The Blues Goes On But When Does It Stop? Public Goods Experiments with Non-Definite and Non-Commonly Known Time Horizons

A robust finding of repeated public goods experiments is that high initial contribution rates sharply decline towards the end. This paper reports on an exploratory experiment designed to discover whether such a decline is simply triggered by the usual experimental practice of publicly informing participants about the exact number of periods to be played. The experiment compares punctual to interval information about the number of repetitions, whereby interval information can be privately or commonly known as well as symmetric or asymmetric. The results indicate that, while the overall average contribution levels do not change significantly across treatments, asymmetric information about the time horizon reduces the frequency of end-game effects.