Promotion by Lot

Posted on October 29, 2010 by peterstone

Haven’t had a chance to read the study described here yet…

Study: Most Efficient Organizations Grab Random Employees, Promote Them

…but it does deal with a fascinating problem. If you promote the best people, the argument goes, you will keep promoting people to tougher and tougher jobs until they no longer excel at them. The result will be an organization full of people stuck in positions for which they’re not particularly qualified. So says the Peter Principle, for which I can claim no credit. I’d be curious of the details as to how exactly the argument works, but the implications are striking. If you randomized the process of putting people into more difficult positions, it would seem odd to call it “promoting” them anymore. The latter term seems inherently related to merit or desert. It would then seem better just to say that the more difficult jobs (i.e., jobs requiring higher levels of competence) are reassigned by lot. (Should this happen periodically? Good question, but one I cannot answer until I actually get around to reading the study.)

Filed under: Distribution by lot |

5 Responses

Yoram Gat, on October 29, 2010 at 5:05 am said:

> If you promote the best people, the argument goes, you will keep promoting people to tougher and tougher jobs until they no longer excel at them.

I think the argument actually goes like this:

If you change the jobs of the best-performing people (through promotion or through any other mechanism) you will keep changing people’s jobs until they have a job they no longer excel at.

> If you randomized the process of putting people into more difficult positions, it would seem odd to call it “promoting” them anymore. The latter term seems inherently related to merit or desert.

I think “promotion” is used here (and in common corporate parlance) simply to denote a move into a job that carries more authority and more (material) rewards. Merit or desert are to a large extent incidental and may or may not hold in any particular case.

The higher authority and higher reward job may not be more difficult in any meaningful sense, but it is likely to require different skills to some extent.

LikeLike
Conall, on October 29, 2010 at 4:52 pm said:

Two things make this a dubious proposal:

1. It’s based on simulated data, not real experience.

2. It’s had an ‘IgNobel’ award, which suggests it’s an object of mockery.

Having said that, I wholly agree with the sentiment – promotion by lot. The alternative is Merit, which begs the question: How can Merit be measured reliably? I present evidence in my book (Lotteries for Education, 2010) that Human Judgement is fatally flawed – see p131 et seq.

(The one exception that I read about is that employees in a section were very good at picking an effective section-leader. It’s the underlings, not the bosses who really know what’s going on!)

Still, organisations need to establish some criterion for promotion: I’d like to see Short Lists of 6 ‘promotables’ drawn up, and then the winner chosen by a roll of a Die.
Valid Merit + Lottery.

LikeLike
conallboyle, on October 31, 2010 at 6:12 pm said:

and for those of us who have to pay to see academic papers, here’s my backdoor free version of Pluchino et co’s paper:

Click to access The_20Peter_20Principle_20Revisited2010.pdf

LikeLike
Harald Korneliussen, on November 2, 2010 at 10:26 am said:

The Guardian picked up this story, and treated it surprisingly well, I thought.

These days, the IgNobel prize isn’t a mere object of mockery. They used to be, but these days they strive to promote science that “first makes you laugh, then makes you think”. While this was based on simulated data, it’s worth noting that in the prior study that the Guardian mentions, the result caught the experimenters by surprise.

LikeLike
More from Pluchino et al. « Equality by lot, on March 16, 2011 at 2:59 am said:

[…] new article by Pluchino et al. is linked to in a recent edit to the Wikipedia entry for sortition (possibly by Pluchino […]

LikeLike