A Structural Model of Advertising Signaling and Social Learning: The Case of the Motion Picture Industry

When new products are introduced to the market, normally, there exists information asymmetry between consumers and firms. Being uncertain about new products' attributes, consumers are motivated to learn from both from consumer-generated information (e.g. the Word-of-Mouth (WOM) communication among consumers) and firm-generated information (e.g. advertising). When deciding optimal advertising strategies, firms have to be aware how advertising interacts with consumer learning to create an asymmetry in the returns for products with different quality levels.
This dissertation analyzes how social learning among consumers shapes the optimal strategies of firms in the motion picture industry for signaling product quality through advertising. I analyze the distribution of advertising spending over time with a structural equilibrium model that incorporates both pre-release information asymmetry and post-release consumer learning. Studios need to decide the pre-release advertising and post-release advertising spending, knowing that the pre-release advertising plays a dual role of informing consumers about a new movie as well as signaling the movie's unobservable quality. Consumers who are reached by advertising and then enter the market at different time have different information sources. Those who enter the market in the opening week, use advertising as the main information source to infer a movie's quality. While those who enter the market in post-release weeks enjoy the extra benefit of information from WOM.
I estimate the model using weekly data on advertising spending, box office performance and movie characteristics from movie theater admissions in the United States. By estimating studios' equilibrium advertising policy function, I demonstrate that advertising does play a quality signaling role as well as reaching consumers in the movie industry. I also evaluate the pre-release information uncertainty for firms and consumers, respectively, and how the information asymmetry is reduced separately by the signaling effect of advertising and by Word-of-Mouth.
Since advertising can be used to reach consumers as well as to signal product quality, I use counterfactual experiments to distinguish the amount of money that is used for purposes of signaling as well as reaching. Counterfactual experiments suggest that around 27% of advertising spending on the movies in my sample is for a signaling purpose, while 73% of advertising money is spent to reach consumers. When signaling movie quality through advertising, studios with high-quality movies tend to spend more and studios with low-quality movies tend to spend less in pre-release weeks than the case when advertising is only used to reach consumers. Information revealed by both advertising signaling and social learning even prevents movies with very low quality from entering the market. When word-of-mouth communication has lower cost and become more efficient, less advertising spending is required to be "burned" for signaling purposes.