In many countries, insurance companies must not discriminate on some given policyholder characteristics. Omission of prohibited variables from models prevents direct discrimination, but fails to address proxy discrimination. To this end, multiple fairness methodologies exist but lead to different results. In this talk, we review causal inference notions and introduce a causal graph tailored for fairness in insurance. Exploiting these, we discuss potential sources of bias, formally define direct and indirect discrimination, and study the properties of fairness methodologies. A novel categorization of fair methodologies into five families (best-estimate, unaware, aware, hyperaware, and corrective) is constructed based on their expected fairness properties. A pedagogical example illustrates our findings on the interplay between our fair score families and sources of discrimination.