Working Papers

• The Impact of Banning Online Gambling Livestreams: Evidence from Twitch.tv (with Jasmine Yang and Andrey Simonov). Major Revision at Marketing Science
  • Abstract
  • Can platforms curb harmful content by banning only part of it? We study the short-run effects of Twitch's October 2022 ban on livestreams of unlicensed gambling websites, using a novel panel of the top 6,000 streamers in which we detect banned content from video clips, stream titles, and in-stream chats. Difference-in-differences estimates show that streamers of banned content cut weekly gambling streaming hours by 65% and overall streaming by 45%. Streamers of unbanned content also reduced gambling and overall content, indicating spillovers to unbanned gambling content - a response concentrated among popular streamers, in line with reputational concerns. On the demand side, viewership, low-tier subscriptions, and engagement fell, while high-tier subscriptions remained unaffected. Gambling exposure also fell among viewers likely to be minors. A narrow ban thus reduced gambling content well beyond its formal scope, but at a cost in platform activity.
• Collaboration Among Content Creators (with Jasmine Yang and Kinshuk Jerath)
  • Abstract
  • We study content collaboration in the creator economy, in which competing creators mutually agree to collaborate on joint content and negotiate on content production and revenue sharing. Using a game theory model with creators competing for consumers on a Hotelling line, we show that collaboration allows creators to use the jointly-produced content to moderate competition, while using their individual content to expand into new audiences. This increases content diversity but also leads to increased monetizability of content. In general, collaboration among creators has an effect of increasing the profits of creators while reducing consumer surplus. When creators create content with heterogeneous entertainment values, the creator producing content of lower entertainment value has an incentive to free ride on the collaborative content. This free riding may increase surplus for consumers (who without collaboration would watch content of low entertainment value), thereby improving creators’ profits as well as consumer surplus. Our results provide guidance to content creators, to platforms designing tools to facilitate collaborations, and to policymakers.
• Impulse Response Inferences With Existence of Repeated Roots
  • Abstract
  • Vector Autoregression (VAR) and local projection (LP) are the two main methods of estimating and conducting inferences of the impulse response functions (IRFs) in macroeconomic studies, allowing researchers to choose between them based on the subjects of interest. This paper extends existing works on the comparison between AR inferences and LP inferences, by considering data generating processes with repeated roots. Consequently, the autoregressive estimation of impulse responses will converge to a special type of real-valued random variable, and the bootstrap Efron confidence interval of lag-augmented AR will always be conservative, even if the roots are away from the unit circle. This problem is more severe when the time series is highly persistent and at both intermediate and long horizons. The results are supported by Monte Carlo simulations with different values of roots in AR(2), AR(3) and VAR(1) models.

Work in Progress

• Discrete Choice Models with Incomplete Nest Structures
• Abstract
• Nested logit model is one of the most widely applied tools in discrete choice analysis, due to its ability to capture rich substitution patterns in market data. However, the researcher needs to define a nest structure ex ante, as the chosen nest is usually based on prior knowledge of the market and will be applied to all individuals. This paper discusses the identification of individual preferences, while relaxing this unique nest structure assumption and allowing for heterogeneity in individuals' recognition of the nest structures in the same market. I characterize the sharp identification region of parameters in the nested logit model, based on the coexistence of a given set of nest structures. I show in a series of Monte Carlo simulations that misspecification of the nest structure may result in the identification region not covering the true parameter, whereas the identification region solves the misspecification problem and partially identifies the parameters by allowing for multiple nest structures.
• Presentation at BU Econometrics Seminar, Spring 2023
• Optimal Achievement System Design on Video Game Platforms