I document that firms in online marketplaces use price promotions to facilitate transitions to better review tiers (similar to the number of stars on Amazon.com). I find that firms close to going one tier up are 4-9% more likely to discount. I theorize that two effects could be at play. First, a selection effect arises because customers who buy during a discount could be different from the regular ones, and could potentially leave more positive reviews. Second, a variance effect reflects the idea that positive reviews could help the firm move up a review tier, while negative reviews would keep the tier unchanged, minimizing the downside risk of giving a discount. To test my hypotheses, I estimate a simple structural model of demand and reviewing behavior, and develop a new approach to estimating demand from data on product usage. I do not find evidence that the selection effect is positive. I find evidence of the importance of the variance effect: when the selection effect is controlled for, firms close to downgrading their review tier are 6% less likely to give a discount, consistent with preferring less variance. I also find that consumers are significantly more likely to leave reviews during a discount. Additional findings include estimates of the causal effect of reviews on sales and equilibrium discount elasticities in an important market not previously studied.
For more information and to register for this event, please contact the organizers.
Organizers: Daniel Waldinger (firstname.lastname@example.org) and Sharon Traiberman (email@example.com).