Internet-based distribution technologies have dramatically reduced storage costs of many entertainment products allowing digital retailers and distribution platforms to make a vast number of titles available to consumers.

Large content catalogues create an interesting trade-off as the same firm often offers old and new content simultaneously. By introducing new products to its catalogues, in addition to the fixed costs of adding new content, the firm may cannibalize sales of existing products. This can potentially reduce incentives to innovate, leading to less product variety on the market.

Additionally, large content catalogues commonly combine both High Definition (HD) and Standard Definition (SD) versions of the same movies. Although all consumers prefer the HD version, they differ in their willingness-to-pay for HD movies. Providers may find it profitable to offer SD versions, so that they can segment consumers and extract a higher surplus from them. Whether or not this is an optimal strategy for the provider depends on the distribution of consumers’ willingness-to-pay for HD.

A digital retailer’s optimal content strategy will depend on the expected revenue generated by the entire catalogue, which is often a complex function of the business model. To advance our knowledge of how VoD and SVoD providers can better manage their content catalogues, we plan to address the following research questions:

• 1. How to estimate the expected revenue of new content?
• 2. How to estimate the expected revenue of incumbent content once new content is introduced?

In order to address these research questions, we will use data from a randomized field experiment that will be run by our IP. The experiment will disproportionally advertise old movies to a random set of consumers. This variation will be used to analyze how sales of old content react to the introduction of new content. We will also investigate whether incorporating aggregate or individual-level viewing data can help predict VoD demand.

• 3. Is it profitable for a VoD provider to engage in second-degree price discrimination by offering HD and SD versions of the same content?
• What is the impact of second-degree price discrimination on consumer’s welfare?

In order to address these research questions, we will use observational data on movie-rentals from our IP’s VoD service. This data will be used to estimate consumers’ willingness to pay for HD movies and investigate whether it is optimal for the firm to offer HD and SD movies, the optimal price premium for HD movies, and how it varies according to movies’ characteristics.