A cornerstone of economic theory is that rational agents are self-interested. When making choices about how to allocate resources, agents behave rationally when they choose to maximize benefits to themselves.

But here’s the rub: Over a decade of research in experimental economics has shown that people do not always seek outcomes that maximize benefits to themselves. Instead, we evaluate choices with respect to outcomes to ourselves but also with respect to outcomes to others. A simple way to put this is that our choices appear to be equally motivated by a desire for fairness as well as self-interest.

But here's another rub: What constitutes fairness? A growing body of research indicates that we do not hold people of different social status to the same standards: What counts as fair for a high-status individual does not necessarily count as fair for a low-status individual.

For example, when asked to monitor compliance with social rules, people are more likely to seek out possible instances of cheating if they adopt the perspective of a high-status individual who must monitor compliance of lower-status individuals than vice versa. So it would seem that we hold those of lower status to higher standards of behavior.

Yet the story is more complicated than that. My colleague, Lawrence Fiddick, and I conducted a series of studies across seven countries (Australia, Canada, Germany, Japan, Singapore, UK, and USA) in which people were asked to indicate their willingness to maintain a simple carpooling arrangement in which they drove another person in exchange for their paying for gasoline. Their carpool partner’s compliance with the rule varied from perfect compliance (100%) to rare compliance (ponying up gas money only 25% of the time). Some people were asked to adopt the perspective of a boss driving an employee, while others were asked to adopt the opposite (an employee driving the boss).

As you might expect, willingness to continue the arrangement declined dramatically as compliance dropped. But what you might not expect is that people were far more tolerant of employee non-compliance than boss non-compliance—even when the employee was described as making more than the boss (due to a home-based side-business). Compared to participants who adopted an employee perspective, those who adopted a boss perspective were far more willing to continue the arrangement despite significant non- compliance, were more likely to feel they had been treated fairly, felt less animosity towards their cheating partners, and believed they got the better deal because they felt they bore less cost and received higher value from the arrangement. These results were stable across seven countries that differed markedly in their social values. Countries scoring high on collectivism (Japan and Singapore) were as likely to exhibit this pattern as those that scored high on individualism (Australia, Canada, Germany, UK, and US).

The results are also consistent with an ethnographic study of small-scale horticultural, foraging, and pastoral cultures.
We refer to this pattern of behavior as noblesse oblige, a social norm that obliges those of higher rank to be honorable and generous in their dealings with those of lower rank. Another way to put this is that status has social utility. High status confers greater satisfaction than low status, which shifts the balance of cost/benefit ratios accordingly.

The question is why we behave this way. Noblesse oblige is not adequately explained by standard game theoretical analyses nor by economic theories that incorporate norms of fairness. One promising explanation, however, derives from Costly Signaling Theory. Briefly, this theory states that costly, altruistic acts may benefit an altruist indirectly, by establishing a reputation that enhances mating opportunities or alliance formation. In order for behavior to count as costly signaling, it must be (1) beneficial to others, (2) observable by others, (3) costly to the signaler in ways that cannot be reciprocated, and (4) associated with signaler fitness. This is plainly the case in noblesse oblige where high-status signalers willingly bear a greater cost to confer a benefit to low status receivers.

Another way to put this is that those who practice noblesse oblige acquire prestige within their groups based on generosity. Prestige status contrasts with other types of status that are based on a sense of acquired or conferred entitlement. In a number of classic experimental economic studies, people were either told they had earned the right to be a dictator or proposer based on their performance on a trivia test or that the roles had been arbitrarily assigned. Dictators unilaterally decided how to divide a sum of money between self and partner. Proposers suggested a division which their partners could accept or reject; if the proposal was rejected, no one got any money. High scoring quiz performers earned significantly more, meaning that low status partners were offered less, made higher offers to high status partners, and were willing to accept lower offers. Results like these suggest that status leads people to behave exploitatively towards losers in ranked competitions, neutrally toward others when status is arbitrarily conferred, and generously towards lower ranking individuals when doing so will increase prestige.

The take-home message from this body of research is this: What counts as a fair transaction depends in large part on perceived relative social status—where the parties stand in the perceived hierarchy and how they believe they got there.

Denise Dellarosa Cummins is the author of Good Thinking: Seven Powerful Ideas that Influence the Way We Think (Cambridge University Press), and The Other Side of Psychology: How Experimental Psychologists Find Out the Way We Think and Feel (St. Martins Press). She is also co-editor of two scholarly volumes: Minds, Brains, & Computers: The Historical Foundations of Cognitive Science (Blackwell Publishers), and Evolution of Mind (Oxford University Press).