IN 2001 I WAS APPOINTED PROFESSOR AT THE EUROPEAN UNIVERSITY Institute (EUI) in Florence, and I took up residence in that beautiful city. The EUI was founded by the European community for training the best minds on the continent in doctoral studies and academic research in the social sciences. Each member of the European Union is allotted a certain number of students that it can enroll at the institute, and as a result the institute is comprised of students of many different nationalities in equal proportion—a wonderful thing for the university, and also, coincidentally, a fantastic group to use for research. Most of the students speak at least three European languages and have lived in more than one country in the European Union. The EUI openly expresses its intention of being the ideological center of the European Union.
In late 2001 the German foreign minister, Joschka Fischer, convened the entire social sciences faculty of the EUI, about thirty professors, and charged us with the strange task of drafting a constitution for “the United States of Europe.” When I was subsequently awarded a large research grant from the European Union, I chose, along with several colleagues, to use part of the money to study trust and trustworthiness in the European context.1 We decided to conduct an experiment using a game based on a “market of favors.”
Students from different parts of Europe who had just arrived at the EUI (and therefore had not had any time to become acquainted with each other) were the subjects of the experiment. They were divided into groups, each with five students. The members of each group did not see each other directly; all their interactions took place through computer monitors.
At the start of the experiment a brief description of each participant was circulated among the members of the group. The most important detail for us (the experimenters) was the participant’s country of origin, but we also included age, academic interests, and other minor details in the circulated descriptions. Each participant was then given fifty euros and instructed that he could give any amount of money from that sum to any other member of the group. As in the trust game, any money that was thus transferred from one participant to another was tripled. The recipients of this initial generosity were given an opportunity to repay the proposers who had given them the money, in any amount that they saw fit, again just as in the trust game.
This was repeated in each group for six rounds. The resulting effect was the creation of a dynamic market of favors in which individuals chose others as recipients of their largesse, with the expectation that they would be rewarded for this generosity, either in the same round or in a later round.
Our aim was to compare the extent to which people are willing to trust individuals from northern Europe compared to those from southern Europe. For the purpose of this study we defined northern Europe to include Denmark, Sweden, Finland, Great Britain, Germany, the Netherlands, and Belgium. We considered Italy, Spain, Greece, Portugal, and France (most of whose citizens live in the southern part of that country) to be southern European countries. Not coincidentally, the geographic line dividing northern Europe from southern Europe is also the cultural dividing line between Latin culture and Anglo-Germanic culture.
Given the backgrounds of the subjects in this experiment, who were all young intellectuals with resumes that included a significant amount of international and multicultural interactions, one might expect that country of origin would have no effect on the trust levels exhibited in the game. But that turned out to be a wrong assumption. Southern Europeans were significantly discriminated against relative to northern Europeans. Northern Europeans exhibited distrust of southern Europeans. The discrimination was evident both in the identities of the partners chosen to be recipients of money and in the amounts of money given to them by the other players. Southern Europeans were chosen relatively infrequently, and when they were chosen, they received less money compared to northern Europeans.
The dynamic aspect of the game gave us an opportunity to follow how discrimination emerged in round after round of the game. We had expected expressions of discrimination to diminish as the game progressed, but to our surprise the opposite happened. A careful analysis of the data revealed the secret of growing discrimination: in the first round there was some expression of mistrust against southern Europeans, but it was small and marginal. The objects of this mistrust then responded with their own measure of distrust in the second round; mistrust naturally fosters a mistrustful reaction. This was interpreted as justifying discrimination against the southerners, leading to even greater discrimination against them in the next round, further entrenching mistrust, and so on in a growing spiral of discrimination and mistrust. A small and marginal initial grain of unjustified discrimination ballooned out of proportion before our eyes.
The small mistrust exhibited at the start of the game was a self-fulfilling prophecy that by the end of the game became full-blown discrimination. We concluded that if young, sophisticated intellectuals enrolled in the elite EUI could act this way, the phenomenon should be prevalent throughout Europe.
Publishing the paper proved difficult because some reviewers, unfairly in my view, regarded it as being provocatively accusatory. To his credit, the editor of the most important journal in behavioral economics, a Spaniard from a German family, recognized the importance of the paper and agreed to publish it.
Both southern Europeans and northern Europeans bore some measure of responsibility for the growing mistrust that emerged in our experiment. Many interpersonal communication failures apparently stem from such self-fulfilling mistrust. An employer expressing lack of confidence in the abilities of an employee limits that employee’s chances of successes. If that employee then fails as a result, the employer will feel that his initial expectations were confirmed. On the other hand, an employee who from the start expects any job success to be summarily dismissed by his or her employers is inviting the very lack of credit and respect that he or she expected. Fear of being hurt or disappointed in a romantic relationship can itself doom the relationship.