L'utile et le juste de la discrimination dans la sélection, la classification et la tarification des risques assuranciels

This thesis addresses the complex issue of risk classification in the field of insurance. Prior to accepting risks, insurance companies must first be able to evaluate those risks. Accordingly, they seek to collect the most information possible from, amongst other sources, the insured, so as to gage relative risk and evaluate whether to insure or not, to what degree and at what rate. In due course, the insurer will use this information on conjunction with statistical and actuarial calculations to draw hypotheses on the degree, probability and cost of risk. In selecting relevant risks for analysis, insurers will utilise set variables based on the area of insurance in which they operate. However, said variables are highly susceptible to being discriminatory. Notably, one thinks of sex and age which are contentiously considered by practitioners and scholars whether or not they operate in the field of insurance. This dissertation will examine exhaustively the normative framework in place in order to determine to what degree, if indeed at all, insurers can legitimately and legally utilize certain classifications such as age and sex in order to select, categorise and fix the price for the various risks offered to them. The question shall arise, to what degree less or all-together non-discriminatory criteria should be favoured over criteria, sometimes considered, prohibited. In order to answer all these questions to better address the issue, we must first examine certain essential notions of insurance. Thus, in the first section, we will describe the relevant logistic practices in insurance industries. We shall focus on the decision process at various levels where potential discriminatory practices may arise. We will see that certain schools of thought on insurance classification are at odds, some times diametrically. We will, incidentally, favour the 'fair discrimination' doctrine over its traditional theoretical rival: 'anti-discrimination'. Our research shows that potentially discriminatory classification occurs at several stages of the ex ante and ex post contractual relationship, stages we will examine one at a time. In the second portion we will cover the general juridical regime of the right to non-discrimination in contracts at the international, national and provincial levels. Special attention will be paid to specific rules which allow some limited derogation to the constitutional rights against discrimination. We shall highlight that the legislative authority granted by the Quebec Charter does have limitations. What's more, certain guidelines recently established by the Supreme Court of Canada regarding application, must take precedence over various classification criteria pertaining to insurance which find their root in article 20.1 of the Quebec Charter. Ultimately, we will concentrate on what is just, which is to say the legitimacy of discrimination in a field that takes it for granted while seldomely questioning its foundations. We will come to apply a new measure for insurance discrimination. We will test this new measure in two specific fields: life insurance and automobile insurance. Overall, this thesis will allow us to determine how discriminatory classification can, at times, be legally employed (mostly in pre-selection and segmentation) in the above mentioned fields. We will conclude by proposing a new operating model which seeks to limit classification procedures that circumvent rights to privacy and non-discrimination.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.