Media Briefings


  • Published Date: December 2013


What measure should we trust to measure trust? Research by Professors Paola Sapienza, Anna Toldrà-Simats and Luigi Zingales suggests that the best indicator is the beliefs about the trustworthiness of others expressed by ‘senders’ in the ‘trust game’. Their study, published in the December 2013 issue of the Economic Journal, explains that while this game is a widely used way of assessing the level of trust in a country, the result that is typically taken from it can be misleading.

In a trust game, the ‘sender’ is endowed with some money and decides how much of that money to send to the ‘receiver’. Any amount sent is multiplied by three, and the receiver then decides how much to return to the sender. The amount of money sent is the frequently used measure of trust.

The new study argues that using the amount sent can be misleading. Sending money is an action that involves the preferences of the sender, such as her attitude towards risk or altruism, in addition to the belief in return, which by definition is the true measure of trust (the belief in the probability with which someone will perform an action that is beneficial to us).

In their experimental setting, the researchers run the standard trust game but in addition, they ask the sender to report her beliefs about the receiver’s behaviour. By doing so, they are able to separate senders’ beliefs from their preferences, for which they also have experimental measures.

Using a sample of 550 students from the full-time MBA program at the University of Chicago Booth School of Business, they show that the quantity sent in the trust game is a mix of two components: the beliefs and the preferences of the sender.

They also show that people’s beliefs measure trust: individuals with stronger beliefs in the trustworthiness of others send more; and an individual is more likely to send more when she has particularly high expectations about the amount the receiver will return.

An alternative measure of trust used by many researchers is the answer to a question in the World Values Survey (WVS): ‘Generally speaking, would you say that most people can be trusted?’ The WVS question mostly captures the beliefs expressed in the trust game. But due to the dichotomous nature of the answer to this question, it might be an imprecise measure of trust.

The researchers conclude that the best measure of trust is the belief in return for large amounts sent obtained from the trust game, since this variable is the least contaminated by other considerations.


Notes for editors: ‘Understanding Trust’ by Paola Sapienza, Anna Toldrà-Simats and Luigi Zingales is published in the December 2013 issue of the Economic Journal.

Paola Sapienza is at Northwestern University. Anna Toldrà-Simats is at Universidad Carlos III de Madrid. Luigi Zingales is at the University of Chicago.

For further information: contact Anna Toldrà-Simats via email:; or Romesh Vaitilingam on +44-7768-661095 (email:; Twitter: @econromesh).