Here is evidence that collective decision making is superior to individual decisions:
This raises a question: if collective decision-making is superior to individual decision-making, why do we have a cult of the individual when it comes to appointing chief executives and party leaders? Why not have more collective leadership?
I can think of several answers:
1. My presumption is wrong. Many business and political decisions, though nominally attributed to CEOs or party leaders, are taken by a team in which the leader might be only primus inter pares.
2. Groups can get things as badly wrong as individuals. Laboratory experiments don’t show this, as they exclude the conditions - such people becoming close over time - which generate groupthink.
3. Having a single individual to praise or blame can sharpen incentives and so improve performance. I’m not sure about this, as high-powered incentives can often backfire.
4. Leadership is not just about decision-taking. It’s also about inspiring and motivating people. Individuals do this better than committees.
5. The cult of the individual is founded not in empirical data, but in ideology. The idea that some super-humans are able to take complex decisions is just a myth used to justify what is in fact rent-extraction by bosses.
Should we really put a non-zero weight on (5)?
Our results indicate that teams (1) learn faster the solution to a reasoning problem than individuals; (2) need less specific (and weaker) performance feedback in doing so; (3) might even perform at a higher level than the best individuals.See also this pdf and this.
This raises a question: if collective decision-making is superior to individual decision-making, why do we have a cult of the individual when it comes to appointing chief executives and party leaders? Why not have more collective leadership?
I can think of several answers:
1. My presumption is wrong. Many business and political decisions, though nominally attributed to CEOs or party leaders, are taken by a team in which the leader might be only primus inter pares.
2. Groups can get things as badly wrong as individuals. Laboratory experiments don’t show this, as they exclude the conditions - such people becoming close over time - which generate groupthink.
3. Having a single individual to praise or blame can sharpen incentives and so improve performance. I’m not sure about this, as high-powered incentives can often backfire.
4. Leadership is not just about decision-taking. It’s also about inspiring and motivating people. Individuals do this better than committees.
5. The cult of the individual is founded not in empirical data, but in ideology. The idea that some super-humans are able to take complex decisions is just a myth used to justify what is in fact rent-extraction by bosses.
Should we really put a non-zero weight on (5)?
Studies come and go on this a bit, evidently. The last I'd heard, the best results suggested that decisions made by groups were very heavily groupthink-dominated and that it was better to have individuals making the decisions because they were less prone to egg themselves on into a more and more extreme position.
Posted by: Philip Walker | July 30, 2010 at 03:19 PM
Because when everyone is responsible, no one is responsible.
If you have a committee in charge, and it all goes t*ts up, who do you sack? All of them? None of them? One or two at random 'pour encourager les autres'?
Or is it better to have one person in charge who takes full responsiblity for their own decision making?
Posted by: Jim | July 30, 2010 at 03:20 PM
My experience has always been that even if a team is "flat" there are still leaders within the team environment who push the team in directions they'd like to see it go. Bullying and/or consensus building need individuals to push them.
Posted by: James Gradisher | July 30, 2010 at 03:37 PM
I don't know exactly what weight (5) should be given but that it should be non-zero seems trivial.
Even a primus inter pares is presumably primus because they are held in some sort of particular esteem. That they will tend to use this esteem to leverage more esteem and produce a Matthew Effect-type feedback loop seems inescapable.
Posted by: twitter.com/matt_heath | July 30, 2010 at 03:53 PM
@Jim: I think this is correct. Many years ago I worked as Chief Assistant to - well, let's just say a charismatic leader of a major organisation. He dealt with operational decisions in a very simple way. The manager responsible would be asked to prepare a short note outlining the problem or situation, and to end with a note recommending the best and most appropriate way forward in his or her view. Then my boss would then immediately write "OK" at the bottom and send it back. I was much struck at the time by the fact that was (is) much admired for his speedy and unambiguous decision-making and leadership, though I knew that most of the time he wasn't actually making any decisions at all. Indeed, he often didn't bother to read the note before writing his usual "OK". This system worked extremely well in business terms because the managers were invariably better informed and had greater expertise in their own areas than my boss could possible have, and had also taken the opportunity to consult their colleagues for their views before "going to the top for a decision".
I thought then, and think now, that this mechanism, although ultimately pointless, was primarily there to satisfy a general need within the organisation to feel that there was "someone in charge" and "taking responsibility" (though a need for approbation may also have played a part).
Posted by: Churm Rincewind | July 30, 2010 at 03:58 PM
@Jim - "noone is responsible" is a good description of what happens when boss-centric management fails. Fred Goodwin escaped astonishingly lightly when RBS went tits up.
Posted by: chris | July 30, 2010 at 05:45 PM
quite a good blog..however i don't think we shoul put a non-zero weight on (5)...
http://6b0c30z86ci7gqcfn2jfoery5y.hop.clickbank.net/
Posted by: mil | August 06, 2010 at 12:58 AM