The Motivating Power of Under-Confidence: 'The Race is Close But We're Losing'

43 Pages Posted: 22 Nov 2014

See all articles by Todd Rogers

Todd Rogers

Harvard University - Harvard Kennedy School (HKS)

Don A. Moore

University of California, Berkeley - Haas School of Business

Date Written: October 1, 2014

Abstract

Should political campaigns in close races communicate that they may win (over-confidence) or that they may lose (under-confidence)? In six studies (three survey experiments, two field experiments, and one archival study) we demonstrate the motivating power of under-confidence. While uncommitted voters show bandwagon effects (prefer candidates who are barely winning as opposed to barely losing), supporters show the opposite (greater motivation when their preferred candidate is barely losing as opposed to barely winning). Two fundraising email field experiments (1M observations) show a large effect size: emphasizing polls that show that a preferred candidate was barely losing raised 55% more than emphasizing polls that show that he was barely winning. The 2012 Obama and Romney campaigns’ emails reflect this insight: they were more likely to send emails reporting that they were barely losing than that they were barely winning. Sometimes leaders are more effective appearing under-confident rather than over-confident.

Keywords: Political Campaigns

Suggested Citation

Rogers, Todd and Moore, Don A., The Motivating Power of Under-Confidence: 'The Race is Close But We're Losing' (October 1, 2014). HKS Working Paper No. RWP14-047, Available at SSRN: https://ssrn.com/abstract=2528690 or http://dx.doi.org/10.2139/ssrn.2528690

Todd Rogers (Contact Author)

Harvard University - Harvard Kennedy School (HKS) ( email )

79 John F. Kennedy Street
Cambridge, MA 02138
United States

Don A. Moore

University of California, Berkeley - Haas School of Business ( email )

545 Student Services Building, #1900
2220 Piedmont Avenue
Berkeley, CA 94720
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
121
Abstract Views
1,391
rank
261,135
PlumX Metrics