Why ranking employees by performance backfires

   The ‘forced distribution’ model used by many companies is an outdated
   idea that should be scrapped
   [141]Sarah O'Connor

   Microsoft threw out the model in 2013, while KPMG says it plans to
   ‘move away’ from the system to ‘allow much greater flexibility’ ©
   Timothy A Clary/AFP/Getty

   [152]Sarah O’Connor
   April 6, 2021
   [153]Jump to comments section [154]Print this page

   When Bill Michael, the former chair of KPMG, told staff to [155]“stop
   moaning” in a virtual meeting in February, one of the issues they were
   complaining about was the “forced distribution” model used to assess
   their performance. This way of appraising people is a zombie idea. No
   matter how many times it proves disastrous for a company’s culture or
   morale, it refuses to die.

   Generally speaking, “forced distribution”, or “stack ranking”, methods
   divide employees each year into a certain percentage of top performers,
   average performers and underperformers. In the UK’s senior civil
   service, for example, the proportions [156]were fixed at 25, 65 and 10
   per cent respectively, until the system was reformed in 2019.

   The idea is to avoid “grade inflation” and force managers to have
   honest conversations with people who are sub-par. Jack Welch, former
   chief executive of General Electric, [157]said in 2013 he couldn’t
   understand why people thought this cruel. “We grade children in school,
   often as young as nine or 10, and no one calls that cruel. But somehow
   adults can’t take it? Explain that one to me.”

   There are circumstances where it works. One former accountancy trainee
   said he had only been willing to tolerate the long hours and grinding
   work if he was likely to be promoted. “In that context you really need
   to know if you’re in the 25th percentile or 75th percentile,” he told
   me.

   But there are many more examples where the method fails, even on its
   own terms. The big problem is that it mixes up someone’s absolute
   performance with their relative performance against their peers. You
   might be meeting all your objectives, for example, yet still be ranked
   bottom and labelled “underperforming” in a strong team. Sarah Nickson,
   a researcher at the Institute of Government think-tank, said the UK
   government’s “deep dive” into the forced distribution system for civil
   servants found that “a lot of the people in the bottom 10 per cent were
   not underperforming”.

   This not only feels unfair for the employee, it is deeply uncomfortable
   for the line manager. “What was difficult was when you had people
   meeting expectations, graded 3, but because there were so many 1s, 2s
   and 3s, there was pressure to give them a 4 and put them on an
   improvement plan,” one former manager in a big accountancy firm told
   me.

   In many organisations, line managers assign provisional grades then
   thrash out the overall distribution in “moderation” meetings with other
   managers. But it is hard to objectively rank people in white-collar
   jobs doing different things. “You’d have people sat in a room who
   barely knew each other, comparing apples with pears,” said a second
   manager at a different firm.

   A number of line managers told me they gamed the system. They would put
   people who had just joined the team in the bottom bracket, because they
   were easier to sacrifice. Or, perversely, they tried to hang on to poor
   performers so they could put them at the bottom and protect everyone
   else.

   The system can also disincentivise teamwork. Microsoft’s forced
   distribution ranking system (since scrapped) [158]was blamed for
   creating a toxic culture in the early 2000s that stifled innovation.
   Good performers reportedly avoided working together for fear of
   suffering in the rankings. People would quietly sabotage their
   colleagues.

   Finally, these systems are often corrosive for morale, which damages
   the very performance levels they aim to improve. [159]Research shows
   that feedback has a moderately positive effect on performance on
   average, but in a third of cases it decreases performance. What is key
   is whether people feel the feedback is fair.

   Employees who express positive emotions after feedback tend to perform
   better in the future, while those who express negative feelings go on
   to perform worse. In a forced distribution, only those ranked better
   than average are likely to feel particularly pleased. As Nickson points
   out, that is by definition only a minority. “It’s nice for your ego
   when you’re told you’re a high performer, [but] I think the question
   is, does that benefit outweigh the downsides for potentially everyone
   else?”

   Microsoft scrapped forced distribution in 2013. The UK’s senior civil
   service followed suit in 2019. KPMG told me it planned to “move away”
   from the system to “allow much greater flexibility” too.

   Human resources departments have had to rethink plenty of old notions
   since the pandemic hit. The counterproductive pseudoscience of forced
   distribution ought to be one idea that finally stays dead.

   [160][email protected]

References

   Visible links:
    (...)
 141. https://www.ft.com/sarah-o-connor
 152. https://www.ft.com/sarah-o-connor
 153. file:///home/km/afg/ft/0691002c-2200-4583-88c9-9c942d534228#comments-anchor
 154. javascript:void(0)
 155. https://www.ft.com/content/5cb42741-6e9e-4996-8e32-032bd09ea28c
 156. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/971013/SCS-PM-UpdatedGuidance-April2021_1.0.pdf
 157. https://www.wsj.com/articles/SB10001424052702303789604579198281053673534
 158. https://www.vanityfair.com/news/business/2012/08/microsoft-lost-mojo-steve-ballmer
 159. https://www.cipd.co.uk/knowledge/fundamentals/people/performance/what-works-in-performance-management-report#gref
 160. https://mail.google.com/mail/?view=cm&fs=1&tf=1&[email protected]