“Trust is not a moral luxury”

Ernst Fehr, established economic theory has long assumed that humans largely focus on maximizing their own benefit. Your research has shown that people can also be altruistic, fair and cooperative. Is the economic model of the homo oeconomicus wrong?

Ernst Fehr: We have to differentiate between the homo oeconomicus as a model and as a description of real people. As a model, homo oeconomicus was extremely constructive, because it provided clear, verifiable forecasts. It only became problematic when economists declared self-interest to be the sole human motive. Self-interest is relevant for all humans, but it certainly isn’t the only motive. Our experiments show that many people are prepared to invest their own resources for fairness, cooperation and the well-being of others. But this isn’t unlimited – the more costly an altruistic act becomes, the fewer people perform it. People are thus neither exclusively egoistic nor unconditionally altruistic. They combine material self-interest with social motives.

How big is the share of people who don’t behave in an exclusively egoistic manner?

Fehr: In a study with around 2,000 participants selected to provide an accurate profile of the Swiss population, we identified three types – those averse to inequality, altruistic people and predominantly self-serving individuals. Around 45 percent are averse to inequality, about 35 percent are altruistic and roughly 20 percent egoistic – with some of the egoist group also displaying an inclination to envy. In US and Danish data, the share of egoists is higher and can account for up to a third of participants. What is crucial, however, is that the egoists never make up the majority. Economic theory had made the exception into the rule. Our research turned this upside-down picture the right way up.

When did it become clear to you that the economic perception of humans was too narrow?

Fehr: Even as a student I had great doubts. I was interested in historic mass movements, especially the workers’ movement. Thousands, even millions of people were prepared to take to the streets at significant risk to themselves to demand democratic rights – not just their own rights but also the rights of others. Political freedom and democracy are some of the most important public goods and they have almost universally been won through battle. This didn’t align with an image of humanity that is based purely on self-interest. It’s one thing to have an intuition, but to scientifically prove it is altogether more difficult. For this, new methods were needed.

You have often and very successfully worked with experiments. How did you arrive at this approach?

Fehr: At the end of the 1980s, my work focused on the job market and the question whether fairness plays a role in setting wages. I had written a theoretical paper on this, which was turned down by every journal. Looking back, this failure was productive. I told myself: what matters isn’t whether a theory is particularly elegant or mathematically complex, what matters is whether it is empirically correct. So I began experimenting. In an article for Nature I later described this turning point as the “productivity of failure.”

What was your first experiment?

Fehr: We conducted a job market experiment to find out whether fairness really does play an important role in wage setting. The experiment later came to be known as the gift-exchange experiment. The standard theory predicts that wages keep adjusting until supply and demand match each other. But we suspected that wages are linked to ideas of fairness. If a wage is perceived as too low, employees respond by making less of an effort. Companies therefore have an incentive to pay higher wages, as this boosts motivation and voluntary cooperation. And the experiment did in fact demonstrate that low wages led to reduced work performance, while higher wages elicited increased performance. The companies didn’t lower the wages as much as the standard theory had predicted, as this would have weakened their own productivity.

What does this finding mean for real companies?

Fehr: The finding primarily applies where work performance can’t be fully monitored. Companies rely on voluntary cooperation. Wages also have a motivating function. This explains why it is very rare for companies to reduce their employees’ wages. A wage reduction is seen as unfair and a lack of appreciation and it can have a strong negative impact on performance. In Switzerland, nominal wages are largely fixed in terms of downward movement. If inflation drops, there are more wage freezes, but wage reductions are rare.

Is there a universal concept of fairness?

Fehr: What is universal is probably the need for justice. But what is considered just is influenced by a person’s culture. We can distinguish between at least three fairness ideals – the egalitarian ideal that emphasizes equality, the meritocratic ideal by which greater achievement also justifies greater returns, and the libertarian ideal which accepts acquired entitlements regardless of whether they were obtained through performance or luck. International studies indicate large differences in the way these ideals are distributed. In Western countries, the meritocratic ideal is often especially strong. In other countries, inequalities resulting from luck are more readily accepted. This doesn’t mean that there’s no desire for justice, it’s simply that justice is defined differently there.

Are social preferences sound indicators of character?

Fehr: In part, yes. In a new study, we compared people’s social preferences with their life goals. People who are more willing to give to prosocial aims in the experiments tend to have other priorities in life. On average, career and income are less important to them. This doesn’t mean that they don’t care about money. But they are more mindful of working in a company that supplies products or services they perceive as meaningful or socially useful. Social preferences are thus not just situational responses in a lab, they are also linked to longer-term life goals.

A core theme of your research is trust. What is trust in economic terms?

Fehr: Trust means providing an upfront payment or service without the return being secured. In the trust game, a person can send for example CHF 10 to a second person. The amount is tripled and the second person decides how much to give back. In sending the money, the first person makes themselves vulnerable. That precisely is trust. Such situations are the basis of many economic relationships – one party first performs a service and has to trust that the other party will meet their obligation. Without trust, many mutually beneficial transactions can’t take place.

You also conducted a trust game in which the first person can choose to trust the partner twice in a row. What were you able to show with this?

Fehr: This experiment allowed us to distinguish between different forms of self-interest. If one partner keeps all the money in the first of the two rounds, this immediately destroys any trust, and they are unlikely to receive anything in the second round. This is a short-sighted or, put bluntly, stupid egoist. A strategic egoist initially cooperates in order to receive another upfront payment in the second round and waits until then to take advantage of the other person. A prosocial person gives back an appropriate amount in both rounds. Such experiments show that behavior which outwardly appears the same can arise from very different motives. In the first round, the prosocial individual and the strategic egoist act in a similar manner. That’s why we have to distinguish between behavior, motive and institutional parameters.

What societal significance does trust have?

Fehr: Many agreements are upheld not because they are legally enforceable but because the parties involved want to continue to be seen as reliable partners in the future. Reputation is economic capital. People who use every opportunity to exploit others may gain in the short term, but they undermine their ability to make agreements in the future. Other current or potential partners will look for alternatives, demand additional securities or discontinue relations. This logic applies to individuals as well as companies or states.

Based on what you learned from your trust research, what can you tell us about current global politics?

Fehr: Situations in real life are more complex than in the lab, but the underlying mechanism remains the same. If you destroy your reputation as a dependable contractual partner, you pay a high price in the long term. For nation states, credibility is particularly important, as there is often no superordinate authority that can reliably enforce agreements on an international level. A government that calls treaties into question whenever it sees fit, uses threats as a normal negotiating tool and puts partners under pressure at every opportunity may extort concessions in the short term. At the same time, however, it creates strong incentives for others to seek new trade partners, alliances and supply chains. This type of power politics only works if the opposite party has no alternative options. In an interdependent world, this is rarely the case. That is why trust is not a moral luxury, it is a prerequisite for stable economic and political relations.

You say people’s willingness to cooperate is often underestimated. What does that mean for political dealings?

Fehr: Politicians often rashly assume that people will only cooperate if they’re being monitored, financially rewarded or punished. Our research shows that people’s willingness to cooperate is often significantly greater. But it has to be activated and protected. Many people cooperate conditionally – they are prepared to do their bit if they can expect others to contribute too and if they know freeloaders won’t benefit without being punished. Good institutions therefore create transparency about the behavior of others, make shared standards public and enable credible sanctions. If, by contrast, prosocial individuals are systematically taken advantage of, their willingness to cooperate will plummet. This is why political policy shouldn’t just create incentives for egoists, it should also avoid destroying the trust of the cooperating majority. Rather than merely focusing on efficiency, politicians must also consider perceptions of fairness. Ignoring these means underestimating a central aspect of human motivation.

You emphasize that institutions have a strong influence on human behavior. What’s new about this?

Fehr: One important insight concerns the distinction between motive and behavior. People can have strong fairness motives and still behave as though they were egoists. Imagine an ultimatum game between two individuals, person 1 and person 2. Person 1 is given CHF 100 and can share this with person 2 in any way they wish by making one single splitting proposal. If person 2 accepts this proposal, it is implemented, if the proposal is rejected neither person receives anything. In Western societies, person 2 in such a game typically rejects any proposals under CHF 30 because they perceive these proposals as unfair. In a second step, consider an ultimatum game involving 10 competing recipients who all have to decide whether to accept or reject the proposal at the same time. In the competitive situation, each person believes another will accept the offer. This means a person’s individual rejection can’t prevent the unfair result and they therefore accept it too. In the end, the result creates the impression that the recipients don’t care about fair distribution, even though they have fairness preferences. Institutions can make social motivation visible, suppress it or turn it into the opposite.

What conclusions do you draw from this regarding the organization of institutions?

Fehr: Institutions have to be judged by the behavior they enable and by the shared knowledge they create. Prosocial behavior becomes more likely when people can rely on others to participate as well. Trade unions, for instance, fulfill the function of organizing solidarity, creating commitment and signaling to the employees that they are acting collectively. A similar concept applies to employer associations. Institutions solve problems of collective action by stabilizing expectations and creating credible obligations. Without this knowledge about the behavior of others, even a group of prosocially motivated people can end up becoming unbalanced.

Looking back on your long career as a researcher, how has your work changed economic theory?

Fehr: The most important impact is the lasting influence on current and future research. Social preferences – fairness, reciprocity, altruism and aversion to inequality – are established economic topics today. Forty years ago, they were still largely ignored in economic theory. Today, many researchers work on measuring these motives, creating theoretical models for them and examining their effects in markets and institutions. Behavioral economics has broadened the way economics views human behavior. It hasn’t simply abolished the homo oeconomicus. Instead, it has shown when its predictions are accurate, when they fail and which additional motives need to be considered.

Has your own perception of humanity changed in the course of your research?

Fehr: My research has given me a more nuanced picture. What particularly surprised me was the aversion to inequality. Together with Klaus Schmidt, I developed a theory whereby a significant proportion of people rejects inequality. Initially, I mainly regarded this as a mathematically simple approximation to reciprocity. It was only much later that our evaluations convinced me that aversion to inequality is in fact a distinct motive in many people. This also shows how important it is to test theories empirically – even one’s own.

What do economists know about people that they didn’t know 40 years ago?

Fehr: We know much more about social preferences, openness to risk, patience and limited rationality. This increased understanding is largely based on empirical research and improved methods, in continual interaction with theory. The homo oeconomicus model was particularly important for this, as it provided clear, falsifiable predictions. A theory that can be disproved is scientifically valuable.

So will the homo oeconomicus stay relevant after all?

Fehr: Yes. There are situations in which people behave exactly the way the model predicts, though this doesn’t mean they are exclusively self-serving. An institution can cause people to behave like a homo oeconomicus. Competition can block fairness motives if nobody believes they can achieve a fair result on their own. That’s why the standard model continues to be an important point of reference. But we have to distinguish whether a result has come about due to egoistic preferences or due to institutional rules that render prosocial behavior ineffective. This distinction changes how we assess markets, organizations and politics.