|
|
[Home]
[Databases]
[WorldLII]
[Search]
[Feedback]
Human Rights and Equal Opportunity Commission |
HUMAN RIGHTS AND EQUAL OPPORTUNITY COMMISSION
SEX DISCRIMINATION ACT 1984 (CTH)
Matter No. H97/237
Between:
D
Complainant
And
A Registered Life Insurer
Respondent
REASONS FOR DECISION
OF COMMISSIONER KATHLEEN
MCEVOY
Location of Hearing: Adelaide
Date of Hearing: 25, 26 and 27 March 1998
Appearances: Mr Jay Weatherill instructed by the Women’s Legal Service for the complainant.
Mr Richard Tracey QC for the respondent instructed by PW Turk and Associates.
Date of Decision: 11 February 2000
These are the reasons for a decision relating to an inquiry into a complaint by the complainant against the respondent, a registered life insurer.
In a letter dated 3 January 1997 the complainant made a complaint to the Equal Opportunity Commission of South Australia (“the South Australian Commission”) against the respondent alleging that she had been treated unfairly because of her sex and as a consequence had incurred detriment. The complaint was received by the South Australian Commission on 16 January 1997 as agent of the Human Rights and Equal Opportunity Commission (“the Commission”).
The complainant’s complaint was that she had been discriminated against on the ground of her sex in the provision of goods, services and facilities, namely the provision of an insurance policy, contrary to section 22 of the Sex Discrimination Act 1984 (Cth) (“the Act”). She complained that the discrimination took the form of a 24 per cent “loading” on her income protection policy which had been issued by the respondent. The respondent denied the allegation of discrimination and further relied on an exemption contained in section 41 of the Act.
The Equal Opportunity Commissioner of South Australia, as a delegate of the Sex Discrimination Commissioner, inquired into the complaint and then provided a copy of the complaint file to the acting Sex Discrimination Commissioner (“the Commissioner”) in April 1997. The Commissioner was of the opinion that the nature of the matter was such that it should be referred to the Commission and did refer it for an inquiry pursuant to section 57(1)(c) of the Act on 24 October 1997.
Pursuant to section 59(1) of the Act an inquiry was conducted by me in Adelaide on 25, 26 and 27 March 1998 when evidence was heard from the complainant and her witnesses and on behalf of the respondent. Subsequently, written submissions on behalf of both the complainant and the respondent relating to the law and the evidence were received by me on 2 July 1998. I take these written submissions into account along with the evidence presented at the Inquiry.
At the commencement of the inquiry, pursuant to section 67 of the Act, I made a direction that the names of both the complainant and the respondent be suppressed from publication and that direction continues.
The relevant legislative provision relating to the complaint is section 22 of the Act.
22(1) It is unlawful for a person who, whether for payment or not, provides goods or services or makes facilities available, to discriminate against another person on the ground of the other person’s sex, marital status, pregnancy or potential pregnancy –
(a) by refusing to provide the other person with goods or services or to make those facilities available to other persons;
(b) in the terms or conditions on which the first-mentioned person provides the other persons with those goods or services or makes those facilities available to the other person; or
(c) in the manner in which the first-mentioned person provides the other persons with those goods or services or makes those facilities available to the other person.
Section 4 of the Act includes within the definition of “services” “services relating to... insurance...”.
Section 5 of the Act defines sex discrimination as follows:
5(1) For the purposes of this Act, a person (in this sub-section referred to as the “discriminator”) discriminates against another person (in this sub-section referred to as the “aggrieved person”) on the ground of the sex of the aggrieved person if, by reason of –
(a) the sex of the aggrieved person;
- (b) a characteristic that appertains generally to persons of the sex of the aggrieved person; or
- (c) a characteristic that is generally imputed to persons of the sex of the aggrieved person,
the discriminator treats the aggrieved person less favourably than, in circumstances that are the same or are not materially different, the discriminator treats or would treat a person of the opposite sex.
5(2) For the purposes of this Act, a person (“discriminator”) discriminates against another person (“aggrieved person”) on the ground of the sex of the aggrieved person if the discriminator imposes, or proposes to impose, a condition, requirement or practice that has, or is likely to have the effect of disadvantaging persons of the same sex as the aggrieved person.”
In its response to the complaint the respondent relied on section 41 of the Act which exempts discrimination in relation to insurance in certain circumstances. Section 41 is as follows:
41(1) Nothing in Division 1 or 2 makes discrimination by one person (in this sub-section could the “insurer”) against another person (in this sub-section called the “client”) unlawful if:
(a) the discrimination is on the ground of the client’s sex;
(b) the discrimination is in the terms of which an insurance policy is offered to, or may be obtained by, the client; and
(c) the discrimination is based on actuarial or statistical data from a source on which it is reasonable for the insurer to rely; and
(d) the discrimination is reasonable having regard to the data; and
(e) if the client gives the insurer a written request for access to the data – either:
(i) the insurer gives the client a document containing the data; or
(ii) the insurer:
(A) Makes a document containing the data available for inspection by the client at such time or times, and at such place or places as are reasonable; and
(B) If the client inspects the document – allows the client to make a copy of, or take extracts from, the document.”
41(1A)...
41(2) In this section:
“Insurance policy” includes an annuity, a life assurance policy, an accident insurance policy and an illness insurance policy.”
At the inquiry evidence was given by and on behalf of the complainant in support of her complaint and also on behalf of the respondent. Further, at the commencement of this inquiry and with the consent of both parties, I accepted into evidence the contents of the Report the Commissioner provided to the Commission when the matter was referred for inquiry. Accordingly, I take this evidence into account along with other documentary material presented in the course of the inquiry.
3.1 Evidence provided by the complainant
At the inquiry the complainant gave evidence on affirmation. Essentially, she relied on the information contained in her letter of complaint of 3 January 1997. With that letter she had also provided information to the South Australian Commission in support of her complaint. This included a copy of the 1989-90 National Health Survey Health Status Indicators Australia by W. McLennan, Acting Australian Statistician, published by the Australian Bureau of Statistics. The preface to this publication by Ian Castles, Australian Statistician, is as follows:
This publication contains detailed statistics about:
These statistics were compiled from information collected in the 1989-90 National Health Service (NHS) conducted by the ABS and represent a selection of those available.
Information about recent and long-term conditions is as reported by respondents and is not necessarily based on medical diagnosis or advice...
The preface is dated January 1992.
The complainant also provided a publication entitled Women’s Health by Richard Madden, Acting Australian Statistician, published by the Australian Bureau of Statistics. She provided extracts from this publication which included some details relating to the main types of: disabling conditions by sex; life expectancy by sex; and death rates by age and sex. The information contained in the extracts provided by the complainant appear to relate to 1993.
The complainant also provided a copy of her Total Care Plan Policy documentation. This comprises 19 pages setting out the policy and associated information. It also includes policy details relating to the complainant. It was issued by the respondent on 27 October 1993.
3.1.1 The complainant’s evidence
The complainant’s evidence is essentially as set out in her initial letter of complaint dated 3 January 1997 to which she attested in her evidence at the inquiry. In her letter of complaint she states that she believed she had been treated unfairly because of her sex, and had incurred detriments. She said a 24 percent loading had been placed on her current income protection policy due to her sex, and she had been given figures which showed that this loading could be as high as 36 percent. She said that she had not been advised of the loading on her policy as a consequence of her sex, but had discovered this from information she received from another insurance company. She alleged “personal insult and intimidation” and “possible manipulation of statistical data and false representation of that data to justify the loading of female income protection insurance policies.”
The complainant states that she had been informed by another insurance company in September 1996 that income protection policies were generally 15 percent more expensive for women than for men, based solely upon sex. She said that on receiving this information she contacted the person from the respondent’s company with whom she had previously dealt and he told her that there was a difference between male and female policies but could not tell her how much as he did not have the software to enable him to make that calculation. She made further inquires and was told by the respondent’s “Head Office in New South Wales” that there was no difference between male and female policies. She asked for this in writing and was again advised that there were no discernible differences. She sought further information and received documentation indicating that for a 29 year old professional female there was a 20 percent greater premium based on sex. Again she made further inquiries and was told that “there was an ‘across the board’ 50 percent loading on female policies due to statistical data”.
The complainant said in her complaint that the person from the respondent’s business with whom she dealt made comments which she found “unnecessary and intimidating and not very helpful”. She said she believed the information with which she had been provided was misleading and “highly questionable”.
The complainant said that had she been aware of the loading on her policy she would not have agreed to have taken it out and she sought a full refund of her policy premium dating back to October 1993.
At the inquiry the complainant did not pursue her complaint of “personal insult and intimidation” which she made against the individual with whom she dealt. However she did give evidence that she felt angry, intimidated, dis-empowered and patronised by this behaviour, in particular, and the existence of the premium on her policy imposed by the respondent.
3.1.2 The evidence of Peter Gerrard
Mr Gerrard, a consultant actuary, gave sworn evidence in support of the complainant’s complaint. As part of his evidence he referred to a letter of 23 March 1998 to the complainant’s solicitor in response to information provided by the respondent to the Commission.
Mr Gerrard told the Commission that a “guaranteed renewable policy” was, in his view, one where the premium can be changed over time for policy holders as a group. He said that such a policy was ‘constructed’ rather than a fixed policy because the data upon which such premiums were fixed is limited and is variable over time. The insurer will be concerned about the level of risk and so wish to constantly monitor premiums through the monitoring of risks. Mr Gerrard said he understood that the complainant had been placed in the “group A” category as a white collar professional. He suggested that for men likelihood of accident was an important part differentiating between levels A, B, C or D classification for insurance purposes, but that this was not so for women: he believed that while men had an increasing likelihood of accident depending on where they were placed in the A to D classification, for women the greater differentiation was in respect of the incidence of genito-urinary illnesses. He said that he also assumed that teaching and nursing fell into category A, and it was possible this may skew the risk classification for women as these professions, where there was a predominance of women, were probably professions of higher stress than many others.
Mr Gerrard told the Commission that he believed that the respondent had relied to a very great extent on pre-1995 US data. He said that he considered this somewhat inappropriate in assessing relative female to male risks as the data was old and the rates of claim in the US were twice as high as those currently existing in Australia. He agreed that there may be some value in the US data but he considered it was very limited as it related to a different country with a different system, in particular in respect to workers compensation and private health insurance. He suggested that the age of the data might also be relevant as medical techniques had developed very significantly since that time.
In cross-examination Mr Gerrard conceded that he had no direct experience in constructing life insurance premiums with respect to life or disability policies but he said he had significant other general experience from which he could extrapolate. He described to the Commission the way in which a premium for a disability policy could be constructed and he emphasised that this was a very difficult process because the data was so variable. He agreed that to construct a risk premium involved an extensive process of collection and analysis of data and the end premium constructed was a matter of assessment and judgement. He agreed that different actuaries might well in good faith come to different figures although he assumed that they would be somewhat similar, and that such collection and analysis of data might very well lead to different male and female rates of premiums. Mr Gerrard also agreed that a basic collection of data was to be obtained from the Disability Committee of the Institute of Actuaries of Australia, which publishes regular reports that are relied on extensively within the industry. He said that he would expect annual review of premiums that would take into account the reviews issued by the Institute of Actuaries and similar UK and US reports.
Mr Gerrard also conceded that he was in fact unaware as to what occupations fell within the A to D classifications, and that he assumed the respondent had proceeded on the basis of applying unmodified US statistics in Australian conditions. He said he believed this was inappropriate but agreed that overseas data was relevant and how much weight was given to it was a matter of judgement as Australian data was as yet inadequate on its own and perhaps always would be. Mr Gerrard agreed that with respect to office premiums there was also scope for variability and judgement. He said he considered that it was probable that most office premiums were similar for male and female policy holders, but nevertheless agreed that there might well be a loading in an office premium which related to the risk premium, and that part of an office premium might also contain an element for commission. He agreed that commissions were variable between insurance companies and indeed some companies may not operate on a commission basis. He also agreed that where a commission was payable it might be the case that the value of the commission was linked to the risk premium payable by the policy holder and that if there was a difference between a male and a female policy holder’s premium this would be reflected in the commission payable and therefore in the office premium.
3.1.3 Dr Trevor Mudge
Dr Mudge, a practising obstetrician and gynaecologist, gave affirmed evidence in support of the complainant’s complaint.
Dr Mudge told the inquiry that the majority of gynaecological or urological conditions in women would not establish disability for more than a fortnight or a month. He told the Commission most gynaecological and urological conditions were able to be dealt with by surgery and he believed that with the increasing tendency to minimal access surgery in these areas, time off work had in recent years been halved: he believed that in the last 7 or 8 years time off work following even significant gynaecological surgery had been halved from 6 weeks to about 3.
3.2 The respondent’s evidence
Sworn evidence for the respondent was given by two actuaries, Bernard Smith and Christopher Seddon.
3.2.1 Mr Smith’s evidence
Mr Smith has been the Convenor of the Disability Committee of the Institute of Actuaries since 1996 and a member of that Committee since 1983. He gave evidence that the Reports of the Committee provided a starting point for the pricing of disability policies and indicated to the inquiry how the Reports issued on an annual basis by the Committee were increasingly based on Australian data as the Australian market grew, and consequently available figures became more statistically significant. However, he emphasised that while the figures were statistically small, the Committee still needed to use data from other equivalent communities such as the United States to “fill the gaps”.
Mr Smith said that the function of the Disability Committee was to investigate the experience of disability policies in the Australian market and generally to maintain an awareness of market developments. He said the Committee’s Reports were submitted to the Council of the Australian Institute of Actuaries and its members, and were made available to the public. The main purpose of the Committee’s Report was to provide a starting point for the pricing of disability policies and the Report provided the most authoritative source in Australia of such information. Mr Smith explained that the process of the collection of the statistical material for the Report was through the collection of individual policy data from registered life insurance companies and then the review and analysis of that data in order to produce the Report. He said that since 1985 the Reports have included data suggesting a differential claims incidence between male and female holders of disability policies. Mr Smith said that this information first became apparent from overseas statistical information, in particular from the United States and the United Kingdom. He said that it was relevant to use such overseas information because the Australian market was fairly small particularly in the 1980’s, and accordingly the Committee compared the Australian experience against the overseas tables. There is less reliance on this comparison now as a larger number of claims and a greater depth of Australian experience is available.
Mr Smith also gave evidence concerning the classification of policy holders into occupational classes referred to in other evidence as A, B, C and D classifications. Mr Smith said that the Committee used four categories of occupational classes, the purpose of which was to minimise the heterogeneity or maximise the homogeneity within each class. He said that the Committee’s concern was to ensure that the groups were reasonably consistent in their make up.
In reference to Mr Gerrard’s evidence Mr Smith said that he believed that in relation to disability and yearly renewable term policies flexibility was the norm rather than the exception, disagreeing with the comment Mr Gerrard had made that “flexibility is unusual in life insurance”. He considered that this notion of flexibility was more applicable to ‘whole of life’ and endowment assurance policies. However, Mr Smith said that in general terms he agreed with the methodology described by Mr Gerrard as the means by which he would construct a premium for a disability policy: in particular, he said that one would look at reports of the Australian experience, the overseas experience, and an individual company’s own experience with similar policies in the recent past.
3.2.2 Christopher Ross Sumner Seddon
Mr Seddon is an actuary who has been employed by the respondent since 1975, and is presently the respondent’s Development Manager. He gave sworn evidence. Mr Seddon was responsible for the responses provided by the respondent to the complainant in relation to this matter.
Mr Seddon stated that he agreed in general with the methodology described by Mr Gerrard in relation to the setting of premium costs. He described the way in which the rates for 1993 had been fixed: they had been set in 1992 based on a detailed study of 1991 figures available. All data had been considered but there had been particular emphasis on US data and on the 1991 Disability Committee Report. The Report had been considered closely to consider the Australian experience and how to relate it to the US data which provided the bulk of the figures. Once the issue of risk in relation to incidence and duration of claims had been considered, then the “office premium” relating to investment rates and commercial rates was developed. The rate differential between men and women varied and depended on age and waiting period and other matters: there was not a single factor across the range which determined the differential. An appropriate premium for male and female rates was developed taking into account a wide variety of factors. Mr Seddon said that both recent United States and Australian figures, as available, were taken into account in reaching these figures.
Initially a “risk premium” would be set taking into account the incidence rate of claim as well as the “termination rates” which relates to the duration of a claim. On the basis of these considerations a model is projected, and this leads to the establishment of a premium rate. Mr Seddon said the rate set as a consequence of these considerations included both the risk premium and the office premium. The system did not notionally create a risk premium and then “add on extra features” built in to provide a final figure. The office premium would include office expenses and perhaps commission rates and other factors. The percentage adjustments built in to the formula vary according to age.
Mr Seddon said there were reviews of the premium rates charged to the complainant in 1995 and 1997. In striking the rates as a consequence of those reviews the most recent Reports from the Disability Committee would have been the primary source for the basis of the reviewed rate. As a consequence of those reviews for persons in the complainant’s category the differential between male and female rates had increased. Mr Seddon said this was on the basis of all the information available from the Australian experience.
Mr Seddon rejected the appropriateness of the 1989-90 National Health Survey Health Status Indicators in Australia referred to by the complainant as the Women’s Health Report. He stated that in his view it did not provide an appropriate basis for striking premiums and said he would never use the information contained in such a report, and believed no other actuaries would do so either. In particular, he said that the publication in its own terms was a subjective narrative piece of analysis: in particular, the publication itself contains a disclaimer that it “contains detailed statistics about people’s assessment of their overall health status”. Mr Seddon said that such a personal and subjective analysis of an individual’s own assessment of their condition made it entirely inappropriate as the basis for an assessment. Further, the Report provided no indication about the extent to which any particular condition might have resulted in time off work or have led to a claim under a disability policy. Mr Seddon said that an examination of the conditions listed in the Report suggested to him that many of the conditions referred to would not lead to any claim at all under a disability policy: he referred in particular to hayfever which was one condition referred to by a substantial number of people whose assessments were contained in the Report. However, Mr Seddon said that hayfever was most unlikely to result in a person being unable to work totally for the sort of period of time a disability policy covered. Nor did the Report contain any indication or assessment of whether a person suffering from any particular condition would otherwise meet the conditions of a particular policy. Mr Seddon further indicated that an insurance company underwriting a person taking out a policy would typically require a certain amount of medical evidence or a medical examination in order to establish the basis upon which such an applicant for a policy might be accepted. This consideration was not apparent in the National Health Survey either.
Mr Seddon also referred to Dr Mudge’s evidence. He said that the anecdotal evidence given by Dr Mudge would not normally be taken into account unless it was actually borne out through the claims experience as reported by the Disability Committee. He emphasised that once the shortened period of time women undergoing genito-urinary procedures might be disabled was reflected in their claims on disability policies, then that would be taken into account by actuaries in setting premium rates. However, in Mr Seddon’s view that was not yet reflected in the material before the Disability Committee.
In cross-examination Mr Seddon gave evidence of the details of the calculation of the complainant’s policy. He was able to look at the actual premium and determine the premium that would have been charged to a male in the same area of work at that time. He was also able to look at premiums that would have been charged in the next two years but was not able to go further than that as then the premium scale altered. He gave evidence that, disregarding the stamp duty and a $40.00 policy fee at the time the policy was taken out, the premium for a woman was $487.00 and for a man was $500.00. Under those circumstances the women worker was charged about 2.5 percent less than a male. In the following year (aged 27 next birthday) the female premium was $508.00 and the male premium $505.00: the female premium was 1 percent higher than the male premium. In the following year the female premium was $542.00 and the male premium $517.00 (the female premium being about 5 percent higher than the male premium). These figures reflect the total premium paid, including the “office premium”. These policy figures related merely to the disability policy not the additional life insurance policy the complainant took out. Mr Seddon conceded that at the time he prepared his initial response to the complainant’s complaint in August 1997 the differential between male and female premiums may have been about 24 per cent: that is, the differential between male and female premiums had increased as the complainant grew older and the premiums have been revisited. Mr Seddon said he calculated that premium on the basis of the respondent’s rate book which applied in 1992 which sets out the basic rates to be charged to various factors by sex or by age. He said that there are a series of factors applied to the basic set of rates which establishes a formula to apply in individual cases. These included factors for the occupation group into which the complainant fell; a factor for non-smokers; a factor for the size of the policy; and a factor in relation to the business overheads benefit which was taken out as part of the policy. Mr Seddon said that calculating the risk premium for an identical male would have involved the same consideration. He emphasised that the assumptions the respondent made in the profit testing for expenses and for discontinuance (that is, the duration of the risk) and commission do not vary according to gender.
He indicated further that the 1997 Report of the Disability Committee reported that across all ages for the occupation class A (into which the complainant fell) where there was a one month deferment or waiting period, females claimed 77 percent more than males: however the respondent’s figure comparable to this (177 percent figure) was 218 percent, which indicated that the respondent’s experience was significantly worse than the industry average with females claiming 118 percent more than males. Mr Seddon indicated that this information would have been taken into account in setting the premium rates. He said that the 1997 Report obviously would not have been available in setting the rates for 1996, but the raw data would have been available within the respondent’s own records and would have been taken into account in restructuring the premium.
Both parties provided written submissions subsequent to the inquiry, received by me on 3 July 1998. They included additional written material sought by the complainant from the respondent, and a further affidavit by Mr Seddon dated 9 June 1998.
4.1 The complainant’s submissions
The complainant’s primary submission is that she was charged more for her income protection policy with the respondent than a male in a otherwise identical circumstances would have been charged. It was submitted that as at 1996 the complainant was being charged 24 percent more in total premium than an otherwise identical male worker. Under these circumstances the complainant’s submission was that this constituted direct discrimination contrary to section 5 of the Act. The complainant noted that the respondent had not pursued any argument that the two policies ought to be considered together.
The complainant referred to section 41 of the Act setting out an exemption relating to insurance policies: it was submitted that the insurer (the respondent) had the onus of demonstrating that it came within the exemption set out, and in particular within the four criteria set out in section 41. In particular the submission focused on section 41(d) which requires that “the discrimination is reasonable having regard to the data”: that is, in particular, “actuarial or statistical data from a source on which it is reasonable for the insurer to rely” (see s.41(c) of the Act). Mr Weatherill’s submission was that section 41(d) is concerned with the reasonableness of the data itself, rather than other obligations or interests which the respondent may have had. And that an amendment to section 41(d) of the Act in 1984, excluding the phrase “and any other relevant factors”, indicated that the consideration of discrimination must be limited solely to the reasonableness of the actuarial or statistical data.
In relation to the evidence concerning the calculation of the premium, the complainant conceded that there was some evidence contained in the various Disability Reports, particularly of 1995 and 1997, to suggest that there is a difference in incidence rates between females and males. Further, the complainant conceded that it was reasonable for this difference to be reflected to some extent in the risk premium component of the policy provided that the extent itself is reasonable having regard to the data. However, the complainant’s contention was that the way in which the total premium was calculated, in particular where some items were calculated on the percentage of the premium, resulted in women being charged more for their policies than is supported by the data. Mr Seddon’s evidence was that the commission on the complainant’s policy for the first year was 50 percent, and 10 percent thereafter. The complainant’s submission was that the calculation of the commission as a percentage of the total premium resulted in women being charged more for their policies than might otherwise be reasonable. The commission does not relate in any way to the risk component of the premium and it was submitted that it is unreasonable and not supported by the data to calculate commission on a percentage basis.
In addressing the issue of the model of profit testing, the complainant submitted that it was not reasonable that the profit margin for the respondent in relation to female clients should be significantly higher than in relation to male clients. The complainant submitted that the evidence given by Mr Seddon at the hearing and the additional information provided at the complainant’s request subsequent to the hearing, indicated that the profit component was three times higher in relation to females than in relation to males in 1995 as a percentage of the total premium: in 1993, the male and female profit ratio had been similar. This differential occurred at a time when the premium rate for a male aged 29 reduced, but the rate for an equivalent female increased, which significantly increased the differential between the cost of male and female policies. The complainant’s submission was that, as at this time women were providing the respondent with three times more profit in relation to policies than men, the way on which the policies were calculated discriminated against women due to the fact that women were initially charged a greater amount as a consequence of the assessment of risk. The submission was that men paid a lesser amount for their premium because of the lower assessment of risk, and then effectively paid a reduced amount in commission, administrative costs, and profit, which did not relate to risk. The complainant submitted that such method of calculating the premium and the end result in terms of the differential and the amount charged to women, is not supported as reasonable with reference to the data, as the data relates only to risk and incidence. The complainant submitted that she had not been provided with the actual calculations for the risk component of the premium or the other components, but rather a breakdown in percentage terms, and she was therefore unable to compare the actual difference in the risk component of the premiums.
Further, the complainant submitted that the respondent did not provide any evidence to support the data relating to incidence and duration of claims for females which would support the extent of the differential that was charged to the complainant. She submitted that it is not reasonable to support a difference based on incidence and, to a more limited extent, duration, without looking at actual claims cost.
The complainant’s final submission in relation to the data relied on by the respondent was in respect of the use of data from the United States. The complainant conceded that it was reasonable for the insurer to have regard to overseas experience. Nevertheless, she submitted that it was not reasonable to rely on this data if it was not tested, scrutinised, and used cautiously in an Australian context.
The complainant also submitted that at the time she took out her policy she was not informed that she would be charged a different amount than that charged to an equivalent male. Nor was she informed that even though there was no substantial differential at that time (in fact she was paying slightly less than an equivalent male) that a differential would occur in the future and could become quite substantial. There was nothing in terms of the policy indicating that sex was a factor in determining the charge for the policy. She was provided with no information by the respondent as to the existence of the differential, or the size, or reason for the size, of the differential between her policy and that charged to an equivalent male. When she discovered this differential she had considerable difficulty in obtaining the information she requested from the respondent. The complainant’s submission was that the discrimination was not in the terms of her policy, but related to the way in which she was treated by the respondent. Accordingly, section 41(b) of the Act, which provides that discrimination by an insurer against another person is not unlawful if the discrimination is in the terms on which an insurance policy is offered to or may be obtained by that client, was not relevant in the complainant’s circumstances.
The complainant in her own evidence had referred to the fact that she felt patronised and intimidated by the respondent when she made inquiries about the issues relating to her policy and this caused her anxiety and injury to her feelings. This aspect of the complainant’s relations with the respondent is relevant to the declarations sought from the Commission.
In her submission the complainant indicates that she seeks: a declaration that the respondent had failed to satisfy the prerequisites of section 41 of the Act: compensation in relation to injury to her feelings; a declaration that the respondent disclose to all policy holders at the time of the taking out of a policy if there is a difference to be charged between males and females, and for the extent of this difference payable at given ages to be disclosed to clients, and the data on which it is based to be provided; and a refund of the policy amounts paid by the complainant.
The complainant’s final submissions were related to the failure of the respondent to satisfy the conditions of section 41 of the Act. In particular, the complainant submitted that the requirement of section 41(b) is not satisfied as the discrimination did not form part of the policy; the requirement of section 41(d) of the Act is not satisfied as it is the data itself that must support the discrimination: any other factors, such as the insurer’s obligation under other legislation, and market factors, are not relevant. In particular in this regard, the complainant submitted that the extent of the discrimination is unreasonable having regard to the data, because the respondent had failed to indicate what was the differential amount charged for the risk premium component of the premium; and that it is not reasonable to include as a percentage of the total premium other elements such as administrative costs, profit, and commission, because the data referred to in section 41(d) must relate only to the risk.
Finally, the complainant submitted that the data upon which the respondent’s formulation of the premium was based, and to which section 41(d) of the Act must apply, is limited and inadequate in that: the United States CIDA statistics are unreliable due to their age and the marked differences in the Australian experience; the data relates primarily to incidence rates and duration of claims, and this does not reflect actual claims cost; the duration of claims tables provided are unreliable; and the occupation class categories may not be as relevant or sensitive to women’s experience and disability, which primarily relates to genito-urinary claims, whereas the male claims experience is more generally linked to accidents.
4.2 THE RESPONDENT'S SUBMISSIONS
The respondent prefaced its written submissions with an issue raised at the commencement of the hearing. The respondent submitted that it had laboured under several disabilities in preparing its response, and in particular difficulties related to the imprecision of the complainant’s complaint. The respondent submitted that the complainant had failed to identify: the precise time or times at which the discrimination was alleged to have occurred, and in particular discrimination contrary to the Act is alleged to have occurred; the quantum of any monetary loss allegedly sustained by the complainant as a result of any unlawful discrimination; and the statistical basis upon which any such monetary loss had been calculated.
At the outset of its submission, the respondent submitted that the complainant’s submissions relating to section 41(1)(b) and (d) ought to be dismissed. In relation to section 41(1)(b), the respondent submitted that any discrimination alleged by the complainant arose from differential rates of premium charged to males and females for policies which were not otherwise materially different in their terms. The premium charged however was prescribed by reference to a term in the policy. Under those circumstances, the respondent submitted any relevant discrimination was attributable to a term on which the policy was offered or could be obtained, and the respondent is entitled to claim the benefit of section 41(1)(b) of the Act.
In relation to section 41(1)(d), the respondent made a number of submissions in response to the complainant’s argument.
The respondent submitted that the evidence of Mr Seddon was uncontradicted in relation to the calculation of the premium. Mr Seddon’s evidence was that it was not possible to disaggregate the various components of a premium so that components other than the risk component could be identified, quantified and compared, with a view to determining whether those other components discriminate between men and women. Mr Seddon’s evidence was that premiums are constructed on various models which are designed to ensure the profitability of the business, and which entail an element of commercial judgement. The models incorporate relevant actuarial data and take into account the peculiar claims experience of an insurer. The respondent submitted that under these circumstances, the complainant’s assertion that the extent of discrimination is unreasonable because the respondent had failed to indicate the differential amounts charged for the risk premium component, should be rejected.
The respondent also made submissions relating to “statistical data” referred to in section 41(1)(c) of the Act. It submitted that the complainant’s case appears to proceed on an assumption that premiums must be fixed exclusively by reference to actuarial data. However section 41(1) of the Act by reference to “statistical data” clearly refers to a wider ranger of data than “actuarial data”, and might include for example statistical comparisons between commission rates paid by various life offices to agents and other comparisons. Further, the respondent submitted the complainant’s submissions seek to preclude the exercise of commercial judgement based on the statistical data. The respondent’s final submission on this point refers to the 1997 Report of the Disability Committee which included (in Table E) a comparison of actual female claims with the number expected using male claim incidence rates. The respondent submitted that “this table would support a premium rate differential between men and women who were seeking an otherwise identical policy of 64 percent in favour of males. On no occasion during the period relevant to this case (1993 to the present) has the male/female premium differential been of this magnitude or even approached such a degree of asymmetry. The highest differential on which the complainant relies is 24 percent in 1996. The 1996 actuarial data would have supported a differential of 61 percent”.
With reference to the submission by the complainant that the data does not support the calculation of disaggregated elements of the premium such as administrative costs, profit, and commission on a percentage basis the respondent’s submission is that it is not possible to disaggregate the components of the premium; the relevant data is not confined to data relating to risk; and in any event even when those non risk factors are taken into account, the differential premium rates between men and women is far lower than would be supportable by reference exclusively to actuarial data.
The complainant’s third point on which the respondent made submissions, related to the alleged limitation of the data upon which the respondent relies. The complainant submitted that the US statistics are unreliable due to their age and the marked difference in Australian experience; the data does not reflect actual claims costs; the duration of claims tables provided are unreliable in that they do not immediately reflect current medical changes; and the occupation class categories might not be relevant or appropriately sensitive.
In relation to the use of US statistics the respondent submitted that the statistics had been relied on exclusively until 1985 in the absence of any comparable Australian data, but since that time data reflecting the Australian experience has progressively become available and has been incorporated by the Institute of Actuaries of Australia in its various reports which were before the Commission. Evidence was given at the inquiry that the respondent as well as having regard to this actuarial data has regard to its own claims experience when it sets its policy premium. The respondent submitted that the actual claims cost constitutes statistical data on which the respondent was entitled reasonably to rely. In relation to the duration of claims tables, the respondent submitted that the evidence before the inquiry is that the tables reflect the most recent available evidence and are more appropriate and reliable than anecdotal evidence. The respondent submitted that if factors such as recovery rates from certain types of surgery alter, those changes will eventually be reflected in actuarial tables and will be taken into account then. The respondent submitted that there was no evidence to suggest that claims made in relation to occupation class categories were unreliable or inappropriate.
The respondent also made submissions relating to the relief sought by the complainant. The respondent submitted that the Commission had no power pursuant to section 81 of the Act to make the declaration sought by the complainant in relation to the disclosure of certain matters to all policy holders. Further such a declaration would not be confined to a variation of the complainant’s policy terms “but would operate as a quasi-statutory provision binding on the respondent in respect of all policies which it issues”. Further, it was submitted, such disclosures would be both costly and impractical in their operation. The respondent also submitted that the duty and the extent of disclosure required of life insurers is dealt with by the relevant industry regulating body, and all the requirements so prescribed since 1996 have been complied with by the respondent.
The respondent states that the application which the complainant completed was contained in a brochure which included a heading “how much will it cost?”, under which it was expressly disclosed that the premium cost of such a policy was likely to increase each year and that there were various factors upon which the cost would depend including sex. Under those circumstances, the respondent submitted that at the time the complainant applied for her policy it had disclosed to her that premium might vary according to gender. Such information is provided by the respondent in its current brochure, and customers are further advised that they can obtain premium tables for the product on request. The respondent submitted that a declaration of a kind sought by the complainant would entail information to be provided to each prospective customer which “would be largely meaningless to consumers, costly to the respondent to produce and may be contrary to the regulatory body’s requirements in relation to promotional material in terms of that material being confusing to consumers”. The respondent referred to Mr Seddon’s evidence that the alternative rates for both males and females taking into account all the various combinations of waiting periods, types of benefits and benefit periods across all ages might contain in the order of 13,000 separate cells of data.
In relation to the complainant’s submission that she “seeks a refund of her policy from the date that she obtained it”, the respondent submitted that “the complainant’s lack of precision has made it impossible for the respondent to make a comprehensive rejoinder”. The respondent submitted that the complainant could not be entitled to a refund of the premium initially paid by her as she had accepted that when that payment was paid any differential in premium rates favoured females over males. The complainant became aware of differential rates in premiums in 1995 but chose not to cancel her policy. Consequently, she has had the benefit of protection offered by the policy in the subsequent years. The respondent has assumed and borne the risks associated with the policy during that time and under those circumstances, it was submitted there was no proper basis for making such a declaration.
The complainant’s final claim in respect of relief sought was for compensation for anxiety and injury to feelings. The respondent submits that there is no basis for a claim that it was in any way responsible for her hurt feelings, as the complainant’s own evidence indicates that these subjective feelings arose from her own dealings with an agent of the respondent. The complainant had expressly agreed that no claim arising from the actions of that agent would be advanced at the hearing and her relevant evidence in relation to the matter was the agent provided information to her about the terms of her insurance policy. As a consequence of this assurance from the complainant, the respondent did not call any evidence from the agent relating to his dealings with her. The respondent submitted that under those circumstances it is not open to the complainant to press this claim for compensation in relation to the alleged behaviour of the agent.
The respondent’s primary submission was that the complaint should be dismissed.
5.1 Findings of Fact
On the basis of the evidence placed before me in the course of this Inquiry I make the following findings of fact:
5.2 The Law
The legal issues before the Commission in the course of this inquiry were not well identified by the complainant. Essentially, her case depended on an assertion that because there was a significant differential between the cost of the premium to her compared with that which would have been charged to an equivalent male, she was subjected to direct discrimination as defined in section 5 of the Act; and that the discriminatory differential in premium was based on inappropriate and out of date information and assumptions upon which it was not reasonable for the respondent to rely in costing its premiums. In this respect the complainant’s case proceeded on four assertions:
(1) That the premium rate charged to her was based primarily on CIDA figures from the United States that were old and likely therefore to be out of date and inappropriate as they reflected conditions and circumstances of claims in the United States which were not transferable to Australian conditions.
(2) That the occupational group A premium which was charged to her had been inflated by the inclusion within that group of teachers and nurses, who the complainant asserted probably had higher claims rates than a person in an occupation such as hers. In addition the complainant appeared to be asserting that teachers and nurses, whom she assumed were included in group A, were likely to be disproportionately (compared with other professionals within group A) women, so that the high risk of claims from this group of workers was not reflected in male workers within group A.
(3) The complainant asserted that in making an assessment of likely claims from women, the respondent should have relied on information such as contained in the Women’s Health Survey.
(4) The complainant asserted that the assumptions made by the respondent relating to the amount of time away from work likely to be incurred by women as a consequence of genito-urinary disabilities was incorrect.
In considering the evidence presented at the inquiry I formed the view that all these assumptions asserted by the complainant are either disproved on the evidence, or are not established by her.
In relation to the CIDA figures which the complainant asserted were out of date and inappropriate as they relate to the US rather than the Australian experience of claims, I accept the evidence presented by the respondent that these figures are only used to “fill the gap”: I accept that they are not used in an unmodified way. This was conceded by Mr Gerrard in his evidence and made clear by Mr Seddon in his evidence. I accept that as an increasing amount of Australian statistical information becomes available the CIDA figures will become statistically less relevant, but as the Australian market is significantly smaller it will take some time for Australian figures alone to present the statistical relevance and significance which is required for actuarial purposes.
It was clear from the evidence that the respondent does not include most nurses within the occupational group A, and teachers fall within both group A and group B for occupational classification purposes, depending on the nature of the work which they undertake. In any event I am satisfied that the figures that appear in the 1997 Report of the Disability Committee of the Institute of Actuaries of Australia (tendered by the complainant at the inquiry), demonstrates a parity of claims within group A across all ages. In relation to the complainant’s assertion that the respondent should rely on information such as that contained in the Women’s Health Survey, I accept Mr Seddon’s reasons for the inappropriateness of taking such information into account: these included the subjective and non-statistical nature of the information; the fact that the survey gave no indication of how the conditions which were discussed might lead to any claim under such an insurance policy or how much time off work such conditions may have precipitated; and that many of the conditions discussed in the survey were unlikely to lead to any claim under a disability policy. Mr Seddon’s evidence was that there was nothing in the Women’s Health Survey to assist the insurer in making an assessment as to whether a person might meet the conditions of a policy which is an essential aspect of costing a premium.
In relation to the complainant’s final assertion concerning the assumptions she considered the respondent made about time off work for women suffering genito-urinary disabilities, Mr Seddon’s evidence, which I accept, is that when the circumstances described by Dr Mudge flowed through to claims made to the respondent, they would be factored into the respondent’s considerations in setting a premium. Dr Mudge’s evidence was that the time required to be taken off work by women suffering from genito-urinary disabilities had in recent times dropped by 50 percent: Mr Seddon accepted that this was likely to be the case at the level at which it was presented to the inquiry and prior to that by the complainant. It was anecdotal only. Such a consideration will only become a relevant one when it is able to be assessed as part of the statistical information before the respondent.
I am satisfied that all these assumptions upon which the complainant’s case proceeded have been disproved or not established on the evidence, and under these circumstances I am satisfied that the complaint should be dismissed. However, I turn to the specific legislative provisions to which the complainant’s case was directed. These are set out above, and in particular are contained in sections 22(1)(b) and 41 of the Act.
5.2.1 “Services”
The complainant’s allegation was that she had been discriminated against on the ground of her sex in relation to the provision of services. That is, in the way the terms or conditions on which certain services were provided to her by the respondent were made available to her. Specifically, her allegation was that she was charged more for the service provided by the respondent than would have been charged for the same service to an equivalent male. Her allegation was that the respondent’s alleged discrimination was contrary to section 22(1)(b) of the Act. Section 4(1) of the Act defines “services” to include those services relating to insurance, and I am satisfied (and it was not contested) that the services provided to the complainant by the respondent were services which are subject to the Act.
The complainant relied on section 5(1) of the Act in alleging discrimination: section 5 defines discrimination on the ground of sex, and section 5(1) defines “direct discrimination”. The complainant alleged that in the provision of services to her the respondent had discriminated against her on the ground of her sex, as the respondent had treated her less favourably than, in circumstances that are either same or not materially different, the respondent would have treated a person of the opposite sex.
I am satisfied on the basis of the evidence before me that there was an element of discrimination in the way in which the service of insurance was provided to the complainant, compared with the terms on which it may have been provided to an equivalent male person. It is clear on the evidence that there were differentials between the premiums charged to the complainant and those which in different years would have been charged to an equivalent male person: I note that in the first year in which her policy applied the complainant was in fact paying less than that which would have been payable by an equivalent male person and under these circumstances there can of course be no issue of discrimination at that time against the complainant pursuant to section 5 of the Act. However, in 1994, 1995 and 1996, the differential was less favourable to the complainant, particularly in 1996, when the differential was about 24 percent against the complainant compared with an equivalent male.
However, section 5 makes it clear that mere differential treatment is not the only factor to be taken into account in considering whether there has been unlawful discrimination. Section 5 qualifies the issue of less favourable treatment by requiring the less favourable treatment to occur “in circumstances that are the same or are not materially different” from the way in which a person of the opposite sex would be treated. In this case I am satisfied that there were material differences between the circumstances of an equivalent male seeking insurance with the respondent and the complainant. I am satisfied that those material circumstances are those identified in the Report of the Disability Committee of the Institute of Actuaries of Australia upon which the respondent relied, and the respondent’s own records of claims. These material differences between male and female workers seeking disability income insurance are set out in actuarial and statistical data upon which the respondent relied in reaching its decisions concerning premium charges.
Under these circumstances I am satisfied that there was no unlawful discrimination within the meaning of section 5(1) of the Act.
5.2.2 Section 41 - exemption
The respondent in its submissions relied principally on section 41 of the Act. Section 41 of the Act provides a general exemption in relation to insurance if certain criteria set out in section 41(1) are fulfilled. The complainant in her submission had referred to this provision and submitted that the requirements to bring the respondent within the general exemption provided by section 41 had not been made out.
It is clear that as section 41 operates as a general exemption the onus must be on the respondent to bring itself within the required terms of section 41. Although I have made a finding above that there was no unlawful discrimination within the meaning of section 5(1) of the Act, as the parties referred extensively to section 41 I see it as appropriate that I deal with the issues raised therein. I am satisfied that if the respondent’s differential treatment of the complainant, compared with an equivalent male insured, amounted to discrimination within the meaning of section 5(1) of Act, nevertheless the respondent would be clearly within the terms of section 41 of the Act.
There was no contention that any discrimination was on the ground of the complainant’s sex (although it also depended on other factors besides gender); this fulfils the requirement of section 41(1)(a).
The complainant submitted that any discrimination was not in the terms of which the insurance policy was offered to her, but the respondent submitted that this was in fact the case. I accept the respondent’s submission on this: I am satisfied that any discrimination which applied in the complainant’s circumstances was precisely in the terms on which the insurance policy was offered to the complainant, as it was contained in the assessment of premium payable by her in relation to that policy. The premium payable by her was prescribed by reference to a term in the policy. Under those circumstances any relevant discrimination would be attributable to a term on which the policy was offered to the complainant. I am satisfied this requirement of section 41(1)(b) is fulfilled.
Further, I am satisfied that any discrimination is based on actuarial or statistical data from a source on which it is reasonable for the insurer to rely. The respondent has relied on actuarial and statistical data contained in the Report of the Disability Committee of the Institute of Actuaries of Australia, and on material available from its own records of claims. There was no argument before the inquiry that it was not reasonable for the respondent to rely on this data: the complainant’s argument rather was that either other data should have been included (such as that contained in the Women’s Health Survey), or that data on which she assumed the respondent relied was inappropriate. I have dealt with these issues above. I also accept the respondent’s submission that it is not possible to disaggregate the various components of a premium to isolate the “risk premium” from the other aspects of the premium. In addition however, I am not satisfied that such disaggregation would serve any purpose. I am satisfied on the basis on the evidence before me that the actuarial and statistical data upon which the respondent relies in setting its premiums is from a source on which it is reasonable for it to rely. I accept the respondent’s submission that actuarial data is not the only relevant data: section 41(1)(c) refers also to “statistical data”, and I accept that this comprehends a wider range of data than “actuarial data”, including statistical comparisons between commission rates paid by various life offices to agents, comparative premium rates charged for similar policies, and differential profiles of policy holders as between life offices, such as contained in the Report of the Disability Committee of the Institute of Actuaries of Australia. I also accept the respondent’s submission that it is reasonable to include in the consideration of premiums chargeable the general sorts of considerations that are relevant to the exercise of commercial judgement based on the statistical data. Clearly the respondent must be able to make an assessment on this basis. I also note the respondent’s submission that had the respondent in fact directly transposed the information contained in the Report of the Disability Committee to the determination of its premiums, the premium rate differential between men and women might have been very significantly greater: in particular, the 1997 Report suggests that actuarial and statistical information could support a differential of 64 percent in favour of men insureds. The highest differential on which the complainant relied was 24 per cent in 1996.
Section 41(1)(d) of the Act requires that any discrimination must be reasonable having regard to the data. I am satisfied on the basis of the evidence put before me at the inquiry that any discrimination is reasonable having regard to the data and particularly that contained in Report of the Disability Committee of the Institute of Actuaries of Australia.
Section 41(1)(e) of the Act imposes obligations on an insurer in relation to disclosure where an insured person makes a written request for access to the data. There was no evidence nor any submissions made in respect of the application of this aspect of section 41(1) of the Act, and accordingly I make no findings relating to it.
Under these circumstances I am satisfied that, were it the case that the respondent had discriminated against the complainant contrary to section 5(1) of the Act, nevertheless section 41 of the Act would operate to exclude any issue of unlawful discrimination applying. I am satisfied that the respondent’s differential treatment of the complainant, if it amounted to discrimination in the terms of section 5(1) of the Act, would be within the terms of section 41 of the Act which sets out grounds for discrimination which is not unlawful.
6. CONCLUSION
For the reasons given above I am satisfied that it is appropriate to dismiss this complaint.
Pursuant to section 81 of the Act I dismiss the complaint the subject of this inquiry.
I certify that this page and the previous 30 pages are
a true copy of the
Reasons for Decision of Inquiry
Commissioner, Kathleen McEvoy.
Associate:
Dated: 11 February 2000
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/cases/cth/HREOCA/2000/5.html