words can repair trust very effectively, but people have to be able to believe the words.”
While much of his research on deception and trust is in an experimental setting, Schweitzer currently is working in the field studying how insurance companies detect fraud. He has conducted this work with another Wharton PhD graduate, Danielle Warren, who is currently on the faculty at Rutgers. The issue he’s focusing on is the importance of interviews by insurance investigators in detecting fraud.
“Most insurance companies have a lot of people who handle claims, but surprisingly few people who actually investigate potentially fraudulent claims. I’m looking at whether there is a way to focus the attention of investigators,” he explains.
Schweitzer notes that he is finding that in-person interviews are a very important tool in detecting fraud even as investigators increasingly rely upon database searches, physical evidence, and sophisticated investigative techniques like accident reconstruction and surveillance. “Unlike a lab environment, there is a serious self-selection issue in the field where people who decide to tell lies are making a personal choice to try to get more
money by using deception at the risk of getting caught. In this setting, when the stakes are high, people might convey deception by being particularly excited. Or, a claimant might try to avoid an interview altogether. People who avoid interviews are actually communicating a lot of information.”
The message for insurance companies, according to Schweitzer, is to keep investigators on the ground and make sure they have strong interpersonal skills. “They might be more expensive, but they are very valuable to the organization from a detection and deterrent standpoint.”
Feeling and Believing
Schweitzer’s research also shows that emotions can have a great deal of impact on trust judgment. In another recent paper, “Feeling and Believing: The Influence of Emotion on Trust,” coauthored with Wharton PhD student Jennifer Dunn, they found that when a person experiences bad traffic or recently received great personal news, those feelings will be carried with them into a morning meeting at work, influencing the way that person will trust colleagues.
“Most research has looked at trust as a cognitive decision, where people are making cool, rational judgments. However, a lot of our trust judgment—particularly for people we don’t know well—is influenced by emotions. I’m asking myself, ‘Should I trust this person?’ And I’m answering that question by checking to see how I feel. Well, if I feel pretty good then that feeling will influence my trust judgment.”
To test this, Schweitzer and Dunn ran five experiments in which test subjects would do things like watch video clips, engage in writing tasks, or fill out surveys to make them focus on particular emotions like happiness or anger.
How can you make a perfect stranger feel angry or happy? According to Schweitzer, inducing emotion is “shockingly easy.” He points to an experiment they conducted at the train station as an example. “We asked people to list three things that made them really happy or angry and then to write about it in great detail. Sometimes people were actually ripping the page with their pen they would become so mad.” After inducing a particular emotion through these tasks, the subject would rate their trust in a co-worker across various things.
“We found that there is clearly an affective component between emotions and how much you trust people you don’t know that well,” he says, noting that this doesn’t work for people you know well. “For example, I know how much I trust my mother because it’s a recall judgment and not a matter of how I am feeling at that moment.”
He says that in business, you can see how it’s important for teams to bond because when colleagues become friends then judgments are made based on recall as opposed to recent emotion. “Why do business people go out together to sing at karaoke bars? Why do business people go out drinking heavily? It’s doing very important work that is extremely hard to do otherwise. It’s also why sales people tell you funny anecdotes and ask you about your kids to bring you to a happy place before asking you to purchase a product. I think we sort of know that this chit chat (called nontask communication) is important, but this research gives us a sense of why that is the case and what they should be doing while they chit chat.”
Schweitzer adds that bonding through alcohol consumption is perhaps even more important in business than many people might think. In earlier research, he studied the effect of alcohol in negotiations and found that while it leads to inefficient negotiations, the consuming of alcohol can actually cement one’s place in a work group.
“There has been no other work on alcohol in negotiations, but it’s an important part of a lot of business transactions and it’s worth thinking about its functional role,” he says. “Introducing alcohol transforms one kind of meeting into another kind of meeting. It lowers inhibitions and in some cultures, like in Japan, people think they can’t truly know you unless they’ve been drinking with you. It’s a critical form of bonding.”
Another important part of the trust process is the emotion of envy, a “delicious” research topic according to Schweitzer, who is currently studying with Jennifer Dunn and Simone Moran, a professor at Ben Gurion University in Israel, how social interactions can produce envy and how envy impacts subsequent interactions.
“Imagine the following—you are in a work group that works collaboratively together selling books and someone in your group is selected as the employee of the month. She gets her picture on the wall, a bonus, and a special parking spot. And then she goes back to her work group. The question is: how do the coworkers feel that didn’t win? What are they likely to do to that returning coworker, and having won the award, will the coworker trust her colleagues any differently?”
Schweitzer says that he’s finding that “people feel envy when they are outperformed in a self-relevant domain (an area you care about) by a similar other (a peer).” He explains, “It’s the guy in the cubicle next to me who I compare myself to, but when it’s somebody much higher or lower in status then it doesn’t really matter.”
He notes that there are three approaches people might take: you might try to avoid the outperformer, you might try to level up, or you might try to drag him down. “In leveling up, you see a coworker has gotten an award and you work even harder to get the award yourself. In leveling down, you see the coworker win the award and you want them to suffer and to bring them back to your level.”
Under what conditions people will level up or down and how people perceive a path to success is important. “If they see that they can achieve success then they will exert more effort. The employee of the month should motivate other employees. But what is interesting about envy is that it is a socially sanctioned emotion so that when you win, people will always tell you congratulations and that they are happy for you even if they then go home and kick their cat. In reality, most people who feel envy will not say, ‘It’s great that you won that award, but I’m envious and will avoid you for a while until I get over it.’”
Schweitzer points out that most coworkers who experience envy are much more likely to lie to the person who wins the award and engage in actions to undermine that person, such as sharing less information. He notes that he has evidence so far to suggest that the award winner doesn’t expect this. The award winner will feel great and continue trusting congratulatory coworkers just as much if not more than before the award.
So what should companies do to reward good employees without creating such a dynamic? Schweitzer says that in some cases, companies should give awards in private rather than in public as well as think about structuring workgroups that are heterogeneous so that social comparisons won’t be as grating. Also, they should think about reassigning workers if there will be negative competition and make it clear for employees what the path is for success.
“Another option is to think about several smaller awards rather than very few large awards,” he says. “One key problem with envy is that we have a hard time anticipating envy. When we enjoy success, we have to work at trying to imagine how others might feel. My own view is that we’re not very good at that because we are busy enjoying our success.”
Schweitzer’s work also has identified problems with one of the most pervasively used motivation tools: goal setting. Managers need to think very carefully about the goals they set for employees. The intent of goals is to focus and motivate employees to work harder, but what happens when the employees miss the goal, especially if they fall just short of the goal?
According to Schweitzer, they are likely to cheat. “For example,” he says, “a lawyer might be given a goal of billing 2,000 hours. Finding herself just short of the goal, will she report the actual amount she worked? Or will she report having billed 2,000 hours?”
In an article coauthored by Schweitzer, Lisa Ordonez of the University of Arizona, and Bambi Douma of the University of Montana, “Goal Setting as a Motivator of Unethical Behavior,” they found that “people with unmet goals were more likely to engage in unethical behavior than were people attempting to do their best.” And “the relationship between goal setting and unethical behavior was particularly strong when people fell just short of reaching their goals.”
Schweitzer explains that if you set specific challenging goals like to sell 30 cars in a month then people will outperform employees who are given low or vague goals such as “just do your best.”
“In general, people exert more effort and work more persistently to attain difficult goals than they do when they attempt to attain less difficult goals or to ‘do their best,’”
according to the authors. Schweitzer notes that this relationship is so strong that many have “pushed this in management education almost like a religion.”
But what happens when the salesman only sells 29 cars that month? Is he then more likely to back-date the first sale of the following month to make it appear that he met his goal? Schweitzer says that it’s not hard to find examples of how goal setting has resulted in “cooked books” and false sales reports.
Since goal setting motivates unethical behavior when people fall short of their goals, he maintains that managers need to use it cautiously and be ready to monitor performance.
With such a focus on unpleasant subjects like deception, mistrust, and envy, you might wonder what kind of guy Schweitzer is? But his students have rated him so highly that he’s won the David Hauck Award for Excellence in Teaching (Undergraduate Division) in 2002, the Outstanding Teacher Award (Undergraduate Division) in 2002, as well as the Whitney Award for Distinguished Undergraduate Teaching in 2000.
He currently teaches negotiations in both the MBA and Executive Education Divisions and was one of the first faculty members to develop a simulation through the Alfred West Learning Lab, which was established at Wharton in 2001 after a $10 million gift from Alfred P. West, Jr., WG’66, chairman and CEO of SEI Investments.
Schweitzer developed the oil pricing simulation called OPEQ for his negotiations classes to teach issues involving shared resources and incomplete information. Students participating assume the role of an oil-producing country and set production levels in a competitive environment with little or no knowledge of their competitor’s intentions.
“I really believe in the Learning Lab,” he says. “Teaching is an important part of our mission at Wharton.”
The Los Angeles native earned tenure last year and lives in Bryn Mawr with his wife and three young daughters. When he’s not at Wharton, he’s often coaching soccer for his oldest daughter.
Schweitzer—who earned his undergraduate degree in economics at the University of California at Berkeley and both his Masters and PhD at Wharton in the Operations and Information Management Department—got his first job at the University of Miami. He came to Wharton as a visiting professor and joined the faculty in 1998.
Although he sometimes misses the sunny weather of Miami and L.A., he says that Philadelphia is an attractive city because of the strengths of the Wharton students, staff, and faculty. “We have a collection of very smart and interesting people here in West Philadelphia. So smart and interesting, in fact, that smart and interesting people from all over the world come here too—it’s not because of the weather or the view.”