Cheap talk efficacy under potential and actual Hypothetical Bias: A meta-analysis

Abstract

Cheap Talk (CT) is a common technique employed in stated preference methods of nonmarket valuation to reduce Hypothetical Bias (HB). However, multiple studies have documented its mixed usefulness. Using meta-analysis of 67 studies identified after reviewing over 400 articles, we first show the efficacy of CT while controlling for publication bias. We then investigate when CT is likely to be effective. Our results indicate that on average CT is significant in reducing estimated economic values by about 20% compared to the baseline treatment without implementing CT. Further analysis demonstrates that using CT in the context of public goods, using a budget/substitute reminder, and using CT in conjunction with other HB mitigation strategies improves its efficacy. In addition, we show that CT works better when the actual HB is larger thus one potential reason for previous studies failing to identify positive reduction in potential HB through CT may be because actual HB is small instead of CT being ineffective.