Media Briefings

THE POWER OF INDIRECT RECIPROCITY: Evidence from a natural field experiment on what drives human kindness in everyday interactions

  • Published Date: June 2018

We are twice as likely to cooperate and act generously towards strangers if we ourselves have been helped in a similar way. What’s more, it doesn’t matter whether we are young or old, male or female, rich or poor, whether we have high or low time costs, or whether we are on our own or with other people – this important boost to human kindness still persists.

These are the key findings of research by Redzo Mujcic and Andreas Leibbrandt, published in the June 2018 issue of the Economic Journal. Their study uses a natural field experiment to test the power of indirect reciprocity in everyday social interactions.

The researchers carefully study the likelihoods with which car drivers give up their right of way and stop to help others in two experimental treatments:

• In the indirect reciprocity treatment, the authors measure the rate at which random drivers give way to an experimenter after another experimenter yields right of way to them – which is found to be 32.1%.

• This outcome is then compared with the baseline treatment, where the likelihood with which drivers voluntarily give way to an experimenter is observed to be 14.6%.

• From this, the authors are able to infer the relative importance of indirect reciprocity in human behaviour relative to unconditional generosity (with an overall increase of about 120%, or more than double the baseline rate). Figures 1 and 2 below illustrate the experimental set-up and the main result.

In contrast with other explanations of pro-social behaviour, which suggest that it results from social comparisons, ‘warm glow’, intentions or beliefs, the idea of indirect reciprocity assumes that it is past encounters that affect pro-social behaviour.

More precisely, the idea is that you exhibit pro-social behaviour because somebody else has exhibited pro-social behaviour towards you (upstream indirect reciprocity) or that you receive pro-social behaviour because you have exhibited pro-social behaviour towards somebody else (downstream indirect reciprocity).

The new study is one of the first to capture the simple idea of upstream indirect reciprocity in natural social encounters between strangers who are able to experience the kind actions of others explicitly and always in real time – and then immediately have an opportunity to reciprocate the same act of kindness towards someone else.

What can we learn from this study? First, it shows that indirect reciprocity can reduce waiting times and affect traffic flow and thus social welfare.

Second, it suggests that indirect reciprocity can account for a large share of pro-social behaviour that may have otherwise been imprecisely labelled as unconditional generosity.

The findings are consistent with the idea that received acts of generosity affect human emotions, which in turn increase the likelihood of generosity.

Such observed generosity in one-shot anonymous interactions can be reconciled from an evolutionary perspective as a consequence of natural selection for reciprocal cooperation, which has left a mark on human emotions.

Hence, while people sometimes only encounter each other once and do not have any strong concerns about their future reputation, they still feel indebted, and are arguably shaped, to pass on such pro-social favours to others in society.

***************************


Indirect Reciprocity and Prosocial Behaviour: Evidence from a Natural Field Experiment’ by Redzo Mujcic and Andreas Leibbrandt is published in the June 2018 issue of the Economic Journal.

Redzo Mujcic is at Wirtschaftsuniversität Wien (WU). Andreas Leibbrandt is at Monash University.

For further information: contact Redzo Mujcic on +43 1 31336 6024 (email: redzo.mujcic@wu.ac.at; Twitter: @RedzoMujcic); Andreas Leibbrandt via email: andreas.leibbrandt@monash.edu; or Romesh Vaitilingam on +44-7768-661095 (email: romesh@vaitilingam.com; Twitter: @econromesh).



Figure 1: Field experimental design for baseline generosity and indirect reciprocity treatments

Notes. Baseline treatment (two-person interaction) is illustrated in the top panel (a). Here, the left-hand side subfigure illustrates the initial interaction between the Subject and Experimenter A. The right-hand side subfigure illustrates the outcome following pro-social behaviour, in which the Subject stops and helps by giving way to Experimenter A. Indirect reciprocity treatment (three-person interaction) is illustrated in the bottom panel (b). Here, the left-hand side subfigure illustrates the initial interaction between Experimenter B, the Subject, and Experimenter A. The right-hand side subfigure illustrates the outcome following upstream indirect reciprocity, in which Experimenter B first stops and gives way to the Subject, and then the Subject (recipient of the kind act) stops and helps by giving way to Experimenter A (a third party not involved in the initial interaction)


Figure 2: Generosity rates observed in the baseline and indirect reciprocity treatments

Notes. The light (white) bar shows the generosity rate (vertical axis) in the two-person interaction (Baseline treatment). The dark (grey) bar shows the generosity rate in the three-person interaction (Indirect Reciprocity treatment). The observed treatment difference of 119% is statistically significant at the 1% level (Fisher’s exact test).