Webinar Series

|

Innovation, Productivity and Challenges in the Digital Era: Asia and Beyond

|

Dec 2022

Crowdsourcing Peer Information to Change Spending Behavior

We isolate and quantify the information channel of peer effects using a unique consumption setting that by construction excludes any scope for common shocks or social pressure—a transaction-level panel dataset of spending paired with crowdsourced information about the spending of anonymous “peers” elicited at a different time than when users make their consumption choices. All consumers converge to their peers’ spending and more so when facing more informative peer signals, with the effect building up over time. The spending adjustments, though, are substantially larger for the overspenders, who close 37% of their spending gap within 12 months of using the platform. The effect for underspenders is 9% over 12 months. For identification, we exploit consumers’ quasi-random assignment to peer groups in an instrumental-variable strategy. Similar evidence from on a non-selected population provides external validity.
Keywords: Social Learning, Beliefs and Expectations, FinTech, Robo-advising, Social Finance, Information Econo
  • View
  • Download
  • Bookmark
  •    |