Motivating Agents: How Much Does the Mission Matter?

Motivating Agents: How Much Does the Mission Matter? Economic theory predicts that agents will work harder if they believe in the “mission” of the organization. Well-identified estimates of exactly how much harder they will work have been elusive, however, because agents select into jobs. We conduct a real effort experiment with participants who work directly for organizations with clear missions. Weeks before the experiment, we survey potential participants for their organizational preferences. At the experiment, we randomly assign workers to organizations, creating either mission matches or mismatches. We overlay performance incentives to test whether they can make up for the motivational deficit caused by a mismatch. Our estimates suggest that matched workers produce 72% more than mismatched workers and that performance pay can increase output by 35% compared to workers who only receive a base wage. Considering matches and mismatches separately, we find that performance pay has only a modest effect on matched workers (a 13% increase) while it has a large effect (a 86% increase) on mismatches, an effect that erodes much of the performance gap caused by the poor match. Our results have broad implications both for those organizations already with well-defined missions (i.e., compensation and screening policies) as well as for those organizations strategizing about strengthening or clarifying their missions. NON-TECHNICAL SUMMARY How much harder do people work when they believe in the mission of their organization? Perhaps as important, how much less do they work when they disagree with the mission? We conduct an experiment to estimate the effect of missions on worker productivity. During the 2012 US Presidential election, we randomly assign individuals to work for either the Obama or Romney presidential campaigns thus creating both mission “matches” and “mismatches.” Compared to people who do not care intensely about the candidates, we find that matched Democrats or Republicans assigned to work for their preferred candidate have a 27% increase in productivity but mismatched workers assigned to work for the other candidate work 43% less. We also find, however, that workers who were assigned to work for their opposition could be “bought” – performance-based incentives for these workers greatly mitigate the productivity gap. This work has implications for how organizations define their mission, screen workers, and provide compensation. JEL Classification: C91, J22, J33, M52