Consumers, experts, and online product evaluations: Evidence from the brewing industry

The growth of the Internet has led to a dramatic increase in the number of consumer or “user” product ratings, which are posted online by individuals who have consumed a good, and are available to other individuals as they make decisions about which products to purchase. These ratings have the potential to substantially improve the match between products and consumers, however the extent to which they do so likely depends on whether the ratings reflect actual consumer experiences. This paper evaluates one potential source of bias in consumer ratings: mimicry of the reviews of experts. Using a rich dataset on consumer product ratings from the brewing industry and a regression discontinuity empirical framework, I show that expert reviews influence consumer ratings. Consumer ratings fall in response to negative expert reviews and increase in response to positive expert reviews. The results are most pronounced for strongly negative or strongly positive expert reviews. This mimicry limits the extent to which information on product quality from actual consumer experiences diffuses to the population. I suggest that “nudges” could be implemented to limit the extent to which mimicry affects ratings.