Opinion Dynamics in Social Networks and Cryptocurrency Prices

Abstract

I propose a model of the cryptocurrency market wherein a set of heterogeneous investors with different methods of communication rely on a social network to speculate and update their beliefs about the fundamental value of a cryptocurrency. This model indicates that the beliefs of stubborn investors, including both rational learners who use all available information and partisans who hold extreme views, prevail over other views. These social dynamics can cause persistent disagreement among investors, which is reflected in cryptocurrency prices. Using the Bitcoin market and data from Twitter posts between 2019 and 2021, I test the model’s predictions about price and find supportive evidence. Empirical analysis suggests that Bitcoin prices 1) are positively related to its fundamental characteristics, such as its blockchain-network size, 2) have a direct association with investor beliefs and disagreements, and 3) are influenced more by investor beliefs and disagreements than by its fundamental characteristics.