Why Performance Rankings Can Backfire for Promotions

Ranking workers against one another can be good for business — except when they’re competing for a promotion.

Based on the research of Eric Chan

It’s an idea as old as the Olympics and as modern as a sales contest: Competition motivates us to work harder. When we compare ourselves with our peers, we push past our old limits.

In today’s business world, that concept takes the form of relative performance information. Several studies have shown that when workers are ranked against others, their output increases.

But performance feedback can sometimes backfire, finds new research from Eric Chan, assistant professor of accounting at Texas McCombs. The problem arises when it’s used for promotions. It can create pressure to choose the person with the biggest numbers — regardless of how qualified they are.

“When you promote the best performer at the current job, they’re not necessarily going to be the best at the next level, because there might be a mismatch between their abilities and the new job,” says Chan.

The reason, he says, goes back to another classic management concept: the Peter Principle.

One Promotion Too Far

In the 1969 bestseller “The Peter Principle,” education professor Laurence Peter offered a fresh explanation for the age-old nuisance of bad bosses. People who do well in one job get promoted. That’s fine, until they reach their “level of incompetence” — a job they can’t handle. At that point, they tend to stay put, harming the whole organization.

Chan was reminded of the Peter Principle by a friend in retail sales at a global education company who was facing a tough decision: which salesperson to promote to manager. One was less experienced but better at organizing a team. The other lacked leadership skills but worked hard and posted the best sales. “My friend felt like he should reward that person,” Chan recalls.

The conversation got him wondering: Could performance-based promotions actually hurt a business in the long run?

To find out, Chan set up an experiment with 220 volunteers, grouped into 44 mock companies. Each firm included one boss and four workers, who were paid for a simple computer task: moving a slider to the correct position along a scrollbar.

Within each company, bosses ranked workers by output. But only half the firms shared those rankings with workers. That setup allowed Chan to compare the effects of providing workers with relative performance information against not providing it.

The experiment was divided into eight separate work periods lasting 90 seconds each. After two initial work periods, the boss announced a chance at promotion, to a higher-level job with better pay. As workers competed for it, over the next three work periods, Chan found their average output jumped 14 percent. Low-performing workers showed the biggest improvement.

Chan believes they worked harder because they knew the boss was watching. “If they can win the heart of their bosses, they increase their chances of promotion,” he says.

Who to Reward?

Bosses were being measured, too, on how they made promotion decisions. Besides being ranked for how well they did the entry-level job, workers had been tested on their aptitude for the higher-level role. That position involved performing a more complicated slider task that required math skills as well as speed. Would bosses pick the hardest worker or the one who had tested the best?

The answer, Chan discovered, depended on whether workers knew their performance rankings. In situations where workers didn’t know their performance standings, 83 percent of bosses selected the one with the highest test score.

But where workers knew their rankings, the opposite was true. Bosses selected the top entry-level performer 63 percent of the time. Their workers, when surveyed, agreed it was the fairest way to make promotions.

“Sharing relative performance information creates the expectation that whoever has the highest performance deserves some reward,” Chan explains. “Even among those who are not promoted, it seems fair for the best person in the current job to be promoted.”

But while such decisions were good for morale, they were bad for business. After their promotions, as they worked for three more work periods, the best performers at the old job were the worst performers at the new one. They produced 28 percent less output than the workers who tested well for the task, and their firms earned 33 percent lower profits.

Those results, says Chan, offer some of the first experimental support for the Peter Principle. “Initially, the Peter Principle was viewed as sort of a satire,” he says. “It’s taken awhile for there to be serious studies. This study provides direct evidence on how bosses, despite their good intentions to satisfy workers’ expectations, end up promoting people who become bad managers.”

One way to sidestep the Peter Principle, he suggests, might be to set up a two-track promotion system. Instead of becoming managers, top salespeople could get elevated to higher sales jobs. Says Chan, “They can get promoted within their area of expertise rather than to their level of incompetence.”

In most cases, though, employers need to weigh trade-offs. “If you just promote the best workers in the current job, it may not lead to the best promotion outcomes,” he says.

“If you just promote the people who will make the best managers, it can reduce the incentives for people to work hard in their current jobs.” — Eric Chan

“The dilemma always exists,” Chan says. “It’s not so much about solving it as understanding the consequences.”

“Promotion, Relative Performance Information, and the Peter Principle” was published in May 2018 in The Accounting Review.

Story by Steve Brooks