I've got a free moment so going to post a couple of interesting things I've read in the past month or so.
First, I remember Dad telling me when I was a kid about the "Peter Principle", neatly summarized as "Everyone rises to their level of incompetence". I remember getting a warm feeling about this, knowing that if I was good at what I was doing, I could then "rise" to do something that was somehow better or more important. Peter himself, however, seems to have taken less solace from his maxim. In his view, this meant that all people, even the most capable among us, are either terrible at their jobs or soon will be.
Meanwhile, the 2010 Ig Nobel prizes, which celebrate strange research, were given out recently. One winner tried to get at some of the consequences of the Peter principle using computer simulations. Here it is:
http://arxiv.org/PS_cache/arxiv/pdf/0907/0907.0455v3.pdf
The upshot of the study is that, if the job one level up from you is not terribly related to your current job, it is highly detrimental to the organization to promote people who are good at their jobs, and it is in fact much better to promote people completely at random. This is because, under the "best get promoted" scheme, people who are doing a job badly tend to keep doing it, as they never get promoted. People who do a job well get promoted up out of the job until they find one that they're terrible at, at which point they stay there for the remainder of their careers.
For this to be true, the authors assume that your competency at your current job is only slightly related to your competency at the job one level up. Clearly, in some situations this isn't true, but it seems to me that it is true more often than not. Think about the skills needed to stock shelves extremely well at a grocery store. Now think about the skills needed to manage a grocery store. Why on Earth should we promote the best shelf stocker, especially if he or she is so good at it?
While no one is doing random promotions yet, this revelation has given some businesses pause. They are doing things like conditionally promoting the best people to see if they're any good, and if they aren't they just get a raise to stay in their current job. Hooray for science!
2 comments:
I haven't had a chance to read the entire study yet, but I do like the conversation begun by Charlie.
Interesting stuff to be sure, but employing a "random promotion" system would certainly have consequences. The first one that comes to mind is the principle of motivation. Maybe that shelf-stocker is putting in the extra effort in the hope of getting noticed and moving up the career ladder, and is not simply motivated by the potential for a monetary raise. Take away the potential to move up the chain, and you may see a quick decrease in the quality of his performance. There are other principles of motivating work such as skill variety, challenge, and meaning that may also contribute to longer term poor performance of a person stuck in one position.
In some fields, being effective in the top position may require the kind of thorough understanding of all the positions beneath one can only get from experience in those positions.
People have come into the shop...more later.
Brian - I also thought of the motivation aspect and totally agree with you that this makes random promotion a bad idea. This is why I like the "conditional promotion" idea. You only get the chance to be promoted if you do a good job, and if your new position doesn't work out, you just go back to your old one with a raise. Handily, the incentive remains even if you go back to your old job as you can always get another raise.
You are also right in your second paragraph. Clearly the assistant pro->head pro promotion is not a promotion to an unrelated job, and so strictly merit-based promotion is a great idea. But in my job, my boss, and his boss are an amazing computer scientist and chemist, respectively, but who the hell cares? I think a few years work as a kindergarten teacher would probably have been better preparation to wrangle such an unruly mob as us.
Post a Comment