| Abstract: |
Although fake online customer reviews have become prevalent on platforms such
as Amazon, Google, and Facebook, little is known about how these reviews
influence consumer behavior. This paper provides the first experimental
estimates of the effects of fake reviews on individual demand and welfare. We
conduct an incentive-compatible online experiment with a nationally
representative sample of respondents from the United Kingdom (n = 10, 000).
Consumers are asked to choose a product category, browse a platform resembling
Amazon, and select one of five equally priced products. One of the products is
of inferior quality, one is of superior quality, and three are of average
quality. We randomly allocate participants to variants of the platform: five
treatment groups see positive fake reviews for an inferior product, and the
control group does not see fake reviews. Moreover, some participants are
randomly selected to receive an educational intervention that aims to mitigate
the potential effects of fake reviews. Our analysis of the experimental data
yields four findings. First, fake reviews make consumers more likely to choose
lower-quality products. Second, we estimate that welfare losses from such
reviews may be important—on the order of $.12 for each dollar spent in the
setting we study. Third, we find that fake reviews have heterogeneous effects.
For example, the effect of fake reviews is smaller for those who do not trust
customer reviews. Fake reviews also have larger effects on those who shop
online more frequently. Fourth, we show that the educational intervention
reduces the adverse welfare impact of fake reviews by 44%. |