Do not follow this hidden link or you will be blocked from this website !

Why Is the Diversifiable Risk Still Priced?

Philippe Trainar * Professeur titulaire de chaire assurance, CNAM (Centre national des arts et métiers) ; directeur, Fondation SCOR pour la Science ; membre, Conseil scientifique, ACPR (Autorité de contrôle prudentiel et de résolution). Contact : L'auteur remercie particulièrement Vivien Levy-Garboua et Patrick Thourot pour leur lecture attentive de la première version de l'article et pour leurs remarques critiques et leurs suggestions qui ont permis de bonifier ce travail dont l'auteur reste cependant seul responsable des imperfections et des erreurs.

Economic theory teaches that the price of diversifiable risk is zero. However, this is not the case in practice: the price of the best diversifiable risks covered by insurers in the context of a competitive market, such as auto insurance, does not tend towards zero and even not tend to decrease in the long run. This paradox can be explained by the combination of many factors that most often have no value in themselves and must be combined to arrive at a more or less accurate picture of reality. The purpose of this article is to provide food for thought to better understand the determinants of the actual market situation.

In the first part, we will come back to the notion of risk, diversifiable risk and diversifiable risk prices, in order to make sure of what is really supposed to tend towards zero. In a second part, we will examine the different causes of this deviation and seek to validate some of them, without however resorting to the rigor of the quantitative analysis which exceeds the limits of this article. In the third part, we will ask ourselves why we do not observe, at the very least, a downward trend in the price of diversifiable risk. In the fourth part we will examine the ambivalent role of prudential regulation of insurance in this respect, the term “ambivalent” being strictly factual, without any negative connotations.

Download PDF Format


1   Pour une analyse synthétique du sujet, voir Eeckhoudt et al. (2005).

Share email Share on Facebook Share on Twitter Share on Google+