Media Briefings

The Impact Of Favouritism On Team Performance: Experimental Evidence From The School Playground

  • Published Date: December 2011

Favouritism – treating certain people better on the basis of friendship ties rather than merit – is not necessarily damaging to the success of a team. That is the conclusion of research by Michèle Belot and Jeroen van de Ven, who have conducted an experiment in nine primary schools, with two different age groups – younger children (6-8 years old) and older children (10-12 years old).

As might be expected, their analysis of an experimental ball game finds strong evidence of favouritism in the school playground. But the study also finds that children who are chosen by friends improve their performance in the game, so that favouritism does not have a negative impact on group performance.

In the experiment, the children compete in teams in a tournament within their class. They simultaneously perform a simple task – bringing balls from one basket to another for 30 seconds – with a clear performance measure. After the first round and after observing all team members' performance, children indicate which team member they would prefer to do the task in the second round, for the benefit of the team.

The main results, published in the December 2011 issue of the Economic Journal, are as follows:

  • First, friends are substantially more likely to be chosen, conditional on their performance in the first round. Best friends are 30 to 45 percentage points more likely to be selected than others.
  • Second, the weight given to past performance is very different for younger and older children. It plays little role for the younger ones, while it plays a large and significant role for older children. Importantly, performance becomes a distinctive selection criterion in addition to friendship and not a substitute.
  • Third, and perhaps most surprising, there is a flipside to favouritism. Specifically, children who are chosen by friends substantially improve their performance compared with non-friends. On balance, favouring friends is not detrimental to efficiency.

A plausible explanation for the positive flipside of favouritism is that these children reciprocate the favour by increasing their efforts. This underlines the importance of measuring performance after the selection decision.

At the time of the selection decision, choosing friends who are not the best performers can easily be prematurely classified as discriminatory and inefficient. But taking account of the fact that these children increase efforts afterwards, it cannot be concluded that favours are detrimental to efficiency.

The researchers begin their study by noting the widespread belief that social connections can make a crucial difference to success in life, particularly in the labour market – ‘success starts with knowing the right people’.

This interplay between friendships and success potentially starts early on in life too. For example, children are regularly put in situations where they are asked to choose between their peers and form teams to compete against each other – typically in physical education lessons. Favouritism could already arise then.

But the implications of favouritism for efficiency are not straightforward. While it is evident that not selecting the most qualified person can be detrimental for efficiency, it could equally be efficiency enhancing, depending on how people respond to favouritism. For example, friends may be more loyal and put more effort into the job if the manager is their friend than otherwise.

This study presents novel experimental evidence on the existence and efficiency implications of favouritism among children. The researchers find that children value friendship ties early on in life. But since favouritism is not detrimental to group performance, this may explain why it continues over time and carries over into adulthood. The positive flipside of favouritism may be the very reason why it survives.


Notes for editors: 'Friendships and Favouritism on the Schoolground – A Framed Field Experiment' by Michèle Belot and Jeroen van de Venis published in the December 2011 issue of the Economic Journal.

Michèle Belot is a research fellow at the Centre for Experimental Social Sciences at Nuffield College, Oxford University. Jeroen van de Ven is at the University of Amsterdam.

For further information: contact Michèle Belot on +44-1865-278548 (email:; or Romesh Vaitilingam on +44-7768-661095 (email: