On 20th January 2020 (1-2pm) Georg Weizsäcker, Professor for Microeconomic Theory and Applications at the Humboldt University of Berlin, will give a presentation about “Do we talk too much?” in Q4.245. Afterwards, Mr. Weizsäcker will be available for questions and discussions. His presentation is part of: https://wiwi.uni-paderborn.de/dep1/me/research/discussing-research/seam/
We consider the trade-off between talking and listening in a laboratory experiment that records a person's decision to claim a limited information channel for herself. Each of two team members receives a private signal about the state of the world and decides whether to "talk'' and share her own signal with her team mate or to "listen'' and obtain knowledge of the team mate's signal. The goal of the team effort is to correctly predict the state of the world in a subsequent guessing stage. Building on the literature on overconfidence, we hypothesize that a player's talking decision is negatively correlated with the task difficulty: the "hard-easy effect'' suggests that players tend to talk more in situations where information is relatively precise---but also for the listener. In our game, this intuition can be formalized through introducing overconfidence about one's signal prediction. Indeed, we find that a more precise information structure leads to a higher talking frequency, with a difference of about 5 percentage points, relative to a baseline of 48 percent. In treatments where a natural context for the conversations was used, the talking frequency is, likewise, significantly higher if the context is easier.
Foto: Matthias Heyde