The monkey handed a small granite rock back into Sarah Brosnan’s open palm, got a quarter of a cucumber slice, and ate it. She was a brown capuchin at the Yerkes primate center in Atlanta, and she had been trading like this for at least two years.
Next to her, behind a mesh partition, sat a second monkey doing the same job. When both were paid in cucumber, Emory’s report on the 2003 study said, they completed 95% of their trades.
Then the experimenter began paying the neighbour in grapes, for the same rock.
Completed trades fell to 60%. The study describes monkeys tossing the rock, or the cucumber, out of the test chamber, giving up food they would eat under almost any other circumstances. Nothing about the cucumber had changed. Something next door had.
People judge their pay by what others get for the same work, not by the pay alone.
A ratio nobody writes down
Forty years before the monkeys, J. Stacy Adams put that idea into a theory. In “Towards an Understanding of Inequity” (1963), he treated unfair pay as a special case of Leon Festinger’s cognitive dissonance: an uncomfortable mismatch that people feel pushed to resolve.
The mismatch is in a ratio. On one side is what you get out of a job: pay, status, benefits. On the other is what you bring to it: effort, education, experience, as you yourself see them. You set your ratio next to someone else’s. When the two match, the deal feels fair, even if the other person earns twice as much for twice the experience. When they don’t, Adams argued, the tension starts working.
He listed the ways out in a 1965 chapter, as a later review summarizes them. You can put in less. You can push for more. You can tell yourself a different story about what each side contributes. You can leave. You can lean on the other person, or start comparing yourself with someone else.
A raise feels good until the payroll spreadsheet leaks. Your salary is the same number it was yesterday. What changed is the line above it.
When the plant cut pay
The management researcher Jerald Greenberg measured employee theft in manufacturing plants while pay was temporarily cut by 15%. Groups whose pay was cut had significantly higher theft rates than in the periods before and after, and than groups whose pay was unchanged. Greenberg reported that the data supported equity theory’s predictions about how people respond to underpayment.
The same study had a second finding. Where the reason for the cut was explained thoroughly and sensitively, employees felt less unfairly treated, and theft went down too. The cut was the same. The story around it was not.
The generous half
Adams’s theory also makes a stranger prediction. If your ratio is better than the other person’s, that is inequity too, and you may work it off by putting in more.
His own early evidence came from a lab experiment in which college men were hired at $3.50 an hour to interview members of the public. Those led to feel overpaid, Adams and William Rosenbaum reported in 1962, were more productive than those who felt fairly paid. Paid by the interview instead, overpaid men produced less, as the theory predicted.
Adams himself expected people to notice being overpaid less readily than being underpaid. The evidence for this half has been shakier too. In 1968 Edward Lawler repeated the result, but the overpaid group tended to do lower-quality work. Adams had made people feel overpaid by telling them they were underqualified; feeling overpaid “by virtue of circumstance” was not enough, in Lawler’s experiment, to make people produce large quantities. In a review that year, Lawler concluded that the hourly-pay studies did not make a convincing case that equity theory was needed to explain them.
Some later results fit the generous half. In a 1988 field experiment, Greenberg temporarily reassigned 198 employees of an insurance company’s underwriting department to the offices of higher-, lower- or equal-status coworkers while their own were refurbished. Those placed in higher-status offices raised their performance; those placed in lower-status offices lowered it. In one of Dov Eden’s army Pygmalion experiments, trainees singled out at random as high-potential reported feeling overrewarded, and the researchers suggested this probably pushed them to put more into the course.
The new hire’s salary
Say Grace has run billing support for four years. A new colleague joins on the same team, with two years’ experience, and in a careless moment Grace learns his starting salary is higher than hers.
Nothing in Grace’s own deal has changed. The comparison has. She might ask for a raise, or start leaving on time, or decide the new hire must have skills she hasn’t seen, or update her CV. Which of these happens may depend partly on what she is told. A manager who can explain the gap (a market rate that moved, a skill the team lacked) gives her something to weigh. In Greenberg’s pay-cut study, a careful explanation came with less theft.
A grape on an empty seat
The monkeys were not a clean demonstration. In the same study, a grape simply placed where the partner usually sat, with no partner there, drew almost as many refusals as a partner eating grapes. Critics argued the monkeys were reacting to a better reward they could see, not to the unfairness of a neighbour having it, and later experiments with capuchins and great apes found the effect hard to pin down. Brosnan and de Waal stand by it: in 2014 they described protest against receiving less than a partner for the same task as widespread in species that cooperate outside kinship and mating bonds.
For people, the two may be just as tangled. A colleague’s salary is a comparison. It is also news that more was on offer.