Reduced form evidence on belief updating under asymmetric information—consumers’ response to wine expert opinions

We estimate the effect of quality labels on purchases through a retail field experiment. Utilising product-level panel scanner and product characteristic data for both labelled and unlabelled wines we estimate the average and heterogeneous effect on purchases. Consistent with earlier work, we find an average effect that is positively correlated with scores. We advance the consumer belief and product information literature on two fronts. First, higher scores matter more for lower priced products. Second, spillover effects impact sales of untreated wines; these effects can be positive or negative and are impacted by the average score and label converge within brand.