Data Sharing and Website Competition: The Role of Dark Patterns

Abstract: 

Regulations like the GDPR require firms to obtain consumer consent before using data. In response, some firms employ “dark patterns” — interface designs that nudge consumers to share data. We study the causal effects of these designs and how they vary across individuals and firms. To do so, we run a field experiment in which users download a browser extension that randomizes cookie consent interface designs as users browse the Internet. We find that consumers accept all cookies more than half of the time in the absence of dark patterns. Hiding consent options behind an additional click is the most effective dark pattern, while designs that only manipulate visual elements have smaller effects. Larger and better-known firms have moderately higher consent rates than other firms, giving them a slight competitive advantage. However, the effects of dark patterns do not vary systematically across site popularity. We find no evidence that more frequent pop-ups result in choice fatigue.

Biography: 

Professor Fradkin is an assistant professor of marketing and economics (by courtesy) at the Boston University Questrom School of Business. His research has been published in a variety of fields including marketing, economics, management, and computer science. He has provided expert input about the digital economy at the President’s Council on Science and Technology and the Federal Trade Commission. Prior to BU, he was a postdoc at the Initiative on the Digital Economy at MIT, worked as a data scientist at Airbnb, and completed his Ph.D. in Economics at Stanford University.