Bad News Turned Good: Reversal under Censorship

Abstract

Sellers often have the power to censor the reviews of their products. We explore the effect of these censorship policies in markets where some consumers are unaware of possible censorship. We find that if the share of such “naïve” consumers is not too large, then rational consumers treat any bad review that is revealed in equilibrium as good news about product quality. This makes bad reviews worth revealing and allows the seller to use them to signal his product’s quality to rational consumers. (JEL D82, D83, L15)

A dynamic model of censorship

“…We also differ from Shadmehr and Bernhardt (2015) in that we emphasize the welfare effect on the evaluator, whereas they, like other works on political censorship, 4 do not provide a framework to study citizens' welfare or presume censorship affects them negatively. Smirnov and Starkov (2022) study censorship on product reviews, but consider costless censorship and assume a seller faces both Bayesian and naive buyers. The model and results are quite different from ours.…”

“…Smirnov and Starkov (2022) study censorship on product reviews, but consider costless censorship and assume a seller faces both Bayesian and naive buyers. The model and results are quite different from ours.…”

Influencer Cartels

“…We study influencer cartels where groups of influencers collude to increase each others' indicators of social media influence. While there is substantial literature in economics on fake consumer reviews (Mayzlin et al, 2014;Luca and Zervas, 2016;He et al, 2022;Glazer et al, 2021;Smirnov and Starkov, 2022) and other forms of advertising fraud (Zinman and Zitzewitz, 2016;Rhodes and Wilson, 2018), the economics of this fraudulent behavior has not been studied.…”

A question about the existence of Russian economic thought

“…See, e.g.,Luca and Zervas [2016] for an exploration of the effects of fake reviews andSmirnov and Starkov [2022] for a model of censorship in product reviews.2 The other common response is "to help or punish the seller after a good or a bad experience, respectively".3 The point that product evaluations produce a positive externality and are hence socially underprovided was made, among others, byAvery, Resnick, and Zeckhauser [1999], who proposed cash payments to alleviate this inefficiency.4 Unlike most papers on experimentation, we do not restrict the model to an exponential bandits setting with binary utility outcomes, allowing instead for a rich heterogeneity in utility realizations.…”