The Peter Principle which states that people are promoted until they end up in a position beyond their capabilities was first proposed by Canadian psychologist Laurence J. Peter in the 1960s. With my name and when, as often happens, I feel out of my depth I often wonder if I am some sort of living case study. (Bob Sutton
has an excellent background article on the Peter Principle.)
It is a paradox of our selection and talent management process that we eventually promote people beyond their capabilities and yet we spend an enormous amount of time, effort and cash on selection and development. So imagine the surprise to find that perhaps we'd be better if we selected our leaders at random.
'Every year, the Ig nobels
honor some of the most entertaining research published in the past year. The research should "first makes people laugh, then make them think." The competition is fierce, and the prizes highly coveted. This year, the winners of the Management category were
Pluchino, A., Rapisarda, A., & Garofalo, C. (2010). The Peter principle revisited: A computational study Physica A: Statistical Mechanics and its Applications, 389 (3), 467-472 DOI: 10.1016/j.physa.2009.09.045
Conventional wisdom suggests that promoting the hardest working, most dedicated or most intelligent employee is the route to success. Not so, suggests the research of this team. Utilizing the Peter Principle, which states that people will continue to get promoted until they end up in a position beyond their capabilities, the team modeled a hierarchical organization and found that a degree of randomness was the most efficient promotion strategy.'
Now some might argue that this policy has already been adopted in their organisation, but I couldn't possibly comment. What I would like to suggest is that as we now have a hypothetical model, can we create an evidence based one and finally banish the Peter Principle. We can then adopt the "random promotion principle". As Bill Bennet
commented on Bob Sutton's blog: 'The ancient Greeks in Athens elected leaders by random, literally drawing names written on shards of pottery out of a large urn. A ruler would be in charge for a year. If they did a good job, the city gave them a farm, if they did a bad job, they were executed. I wonder if modern companies would run better with a similar system?'