Insurance, Big Data and Changing Conceptions of Fairness

Abstract

This paper aims to show how insurance mechanisms that historically propelled a conception of fairness based on solidarity and a collective approach shifted along the 20th century towards an idealistic adjustment to individual risk. Insurance originally assumed that, while individual hazards remained unknown, risk could be measured and managed on the aggregate. An examination of the proceedings of the American Casualty Actuarial Society (CAS) during the 20th century demonstrates the slow crystallization of another conception of fairness, that aims at a scientific adjustment of insurance premiums to actual individual risks. I argue that this conception of fairness deconstructs the one based on solidarity. Big data technologies have further radicalized this shift. By aiming at predictive individual risk scores rather than average costs estimated on the aggregate, the algorithms contribute to replacing fairness as solidarity by the correctness of a computation.