«First prize is a Cadillac Eldorado, second prize a set of steak knives, third prize you’re fired». What Alec Baldwin introduces in a famous Glengarry Glen Ross scene is a particularly crude form of performance ranking and what follows in the movie is a story of cheating, betrayal, and infighting as actors attempt to get ahead in the raking. In the real life, the risks with performance rankings are not too far off. Rankings can help attract and retain top talent who thrive in competitive environments, improve the speed of group decision-making, and have been known to reduce biases in performance evaluations. But, rankings have a dark side–they often enhance competitive pressures, making them potentially problematic for the maintenance of continued cooperation. Despite these potential drawbacks, rankings are still widely used to incentivize employees and successful companies are able to rank employees while managing to achieve high levels of cooperation.
Making use of an experiment, Cassandra Chambers, an Assistant Professor at Bocconi University’s Department of Management and Technology, highlights on the one hand that performance rankings do in fact dramatically reduce levels of cooperation in groups and, on the other, that sharing reputational information (individuals’ histories of pro-social contributions) almost completely offsets the disruptive effect of performance ranks.
In her experimental setting, the introduction of performance rank information reduced the odds that a participant would cooperate to 0.36 times those in the control condition who did not receive any information. However, the odds of cooperating for participants who received reputation information (i.e., how much others gave in the past) alongside rank information were 1.87 times that of participants who just received information about their rank.
In the lab experiment, 592 people (students, lecturers and staff of an American university), divided into 74 groups, were asked to decide whether or not to give points with other participants in an extended period of decision-making. After some rounds, when an organic routine of cooperation was established, a ranking system was introduced and participants were provided with information about their own rank.
The propensity to give to others plummeted, due to concerns about losing one’s rank position or perceptions of unfairness. Furthermore, the drop was larger in the groups that had proved to be more generous in the first stage of the game, suggesting that performance rankings can be particularly disruptive in the most cooperative cultures.
However, it turns out that this disruptive effect of rank can be largely offset by the introduction of information about others’ rates of giving. After a brief disruption in cooperation levels, groups that received both types of information restored cooperation to almost pre-disruption levels.
«Our key finding is that displaying prosocial reputations – giving recognition to helpers – is a mechanism that allows systems of cooperation to withstand disruptive forces created by performance rankings. In a way, managers may be able to have the best of both worlds–a thriving system of cooperation without sacrificing a ranking system that motivates high levels of effort», Prof. Chambers says.
«Put differently», she continues, «our research suggests that managers should be very careful utilizing performance rankings, if they don’t want to disrupt a cooperative culture, but that minor efforts to provide recognition for prosocial activities may greatly shore up cooperative cultures. For example, managers can make a special effort to offer public recognition for employees’ prosocial contributions, use peer-to-peer bonus systems that enable employees to recognize and reward helpers, and create formal performance reviews that explicitly focus on rewarding helpful behaviors.»