Why some people choose to be leaders: The emergence of leadership in groups and organizations
Lead Research Organisation:
University of Kent
Department Name: Sch of Economics
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Cartwright E
(2014)
Strategic delay and information cascades
in Journal of Economics
CARTWRIGHT E
(2013)
LEADERSHIP BY EXAMPLE IN THE WEAK-LINK GAME
in Economic Inquiry
CARTWRIGHT E
(2010)
Imitation and the Incentive to Contribute Early in a Sequential Public Good Game
in Journal of Public Economic Theory
Gillet J
(2011)
Selfish or servant leadership? Evolutionary predictions on leadership personalities in coordination games
in Personality and Individual Differences
| Description | Whenever a group of agents interact a leader-follower relationship almost always emerges. The broad objectives of this project were to improve understanding of (i) who chooses to lead and follow within the group and why, and (ii) does leadership benefit the group. To meet these objectives we analyzed, both theoretically and experimentally, leadership in three different settings. Each setting involved a coordination game in which individuals stood to benefit from coordinating behaviour, but doing this was not easy. The three settings combined provide a good opportunity to see how the emergence of leadership depends on the particular problem faced by a group, and how well leadership lets groups overcome a coordination problem. The first game we considered was one in which individuals needed to invest in either technology X or technology Y. Positive externalities meant that it could be beneficial for individuals to coordinate on a choice of technology. Different individuals might, however, differ in which technology they preferred. If some prefer X and some prefer Y then there is a conflict of interest over which technology to coordinate on. Without leadership we demonstrate that there can be significant coordination failure in the sense that different people choose different technologies and so the full gains to coordinating are not realised. With leadership we demonstrate that coordination failure is much reduced with the actions of a leader allowing players to coordinate. The gains to leadership that we observed in our experiments were not as high, however, as those predicted by the theory. This was particularly the case if the preferences of others were not common knowledge and so leaders were choosing without knowing quite what followers would do. We explain the lower than predicted coordination in the experimental sessions as due to some subjects being too eager to lead. More specifically, we allow different individuals to have different preferences over the technologies, one might really prefer X while another prefers Y but only slightly, and different information about others preferences, one might know the others preferences while the other does not. Such asymmetries in preference and information should have determined who lead and who followed. In practice we found that these asymmetries were good but not perfect predictors of leadership. Some subjects lead more often than they should have and their leading led to less coordination that was possible. We highlight, though, that even with this slight negative effect groups with leadership did significantly better than groups without. The second game considered was a weak link game where individuals must contribute to a group project. Success is determined by the minimum contribution. Most efficient, therefore, is for all to contribute the highest amount. This, however, involves the risk of contributing a lot only to see one other member of the group contribute little. Because of this, without leadership we typically observe low contributions and coordination failure. We found that leadership led to no improvement in initial coordination and groups still suffered from low contributions. The continued efforts of leaders did, however, allow a good proportion of groups (25-50% depending on how we count) to escape from this coordination failure. Given the problems of overcoming coordination failure, well documented in the literature, this is a relatively high success rate, so leadership is beneficial. The final game considered was a threshold public good game where members of a group need to collectively contribute beyond a threshold in order to provide a public good. Coordination requires that the right amount be contributed. This, however, entails some conflict because some might free-ride, contribute little, forcing others to contribute a lot or forgo the public good. We observed considerable coordination failure with the public good only being provided around two thirds of the time. This level of efficiency is more than in settings without leadership but still relatively low. It turns out that success in providing the public good depends critically on the contribution of the leader. The more the leader contributes the more certainty that followers will help reach the threshold. We find that leaders contribute more, and thus leadership is most successful, when their identity will be revealed and there is some strategic uncertainty what others will do. Some of the basic conclusions that we draw from the project at this stage are: (1) leadership does help groups coordinate but does not completely eradicate coordination failure. (2) The actions of a leader are crucial in determining how successful leadership will be. (3) Leaders can expect to get lower payoffs than followers. (4) Who chooses to lead and follow will depend on the strategic environment. The personality traits of leaders varied across the three games considered, reflecting the different type of environment. |
| Exploitation Route | Our findings add to knowledge on effective leadership in the workplace or elsewhere and so are of general interest. |
| Sectors | Communities and Social Services/Policy,Financial Services, and Management Consultancy |
| Description | The results have been published in leading economic and psychology journals. Work is still ongoing and we expect further publications. The results have been disseminated at relevant conferences and workshops including PET09, ESA and European ESA meetings, and the Tiber symposium on psychology and economics. Results were also communicated to a general audience, including an article in New Scientist and Mark van Vugt's book 'Naturally selected: The evolutionary science of leadership'. |