Research

Many platform strategies focus on indirect network effects between sellers through platform expansion. In this paper, we show sellers on the console video game platform generate a positive intertemporal spillover effect and expand the demand for other sellers, holding the set of platform adopters fixed. We propose a novel identification strategy that leverages exogenous variation in the release timing of games exclusively available on a console platform, and examine how this variation affects the sales of games available on both platforms. We find a sizable intertemporal demand spillover effect between games: A 1% increase in total copies sold on a platform leads to a 0.153% increase in the sales of other games in the next month (i.e., an elasticity of 0.153). Additional analysis suggests this demand spillover effect is reminiscent of habit formation on the consumer side, in that past purchases keep end users active on the platform. Our finding provides a potential explanation for recent platform sales events and subscription services that provide free games to consumers every month.

Influencers promote firms’ products by posting content such as videos on social media platforms. For entertainment products, these posts could substitute or complement demand for the original entertainment product. We study video games, the largest entertainment product category comprising 1/3 of YouTube traffic, using a large daily panel data set on thousands of video games. Leveraging a supply shock on YouTube called the "Adpocalypse", we measure the impact of influencer videos on purchase and usage of games. We provide evidence that on average influencer video posts substitute to video games for purchases but complement for usage. We also find that influencer effects differ across firms. Managers can use these results to align the influencer effects they face with their revenue models, such as using in-game purchases or a subscription model when facing complements on usage.

From taking inspiration to outright copying, new apps are frequently developed using the ideas of successful competitors. Mobile app platforms (markets), such as the Apple App Store and the Google Play Store, have policies intended to screen these "copycat” apps. We study the effect these "copycat” apps have on incumbents by leveraging data from multiple Android app markets in China. We identify which apps copied others using a supervised machine-learning approach. Our identification strategy leverages the fact that we observe the same app competing against different competitors on different platforms. We find that on average copycats do not reduce the demand for incumbent apps. However, moderators of this effect show that the largest and most highly rated apps are harmed by the presence of copycats. Furthermore, copycats cause developers of incumbent apps to behave differently. Higher-quality incumbents change their descriptions to differentiate themselves from low-quality copycats. Our results show that the filtering of copycat apps does not confer significant benefits to app developers, which may inform recent discussions on the value that app platforms provide.