Give Us a Little Social Credit: To Design or to Discover Personal Ratings in the Era of Big Data

Abstract

In 2014, China announced the institution of a social credit system by 2020. Social credit ratings of the type being developed by China go beyond existing financial credit ratings in an attempt to project less-tangible personal characteristics like trustworthiness, criminal tendencies, and group loyalty onto a single scale. The advent of Big Data — characterized by a large and increasing volume of personal data and tools like machine learning to detect patterns and generate predictions based on that data — strongly indicates that various kinds of social credit ratings will become a reality in the near future. Supposing that the emergence of Big Data-enabled personal ratings is both a cost-saving adaptation to and general improvement upon traditional forms of signaling trustworthiness, we use both traditional modeling techniques and evidence-based argument to determine whether “optimal” social credit should develop publicly, privately, or in a polycentric fashion.

Keywords: social credit, personal ratings, big data, trustworthiness, public goods, institutions

JEL Classification: D02, E02, O33, O35, P21, P50