Deception is an integral part of life. Unseemly as it may sound, everybody lies – often several times in one day. These range from little white lies and sins of omission to outright deception. And none of this is necessarily a bad thing, says Maurice Schweitzer, associate Professor of operations and information management at the Wharton School of the University of Pennsylvania.

“Deception is more nuanced that you might initially suspect,” says Schweitzer, who specialises in behavioural decision research. “Your mum might exhort you never to lie and, in the next breath, answer the phone and tell the telemarketer she’s not home right now. We lie all the time.

“A lot of lies we tell are social and help us get along with people better,” says Schweitzer. “Deception is extremely functional and is a part of the fabric of our lives.” At the same time, trust is an essential element in all social relationships, including those at work. “Trust is the social glue of the economy. It’s the glue for any transaction,” says Schweitzer. “You can’t contract for everything. Ideally, at the base there is some trust in individuals, groups and institutions.”

Any CIO who has ever shepherded a big project which has been delayed, over budget or that simply under delivered knows just how destructive a broken promise can be to trust between IT and business.

"If you’re a new CIO coming into a situation where there’s no trust in the IT department, it might serve you well to make a very specific promise about how things are going to change"

Maurice Schweitzer, associate Professor, Wharton School, University of Pennsylvania

Although trust is a core construct in management literature, the focus of much of that research has been detecting deception. Precious little examines what happens after trust is broken. You spent a million more than expected on that SAP implementation that you swore would revolutionise the enterprise and the ROI is nowhere to be found. Now what?
So Schweitzer, with Wharton colleagues John Hershey, professor of operations and information management, and Eric Bradlow, professor of marketing, conducted a series of experiments between 2000 and 2004 to uncover what happens at the intersection of deception and trust.

Some of the results surprised even Schweitzer. The bad news? Broken trust, when accompanied by deception, is harder to repair. A simple apology does little to reverse the damage. The good news? Trust is less fragile than most of us think and a promise to change things followed by visible positive actions can go a long way in mending trust. Schweitzer spoke about his findings and what they can teach CIOs about managing expectations, repairing broken trust and making promises you cannot keep.

Why did you explore the issues at the intersection of trust, deception, apologies and promises?

I teach a negotiations class and deception is a chronic problem. That got me to thinking about deception more deeply. A lot of lies we tell are social, along the lines of ‘that dress makes you look terrific’ or ‘what a great haircut’. There’s a whole class of lies that help us get along in a functional way. At the same time, trust is the glue that holds any social relationship together.

If you look at the deception literature, an enormous amount of it is focused on how to catch liars and there has been a great deal written about the ethics of lying. But there’s very little research on what happens once somebody lies.
The common wisdom is that trust, once broken, is impossible to repair. The recovery of trust is slow or difficult and trust never fully recovers. People have always talked about trust as if it were made of glass: easy to break and difficult to repair.

As I began to think about deception, that seemed wrong to me. With some relationships, you violate trust and the relationship ruptures completely.

But in many settings, particularly the office, the relationship continues. And in many cases, trust gets repaired. The question I wanted to answer was how does deception harm trust?

How did you test your theories?

We had to agree on a definition of trust. The meaning agreed upon was ‘a willingness to accept vulnerability based upon positive expectations about another’s behaviour’.
To find out what happens when trust is harmed, individuals were paired with each other in a game involving money. One player in each pair – the ‘odd’ player – was given $6 in each round, which they could either keep or pass to the other person – the ‘even’ player. If the odd player kept the $6, the round ended and the even player got nothing. If the odd player passed the $6 to the even player, it tripled to $18 and the even player could decide how much to return to the odd player.

We used money because it gave participants something they actually cared about. They were trusting this money to someone with the expectation that if they did, that money would increase and the other person would return some of it to them.

Why would I loan you money? You do that with the expectation that you’ll receive something positive from that going forward.
We explored how trust is harmed by untrustworthy behaviour and deceptive behaviour, as well as how apologies, promises and subsequent trustworthy behaviour affected it.
A key aspect of our experiment favoured trust recovery. We didn’t let the relationship rupture. Players had to continue playing even when one acted in an untrustworthy manner. In real life, relationships sometimes rupture. If your spouse violates your trust, you may separate. But if your boss does, you may just have to deal with it. The relationship may or may not recover.

What did you discover?

We went in with the assumption that trust is extraordinarily fragile. Our results suggested this was not true.

We found that trust could be effectively restored when individuals observed a consistent series of trustworthy actions, such as having money returned to them each round. Or when a promise was made to change: “I give you my word. I will always return $9 every round, including the last one.” People were very receptive to that. In fact, we were surprised how quickly trust was restored. In some cases, the trust eventually recovered completely. However, the promise to change only worked if it was accompanied by a delivery on that promise.

When a person’s trust was violated – and that violation included deception – it was much more difficult to restore, even when followed by a series of trustworthy actions or promises.

We didn’t anticipate just how harmful deception would be to trust. We also thought that an apology would be more effective in trust recovery than it was.

The apology – ‘I really screwed up. I shouldn’t have done that. I’m very sorry I tried taking so much these last two rounds’ – did little. We thought that it would be more effective.

It may be that the apology didn’t go far enough. Apology has to be perceived as sincere. It has to indicate remorse and a plan to change. An effective apology can be powerful. But ‘I’m sorry’ just isn’t good enough.

What can CIOs learn from this research?

Be very careful about making promises or commitments you cannot keep. Inevitably there is going to be a time when – intentionally or unintentionally – you let people down. But you should recognise that if a relationship doesn’t rupture completely, chances for rebuilding trust are very high.

When it comes to CIOs, people may have inflated expectations of IT. There may be deadlines that they expect you to meet that you miss. There’s also something called a psychological contract. For example, an IT employee has a psychological contract with the CIO that involves what the worker expects that isn’t written down. Those are almost invariably violated because no one is sure of those expectations. Then there are things a CIO might never have promised – daily backups of data – but people assumed the IT department was doing.
It’s impossible to manage every expectation. So it’s important to figure out what you need to do to repair it after a trust violation occurs. One of the most important lessons from the research is that words can be very powerful in repairing relationships, specifically in repairing trust. But for words to be powerful, they have to be credible. And for them to be credible, you can’t have lied to people in the past. In fact, you can’t have even over promised in the past. For example, if you are going to lay people off, a CIO should not say, ‘I’m going to lay off 200 people and that one action will put us on track and everyone else will be safe’. Making that kind of statement with conviction risks credibility you might need later if you have to come back and say, ‘we need to do another round of layoffs’. Employees won’t buy it.

If you’re a new CIO coming into a situation where there’s no trust in the IT department, it might serve you well to make a very specific promise about how things are going to change.

A promise to change can be very effective. But you need to make sure that the actions you’re taking are clearly observed. Make sure that employees can see your staff coming in on Saturdays to get that project done. Have the IT staff interact more with the business so they know what’s going on. People are usually willing to give you a chance but you have to work hard to follow through on the promises you make.

Does context matter when trust is broken?

A critical question for managers across industries involves the role of trust in their business and the nature of the violation.

When Arthur Andersen committed accounting violations, the firm faced a serious threat to its business. They were selling a seal of approval. When that seal is less credible, it has less value. When Martha Stewart commits an accounting violation, she has not fundamentally threatened her business – aside from her absence – because she is selling style advice. Advice about decorating is not related to her lying.

The message of my research is that trust recovery can happen, but the receiver needs to believe the message and perceive that he hasn’t been lied to previously. Senior managers at HP or Enron would have needed to convince their audience that the untrustworthy act would not be repeated, and they would want others to observe their future behaviour.
This can take the form of voluntary decisions to “open their books” or have independent auditors inspect their work. Greater transparency can be a big help.

Understanding how trust works is key for executives – like CIOs – who work globally.
Yes. One strength of the US economy is that most Americans are trusting. They trust its institutions. You can fly to Cincinnati, sign a contract, get back on the plane and assume the deal you signed is going to happen.

In many developing economies, that’s not the case. Business becomes encumbered by rituals that have been put in place to develop relationships. You can’t meet a total stranger and agree on a large transaction in a short period of time. China is the obvious example. If you’re doing business there, it’s essential to develop relationships and spend a lot of time going to banquets and making toasts and travelling to different events.
The building block for business there is trust in the individual. American companies have a lot of trouble when they swap out managers there every few years or so and the new manager has to start all over to build that relationship and that trust.

How do emotions affect trust?

We looked at the influence of incidental emotions on trust. You get a speeding ticket just before a board meeting. Or you find out you got a promotion before meeting with a new vendor.

We did extensive tests to find out how emotions influence trust judgments. We found out that they have a very big impact. When you go into that meeting with the new vendor, you’ll ask yourself, ‘do I trust them?’ And if you just don’t know, you’ll with your emotions, ‘how do I feel?’ If you feel pretty good, you’ll make a positive trust judgment based on emotions unrelated to the actual situation. It’s important to take that into account when trying to earn or keep trust. That’s why a really good salesperson may tell a joke to try to affect someone’s emotion. If you encounter someone in an emotional state that might negatively influence their judgment, there are three strategies. Change the emotion: Tell a joke or comment on the weather. Recognise the source of emotion: “I was really sorry to hear what happened to your house.” Or harness good emotions: “I heard your kid got into Stanford!” You have to be emotionally savvy.