Can Monkeys Teach Us About Fairness?

In 2003, a study titled ‘Monkeys reject unequal pay’ was published in the leading journal Nature. The authors, Sarah Brosnan and Frans de Waal, reported evidence that capuchin monkeys exhibit ‘inequity aversion’. They conducted an experiment where pairs of monkeys exchanged tokens for rewards with a human. In one condition, both monkeys received cucumber (an inferior reward). In the other, one of the two monkeys received grapes (a superior reward). Brosnan and de Waal found that moneys were much more likely to reject the cucumber in the second condition, such as by throwing it at the human, after seeing their partner receive a grape.

The study soon became famous, and was taken to show that unfairness in economic transactions is so unnatural that even our primate cousins won’t stand for it. For example, the New York Times ran a piece titled ‘What the Monkeys Can Teach Humans About Making America Fairer’. And one of the study’s authors once compared his finding to “the Wall Street protests”.

As a brief aside, ‘inequity aversion’ seems like an inaccurate term to describe what the study found. If monkeys were really averse to inequity, the ones that received grapes should have also rejected their rewards. ‘Being-treated-unfairly aversion’ seems like a more accurate term, although that’s admittedly a bit of a mouthful.

Anyway, a new paper suggests that monkeys don’t have much to teach us about fairness after all. Oded Ritov and colleagues carried out a meta-analysis of 23 studies testing for ‘inequity aversion’ in non-human animals, including the original 2003 study. Importantly, they were able to meta-analyse the primary data from those studies rather than the reported estimates, allowing for greater rigour.

As the authors point out, ‘inequity aversion’ isn’t the only possible explanation for the original finding. Another is disappointment. Perhaps the monkeys who rejected the cucumber were angry that they received an inferior reward when a superior one was available, regardless of what their partner received. Fortunately, there’s a way to disentangle these two explanations: check whether monkeys are more likely to reject the cucumber when their partner is given a grape than when they are simply shown a grape. Some of the studies included in the meta-analysis actually did this.

So what did Ritov and colleagues find? In short, no evidence for ‘inequity aversion’ in non-human animals. When they compared rejection rates in ‘disappointment conditions’ to those in ‘inequity aversion conditions’, they found that they were higher in the former. In other words, merely drawing an animal’s attention to the superior reward has a bigger effect than showing another animal receive it. In fact, rejection rates in ‘inequity aversion conditions’ were not significantly different from those in control conditions (where there was no superior reward), which suggest that the original finding may have been a false positive.

Ristov and colleagues note that while some individual animals may exhibit ‘inequity aversion’, it does not appear to be a general phenomenon outside humans. On reflection, trying to draw lessons about fairness from a study on monkeys was kind of pointless: we already knew that humans dislike being treated unfairly.