3.3 Exemptions and exceptions (a) Insurance exemptions There is recognition in the Discrimination Acts that some discrimination by insurers may be necessary. The ADA, DDA and SDA provide exemptions which mean that discrimination relating to insurance services may not be unlawful in some circumstances. No such exemption applies under the RDA. ADA and DDA exemptions The ADA39 and DDA40 provide that discrimination on the basis of age or disability in relation to provision of insurance by either refusing to offer a product, or in respect of the terms or conditions on which the product is offered or may be obtained, is not unlawful if the discrimination: • is based upon actuarial or statistical data on which it is reasonable to rely, and the discrimination is reasonable having regard to the matter of the data and other relevant factors (the ‘data exemption’) or • in a case where no such actuarial or statistical data is available and cannot reasonably be obtained — the discrimination is reasonable having regard to any other relevant factors (the ‘no data exemption’). An insurer who relies on the exemption must be able to show that the requirements of either the data exemption or no data exemption are met. The data exemption An insurer who relies on the data exemption must be able to show that the data was available and relied upon at the time the decision was made. Insurers should keep accurate records of data relied upon. If a complaint of discrimination is made to the Commission, an insurer may be required to disclose the source of the data as part of the Commission’s conciliation process.41 The data may also need to be produced if the discrimination complaint proceeded to court. Insurers should make ‘an objective judgment about the nature and quality of the actuarial or statistical data’.42 It may not be reasonable to rely on data that is out-of-date, qualified, incomplete, discredited, based on an insufficient sample size, or not directly applicable to the particular situation.43 Data that may be reasonable to rely upon includes: • underwriting manuals with information about the nature and degree of extra risk of insuring particular groups (provided the data used to populate them is complete and up to date) • local data, such as government studies, census statistics, studies reported in medical journals, and insurance studies with data from a reliable source44 • international studies, particularly if local data is insufficient or it can be shown that the international data remains reasonably applicable in Australia.45 Guidance Resource: Artificial intelligence and discrimination in insurance pricing and underwriting • 2022 • 15