UC San Diego

This dissertation comprises two papers that shed light on how digital platforms are shaping consumer preferences and behavior.

Chapter 1 investigates a large card spending panel to understand how ChatGPT adopter changed their spending on other digital services. We employ Coarsened Exact Matching to balance early ChatGPT subscribers—those who began their subscriptions in distinct monthly cohorts from February to June—with those of later ChatGPT subscribers who joined between September and December. Then we use a triple-difference identification strategy to predict counterfactual digital service payments. We find an increase in consumers' spending on other AI products following the adoption of ChatGPT across five monthly adopter cohorts. The estimates rule out market share changes of 2\% for the majority of brands, with a few exceptions.

Chapter 2 examines online tipping in the gig economy, focusing on how digital platforms can boost tipping. We analyze 4.1 million transactions and find that a buyer's nationality and satisfaction levels are key tipping factors. We also run two field experiments to test if psychologically-based messages about norms and reciprocity could lead to more frequent tipping. The results show that informing buyers about social norms effectively increases tipping. Additionally, in a MTurk survey, we test various normative messages and find that descriptive norms, like highlighting high user tipping rates, effectively raise tipping intentions. This paper highlights the importance of understanding and shaping online tipping practices, not just for helping consumers navigate tipping decisions but also for improving gig workers’ financial stability.