The Impact of Online Product Reviews on Retailer's Pricing and Return Policy Decisions

After a purchase, customers increasingly feel more comfortable sharing their experience in the form of online product reviews, which help resolve future buyers' valuation uncertainty for that product. On the other hand, retailers offer lenient return policies that also help with customers' ex-ante valuation uncertainty. We consider a monopolist retailer that sells a new experience good over two periods and makes pricing and refund decisions. Because returns for these types of products are generally associated with lack of or inadequate functionality, a customer review mentioning a return experience generally creates a negative sentiment. Consequently, in this environment, a higher refund not only increases the volume of returns which need to be resold at a discounted price, but also creates more reviews that will have a negative sentiment. Similarly, a lower refund (and the associated lower price) would decrease the return volume and increase the fraction of customers writing positive reviews. Using a framework in which the negative reviews are weighed more heavily compared to the positive ones, surprisingly, we show that the retailer offers even a more generous refund compared to the benchmark setting with no reviews. In a duopolistic competition, the overall common sentiment of the on-line reviews are influenced by the refund and pricing decisions of both retailers; in other words, a customer with the same valuation purchasing the same product at different price and refund values will have a different ex-post utility and experience. Interestingly, we show that the retailers make their returns even more lenient compared to not only the benchmark setting with no reviews, but also the monopolist retailer with the same salvage value when their decision collectively determine the review sentiment. In both settings, we show that the refund increases with salvage value, valuation heterogeneity, second-period demand and decreases with first-period demand. We show that our results remain robust when i) consumers incur hassle cost ii) consumers have ex-ante valuation heterogeneity iii) the second-period valuation uncertainty is partially resolved with reviews.