Narcissists, especially the ones who have taken it to high-art levels, can be an annoying bunch. But empirical work on the trait has, unfortunately, consistently shown that it correlates positively with "leadership emergence," or becoming a straight-up boss. Where the research has been inconsistent, though, is with regard to whether these people are effective at leading their teams.
A newly completed meta-analysis of previous studies, set to appear in Personnel Psychology, suggests that "there exists an optimal, midrange level of leader Narcissism." (Hilariously, the analysis also indicates that "self-reported leadership effectiveness ratings are positively related to Narcissism.") In an email, University of Illinois-Urbana-Champaign psychology researcher and lead author Emily Grijalva, whose work focuses on the "impact of aberrant personality traits and leadership," explained the new wrinkle:
The major finding is basically that there is an inverted U-shaped relationship between narcissism and leadership effectiveness. So as a leader goes from having very low narcissism up to having average levels of narcissism, leadership effectiveness increases. But as a leader increases from average narcissism up to above-average narcissism, leadership effectiveness declines.
So the "point" at which leadership effectiveness starts to drop off is right around the population's average level of narcissism. In other words, it starts to drop off right after surpassing the average American's level of narcissism.
When your boss goes from casually self-impressed to staring at his or her reflection on the computer screen instead of shooting out emails delegating duties, that's when things start to go down hill. My very unscientific graphical interpretation: