Articles & stories

"Trust is the key to the private sphere"

Iris Bohnet, a Swiss-born professor at Harvard University, argues against relying on intuition, explains how to conduct a job interview and warns of the dangers of social media. 

Ms. Bohnet, let me start with a simple question: What drives people?

You may not mean that seriously, but there is in fact a short answer to your question: passion.

Is it that simple?

I might also have said "a thirst for knowledge" or "the meaning of life." But passion goes deeper. It is what impels us to work, love, go jogging in the morning, buy flowers, do our best in the workplace – but it also motivates us to put up with an unpleasant job so that we can feed our families, or to survive incarceration.

You have a degree in economics.

Yes. So did you expect me to say that money is what drives people?

Perhaps, or more generally that we are motivated by a desire to maximize utility.

Status, income, wealth, power – all of these things are external influences. But what truly drives us is not these things, but factors within ourselves. Passion is a good shorthand term for these internal forces, which might also be referred to as intrinsic motivation.

Your research examines trust from the perspective of behavioral economics. What does trust have to do with the private sphere?

The better I know someone, the more I'm able to trust him or her. And the better I know someone, the more inclined I am to let that person get close to me. The key to the private sphere is trust.

There's a saying that trust is good, but control is better.

Granted, a great deal can be achieved through control. But it comes at a high price. If you check the ticket of everyone on the subway, fare evasion will no longer be a problem – but is that an efficient approach?

Kenneth J. Arrow, the recently deceased Nobel Laureate in economics, once said, "It saves a lot of trouble to have a fair degree of reliance on other people's word." Political scientist Francis Fukuyama argues that trust is a significant factor in a country's or company's prosperity. Do you agree?

I went shopping recently, and when it came time to pay I discovered that I had forgotten my wallet. I asked the customer in line behind me if he would lend me the amount I needed – and he agreed, giving me the money along with his name and address. That saved me the trouble of going home and coming back again. Along with my check, I sent him a box of chocolates – as interest, you might say. Trust clearly increases efficiency; it makes the pie larger.

Reputation is also a factor in determining whether you trust someone.

Sometimes your own opinion is less important than what others think. That is the power of reputation. And that power is increasing as technology advances. We look at the reviews before making a purchase. Amazon, Ricardo and TripAdvisor have made it easier to navigate the world and led to an increased focus on the customer. But as with progress of any kind, there are certain dangers.

A stream of unrestricted commentary?

I've recently found myself thinking about Heinrich Böll's novel "The Lost Honor of Katharina Blum." Because of her friendship with a criminal, an innocent woman is publicly vilified and driven to actions that end in tragedy. Böll's novel is an indictment of the tabloid media – and that's nothing compared with social media. How can we possibly determine whether all the things we read there are true? It's no wonder that we're living in an era of "fake news." Protecting privacy is more crucial than ever before.

Is trust becoming more important?

Its significance is changing. There is a fundamental need for trust when information is distributed asymmetrically. If everyone knows the same things, trust is unnecessary. In the world of finance, for example, recent technologies have led to greater transparency and more information sharing. With an iPad, it's possible to help determine your investment strategy and review it on a regular basis. When you meet with a client advisor, you have all of the data and can simulate various scenarios. Clients now have more information. But the situations they face have become more confusing and complex, and they have more providers to choose from. As a result, trust and expertise have once again become enormously important.

Some of your research has shown how sensitive people are to an abuse of trust, a phenomenon you call betrayal aversion. Is that comparable to risk aversion?

No. Let me give you an example that illustrates the difference. Let's say you go to a casino and play a game that offers you the chance to double your stakes. How high does the probability of winning have to be for you to decide to play?

If the odds are slightly better than one in two, we're in.

Okay, so about 55 percent. Now suppose that a colleague you don't know very well asks to borrow some money – saying that he'll pay back double that amount at the end of the week. For you to agree to lend him the money, how high would the likelihood of repayment have to be?

It would have to be quite certain. Maybe 75 percent?

So you see, your risk aversion is lower than your betrayal aversion, by 20 percentage points. That's true of most people. We simply don't like being cheated.

You also point out that there are significant cultural differences in how people respond to betrayal.

Let's say I run a gallery, and you have purchased a painting from me for 1,000 Swiss francs, but you haven't picked it up yet. Now another customer comes in who is even more enthralled with the picture, and she offers me 2,000 francs. In the United States, it is quite possible that I will sell the picture to her and refund your 1,000 francs, giving you a little extra as compensation. After all, I have caused you minor emotional damage. This is called "efficient breach." The idea is that we should break contracts when it's efficient to do so, but we should also compensate the person who has lost out.

That would be unthinkable in Switzerland. Here the system relies on the presumption that a handshake is binding.

Correct. The Swiss system is based on the legal principle of "pacta sunt servanda": Agreements must be kept. In this case, therefore, it is very likely that I would have to give the painting to the first buyer.

From a moral perspective, this seems right.

Yes, but the American system, too, promotes trust. You will be compensated if an agreement is broken. And the system doesn't discriminate as much. You don't have to think about whether you can trust someone or not, since the system ensures that the costs of a breach of contract are very low. It's much like how insurance works. In countries where it is considered extremely important not to be cheated, on the other hand, you do business only with people you can absolutely trust – your family or clan.

What cultures are you referring to?

Conducting our experiments all over the world, we found that in many Middle Eastern countries, trust is created by eliminating as many instances of breach of contract as possible. In that region, betrayal has a significant moral component. You also lose face if you're cheated, since it suggests that you haven't done enough to make sure that the person you were dealing with was trustworthy. Punishments are correspondingly draconian.

What countries are at the opposite end of the spectrum?

At the extreme end is China, where we found the lowest level of betrayal aversion. It was also very low in Brazil. In those countries, the levels of risk and betrayal aversion are practically identical. Trusting someone is viewed as similar to gambling. Sometimes it works out, sometimes it doesn't.

Recently you have been looking more closely at intuition and how it leads us astray. But isn't it that famous "sixth sense" what makes us human?

A great deal of research has shown that intuition is deceptive, as it's based on prejudices and stereotypes. Particularly when making major decisions, such as hiring someone, you want to be sure that you're choosing the very best candidate.

And you're saying that means that you should ignore your intuition?

Yes. Intuition isn't the best source of guidance when assessing a job applicant's performance and skills. If you know the age of a job candidate, or that she has two children or graduated from a specific university, that triggers certain associations that may not be accurate. Here's a question for you: When you think about people who live in Florida, what image comes to mind?

Retirees who want to enjoy the sunshine.

But the truth is that 84 percent of the population of Florida is under the age of 65, which is only slightly lower than the percentage for the United States as a whole. So while Florida's population is slightly older than the national average, there is no reason to assume that just because an applicant is from Florida, that person is older. This is a classic case of bias, a common phenomenon that we focus on in behavioral economics.

What should you base your hiring decisions on, if not on your intuition?

We know from numerous studies what works and what doesn't. First, you need to advertise the position in a way that appeals to the right candidates – that's not a simple matter, but algorithms can help find language that is free of unintentional bias. Resumes should be anonymous; they should not include the candidate's name or address – nor, of course, a photograph. Our research has demonstrated that such information is not helpful. In fact, it leads us astray.

What about the job interview itself?

We know more about job candidates than ever before. Yet we continue to conduct unstructured interviews, where interviewers are at the mercy of their prejudices. Google has conducted studies to determine the optimal number of interviewers – four – and to find out which questions actually predict future success in the company. That's the power of big data. It's also important to ask candidates to solve problems that have something to do with the job they are applying for. Such tests are a much better predictor of future success than interviews.

You say that you shouldn't base your decision on whether or not you find someone sympathetic. Why is that?

In recruiting a new employee, our aim should not be to duplicate ourselves. After all, diverse teams have a higher level of what we refer to as "collective intelligence." Hiring someone you like often means hiring a clone of yourself.

Not all prejudices are false; for example, the average person in Switzerland is probably more punctual than the average person in India.

That's correct. Intuition is like a rule of thumb – it helps us get through the day efficiently. But like all such rules, it's sometimes wrong. When you're making an important decision, do you really want to rely on a system that has been shown to be error-prone, especially when there are better options?

You're trying to eliminate the role of prejudices as much as possible. But the world is moving in a different direction.

Yes, unfortunately. Historically, people have retreated in times of rapid change and uncertainty. That is what we're experiencing right now. We're seeing a retreat in favor of our own nation, our own skin color, our own gender, our own political party. Many Western countries have experienced growth at the expense of the lower middle class, and we're now paying the price. The gap between rich and poor has widened, and for the first time not all of society's winners are white men. Some of that group are now rejecting the establishment and the advance of globalization.

Some are also lashing out against people like you – a member of the elite, a professor who argues using facts and figures.

Yes, trust in the so-called elites is declining. I found it very shocking that prejudices were expressed so casually during the American presidential campaign. Only a few years ago an open display of racism would have been unthinkable.

The situation is somewhat different in Switzerland. According to Credit Suisse's Worry Barometer, the Swiss government continues to enjoy very high levels of trust, higher than the governments of other countries. Why is that?

I think it has something to do with direct democracy. Our political system is very responsive to the people. There is a widening gap between the political sphere and the people in many countries – but not in Switzerland. Although some would disagree, I see no indication that there is an isolated "political class" in Switzerland.

Finally, a personal question: How did you make what I presume was the most important personnel decision of your life – "recruiting" your husband?

(She laughs.) Okay, you've caught me there – it wasn't a very systematic process. I simply fell in love with him. But before we got married, we had some matter-of-fact and not particularly romantic discussions about the important things in life: Do we want children? Do we want careers, and how much do we want to work? We didn't sign our marriage contract until we had reached agreement on all of those questions. I would encourage every couple to do the same.