Overconfidence due to preference for control

A large body of research suggests that people tend to be overconfident in their own abilities in a variety of domains. At the same time, people are reluctant to delegate decisions even when they understand that it would monetarily benefit them (Botti et al., 2004; Botti and McGill, 2006; Botti et al., 2009) and they take over tasks from others (Verrey, 2019) that then lead to suboptimal group outcomes. We propose that beliefs about own and others’ performance are motivated by intrinsic preferences for control (Owens et al., 2014). Specifically, in contexts where one can decide whether to rely on own performance or the performance of someone else, people want to choose the better performance and self-rely at the same time. When learning about potential performances, one could engage in motivated reasoning to get to the desired conclusion, with high certainty, that self-reliance is indeed the money maximizing choice. In an online experiment we test whether the described belief mechanism is indeed present. We let people work on a real-effort task with a piece rate scheme. To cleanly identify the relationship between delegation and beliefs about one’s own and others’ performance, agents don’t work directly for principals. Instead, we use their piece rate performance for later sessions. Additionally, principals only learn about the option to delegate after completing the task themselves. Therefore, delegation simply translates to being paid after another participant’s prior performance. Thus, the only instrumental information for the delegation decision is principals’ beliefs about their own and the agent’s prior performance. The treatment exogenously varies whether the principals know about the opportunity to delegate before or only after receiving noisy information about the agents’ prior performance. The rationale behind this manipulation is that when the principal is aware of the opportunity to delegate, he can engage in motivated reasoning (Kunda, 1990) when processing the noisy information about the agents’ performance. If his desired outcome is not to delegate, he potentially inflates his beliefs about his own and deflates his beliefs about the agent’s performance. That is, we predict that principals are relatively more optimistic about their own performance — compared to the agent’s performance — when they know about the opportunity to delegate ahead of time, at the same time, they are more reluctant to delegate. Primary research questions: • Are people more likely to end up with a posterior, believing they are better than their counterpart, when there is more opportunity to engage in motivated reasoning? • Are people more reluctant to delegate when there is more opportunity to engage in motivated reasoning? Auxiliary questions: • When learning about the performance of others, does the opportunity to engage in motivated reasoning make people revise their beliefs about their own performance more frequently? • Can payoff considerations fully explain delegation decisions?

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.