First published in ITIJ 98, March 2009
New and proposed anti-discrimination laws could affect the way UK travel insurers write business. But Katie Tucker from Pinsent Masons LLP asks whether insurers really are being ageist and sexist in the way they assess risk and to what extent differential treatment can still be justified
Prior to the introduction of the Gender Directive, the UK already prohibited insurers from discriminating on the grounds of gender under the Sex Discrimination Act 1975.
Under the Act, UK insurers are deemed to discriminate directly against a person if they treat them differently on the grounds of their gender. This might include refusing to provide cover, charging different premiums or providing different policy benefits. Indirect discrimination – where there appears to be equal treatment (because requirements or conditions apply equally to men and women), but that treatment adversely affects more women than men (or vice versa) – is also prohibited, unless it can be justified as a proportionate means of achieving a legitimate aim.
However, recognising that risks vary between the sexes, the Act provided an exemption to this general prohibition for insurers so that differential treatment would not be unlawful if it was based on actuarial or other data.
The Gender Directive follows a similar pattern to the Act. It is unlawful to discriminate on the grounds of a person's gender when providing insurance services, subject to a specific exemption for insurers. But it differs from the Act in two key respects: the exemption, while alike, is more stringent, and there is an absolute prohibition on less favourable treatment of women for reasons of pregnancy and maternity.
To implement these changes, the Act has been amended by the Sex Discrimination Act 1975 (Amendment) Regulations 2008, which came into force on 6 April 2008. All insurance contracts entered into on or after that date must comply with the revised position. The pregnancy and maternity provisions were deferred and came into force on 22 December 2008.
it is now explicitly unlawful to treat a woman less favourably on the grounds of her pregnancy or maternity
It is not easy for insurers to apply the revised Act to their policies and procedures. This is new legislation, amending relatively untested existing legislation, and the wording is ambiguous in places. Ultimately, it will be for the English court to determine the correct interpretation. This is further complicated because the UK legislation derives from EC law, and therefore any interpretation of it should take into account the purpose of the Gender Directive.
Insurers may therefore want to take a cautious approach in assessing the Act and deciding how to apply it. They may also want to take into account that the discrimination burden of proof rules make it easier for a claimant to show an insurer has unlawfully discriminated against them and that equality bodies, such as the Equality and Human Rights Commission, may back an individual making a claim. The Commission has the power to assist complainants and could take on test cases in order to clarify the law.
The Insurer Exemption
Under the revised Act, UK insurers may continue to use gender as a factor in assessing a risk if: they not only have relevant and accurate actuarial and statistical data to base this upon, but also compile and publish the data in accordance with guidance issued by the Treasury.
Any differences in treatment must also be proportionate, so there is a close relationship between the underlying data and the assessment of risks reflected in premium and benefits. While the Treasury guidance envisages some tolerance around the degree to which differential treatment may be held to be proportionate, proportionality is likely to be a difficult requirement for insurers to get right; particularly given it is a subjective concept.
The Treasury guidance sets out fairly prescriptive provisions regarding how the data should be published, including minimum requirements for content and form, who vouches for the data and how regularly it must be reviewed and updated.
In most cases, published data will be in the form of a table or chart, which illustrates the differences in risk between men and women. It is likely to be aggregated and simplified to ensure it can be easily understood by consumers and avoids releasing information that is price sensitive or may facilitate anti-competitive practices. It does not need to be the detailed underwriting data relied upon by the insurer, although this should underlie the published data and be used in underwriting the risk.
indirect discrimination … is also prohibited unless it can be justified as a proportionate means of achieving a legitimate aim
One of the main rationales for requiring that data is published seems to be transparency. Customers should be able to understand the relationship between their gender and the premiums and benefits offered to them. Specifically, the Treasury guidance requires the published data to be in plain English and intelligible to someone who is not an insurance expert.
But, in reality, is this aim achieved? Actuarial techniques are complex and insurers will take other factors affecting the customer’s risk assessment into account. And, as the published data will be simplified and may represent an industry average, it is unlikely to reflect variations in an individual insurer’s own data. The guidance even recognises that the published data is unlikely to present a direct correlation with the premiums charged or the benefits obtained in individual cases.
Consequently, while the published data may demonstrate the case for differing treatment, it seems unlikely that the average policyholder will understand from it the effect their gender has on the premium and benefits offered to them and whether these are proportionate to the underlying data.
Pregnancy and maternity
The most controversial provision from the UK insurer’s point of view is the prohibition on discrimination for reasons of pregnancy and maternity.
Under the Act it is now explicitly unlawful to treat a woman less favourably on the grounds of her pregnancy or maternity. The Gender Directive does not define ‘maternity’, so the UK government has adopted a period of 26 weeks from the day on which the woman gives birth. After the maternity period, women can still pursue a claim for discrimination on the grounds of gender in the normal way.
This is an area that is likely to have a major impact on travel insurers. Travel policies often exclude benefits (such as cover for cancellation, curtailment and medical expenses) after a specified number of weeks of pregnancy or exclude cover for medical complications related to previous pregnancies. Or, where such exclusions do not apply, load the premium to cover the risk.
Such policies and procedures need to be carefully considered to determine whether they are lawful: is the woman being treated less favourably and, if so, is the treatment because of her pregnancy or maternity? Does this mean insurers will be required to pay for medical costs arising from childbirth abroad? Is there a risk that women travel in order to obtain the benefit of such cover?
When carrying out this exercise, it is worth bearing in mind that the purpose of the pregnancy and maternity provisions in the Gender Directive are to ensure the costs of pregnancy or maternity do not fall solely on one gender. As commented earlier, the purpose of the Gender Directive should be taken into account in interpreting the Act.
There is an exemption to this ban, which applies where there could be a risk to a pregnant woman's health or safety. This was introduced to address the aviation industry’s concerns that airlines’ policies of refusing to carry late-term pregnant women for health and safety reasons would breach the Act. It is unlikely insurers could rely on the exemption, but it may provide travel insurers some comfort as it should limit the activities carried out by pregnant women and therefore the risks associated with them.
it could become unlawful to refuse to provide travel insurance to certain age groups
It is also unlikely that the insurance exemption can be relied upon. The Act specifically prevents the exemption applying if differences in treatment result from costs related to pregnancy or maternity. It is difficult to see how varying premium and benefits to cover the increased risk of a claim would not be related to cost.
The European Commission has recently adopted a proposal for a Directive aimed at protecting against discrimination outside the workplace. This includes a prohibition of discrimination on the grounds of age in the provision of insurance services. At the same time, the UK government has proposed a similar prohibition under the Single Equality Bill.
Accordingly, it could become unlawful to refuse to provide travel insurance to certain age groups or provide differential premium or benefits due to a person's age.
It is likely, though, that insurers will have the benefit of an exemption along similar lines to that under the Gender Directive. The draft Directive states that, in the provision of insurance services, Member States may permit proportional differences in treatment where the use of age is a key factor in the assessment of risk based on relevant and accurate actuarial or statistical data. And the UK government has said the Bill will not prevent ‘actuarially justified age-based treatment’.
Both proposals are at a relatively early stage and, until the wording is determined, it is difficult to judge their impact. However, the industry has already expressed concerns that, if it is introduced, insurers will be forced to withdraw certain products, thereby reducing competition and increasing prices.
The Institute of Actuaries has said that many insurers do not have sufficient data to show the statistical effect of age on claims experience – either because they do not insure certain age groups or, due to the lack of demand, they rarely do. To illustrate this, they highlight that travel insurers are likely to have plentiful statistics on 40-year-olds taking skiing holidays, but very little, if any, data on 80-year-olds taking skiing holidays.
Given that there are specific insurance products that target and specialise in certain age groups, some insurers have suggested that a more proportionate response to ensure accessibility would be to concentrate on signposting those products, rather than imposing mandatory universal access to all products.
A change in approach?
The Gender Directive has required two key changes in the approach taken by UK insurers. First, the revised insurance exemption must be complied with in order to use gender as a factor when underwriting the risk. Second, insurers cannot directly discriminate against a woman on the grounds of her pregnancy or maternity, even if this is actuarially justified.
In general, though, insurers will be able to underwrite risks much as they previously have done. Their position is also likely to be protected where age is used as a risk factor. Please note that it is not certain that this position will remain in relation to gender discrimination, as the UK government is required to review its decision to allow the insurer exemption at the end of 2012.
many insurers do not have sufficient data to show the statistical effect of age on claims experience
Problems arise where an insurer does not hold the necessary data in order to justify differential treatment and comply with the exemption. In order to lawfully provide their products, insurers may be required to move away from assessing risks based on assumptions related to applicants’ gender or age and focus more on the risk the individual poses.
While this approach arguably represents a fairer way of providing insurance and will result in benefits for some customers, ultimately it could result in detriment for others. If insurers carry out detailed individual assessments, they will probably need to dedicate greater resources to the underwriting process, thereby increasing costs, which are likely to be passed on to all policyholders.