It’s the end of the year, the traditional time for asking for a raise or a wage bonus from your boss. If you’re a friend of his, the better your chances, no?

In fact, researchers at the Hebrew University of Jerusalem together will colleagues at the University of Chicago and the University of California at Los Angeles have found that if you’re asking for an end-of-year salary hike or bonus, it will be more forthcoming if he or she is NOT your friend. Nepotism is a no-no, at least if it becomes public knowledge.

Suppose a manager has to give a bonus to one of two equally deserving employees, one of whom is his friend. Or a judge in a high-school debating competition has to decide which of two finalists to vote for, one of whom is a student from her alma mater. Or a basketball coach has to decide which of two players should start in the championship game, one of whom is the coach’s niece.

It is well known that people show favoritism toward those close to them, such as family, friends or in-group members.

But the new study published in the Journal of Experimental Social Psychology by Hebrew University’s School of Business Administration Prof. Shoham Choshen-Hillel and her colleagues – Assistant Prof. of psychology Alex Shaw of Chicago and Associate Prof. of behavioral science at UCLA – suggests that in contexts where fairness is important, people attempt to avoid the appearance of partiality. Although such efforts to avoid appearing partial can often reduce biases, the researchers argued that, at times, such efforts can actually lead people to be biased against their friends.

In a series of eight studies using workplace scenarios, the researchers found that when managers had to decide whether an employee deserved a bonus, they were less likely to award one to a deserving employee who was also a friend if news of the bonus were made public. They didn’t want to be seen as playing favorites. Instead, they often awarded bonus to a less-deserving employees who were not their friends.

This fear, even when an employee-friend deserves the bonus based on merit, causes bosses to over-correct against their friends. “Bosses systematically treat their friends in a worse manner than they would treat their non-friends,” explained Choshen-Hillel.

“This concern with nepotism, even when there is a merit-based justification, can cause bosses to throw high-performing friends under the bus in an effort to avoid charges of favoritism.”  In private, however, participants were willing to give the bonus to the deserving person whether she was a friend or a non-friend, suggesting that their public behavior was aimed at avoiding the appearance of bias. The researchers found that bosses’ reluctance to give a bonus to a deserving friend is mediated by their beliefs that others would find this behavior to be unfair.

In one experiment, participants acting in the role of “boss” were faced with two employees, “Mark” and “Dan.” Both employees had the same job description and salary, and each was awarded a $200 bonus.  The boss was given the choice of awarding one employee with an extra bonus of $100 dollars.

To make their decision, the boss was asked to rely on input from fellow staff members.  “Mark” received better reviews than “Dan.” When study participants/bosses were told that Mark is their friend and that giving him the extra bonus would be public knowledge, they chose to give him the deserved bonus only 61% of the time. However, when participants were told their bonus decision would be kept private, they chose to give their friend Mark the extra bonus 89% of the time.

But when “Mark” was described just as an employee and not a friend, participants awarded “Mark” the bonus 88% of the time if their decision was made public and 87% when it was kept private.

In another experiment, participants were told their employee-friend had outperformed another employee who was up for the bonus. They were free to award their employee-friend the bonus based on merit or to flip a coin and let chance decide who should get the bonus.

In cases where the bonus decision was made public, participants awarded their deserving friend a bonus only 27% of the time, and the coin-flip the majority of the time. However, when participants were told the deserving employee worker was just an employee and not a friend, they awarded them the bonus more than 60% of the time.

Study participants’ concern over how their bonus decisions would be viewed by others in the office, was warranted. Choshen-Hillel and her colleagues did an experiment in which participants were briefed on the boss’s bonus options and their subsequent decisions. They rated the bosses on a scale of 1 to 7, (1= strongly disagree, and 7 = strongly agree) as whether the boss had acted fairly.

Giving the bonus to a neutral employee was deemed more equitable (5.54) than giving the bonus to a friend (4.57). Not giving anyone the additional bonus was considered fairer when the boss was considering giving it to a friend (5.13) than to a neutral employee (4.11).

Choshen-Hillel and her colleagues note that not all managers “navigate the dynamics between impartiality and favoritism” by over-correcting in a biased way against a friend. In the experiments, many participants did decide to award bonuses based on merit alone. But the researchers’ overall findings “highlight the delicate pressure that bosses face when they are faced with the competing values of fairness and friendship.”