ABSTRACT: We use two experiments to test predictions about the positive and negative impacts of allowing analysts to revise their forecasts in light of the consensus forecasts. We find that such mutual observation not only facilitates information aggregation, but also induces free riding, which offsets the benefits of information aggregation unless incentives for accuracy are high. In our second experiment, we find that participants acting as investors anticipate these effects in the consensus and adjust their own forecasts accordingly. Our study demonstrates that the positive and negative effects of mutual observation are more complex than typically portrayed in the debate about analyst independence, and provides a framework that can be used in future research on the interactive nature of public forecasting.
