34 Pages Posted: 22 May 2019 Last revised: 24 Jun 2020
Date Written: March 26, 2020
Firms and politicians, among others, invest heavily to influence people's opinions. Because peers influence one another, these efforts must account for social networks. Using a model of opinion dynamics with a non-degenerate steady state, I highlight how opinion fluctuations impact influence campaigns. If agents interact less frequently with those holding different opinions, the variance of public opinion decreases. This generates distinctive predictions in applications. I argue that market segmentation based on social networks can create echo chamber effects that make demand more predictable, and I provide a new rationale for electoral platforms that diverge from the median voter's preferred policy.
Keywords: Networks, Influence Campaigns, Voter Model, Stochastic Block Model
JEL Classification: D85
Suggested Citation: Suggested Citation