02.03 – Implications of Performance Measurement Systems

So why do we care about forced performance distributions? What are the benefits? One of the recent very interesting studies compared the performance of groups under forced distributions relative to non-forced distributions or absolute rating systems. And one of the key findings in that study was that groups working under forced performance distribution under forced curve registered, on average, 8% percent productivity gains. So basically, when we know we’re gonna be compared to others, we work harder. In part, because incentives are clearer and more crisply defined. In the absence of a forced performance distribution, managers assign more lenient and less differentiated ratings. And as a result, top performers are not so clearly differentiated from not so top performers.

Under forced curve, in contrast, the ratings are more clearly differentiated and the differences in rewards are clear. And that’s what drives our performance and motivation. But the study had another really interesting and important caveat, which is it turns out we don’t fare very well in switching to forced performance distribution. In other words, it might be problematic to install a forced curve in an organization and a team if your employees are used to more liberal lenient system. For those employees that were switching to forced distribution, initial increase in performance was followed by a sharp decline in performance.

And that decline was so precipitous that performance levels drop well below those that the team had prior to the introduction of forced performance distribution.

What happens here is that initially, we recognize that we need to work harder under a forced curve. But we soon get demotivated because we cannot quite get the same ratings and the rewards we used to get. Now, there are other problems permeating forced peer rankings and comparative systems in general. In a forced peer ranking system, we’re hoping essentially for the effect of a running pack, where we push each other’s performance. We strive to keep up and beat the best. But in reality, we might get something like this, where there are two ways to the top. One is to improve my own performance, and the other is to drag others down.

And the resulting behaviors can range from reduced collaboration, all the way to outright sabotage. Many years ago, a PhD program at an elite university had a system that was similar to GE’s, where they ranked ordered their PhD students. And the bottom 20% of the first year class would be asked to leave the program no matter what the average performance of the class was. That university had to bend in this practice because their library was suffering too much. Students would go in the library and rip out the chapters assigned for seminars, so that they would be the only ones prepared for class.

So there are significant disadvantages of comparative rating systems and forced distributions in particular. Those range from outright sabotage, to cheating, to creative accounting, and other forms of unethical behavior.

Employees are more likely to take extreme risks under forced curves and forced distributions in peer ranking systems, especially if rewards are heavily skewed to the top and we feel like we’re falling behind. Forced distributions can result in perceptions of inequity and low morale. And the primary culprit here is a situation when your forced curve doesn’t quite match the actual distribution of performance. The managers that I interviewed said that one of the most demotivating comments they heard from their senior leaders in performance appraisals is, you know, I’d love to give you five but I have to give you four because of the curve. And finally, enforce distribution systems and comparative peer evaluations.

Performance rating conversations are extremely emotionally charged. They’re very political, where managers were more vocal, more aggressive, more influential, more prepared, have a better chance of building a stronger case for their employees. Take a look at the quote from a former GlaxoSmithKline manager who discusses a meeting of area managers where they assign ratings to employees under the curve. He says, look, it’s like being in court. The more you’re prepared, the better chance your people have of reaching the top of the scale.

And the ensuing perceptions of the zero sum game can severely undermine collaboration and coordination across different units within the organization.