The effect of ambiguity in strategic environments: an experiment

TL;DR

This paper experimentally investigates how ambiguity affects group contributions in a strategic game, finding no overall effect of ambiguity but highlighting the role of individual risk attitudes.

Contribution

It provides empirical evidence that ambiguity does not influence contributions in a large sample, challenging theoretical predictions and informing policy on ambiguous issues.

Findings

No significant average treatment effect of ambiguity on contributions

Risk attitudes significantly increase contributions under ambiguity

Theoretical predictions of higher contributions under ambiguity are not supported empirically

Abstract

We experimentally study a game in which success requires a sufficient total contribution by members of a group. There are significant uncertainties surrounding the chance and the total effort required for success. A theoretical model with max-min preferences towards ambiguity predicts higher contributions under ambiguity than under risk. However, in a large representative sample of the Spanish population (1,500 participants) we find that the ATE of ambiguity on contributions is zero. The main significant interaction with the personal characteristics of the participants is with risk attitudes, and it increases contributions. This suggests that policymakers concerned with ambiguous problems (like climate change) do not need to worry excessively about ambiguity.