The Editor vs. the Algorithm: Targeting, Data and Externalities in Online News

Abstract

We run a field experiment with a major news outlet to quantify the economic returns to data and informational externalities associated with algorithmic recommendation. Automated recommendation can outperform a human editor in terms of user engagement, though this crucially depends on the amount of training data. Limited individual data or breaking news leads the editor to outperform the algorithm. Additional data helps algorithmic performance but decreasing economic returns set in rapidly. Investigating informational externalities highlights that personalized recommendation reduces consumption diversity. Moreover, users associated with lower levels of digital literacy and more extreme political views engage more with algorithmic recommendations.

Keywords: Field experiment; Economics of AI; Returns to data; Filter Bubbles