This podcast from VoxDev discusses how different methods of choosing leaders can lead to different outcomes.
If leaders in this experiment in Uganda were chosen in an open discussion, the researchers found that the most influential people tended to be chosen. However, these influential and loud people weren't exactly the best leaders for the goals of the microfinancing group.
When the leaders were chosen through a silent ballot, less "loud" people in the community ended up with the posts, but those people also ended up helping the group succeed more effectively.