Ingram v QBE Insurance (Australia) Ltd (Human Rights) [2015] VCAT 1936

1. Under section 125(a) of the Equal Opportunity Act 2010 (Vic) (EOA), the tribunal finds that the respondent has contravened sections 44(1)(a) and (b) of the EOA.

1. Under section 125(a)(ii) of the EOA, by 18 January 2016, the respondent must pay to the applicant the sum of $4,292.48 for economic loss.

1. Under section 125(a)(ii) of the EOA, by 18 January 2016, the respondent must pay to the applicant the sum of $15,000 for non-economic loss.

1. Under section 109 of the Victorian Civil and Administrative Tribunal Act 1998 (Vic), the respondent shall pay to the applicant the costs associated with her expert's attendance at the tribunal for one day. The applicant's applications for costs are otherwise dismissed.

Ms C Harris of counsel, instructed by Victoria Legal Aid

Mr A Naylor of counsel, instructed by Gilchrist Connell

1. In late 2011, during the course of her year 11 studies, Ms Ella Ingram and her mother decided she could join a school tour to New York scheduled for 30 March to 11 April 2012. The required deposit and subsequent instalments were all paid. Some of the costs were applied to a travel insurance policy issued by QBE Insurance (Australia) Ltd (QBE) on 8 December 2011.
2. In about January 2012, for the first time in her life, Ms Ingram experienced symptoms of depression. Over the course of the following months she was diagnosed with that illness and received treatment. In consultation with her doctors and mother, Ms Ingram decided not to go on the planned school trip, in the interests of her health.
3. In about April 2012, Ms Ingram's mother made enquiries about claiming the cost of the trip on the QBE policy. The claim for $5,860 was lodged in May 2012 and was denied by QBE on 17 August 2012. The refusal was confirmed by a further letter sent on 4 December 2012.
4. In rejecting the claim, QBE relied upon a general exclusion that said, in summary, there is no cover where the claim arises directly or indirectly due to mental illness. There is no dispute that the definition of mental illness in the policy includes depression. In its 4 December 2012 correspondence to Ms Ingram's mother, QBE said that the decision to refuse the claim was based on detailed statistical modelling and analysis of claims arising from a range of causes including mental illness. QBE said that mental illness is excluded from the policy because its statistics demonstrated that in travel policies there is a high risk of cancellation because of mental illness.
5. Ms Ingram said by including the mental illness exclusion in the policy, QBE treated her unfavourably because of her disability and directly discriminated against her in the terms on which it provided the service of travel insurance. Ms Ingram said that conduct was a breach of section 44(1)(b) of the Equal Opportunity Act 2010 (Vic) (EOA). Ms Ingram further said that, by refusing to indemnify her in August and December 2012 on the basis of her mental illness, QBE treated her unfavourably because of her disability and so directly discriminated against her contrary to section 44(1)(a) of the EOA.
6. Ms Ingram has sought a declaration that QBE unlawfully discriminated against her, compensation for economic loss and the damage of hurt and humiliation suffered by her and costs. The economic loss claimed is $4,292.48 which takes account of a flight costs refund. Ms Ingram sought $20,000 in compensation for hurt and humiliation.
7. QBE denied discriminating against Ms Ingram. In its defence it said that it did not refuse to provide insurance to Ms Ingram because of her disability. Rather, the policy wording contained an exclusion for mental illness. [1] Further or alternatively, QBE said that, if it did discriminate against Ms Ingram in either of the ways alleged, that discrimination was lawful because an EOA and/or a Disability Discrimination Act 2004 (Cth) (DDA) exception applied.
8. I was greatly assisted by the parties' detailed written submissions which I have carefully considered in reaching my decision.

1. Paragraphs 10 and 11 are a brief summary of my decision. The full reasons for that decision then follow.
2. I have found:
   • At the relevant points in time, Ms Ingram had a 'disability' within the meaning of that term in the EOA;
   • QBE engaged in direct discrimination in breach of section 44 of the EOA first, when it issued her with a policy which included the mental illness exclusion and second, when it refused her indemnity relying on the terms of that exclusion;
   • QBE cannot rely upon the statutory exceptions to excuse the discrimination. That is because QBE has not produced evidence to prove it is more probable than not that:
     • At the relevant points in time, the acts of discrimination by QBE were based on actuarial or statistical data. I have found I cannot infer otherwise. Accordingly, the exception contained in section 47(1)(b) of the EOA and section 46(2)(f) of the DDA which depend on such data do not apply;
     • It would have suffered unjustifiable hardship if it had not included the mental illness exclusion in the policy issued to Ms Ingram. Accordingly, the exception contained in section 29A of the DDA does not apply;
   • As a consequence, QBE engaged in unlawful discrimination when it included the mental illness exclusion in the policy issued to Ms Ingram and when it denied her indemnity relying on that exclusion;
   • Ms Ingram is entitled to economic loss in the sum of $4,292.48, being the value of her cancelled trip;
   • Ms Ingram is entitled to non-economic loss in the sum of $15,000 for hurt and humiliation; and
   • Ms Ingram is entitled only to the costs associated with her expert being required for one additional day.
3. Ms Ingram applied for a declaration that QBE had engaged in unlawful discrimination. I have declined Ms Ingram's application to ensure that an impression is not given that my decision automatically applies beyond the dispute between these parties and, in particular, to avoid any impression that it applies to all insurers.

1. The EOA is protective legislation that seeks to ensure that persons with named attributes are not discriminated against because of those attributes in relation to specified activities or classes of activities. One of those protected attributes is disability.[2] There was no dispute that Ms Ingram's depressive illness is a disability under the EOA and that, at the time she became ill in around February 2012, she had that attribute. There is a dispute about whether, at the time the policy was issued in December 2011 she had a disability as defined. I will come to that dispute later.
2. Ms Ingram makes claims of direct discrimination, which is defined in section 8 of the EOA as follows:

(1) Direct discrimination occurs if a person treats, or proposes to treat, a person with an attribute unfavourably because of that attribute.

1. An employer advises an employee that she will not be trained to work on new machinery because she is too old to learn new skills. The employer has discriminated against the employee by denying her training in her employment on the basis of her age.
2. A real estate agent refuses an African man's application for a lease. The real estate agent tells the man that the landlord would prefer an Australian tenant. The real estate agent has discriminated against the man by denying him accommodation on the basis of his race.

(2) In determining whether a person directly discriminates it is irrelevant -

(a) whether or not that person is aware of the discrimination or considers the treatment to be unfavourable;

(b) whether or not the attribute is the only or dominant reason for the treatment, provided that it is a substantial reason.

1. Section 10 of the EOA says that, in determining whether or not a person discriminates, the person's motive is irrelevant.
2. The term 'unfavourable treatment' is not defined in the EOA but it is understood to bear its ordinary meaning which includes adverse treatment and humiliation.[3] QBE accepts that, in declining to provide indemnity, it treated Ms Ingram unfavourably.[4]
3. An applicant is required to show a causal link between the unfavourable treatment and the attribute. As set out in section 8(2)(b), it is not necessary for the attribute to be the sole reason for the unfavourable treatment, but it must be a substantial reason.
4. The class of activity relied on here is the provision of services. Section 44 of the EOA says:

(1) A person must not discriminate against another person -

(a) by refusing to provide goods or services to the other person; or

(b) in the terms on which goods or services are provided to the other person; or

(c) by subjecting the other person to any other detriment in connection with the provision of goods or services to him or her.

(2) Subsection (1) applies whether or not the goods or services are provided for payment.

1. QBE admitted that in issuing a contract of insurance, it provided services within the meaning of that term under the EOA. Further it accepted that in declining indemnity, QBE refused to provide a service within the meaning of section 44. That refusal first occurred on 17 August 2012 and again on 4 December 2012 when the refusal of indemnity was confirmed.[5]
2. As summarised above, QBE denied discriminating against Ms Ingram but said that, if it is found to have done so, that discrimination is not prohibited because one or more of the exceptions contained in section 47 of the EOA applies.[6] I will come to the detail of those provisions in due course.

1. The travel insurance policy shows that the particular version relied on by QBE was prepared on 1 March 2010 and came into effect on 22 March 2010.[7]
2. Relevant here, a claim could be made under the policy if Ms Ingram was unable to start the trip because of sudden serious illness. QBE's evidence was that the term 'serious illness' indicated an inability to travel.[8]

1. Cover is only provided to a person who was unable to travel on production of proof that the illness required hospitalisation or confinement.[9]
2. The costs that can be claimed include unused pre-paid travel related costs less any refunds the person is entitled to receive.[10] Medical, hospital and ambulance expenses incurred as a result of the illness while overseas may be claimed. No cover for these benefits is granted in respect of an illness if the person became aware of signs or symptoms of that illness before the travel.[11]
3. The policy wording included a series of general exclusions. The opening words of the general exclusions said there is no cover under any section of the policy for any claim arising directly or indirectly because of any of the listed circumstances. The circumstances included if a member of the travelling party (which includes the person issued with the insurance):

suffers any mental illness including dementia, depression, anxiety, panic attack, stress, bipolar, mania, schizophrenia or other nervous disorder[12]

1. The term 'mental illness' was defined in the policy to mean 'a condition characterised by the presence of symptoms such as delusions, hallucinations, disorder of thought form, disturbance of mood, or sustained or repeated irrational behaviour, which impairs, either temporarily or permanently, the mental functioning of a person.'[13]
2. Clause 3 of the general exclusions excluded claims where the illness, injury or death was caused by or exacerbated by or consequential upon an existing medical condition. An 'existing medical condition' includes:
   1. any chronic or ongoing (whether chronic or otherwise) medical or dental condition, illness or disease of which you were aware or should reasonably have been aware and which is medically documented or under investigation in the 12 months prior to the issue of the Certificate of Insurance; or

1. any physical, mental illness or medical condition (including pregnancy), defect illness or disease of which you were aware or should reasonably have been aware, or for which treatment, medication, preventative medication, advice, preventative advice or investigation have been received or prescribed by a medical or dental advisor in the 60 days prior to the issue of the Certificate of Insurance and in the case of the Annual Multi Trip Travel Plan also within 30 days of booking a particular trip.[14]

1. A policy holder could apply to QBE for cover for an existing medical condition and QBE could decide to grant cover if the policy holder pays an additional amount.[15] The relationship between this clause and the mental illness exclusion was not clear. While I do not need to make a finding on the question, having regard to the evidence discussed below as to QBE's approach to mental illness related claims, I anticipate that QBE would not agree to provide cover for a mental illness even on payment of a higher premium.
2. As in this case Ms Ingram was refused indemnity for her cancellation related costs only and not for any medical or other expenses, cancellation cover has been the focus of my reasons.

1. As summarised above, Ms Ingram said that she was discriminated against in two ways.
2. The first claim is that, by including the mental illness exclusion in the policy issued in respect of Ms Ingram and so excluding persons with the disability of mental illness from indemnification, Ms Ingram was treated unfavourably because of her disability. That is said to be a contravention of section 44(1)(b) of the EOA.[16]
3. The second claim is that, by refusing to indemnify her on the basis of her mental health condition, QBE treated her unfavourably because of her disability. The unfavourable treatment was said to be QBE's refusal to provide her the service of indemnity or cover under the policy, in breach of section 44(1)(a) of the EOA.[17]

1. Because Ms Ingram did not have the depressive illness when the policy was issued on 8 December 2011, QBE contended that she did not at that time have an attribute protected by the EOA and so the first claim could not succeed. Ms Ingram contended that the definition of 'disability' extended to her circumstances.
2. Section 4 of the EOA says disability means:

(b) the presence in the body of organisms that may cause disease; or

(c) total or partial loss of a part of the body; or

(ii) a condition or disorder that results in a person learning more slowly than people who do not have that condition or disorder; or

(e) malformation or disfigurement of a part of the body -

and includes a disability that may exist in the future (including because of a genetic predisposition to that disability) and, to avoid doubt, behaviour that is a symptom or manifestation of a disability (my emphasis).

1. I was provided with a copy of the Explanatory Memorandum for the 2010 Bill which introduced this definition. It does not assist in explaining what these additional words were intended to include. QBE referred to the statement of compatibility tabled in compliance with the Charter of Human Rights and Responsibilities Act 2006 (Vic) (Charter) and the final report which led to the introduction of the EOA.[18] Neither made reference to or included a recommendation regarding this part of the definition.
2. Both parties contended that the definition was probably amended to make the EOA definition more consistent with that which applies under the DDA and contains the same words.[19] The Explanatory Memorandum for the Commonwealth Bill explained that the definition:

. . . also includes a disability which presently exists, that existed in the past and one that may exist in the future. This is designed to include persons whose disability is not yet apparent but which may occur at some time in the future.

1. Ms Ingram said that the intention is for the definition to apply notwithstanding the fact that the person's disability may not be apparent to that person or to anyone else. She further contended that the future disability is not limited to disabilities arising from a genetic predisposition as that is included by way of example only. Ms Ingram said that, at the time the policy was issued, she was a person for whom the mental health condition 'may exist in the future' so, at all relevant times, she had the attribute of disability.[20]
2. QBE contended that the DDA materials indicate that the words are intended to capture persons who already had a disability but whose disability was not manifest or was, at the relevant point in time, hidden. QBE noted that the mere fact that a disability onsets at a particular point in time does not mean that it was diagnosable or in existence at an earlier point in time. It said that medical evidence would be required to prove that a person was suffering from an undiagnosed injury or illness at an earlier point in time. QBE submitted that the words are intended to ensure that actual or existing disabilities that are hidden are captured but the words are not intended to capture disabilities that do not exist. It contended that, where there was no evidence that Ms Ingram was suffering from a disability in December 2011 and it is not clear whether her feelings in January 2012 were symptoms of diagnosable depression, it cannot be inferred that she was suffering from that disability at a point in time earlier than when she was diagnosed in early February 2012.
3. QBE referred briefly to the Productivity Commission Report Review of the Disability Discrimination Act 1992 which made a recommendation to include the reference to genetic predisposition in the DDA definition. The report said in the overview:

The broad definition helps avoid genuine complaints of discrimination falling at the first hurdle - determining whether or not the person concerned is covered by the DDA. This helps focus attention on the discriminatory action rather than the person concerned.[21]

1. The report, when summarising the meaning of 'disability' said it included 'disabilities that people have now, have had in the past, might have in the future or are believed to have' (my emphasis).[22] The word 'might' is consistent with reading the definition to refer to a disability which, at the relevant time, may not exist. The latter highlighted words are consistent with the notion that the DDA (and the EOA) provide protection where an incorrect assumption is made that a person has a disability. That concept sits comfortably with the contention that the definition is broader than existing disabilities.
2. QBE acknowledged the need to be mindful that the EOA and the DDA are beneficial legislation and should not be read narrowly, but submitted the proper construction of the words cannot supervene or replace the proper meaning where that meaning is clear.[23]
3. The applicant's submissions drew my attention to the Court of Appeal's recent analysis of the interpretative task required by section 32 of the Charter.
4. As a consequence of section 32 of the Charter I am required to, so far as it is possible to do so consistently with their purpose, interpret the relevant provisions in a way that is compatible with human rights. I must undertake that task having particular regard to the section 8 Charter right which provides that every person is equal before the law and entitled to the equal protection of the law without discrimination and has the right to equal and effective protection against discrimination.
5. In summary, section 32 calls for the following approach to the interpretation of legislation. If the words of the statute are clear then they must be given that meaning. If the words are capable of more than one meaning, they must be given whichever meaning accords best with human rights. Section 32 does not authorise a process different from the established process of interpretation: it may serve as a guide as to which of two possible constructions is to be preferred but does not allow the reading in of words which are not explicit or implicit in a provision, or the reading down of words so far as to change the true meaning. The task is to discern the purpose of the provision in question in accordance with the ordinary techniques of statutory construction as discussed in Project Blue Sky v Australian Broadcasting Authority.[24] The statute is to be construed against the background of human rights and freedoms set out in the Charter. [25]
6. I start with the proposition that, where the meaning of the words of a statute are clear, that is likely to reflect their correct meaning and the meaning Parliament intended.[26]
7. The words 'and includes a disability that may exist in the future' on their face extend to any disability that may exist at any time in the future. The example which follows is not exclusive or limiting. Assuming it can add some meaning to the preceding words, the example would tend to support Ms Ingram's proposed reading of the provision. By its nature a genetic predisposition is not a certainty - it might say something about the likelihood or otherwise of a person developing a disability but does not say anything about whether they currently have the disability.
8. I also note the word 'may' which appears in the phrase 'the presence in the body of organisms that may cause disease'. That phrase has the effect that the mere presence of organisms which may later cause disease is sufficient to establish a disability under the EOA.
9. The Commonwealth Explanatory Memorandum referred to persons 'whose disability is not yet apparent but which may occur at some time in the future'. It seems to me that sentence could be read two ways. On one reading, it could mean that a disability occurred in the future but was not at the relevant time apparent: that is it existed but was not diagnosed. Another reading would give greater emphasis to the words 'may occur', suggesting that there is the possibility of a future disability and that, at the relevant point in time, it is not apparent that it will occur.
10. I have some sympathy with QBE's proposed reading of the definition. However, where there are no extrinsic materials to assist in understanding Parliament's intentions, a Charter consistent interpretation would be to read the words widely and without limitation. That would be in keeping with the EOA objective of eliminating discrimination to the greatest possible extent by ensuring that all persons with disabilities, past, current or future, may rely on its protections.
11. Accordingly, I find that the definition of disability applies to a person at a point in time when a future disability did not yet exist or had not been diagnosed, was not known to the person or others or was not otherwise apparent.
12. On the undisputed evidence that, in around February 2012 Ms Ingram was diagnosed with a depressive illness which falls within the definition of 'disability' under the EOA, I find, as at December 2011, Ms Ingram was a person for whom a disability may exist in the future.
13. I will now turn to the discrimination claim relating to the inclusion of the mental illness exclusion in the policy.

1. It was contended for Ms Ingram that the inclusion of the mental illness exclusion can be compared to policies which mandate differential treatment on the express basis of a protected attribute. She said that, in cases of treatment of that kind, it has been held that it is self-evident that the differential treatment is because of the attribute.[27] The written submissions said that QBE was aware that persons with the relevant attribute may apply for cover under the policy whether the condition applied at the time or in the future. Its express purpose was to ensure that no person with a mental illness could make a claim based on that illness. Given the policy includes a pre-existing condition exclusion, it was said to be self-evident that the clause excluding all claims arising directly or indirectly from mental illness was intended to apply whether or not the policy holder had that illness at the time of applying.[28]
2. QBE's position is that, as on the proper reading of the definition 'disability', Ms Ingram did not have an attribute as at December 2011, the claim of discrimination cannot be proven because there was no treatment or proposed treatment of her because of an attribute. Further, QBE said that, prior to the refusal of indemnity in August 2012, it did not treat Ms Ingram unfavourably and there could be no such treatment earlier than the date the claim was made, 22 May 2012.

1. Having concluded that Ms Ingram had a protected attribute as at December 2011, the issue is whether the inclusion of the mental illness exclusion in the policy issued to Ms Ingram was unfavourable treatment or proposed unfavourable treatment because of that attribute.
2. On the evidence of Mr M De Luca, National Manager - Sales, Distribution and Strategy with QBE, the consequence of the mental illness exclusion is that all claims based on a mental illness, as defined in the policy, are refused. That approach is taken irrespective of whether the pre-existing condition exclusion could also apply. The mental illness exclusion is used as a blanket response to all claims based on a mental illness.[29]
3. The definition of pre-existing condition extends to a range of circumstances which include where a person knew or ought to have known they had a mental illness. If one assumes that a person with no prior history of mental illness is not excluded on the basis that they knew or ought to have known they had a temporary disturbance of mood (see policy definition of mental illness), it is clearly unfavourable for them to find that the entirely unexpected later development of a mental illness is a basis for indemnity being refused.
4. These comments lead on to the question of whether the claim is best characterised as direct or indirect discrimination. I accept Ms Ingram's contentions that the inclusion of the mental illness exclusion can be raised as an act of direct discrimination.
5. Arguably, the mental illness exclusion which applies to all people who take up an offer of insurance based on the particular standard form policy, has the facially neutral characteristics of a requirement, condition or practice which arises under indirect discrimination. However, the point of the mental illness exclusion is that it will be applied to anyone who first develops that particular disability after the policy is issued. The nature of the disability people in that category have is their future mental illness. The exclusion is directed at them and relied upon when a claim is made because they developed that condition. By including the exclusion in the policy, QBE proposes to treat unfavourably any person who in the future, during the life of the policy, develops a mental illness. In a sense, the exclusion is more targeted than the requirements, conditions or practices which are considered under indirect discrimination.
6. In Xiros v Fortis Life Assurance Ltd, Mr Xiros did not know that he had HIV/AIDS at the time he was issued with mortgage protection insurance which expressly excluded claims based on that disability. The issuing of a policy which included the HIV/AIDS exclusion at a time Mr Xiros was unaware of his illness was found to be an act of direct discrimination. The Federal Magistrate found that it was unnecessary to have regard to the alternative claim of indirect discrimination. [30]
7. I accept it might be contended that, at the time at which Ms Ingram did not have the depressive illness, the existence of the mental illness exclusion in the policy could have only a theoretical effect. Had she not developed the depressive illness but instead suffered a physical injury which required her to cancel her trip, the exclusion would not have been relied upon and so, like the many other exclusions, it would have remained a possibility only.
8. I am satisfied and find that, as a person who falls within the definition of 'disability' as she developed a mental illness at a later point in time, the inclusion of the mental illness exclusion was proposed unfavourable treatment of Ms Ingram because of her attribute.
9. I find that the inclusion of the mental illness exclusion was a contravention of section 44(1)(b) because it was an act of discrimination in the terms on which goods and services were provided to Ms Ingram.

1. Ms Ingram's second claim is that, by refusing to indemnify her on the basis of her mental health condition, QBE treated her unfavourably because of her disability. The unfavourable treatment was said to be QBE's refusal to provide her the service of indemnity or cover under the policy, in breach of section 44(1)(a).
2. As noted earlier, QBE accepts that when it declined indemnity to Ms Ingram it refused to provide a service within the meaning of section 44(1)(a) of the EOA and that it was unfavourable treatment. QBE said that was not unlawful discrimination because both or either of the two exceptions relied upon apply.[31]

1. Where there is no dispute that, at the time QBE denied cover, Ms Ingram had a mental illness and QBE relied on that disability as the basis on which the exclusion was said to apply, I am satisfied and find that QBE refused to provide the service of insurance cover and did so because of Ms Ingram's attribute.
2. Accordingly, I find that QBE engaged in direct discrimination contrary to section 44(1)(a) of the EOA.

Section 47 exceptions

1. Section 47 of the EOA allows for discrimination in respect to insurance policies in certain circumstances. Section 47(1) says relevantly:

An insurer may discriminate against another person by refusing to provide an insurance policy to the other person, or in the terms on which an insurance policy is provided, if -

(ii) the Disability Discrimination Act 1992 of the Commonwealth; or

(i) is based on actuarial or statistical data on which it is reasonable for the insurer to rely; and

(ii) is reasonable having regard to that data and any other relevant factors . . .

1. QBE is an insurer for the purposes of section 47 and the travel insurance policy was an insurance policy within the section.
2. Section 46 of the DDA includes an exception to discrimination in respect to insurance framed in largely the same way as section 47(1)(b). Section 46(2)(f) of the DDA says:

This Part does not render it unlawful for a person to discriminate against another person, on the ground of the other person's disability, in respect of the terms or conditions on which:

(c) a policy of insurance against accident or any other policy of insurance;

is offered to, or may be obtained by, the other person, if:

(i) is based upon actuarial or statistical data on which it is reasonable for the first-mentioned person to rely; and

(ii) is reasonable having regard to the matter of the data and other relevant factors . . .

1. As a consequence of section 47(1)(a)(ii), QBE may rely on section 46(2)(f). That section also allows QBE to rely on section 29A of the DDA. That section says, in summary, that it is not unlawful to discriminate if avoiding that discrimination would impose an unjustifiable hardship. The term 'unjustifiable hardship' requires consideration of a number of specific matters I will set out later.
2. QBE carries the burden of proving one or more of the defences it relies upon applies. The standard of proof is the civil standard.

1. QBE made closing submissions regarding the proper interpretation of the exceptions relied on and how they ought to be applied to the evidence before me.
2. In addition and more broadly, QBE contended that the observations made by the High Court of Australia in Australian Mutual Provident Society v Goulden and Ors,[32] as to the legislative policy underpinning the Life Insurance Act 1945 (Cth) are relevant as they 'clearly' informed the exception to discrimination contained in section 46 of the DDA regarding actuarial or statistical data. The observations were to the effect that, the life insurance business of a company is more likely to prosper and the interests of its policy holders are more likely to be protected, if it is permitted to classify risks and fix rates of premium in that business in accordance with its own judgment founded upon the advice of actuaries and the practice of prudent insurers. That was said to be the legislative policy underpinning the Life Insurance Act 1945 (Cth).
3. QBE submitted that, having regard to these matters, the High Court had made it clear that the interests of policy holders other than the insured whose coverage is in question, is integrated with the interests of insurers such as QBE. Further, QBE, as a corporation which exists to serve the interests of its shareholders, cannot serve those interests 'if it cannot honour the terms of insurance policies on issue to persons not in the same category as the insured whose coverage is in issue, because it is forced to make a loss due to the operation of a prohibition on disability discrimination. The interests of policy holders and shareholders must be observed by the Tribunal when it is considering whether [QBE] has proved its statutory defences.' [33]
4. Ms Ingram's written submissions contended that the decision in Australian Mutual Provident Society might have limited effect here given it was concerned with the highly regulated field of life insurance. In any event, she said it emphasised the wording of the actuarial or statistical data exception in that the High Court made it clear that policy conditions were to be based on a rational and considered analysis.[34]
5. I accept that the High Court's observations regarding reliance on the advice of actuaries and the practice of prudent insurers are consistent with the section 47(1)(b) EOA and section 46(2)(f) DDA exceptions.
6. In considering whether section 29A of the DDA applies, I may have regard to all relevant circumstances including QBE's financial circumstances and the estimated amount of expenditure required to be made by QBE. That touches on the interests of other policy holders and shareholders.
7. On the question of interpretation of the exceptions in issue, I have been mindful that one of the objectives of the EOA is the elimination of discrimination to the greatest extent possible.[35] I have also been mindful of the section 32 Charter obligation discussed above.

1. The introductory words of section 47(1) say as follows:

An insurer may discriminate against another person by refusing to provide an insurance policy to the other person, or in the terms on which an insurance policy is provided, if . . .

1. Some preliminary observations may be made.
2. The term 'discriminate' must be read in the light of section 7 of the EOA which, relevantly here, says that discrimination means direct or indirect discrimination. As discussed above, direct discrimination means proposed or actual unfavourable treatment because of an attribute.
3. Two kinds of discriminatory conduct are covered by section 47. The first occurs by an insurer 'refusing to provide an insurance policy to the other person'. The second is by the insurer discriminating 'in the terms on which an insurance policy is provided'.
4. The meaning of the words 'refusing to provide an insurance policy to the other person' was uncontroversial and did not arise here.
5. The words 'in the terms on which an insurance policy is provided' is relevant to the first claim of discrimination.
6. In respect of the second claim, Ms Ingram contended that section 47 cannot apply to the refusal of indemnity because that was neither a refusal to provide a policy or an act relating to the terms on which the policy is provided. The submissions said that, consistent with the words of the section, the point in time when the exception must be proven to apply was the time when the policy was provided. She said that, if it is found that the terms of the policy limb of section 47(1) can apply to acts which post-date the contract date, it must be shown that the specific acts of discrimination can be justified in respect of Ms Ingram personally.[36]
7. QBE said that it is clear from section 47(1) that it concerns an insurance policy which is a contract containing terms at the time it is agreed and which have an ongoing existence for the duration of the policy. It said that it relied on the mental illness exclusion when it refused indemnity and that, prior to the time the claim was made, the exclusion had no work to do. The refusal was conduct that relied upon or was in compliance with a term of the insurance policy. There was no amendment to the policy. It was said to be conduct in the terms on which the policy was provided to Ms Ingram.[37]
8. There have been no decisions on this issue under the EOA and few under the equivalent provision of the DDA. Section 46(2)(f) of that Act is slightly differently worded, referring to discrimination in respect of the terms and conditions on which an insurance policy 'is offered or may be obtained'. This timing issue was not considered or apparently regarded as a bar to reliance on the exception in Xiros.[38] If anything, the introductory words of section 46(2)(f) might be regarded as being directed more to the time the policy is purchased rather than during its lifetime.
9. It seems to me more likely that a proper reading of the introductory words of section 47 would allow for the exceptions to be relied upon in respect of the terms offered and also reliance on those terms at a later point in time given that, by its nature, an insurance policy represents an ongoing relationship between the parties and its terms make provision for future circumstances. As will become apparent, this matter is not determinative because I have found none of section 47(1)(b) of the EOA, section 46(2)(f) of the DDA or section 29A of the DDA can be relied upon by QBE.

1. Section 47(1)(b) of the EOA and/or section 46(2)(f) of the DDA will apply where it is proven that the act of discrimination was based on actuarial or statistical data on which it was reasonable for the insurer to rely and that the reliance was reasonable having regard to the data and other relevant factors. There was no dispute that the reference to the discrimination being 'based' on the relied upon data means that the discriminator must have actually based its decision on that data and so, by implication, the data must have been in existence at that time.[39] The question whether the discrimination is reasonable having regard to the data and any other factors is to be considered giving the words their ordinary meaning and the context of the case.[40]

1. The only actuarial data produced to the tribunal was a report prepared by Associate Professor David Pitt and Dr Timothy Kyng dated 31 August 2015. As that report was prepared for the purposes of this litigation, clearly it was not available to QBE at the time it made decisions regarding the content of the policy and Ms Ingram's claim for indemnity, and so it cannot be relied upon under section 47(1)(b) of the EOA or section 46(2)(f) of the DDA. QBE accepted that it had no actuarial data to rely on in respect of the inclusion of the mental illness exclusion in the policy.[41]
2. QBE relied upon the following statistical and other reports:
   • The National Survey of Mental Health and Wellbeing: Summary of Results based on a 2007 Australian Bureau of Statistics (ABS) national survey (ABS survey summary);
   • A December 2009 Access Economics Pty Ltd paper entitled The economic impact of youth mental illness and the cost effectiveness of early intervention (Access Economics paper);
   • A May 2007 paper written entitled The Burden of Disease and Injury in Australia 2003 (Burden of Disease paper);
   • A beyond blue report entitled beyond blue Depression Monitor Independent Findings from 2004 to 2012;
   • A 2010 guide (apparently published in September 2010) entitled Mental illness and life insurance: What you need to know - a detailed guide (2010 guide);
   • An October 2012 report entitled Opening Eyes and Opening Minds: The Ontario Burden of Mental Illness and Addictions Report (Ontario report);
   • A copy of an Australian Financial Review newspaper article published on 29 June 2013 entitled Life insurers hit by mental health claims;
   • The Australian Institute of Health and Welfare 2014 report Mental Health Services In Brief (the AIHW 2014 report); and
   • Australian Institute of Health and Welfare data regarding services for 2007-2008 to 2012-2013.
3. The Pitt/Kyng report referred to the ABS survey summary and the Ontario report and listed the AIHW 2014 report as a reference. During the hearing I was taken to parts of the ABS survey summary, the Access Economics paper, the 2010 Guide and the AIHW 2014 report. Each of the above documents were annexed to an affidavit sworn on 12 May 2015 by Mr De Luca.
4. Mr De Luca placed particular emphasis in his affidavit on the following statistics:
   • At least 45% or 7.3 million Australians will have some mental disorder at some point in their lifetime; and
   • Nearly a quarter of Australian youth aged between 15 and 25 have anxiety, depression or substance abuse disorders.
5. In his evidence, Mr De Luca referred to another statistic relating to the number of people who reported illness. Mr De Luca said his memory from the ABS summary survey was that 65% of people do not report a mental health condition.[42] In fact the relevant statistic in the ABS summary survey was that, of those surveyed who self-reported mental illness in the preceding 12 months, only 35% had accessed services. In cross-examination, Mr De Luca accepted the proposition that the other 65% might not have accessed services because they were feeling well and did not require access to services.[43]
6. In its closing submissions, QBE set out a number of statistics which come from the three reports which were available as at March 2010, namely the ABS survey summary, the Access Economics and Burden of Disease

1. papers.[44] It submitted that I was obliged to have regard to all of the evidence before me, where there was no disconformity between the evidence and the particulars of defence, relying on the High Court's decision in Dare v Pulham.[45]
2. That case considered the extent to which a jury verdict may be relied on where it was based on evidence adduced at trial which differed from the pleadings. The High Court found that, where the evidence in issue had not and could not be objected to, once the issue of general damages was submitted to the jury, it was a matter for it and the fact the damages which were the subject of the evidence did not correspond with the damages as claimed in the pleadings was not a basis to overturn the jury's findings.
3. I accept that it might be open to me to proceed on the basis that the statistics relied on extended beyond the three set out above and include those contained in the various reports and publications annexed to Mr De Luca's affidavit. However, where the tribunal is not a court of pleadings and where the content of most of the reports tendered was not the subject of any oral evidence or discussion by Mr De Luca or Associate Professor Pitt, I do not consider I am bound to review lengthy documents to which no specific reference has been made in order to locate and give weight to material which assists QBE. This matter was discussed at the hearing.[46]
4. In the closing written submissions QBE acknowledged that 'there is a paucity of evidence to show a direct or specific link between [QBE's] decision to incorporate a general exclusion for mental illness into the travel insurance policy effective from 22 March 2010 and the statistical data referred to by Mr De Luca and exhibited to his affidavits and all the statistical data referred to above.' QBE further acknowledged that there was an absence of evidence which demonstrated that it made the decision to incorporate the mental illness exclusion on the basis of contemporaneous actuarial data.[47]
5. QBE invited the tribunal to infer that the statistical data relied on by QBE was taken into account by it and did provide the basis for its decision to include the exclusion within the travel policy issued to Ms Ingram. QBE said that inference should be drawn for four reasons:
   • First, the ABS survey summary, the Access Economics paper and the Burden of Disease paper were in the public domain and therefore were available to QBE for a considerable period of time before the policy was issued and became effective on 22 March 2010;
   • Second, the ABS survey summary was said to be a seminal document in relation to the prevalence of mental illness in the Australian community. The submissions contended that it is highly likely that such seminal data was relied on by QBE's actuaries and underwriters when it decided to incorporate the exclusion into the policy. QBE contended it is well-known that the respondent and other insurers make use of actuarial and statistical data;
   • Third, Mr De Luca's experience that there was a high prevalence of mental illness in the community was likely to have been informed by the data in the ABS survey summary and the other reports referred to in his affidavit. The three reports mentioned above were published during his time working as a manager in QBE's travel insurance business; and
   • Fourth, QBE lacked a history of mental illness claims and that explains why the exclusion was not based on actuarial data. The absence of actuarial data arises because mental illness claims are excluded from cover and so insurers do not have complete actuarial data of the number of claims made per annum. QBE contended that it was highly unlikely that the mental illness exclusion had no rational basis. Accepting that the exclusion had a rational basis and that it was not based on actuarial data, QBE said the only other viable and reasonable explanation for the exclusion was that it was based on statistical data. If that was correct, QBE said it was highly likely that the relevant statistical data was the ABS survey summary as that report was the most recent, most comprehensive and most reliable data available to QBE at the time the policy was formulated. The submissions said 'it is entirely sensible that these statistics should have informed the mental illness general exclusion.'[48]
6. Mr De Luca gave evidence that he was not involved in giving any instructions as to the content of the policy in this case. He was not involved in the refusal of indemnity.
7. Mr De Luca's evidence about the prevalence of mental illness in the community came from his observations of publicity and media announcements. Mr De Luca read the documents attached to his affidavit in around May 2015. He was unable to give any evidence that the terms of the policy or the refusal of indemnity was based on any data. He explained that QBE has no data because, given the mental illness exclusion, it does not receive claims or, if they come in inadvertently or someone believes they have a claim, it will be denied on the basis of the mental illness exclusion. As a consequence, no payments are made and so there is no actuarial data of the number of the mental health claims made per annum. I asked him whether it was correct that his experience of prevalence of claims and the likelihood of success was all based on impressions because, in fact, neither he nor QBE deal with claims because they are all automatically rejected. He agreed that was correct.[49]
8. Mr De Luca's affidavit also attached copies of policies issued by other insurers. A number, but not all, contained an exclusion like that which is in issue in this proceeding.
9. In the final minutes of the hearing, the tribunal was told that the author of the December 2012 letter which confirmed indemnity would be refused, Ms Genevieve Staff, relied on a document or report when she wrote confirming that indemnity would be denied on the basis of section 47 of the EOA. While that document seemed to be in the physical possession of QBE's instructing solicitor it had not been produced at any time. It was submitted for Ms Ingram that, as Ms Staff is still employed by QBE and was present throughout the hearing, an inference should be drawn to the effect that her evidence would not assist QBE's case.[50]
10. Given that the document in issue was never produced, I am unaware of whether it contained actuarial or statistical data. In those circumstances, the only evidence I had was from Mr De Luca and that was that he was unaware of any actuarial or statistical data other than the Pitt/Kyng Report. I give the possibility that there was a document which was relied on in December 2012 no weight when considering this issue.
11. In the reply submissions, the applicant submitted that there was simply no evidential basis for the inference sought by QBE to be drawn. It was contended that to draw inferences of fact from other facts it must be reasonably probable that a fact exists, not that it was simply a possibility. This was said to amount to an invitation to engage in impermissible conjecture and likely to lead me into error.[51]

1. Curiously, QBE did not produce any evidence about the date the policy or similar polices first contained the mental illness exclusion. Accordingly, I am unaware whether around March 2010 was the first time QBE turned its mind to such a term or whether it had been a term of comparable policies earlier. As a consequence, it is not clear whether the first time QBE decided to include the mental illness exclusion in its standard travel insurance policies was around March 2010. It might have been included in earlier versions and carried over to the March 2010 policy. That raises a question as to whether I could infer that, at the time the mental illness exclusion was introduced into the March 2010 policy, the statistical material relied on was in existence.
2. For the purposes of this discussion, I will assume that in March 2010 the inclusion of the mental illness exclusion was the subject of some consideration by QBE.
3. There was no evidence that QBE separately turned its mind to statistical data at the time it refused Ms Ingram indemnity.
4. I am willing to infer that, at the time QBE decided to include the exclusion in the policy, it had a reason for doing so.
5. Consistent with the High Court's observations in Australian Mutual Provident Society, I would accept that insurers such as QBE rely upon actuarial and underwriting advice. I could infer that advice of that kind was available at the relevant time. I would expect that legal advice might have been sought about either the specific exclusion or the policy generally before it was released. That would be consistent with what would be expected from a large insurer like QBE and the practices of a prudent insurer.
6. QBE asks me to go further, though, and infer that specific information was in fact considered by those advising on the mental illness exclusion in around March 2010, notably the ABS survey summary. As six out of the nine documents annexed to Mr De Luca's affidavit were produced after the policy wording had taken effect, clearly I cannot infer that they had been taken into account when the mental illness exclusion was formulated.
7. It might equally be open to me to infer that, because a number of other insurers had added a general exclusion in similar terms to their policies, QBE did the same for commercial reasons. Of the policies in evidence, half contained a similar mental illness exclusion. QBE described the inclusion of the mental illness exclusion as general industry practice at the time.[52]
8. Alternatively, it might be open to infer that officers of QBE decided to introduce the mental illness exclusion because, like Mr De Luca, they had their own personal perceptions of the prevalence of mental illness in the community.
9. I accept QBE's submissions that it is not required to meet the so-called Briginshaw standard in order to prove the exceptions it relies on. The usual civil standard applies and so it must demonstrate that it is more probable than not that, at the time the mental illness exclusion was being considered, regard was had to the range of statistics relied on. It is another matter however, to ask that I draw an inference of the kind described as the basis for proof of the exception when it is unsupported by any evidence. I accept the applicant's written submissions on this matter.
10. In circumstances where QBE was served with the application on around 3 June 2014 and filed its defence in September 2014, it ought to have been in a position to produce evidence about the drafting and approval of the policy in issue. Instead, it asks me to infer that evidence of that kind existed. There is no sufficient evidentiary basis for the inference to be drawn. Further, it would be a gross unfairness to the applicant to draw this inference given the time this proceeding has been running and the many opportunities QBE has had to produce evidence.[53]
11. The exception contained in section 47(1)(b)(i) of the EOA and section 46(2)(f)(i) of the DDA may be relied on provided that, at the time the relevant policy containing discriminatory terms was offered, that discrimination was based on statistical data on which it was reasonable to rely. Where QBE has not produced evidence to establish that the statistical data relied upon was in existence when the terms of the policy were formulated or that any person involved in the drafting or approval of the policy wording had any knowledge of or regard to those statistics, there is insufficient evidence available on which I could safely draw the inference sought. The same position applies in respect of QBE's decision to refuse to indemnify Ms Ingram. Accordingly, I find that QBE has not proven that section 47(1)(b)(i) of the EOA or section 46(2)(f)(i) of the DDA applies.
12. As both limbs of section 47(1)(b) of the EOA and 46(2)(f) of the DDA must be proven, the defence has not been made out. Accordingly, I find QBE cannot rely on the exceptions contained in those provisions for either claim of discrimination.
13. For the sake of completeness, I will comment on the reasonableness question raised by this defence. QBE contended that, if it is inferred the three reports were relied on by it when deciding on the form of the mental illness exclusion, it is open to also find that reliance was reasonable given that:
    • The three reports were published by reputable sources;
    • The ABS survey summary is a seminal report, reflective of a national survey and highly reputable;
    • The methodology described in the ABS survey summary is reliable and well recognised;
    • The statistical data in the ABS survey summary was the most up to date;
    • The statistics in the ABS survey summary are highly applicable to the question of the provision of insurance cover for mental illness; and
    • There is no evidence that the data has been discredited. [54]
14. QBE contended that the exclusion was reasonable in the light of the statistical data and other factors having regard to the Pitt/Kyng report, Mr De Luca and Associate Professor Pitt's evidence, which was said to show that the travel insurance business would be unprofitable if the exclusion could not be relied upon. The evidence relied on for this limb of the defence included that discussed below as to the likely losses to be incurred if the mental illness exclusion could not be relied upon, the acceptability of using a time frame for analysis which post-dated the events in issue, the role to be played by the pre-existing condition clause if the blanket exclusion was removed and industry practice. Reliance was also placed on the fact that, as Ms Ingram had not had any episode of mental illness at the time the policy was issued, QBE could not make an assessment of the risk of a claim. Further, as QBE had no data on claims of that kind, it could not make use of data to assess risk more generally. In those circumstances, this case was unlike other cases considered under the DDA exception where data was available.[55]
15. As will become apparent from the discussion below, I am not persuaded that the particular methodology and statistics from the ABS survey summary which were relied on by Pitt/Kyng when preparing their report and which were referred to in Mr De Luca's affidavit supported the conclusions I was asked to adopt. Further, as discussed earlier, QBE's contentions on the question of reasonableness depend on the consideration of a range of statistics and matters which were not the subject of direct evidence. I am not persuaded that the material before me was sufficient to meet the section 47(1)(b)(ii) EOA and section 46(2)(f)(ii) DDA reasonableness tests.
16. I will now turn to the exception contained in section 29A of the DDA.

1. QBE can seek to rely on section 29A of the DDA via section 47(1)(a)(ii) because, if the discrimination is permitted by the DDA, it is permitted under the EOA.
2. Section 24 of the DDA contains a prohibition on discrimination in respect of services equivalent to that found in section 44 of the EOA. Section 29A of the DDA provides that provisions included in Division 2 (including section 24) do not render it unlawful for a person (the discriminator) to discriminate against another person on the ground of a disability of the other person if avoiding the discrimination would impose an unjustifiable hardship on the discriminator.
3. Section 11(1) sets out matters to be taken into account when determining whether a hardship would be unjustified. I will come to those later_._
4. Section 11(2) says that the burden of proving that something would impose unjustifiable hardship lies on the person claiming unjustifiable hardship, here QBE.
5. It is apparent from the terms of section 29A that some hardship is justifiable. In order to determine whether that hardship is unjustifiable turns on the nature and degree of the hardship in the context of the section 11 DDA factors and any other relevant circumstances. A financial burden may be justified, given the objectives of the DDA in respect to the elimination of discrimination as far as possible. While the financial burden which may be imposed will be relevant, it is not the only factor to consider. If the financial burden is minor then it is not likely to fall within the exception. If, on the other hand, it is very significant and might lead to the relevant entity not being financially viable, then the exception is more likely to apply. What is required is for an assessment to be made of whether some decision or action might be taken to avoid the discrimination and whether it would impose an unjustifiable hardship.[56]
6. QBE relied on the Pitt/Kyng report to prove that unjustified hardship would be imposed on QBE if it was unable to rely on the mental illness exclusion. I will come to the preparation and content of the report soon.
7. Ms Ingram submitted that section 29A requires the focus to be placed on the actual acts of discrimination and whether avoiding that particular discrimination, rather than 'any general forward looking question of unjustifiable hardship'. She said that it is plain that QBE could not demonstrate it would have suffered unjustifiable hardship had it paid Ms Ingram's claim.[57]
8. QBE contended that the manner in which the alleged discrimination is particularised informs how the question of unjustifiable hardship is answered. Having regard to the claims made, QBE said that the inquiry is not confined to the omission of the mental illness exclusion in the contract between itself and Ms Ingram only. Because the exclusion is of a general nature and was intended to apply to everyone, the question is whether QBE would suffer unjustifiable hardship if the mental illness exclusion was omitted from all policies issued. That requires consideration of the ongoing impact of the removal of the clause. It contended that was how the claim had been made and so the suggestion that the inquiry should focus on unjustifiable hardship if that clause was not relied on in respect of Ms Ingram only was incorrect.
9. QBE further submitted that the section 29A inquiry is not directed towards a hypothesis about the inclusion of a more limited form of the mental illness exclusion. Reference was made to circumstances where the insured had a pre-existing illness which could be the subject of individual and careful risk evaluation including by considering previous claims relating to the same or similar medical conditions. QBE contended that such an approach is not open here because Ms Ingram did not have a pre-existing illness and QBE had no claims history for mental illness against which it could undertake an assessment. The submissions said that '[i]t is not possible to conceive of an hypothesis that involved [QBE] assessing [Ms Ingram's] risk of becoming mentally ill in the absence of a pre-existing mental illness. The only workable hypothesis to be addressed in relation to s 29A involves a consideration of the wholesale removal of the mental illness general exclusion.'[58]
10. I was referred to the Federal Court's decision in King v Jetstar Airways Pty Ltd (No 2).[59] That case considered whether unjustifiable hardship would be caused if, contrary to its existing policy, Jetstar was required to allow more than two passengers who required wheelchair assistance to travel on that flight. The court described in some detail Mrs King's specific requirements and how they might have been accommodated.[60]
11. A time and motion study conducted by Jetstar and relied on in King was more generalised. On the face of the description in the reasons, the Jetstar study looked at passengers who were confined to a wheelchair, passengers who could climb some stairs and others who could not. The study also considered passengers who required assistance to move from the chair to the cabin seat. Averages were used and the study reported on the time taken and assistance required for the range of passengers.[61] The court used those statistics and other material to consider the flow on consequences for flights and flight turnaround times and cost presuming that the two wheelchair dependent passenger limit would be removed. That broader sense of unjustifiable hardship was in issue, not whether allowing Mrs King to travel on the particular flight would have had that outcome.[62]
12. One of the grounds of Mrs King's appeal to the Full Federal Court was that Robertson J had mischaracterised her claim and so had erred in how he assessed whether the unjustifiable hardship exception had been proven. The Full Federal Court rejected that contention, primarily because Mrs King's claim had proceeded on the basis that the two wheelchair dependent passenger limit should be removed. Mrs King had sought a declaration to that effect.[63]
13. I accept QBE's submissions that this claim has proceeded on the basis that the mental illness exclusion ought to be removed from the policy.[64] That is consistent with the particulars of claim, the declaration Ms Ingram has sought and the way the case was run. That is also consistent with the written submissions made: in her closing written submissions, Ms Ingram's written submissions contended that a reason for making a declaration was because the mental illness exclusion may affect many consumers, including where QBE's policy still contains the exclusion.[65]
14. That broader approach is particularly pertinent in this case. The specific act of discrimination against Ms Ingram arose from a general exclusion clause. There were no negotiations about the content of the policy at the time it was issued. Although specific information was sought about Ms Ingram's illness when she made the claim on the policy, there was no suggestion that QBE was at any time contemplating not relying on the blanket clause. At all relevant times, the parties have proceeded on the basis that was how the clause was intended to be used and how it was applied in this case. Given that by its nature the clause would apply to all people who have a mental illness and make a claim on the policy, in order to consider whether unjustifiable hardship would have accompanied QBE not relying on the clause, I consider it appropriate to proceed on the basis that the clause be removed from all policies.
15. Even if I am wrong in that approach, it would not change the outcome of this claim. That is because, if it is correct that the only question is whether QBE would have suffered unjustifiable hardship by paying only Ms Ingram's $4,292.48 claim, the answer would clearly be no. Having found that QBE has not proven to the required standard that unjustifiable hardship would follow the entire removal of the clause, Ms Ingram succeeds in any event.
16. I accept that there is no evidence to show that, at the time the contract of insurance was entered into, a modified version of the mental illness exclusion could have been applied. I accept that, when considering whether section 29A applies, it is appropriate to proceed on the basis that it is omitted from the policy. However, other clauses of the policy must be treated as continuing to apply and might be relevant to the question of unjustifiable hardship.
17. There was a dispute about whether evidence based on events which post-dated the discrimination could be relied on to prove unjustifiable hardship. This question arose because the actuarial report relied on by QBE relied on data for the period 1 January 2013 to 1 June 2015 rather than around 2011 to 2012. I will discuss that issue further below.
18. As QBE's actuarial report was the key focus of the hearing, I will turn to that first before addressing the matters expressly raised in section 11(1) of the DDA.

1. On 21 May 2015, QBE's solicitors first wrote to Associate Professor Pitt and Dr Kyng about the preparation of an expert report. On 4 August 2015, they wrote again saying that QBE had given instructions for preparation of the report to go ahead.[66]
2. Associate Professor Pitt and Dr Kyng are both Fellows of the Institute of Actuaries Australia.
3. Associate Professor Pitt is an associate professor in actuarial studies at Macquarie University, NSW. He holds bachelor degrees in economics and science and is a Doctor of Philosophy having written his thesis in the area of income protection insurance. He has had over 17 years' experience in conducting research, teaching and analysis as an actuary. His expertise is in the statistical analysis of data and insurance and related fields. Prior to joining Macquarie University, Associate Professor Pitt lectured at The University of Melbourne and the Australian National University. Prior to his career in academia, Associate Professor Pitt worked for AMP Limited.
4. Dr Kyng is a senior lecturer in actuarial studies at Macquarie University. He holds a bachelor degree in science, masters of statistics and economics and is a Doctor of Philosophy in applied mathematics. He is also a Fellow of the Financial Services Institute of Australia. Dr Kyng has over 25 years' experience in conducting research, teaching and analysis as an actuary. His expertise is in the statistical analysis of data in insurance, finance and related fields. Prior to joining Macquarie University, he held the position of actuary at Coopers and Lybrand Actuarial Services with responsibility for general insurance and financial risk management consulting projects. Prior to that he worked in the actuarial department of the Commonwealth Bank of Australia and earlier worked at the New South Wales State Government Department and GIO.[67]
5. The report, dated 31 August 2015, was served that day. Although the orders for the report required it to comply with the tribunal's practice note regarding expert reports,[68] it was not served in that form. A report which complied with the practice note was provided to Ms Ingram's legal representatives at the commencement of the first hearing day.[69] The content and date of the report remained the same.
6. The report attached a data sheet which was said to have been relied on when the report was prepared. At the commencement of the second hearing day two further versions of that data sheet were produced. The evidence was that the data sheets were prepared by a QBE employee, Mr J Dong, and that, after considering the first version, Associate Professor Pitt and Dr Kyng asked for some of the assumptions to be altered. That led to the preparation of the other versions. After some confusion it was agreed that one of those versions was relied on for the final report.[70]
7. Ms Ingram's written submissions made a number of complaints about the manner in which QBE produced the Pitt/Kyng report. Some of those matters are relevant to the question of costs discussed below. Of more immediate relevance, she complained that, at the time the report was prepared, Associate Professor Pitt and Dr Kyng were not aware of their obligations as experts appearing in the tribunal as they had not been provided with a copy of the practice note at that time. That led to concerns about whether the experts understood their duties to the tribunal. Further, given only Associate Professor Pitt was called, it was contended that it was not possible to be confident about the degree to which Dr Kyng was aware of his duties. The fact only one of the experts gave evidence was said to reduce the value of the opinions contained in the report. There were other concerns about the instructions given to QBE's Mr Dong in respect of the source data relied on for the report.
8. At the hearing, it became apparent that Dr Kyng had worked for QBE in the past but had failed to mention that in the curriculum vitae attached to the final report. It was suggested that he had deliberately excluded that information. Further criticisms were made and it was submitted that it would not be appropriate to give any real weight to the report.[71] Later, the parties agreed that I could have regard to a letter written by Dr Kyng in which he explained that he had inadvertently forgotten to include reference to a period of employment with QBE for three to four months in 1991.[72] That point needs to be taken no further.
9. Where QBE chose to only call Associate Professor Pitt to give evidence, the extent to which the report has been explained and defended in the face of the criticisms of Ms Ingram's expert, stands or falls on that evidence. I proceed on the basis that, when giving evidence, Associate Pitt was aware of his responsibilities to the tribunal as an expert and that he was mindful of them when doing so.

1. Mr Dong commenced work with QBE in its graduate programme in 2012. He holds degrees in commerce and science, with majors in actuarial studies, finance and psychology. For the last two and a half years he has been working in fraud and data analytics.
2. In about June 2015, he was asked by QBE's in-house solicitor to produce general travel data in different dissections, so that information on age of policy holders and information on travel duration could be assessed. He also incorporated data from the ABS survey summary. His calculations were directed at estimating the impact on QBE's profitability if mental illness claims were paid.
3. Those calculations excluded annual policies and so only dealt with policies for specified periods of travel. The calculations took account of a range of matters including the age of policy holders, the likely number of policy holders who had a mental illness (based on the ABS survey summary), the average number of days of travel and days between inception of the policy and departure date, the risk of a mental illness episode prior to departure and during travel and the likelihood of a claim being made (using ABS survey summary statistics). The calculations also took into account the average cost of a travel insurance policy, the average commissions paid on those policies and the average cost of medical cancellation payments.[73] The calculations sought to quantify the likely number of additional claims made and the impact of those claims on QBE's profitability.
4. At Associate Professor Pitt and Dr Kyng's request, Mr Dong excluded claims based on mental illness which might arise from substance abuse only. That was done because there is a specific exclusion for substance abuse.[74] In the first set of calculations he prepared, Mr Dong assumed that 100% of those who had a mental illness within the travel parameters considered would make a claim and that all would be eligible to do so. He was asked to revisit that aspect of the calculations and to instead use a figure of 55%. The likelihood to claim calculations had regard to statistics about time away from work and levels of disability or restriction for anxiety and affective disorders.[75]
5. Mr Dong found that the net combined operating ratio (which took into account the net loss ratio, the additional claims ratio, the net commission ratio and the average expense ratio) was, on the first set of assumptions, 112% and on the revised set, 107%.[76] His evidence was that the ideal operating profit margin was a ratio of 85%.[77]
6. On the question of age of policy holders, no age data was held for around 75% of the policies issued because the travel insurance was issued as an automatic part of a credit card issued by a third party. The data relating to those policies was excluded and so the data related to 25% of QBE's policies over the stated period.[78] The calculations only considered cancellation costs rather than also including repatriation costs which would lead to greater expenses.[79]

1. Associate Professor Pitt and Dr Kyng were instructed to provide an expert opinion in relation to claims arising out of travel insurance policies.[80] While they were provided with copies of the pleadings and other documents related to the proceeding, Associate Professor Pitt gave evidence that he was not aware of the two alleged events of discrimination or the dates on which they were said to have occurred. His understanding of the question asked of him was general - he referred to 'looking at the exclusion clause for the mental illness incidents'.[81]
2. The letter of instruction dated 21 May 2015, said:

Our client denied cover on the basis of a mental health exclusion clause in the policy. The claimant subsequently made a claim under the [EOA] on the basis that the imposition of the exclusion was discriminatory and a breach of the statute.

Our client's primary defence is that declining cover was not discriminatory as there is actuarial or statistical data or other material that supports the declinature as being reasonable in the circumstances, or that it was reasonable to underwrite the risk conditionally.[82]

1. The report states that its purpose is to 'provide an opinion on the issue of discrimination applied by QBE in its travel insurance policy documents and claims practice relating to cover for mental illness. Specifically, we focus on an analysis of this discrimination and whether it is based upon actuarial or statistical data on which it is reasonable for the insurer to rely'.
2. I pause to note that it appears that the authors had in mind the EOA section 47 exception and its DDA equivalent rather than the unjustifiable hardship exception under section 29A of the DDA.
3. In cross-examination, Associate Professor Pitt confirmed that the report had proceeded on the basis that the question was what the financial effect would be on QBE's travel insurance business if the mental illness exclusion was removed in future.[83] Accordingly, they were looking forward using the most recent available data rather than looking at what the impact would have been from around December 2011, when the policy was issued, to August 2012 when the claim was first refused.
4. The report relied primarily on data from the ABS survey summary. There was no dispute that, in general terms, that was an appropriate data source.[84]
5. The authors calculated a combined ratio for QBE travel insurance policies if QBE were to pay mental health related claims. The combined ratio (CR - apparently the same conceptually as Mr Dong's net combined operating ratio) was derived using the following formula:

1. The report states that a CR in excess of 100% means that the premium income collected is insufficient to cover the costs of claims, expenses and commissions. The analysis estimated the additional claims which would be payable by QBE if it were to cover mental illness claims. To determine that figure, the ABS prevalence rates for mental illness for the surveyed 12 month period were applied to the number of travel insurance policies issued by QBE over a two and a half year period starting January 2013. As the prevalence rates differed according to age, the authors applied the estimates to the age distribution of QBE travel insurance policy holders. As stated above, substance abuse related claims were excluded. The outcome of this part of the analysis was to find that 16.95% of QBE travel insurance policy holders would suffer a mental illness event or events during a 12 month period.
2. The authors next considered the estimated actual number of claims which would be made based on mental illness events. Three inputs were used: the duration of cover provided by the average policy; the average length of a mental illness event; and the probability that a claim would be made.
3. The average duration of cover was estimated as nine days. That figure came from QBE's records for the period 1 January 2013 to 1 June 2015.
4. The average duration of mental illness events were estimated at five, 10 and 15 days. Those figures came from the AIHW 2014 report which provided information on the duration of events being less than two weeks to in excess of five years.
5. The probability of a claim being made was identified as 45%, 55% and 65%. Those figures came from the ABS survey summary which showed that 54% of all mental health disorders are classified as moderate or severe and that people who have days off work have a much higher probability of suffering a mental illness in a 12 month period. Associate Professor Pitt gave evidence that he had relied upon the statistics regarding severity of condition with a focus on whether the condition imposed on the person's 'core activity limitations'. Those relate to the capacity to undertake tasks associated with communication, mobility and self-care. Associate Professor Pitt used these ABS survey summary results on the basis that people in that category would be unlikely to be able to travel.[85]
6. An error ratio of plus and minus two was applied because the ABS survey summary was derived from a sample and was subject to variability.
7. Calculations were undertaken allowing for the range of variables identified. The result was that the average increase in loss ratio was 12%.
8. The report noted that:
   • Some people will have multiple mental illness related events during a 12 month period and that phenomenon would increase the probability of a mental illness event based claim beyond the levels assumed in the Pitt/Kyng report;
   • There is a significant age variation in the mental health prevalence rates reported in the ABS survey summary, noting that anxiety and affective disorders have higher prevalence for younger members of the population; and
   • There are differences between males and females in rates of mental illness prevalence with males having higher rates of substance abuse related illness and females, generally, having higher rates of anxiety and affective disorder.
9. These three factors were said to make underwriting travel insurance policies including mental health illnesses difficult and impractical.
10. The report concluded that it is reasonable for insurers to exclude mental health illness related claims from their travel insurance policies and that the insurer would have to increase the price of the insurance otherwise or bear losses for offering the insurance at the current premium rates.
11. No calculations were done to show what increase would be required to the price of the policies to cover the additional claims.

1. The closing written submissions referred to the various figures relied on for the components of the Pitt/Kyng report calculations and noted that, had the mental illness exclusion not been included in the policy, for the period 1 January 2013 to 1 June 2015, the median loss would have been 12% per annum. The overall loss to QBE's travel insurance business was put at 7% as compared to the existing overall profitability based on a net combined operating ratio of 95%. It was said that the consequence would have been a loss of $3,212,818 per annum. I note these figures were not contained in the Pitt/Kyng report or given in evidence and seem to have been calculated for the purposes of the written submissions.
2. QBE said that, although the start of the period used for the calculations was after the relevant events occurred, that was a reasonable approach because the mental illness exclusion would have been effective from December 2011 on an ongoing basis. QBE noted that the period used commenced soon after indemnity was refused. It contended that it was an appropriately long period of time to provide a reliable basis for predicting the effect of omitting the mental illness exclusion in the future.
3. The closing written submissions set out further scenarios with slight alterations to the parameters producing results either side of the 7% overall loss at 2% or 13% with comparable alterations to the losses.
4. The submissions confirmed that the Pitt/Kyng report assumed that the premium would remain at an average of $92 per policy.[87]
5. QBE contended that the results of the actuarial analysis are more profound than were the case in King. In King, Jetstar would have continued to earn a profit despite allowing any number of wheelchair-bound passengers to travel on the aircraft whereas QBE could make a loss. The Court found the adverse effects on Jetstar to be sufficient to attract the protection of the unjustifiable hardship exception.

1. Ms Ingram engaged Ms Robyn Bateup to provide an expert report in reply. That report was dated 20 October 2015 and formed Exhibit A6.
2. Ms Bateup holds a bachelor degree in science and is a Fellow of the Institute of Actuaries of Australia. She currently works and is the sole director of Bateup Actuarial and Consulting Services Pty Ltd. That company specialises in providing general insurance actuarial advice. She has over 23 years' experience in providing actuarial advice to the general insurance industry, first as an employed actuary with a general insurer and then as a consultant. Her specific experience includes providing advice to general insurers on issues including premium rate reviews, outstanding claim and premium liability valuations, risk margin analyses, budget planning and strategic management reviews. While employed by a general insurer, she held the position of Manager, Pricing and Product Support.[88]
3. Ms Bateup made assumptions in her report about matters such as the parts of the ABS survey summary the Pitt/Kyng report relied on as the report was silent on these. She found she could not replicate some of the calculations in the report and that was explained by the fact that QBE had not provided the updated calculations prepared by Mr Dong.
4. The following matters were relevant to her assessment of the Pitt/Kyng report:
   • The Pitt/Kyng report relied on prevalence data rather than incidence data. The former measures the proportion of subjects who have a specific health related condition where incidence refers to the occurrence of new cases during a time period. Incidence conveys information about the risk of contracting the disease, whereas prevalence indicates how widespread the disease is. The 12 month prevalence is the proportion of subjects that have the condition at some time during the 12 month period and includes people who have the condition at the start of the period and those who acquired it during the 12 month period;
   • The Pitt/Kyng prevalence rates relied on in the report did not discuss or adjust for people who would have been excluded as a consequence of the pre-existing condition exclusion. Some of those included in the ABS survey summary would have been people who were on medication for a mental illness diagnosed a number of years earlier and who experienced symptoms in the 12 months prior to the survey;
   • The use of unadjusted age-based prevalence rates rather than age-based incidence rates adjusted for the impact of the pre-existing condition exclusion was not reasonable because the QBE data used for comparison did not capture the ages of around 75% of policy holders. It appeared that Pitt/Kyng had assumed that the age distribution of those policy holders was in line with the age distribution contained in the ABS survey summary and adopted by them. I note that Associate Professor Pitt's evidence was that any variation in the distribution would be negligible;
   • The calculations did not make adequate provision for variability over time. In Ms Bateup's experience the net loss ratio could vary significantly over time and quarter by quarter making the selected period a key driver of the net loss ratio. She said that seasonality around preferred holiday seasons made the selection of an average over a 12 month period desirable to remove any such seasonality;
   • There was other data in the ABS survey summary which would have been reasonable to take into account. That included: data which would have assisted in respect of the analysis of propensity to claim; data regarding physical conditions which could have been used to eliminate claims where, although there was a mental illness, the physical condition pre-existing condition exclusion could have been relied on; age of onset data; and data regarding use of services including admissions to hospital and consultations with mental health professionals;
   • The approach adopted for estimating the mental illness claim frequency by applying an assumed propensity to claim was not unreasonable. However, the use of rates of propensity to claim based on the expected severity of the mental illness event are not reasonable. The ABS survey summary data shows that, excluding substance abuse disorders, 67.5% of all 12 month mental disorders result in no disability with only 32.5% resulting in some level of disability, including mild. That figure is lower than the 45%, 55% and 65% figures used in the Pitt/Kyng report; and
   • Data on services used for mental health problems could be used to assist with estimating propensity to claim. The data in the ABS survey summary shows that, across ages 16 to 85 years, 35% of all 12 month mental disorders (including substance abuse disorders) used medical services for those problems. Those figures might be used as a proxy for the severity of the illness and propensity to claim. It was contended these figures result in a lower estimate of the propensity to claim than relied on by Pitt/Kyng (45%, 55% and 65%).
5. Ms Bateup commented that an insurer such as QBE could take steps to limit the adverse effect of a change to its policy, such as removal of the blanket exclusion. She suggested that an adjustment could be made to the premium, access to benefits could be restricted by a higher excess or lower benefit being available for mental illness claims or the type of mental illness covered could be restricted, for example by only providing cover where hospitalisation is required.
6. As to the financial impact on QBE if the mental illness exclusion was removed, Ms Bateup set out in her report published profit results she considered to be relevant. Her conclusion was that the Pitt/Kyng estimated loss as a result of being required to cover mental illness related claims, would represent less than 1.1% of the profit earned by the Australian and New Zealand Division of QBE for the six months to June 2015. As at the half year ended 30 June 2015, the overall underwriting result of the division was said to be USD153 million and the insurance profit for the division was said to be USD247 million.

1. Having carefully considered the written submissions, there are five matters which I regard as relevant to determining the weight to give to the Pitt/Kyng report. Of the other contentions raised, I have proceeded on the basis that they either assist QBE's case or are neutral.[89]
2. I have made this assessment mindful that, because it keeps no data on mental illness claims, QBE has been required to retrospectively explain why the defences can be relied upon. That has required it to consider available data and formulate a methodology after the relevant events. Clearly that was a difficult task and one I have found to be unsuccessful.

Consideration of the pre-existing condition exclusion

1. There was no dispute that, when QBE received claims or inquiries about claims on the policy in respect of mental illness, the mental illness exclusion was relied upon rather than the pre-existing exclusion. There was no dispute that the Pitt/Kyng report made no allowances for claims being refused because the pre-existing clause applied.
2. In his evidence on cross-examination, Associate Professor Pitt said that, when preparing the report, he had assumed that, if the mental illness exclusion was removed, there would be a change to the pre-existing condition clause such that it had wider application. He had anticipated that the clause would be changed so that it would exclude not only people who had a pre-existing condition that they disclosed but others including people who did not know they had the condition.[90] While in re-examination further questions were asked about this matter, Associate Professor Pitt was unable to say how the pre-existing condition clause might be changed but he had assumed something would be changed. He confirmed the removal of the mental illness exclusion would mean the policy was operating very differently.[91] The report contained no suggestion this assumption had been made or what changes had been anticipated. It was not clear whether he had in fact made an assumption when preparing the report or was assuming that some change would be made in future.
3. I proceed on the basis that the method applied by QBE's experts made no allowance for how the pre-existing condition clause might apply to claims made by persons who had a pre-existing mental illness (disclosed to QBE or not) where they are unable to travel because they experience a further episode or re-occurrence of that illness. It did not differentiate between people who, like Ms Ingram, had their first episode of illness between the time the policy is issued and their proposed date of travel. Even if I accept that the basis for the other statistics is valid, if these categories of illness are not accounted for, it must be the case that the loss calculated in the Pitt/Kyng report includes claims which could have been rejected on the basis of the pre-existing condition clause. As I understood Ms Ingram's written submissions, this is one of the consequences of using prevalence data rather than incidence data.[92]
4. As the Pitt/Kyng report did not take account of QBE's entitlement to reject claims arising from pre-existing conditions, it is not clear whether the financial consequences would have been large or small.

1. In cross-examination, Associate Professor Pitt was asked about the policy requirements that the relevant illness be serious and require hospitalisation or confinement in order to claim cancellation costs. He agreed that it would not be safe to assume that every time a person with a mental illness had a mental illness episode, it would be serious and one that would require hospitalisation or confinement.[93]
2. As I understood it, Mr Dong and Associate Professor Pitt had regard to tables in the ABS survey summary which described the disability status which attached to all mental disorders experienced in the preceding 12 months.[94] Percentages are attributed to disability status ranked in the following categories: profound/severe; moderate/mild; schooling/employment restriction only; and no disability/no specific limitations or restrictions. There are figures addressing the level of psychological distress and days out of role. Using figures from the relevant table for both anxiety and affective disorders, Mr Dong calculated that there was a 55% chance of a claim being made taking into account the percentages applicable to persons who have one or more days out of their role and some disability or restriction.[95]
3. The glossary to the ABS survey summary shows that the 'days out of role' category referred to the number of days in the 30 days before interview the person had been unable to work or carry out their normal activities due to their health. Disability status referred to the level of impact on core activities but there was no differentiation in the glossary between the rankings set out above.[96] Ms Bateup contended that other material published by the ABS and available for purchase might have assisted to clarify the question of propensity to claim. In particular, she made reference to data which would disclose what services, including admissions to hospital, were accessed.
4. The ABS survey summary glossary shows that references to 'services' in the document in respect of mental health problems includes hospitalisation and consultations with health professionals for mental health. The 'health professionals' include general practitioners, psychiatrists, psychologists, mental health nurses, social workers, counsellors, occupational therapists and others ranging from surgeons to herbalists and naturopaths. The range of help sought included providing information, medication or counselling, social interventions including housing or financial support and skills training.[97] These are the range of health professionals, services and help sought by the 35% of people with a mental disorder in the preceding 12 months. Table 12 of the ABS survey summary shows that 7.9% of people sought services from a psychiatrist. Professional help of that kind ordinarily precedes or is received during a hospital admission for a mental illness.
5. As noted by Ms Bateup in her report, the ABS survey summary shows that 67.5% of all 12 month mental disorders (excluding substance abuse disorders) result in no disability with only 32.5% resulting in some level of disability. Ms Bateup suggested that the use of services statistics reflected in table 12 of the ABS survey summary could be used as a proxy for propensity to claim. [98]
6. Having regard to these matters, I am not satisfied that the likelihood to claim component of the Pitt/Kyng report is reliable. The data provided and its analysis in the Pitt/Kyng report does not establish that the claims considered fell within the policy prerequisites: it was not apparent that a person who has one day out of role (or even a few days out of role) would be regarded as having a 'sudden serious illness'. Further, where a claim may only proceed on production of proof that hospitalisation or confinement was required, it is not clear to me how I can rely on the conclusion that, of all persons who had a mental illness over the relevant 12 month period, 55% were likely to be able to prove that they were hospitalised because of a serious illness. I accept Ms Ingram's submission that I ought not to accept the assumption that a person who falls within the disability status categories of the ABS survey summary and so is unable to undertake core activities would necessarily be unable to travel.[99]

1. As earlier noted, the data used by Mr Dong and applied in the Pitt/Kyng report was for the period 1 January 2013 to 1 June 2015. It was submitted for Ms Ingram that was the incorrect period to demonstrate that, at the time of the discriminatory conduct (December 2011 when the policy was issued and August and December 2012 when the indemnity was refused), not discriminating would have imposed an unjustifiable hardship on QBE.
2. QBE relied on the decision in King where the court had regard to a time and motion study prepared in a period after the relevant time frame. Ms Ingram contended that had only been accepted because it was understood that the conditions at each point of time were essentially the same.[100]
3. If Ms Ingram's contention is correct, then the actuarial report cannot be relied on at all. It is not necessary for me to finally decide this issue given the other significant reservations I have about the report but I make the following comments.
4. QBE contended that the time frame was sufficiently close in time to the period in issue and of sufficient length that it could be relied on to assess the financial impact on QBE during the relevant 2011 to 2012 period, in part relying on Mr Dong's evidence that, over the period, the abovementioned ratios are fairly stable.[101] Even if that time frame was acceptable, the evidence made it clear that all experts agreed that the ratio of net incurred claims to new earned premiums can vary over time, including year to year and quarter by quarter. In respect of the combined operating ratio from year to year, Associate Professor Pitt agreed with the proposition that there could be a variation of a range of 10%.[102] Associate Professor Pitt accepted that, if using actuarial and statistical data, it would be appropriate to use data which is as relevant as possible to the scenario and he also accepted that there could be variation in net loss ratios over time, including from year to year.[103] Evidence of that kind led to QBE's second day application for an adjournment so that new evidence in the form of a revised report based on the earlier period could be prepared.[104]
5. Where:
   • There was evidence that there can be material variation in matters relevant to the calculation of loss relied on by QBE; and
   • The Pitt/Kyng report was prepared looking forward to the financial impact on QBE if the outcome of these proceedings was that the mental illness exclusion could not be relied upon,

I am not satisfied that it is more probable than not that the data produced demonstrates that, as at December 2011 to December 2012, QBE would have suffered a loss if it could not have relied on the mental illness exclusion in respect of Ms Ingram or any other insured's claim.

1. As indicated above, the Pitt/Kyng report appeared to have been prepared on the basis of instructions which focussed on the section 47(1)(b) EOA and section 46(2)(f) DDA exceptions rather than on section 29A of the DDA. Whether that made a significant difference is not clear but it is possible that, had the authors been aware of that, they might have provided some evidence about ways in which the hardship of a loss making product could have been reduced or mitigated, perhaps including by an increase in premiums.
2. In the written submissions, reference was made to Associate Professor Pitt's lack of instructions and knowledge about the background to this case and how that may have impacted on his report. I have mentioned some of those matters above where it appeared they might have a material consequence.[105]
3. There was a large amount of discussion at the hearing and in the written submissions about the use of prevalence data rather than incidence data.
4. As indicated above, it was Ms Bateup's opinion that incidence data would have been of greater assistance in respect of this case in circumstances where there was a pre-existing condition exclusion clause and where the mental illness exclusion operated to exclude claims from people, like Ms Ingram, who have first had an episode of mental illness after being issued with a policy and leaving to travel.
5. QBE's written submissions agreed that incidence rates might be useful but that there was no evidence that they were available at the time or as reliable as the prevalence data from the ABS survey summary.[106] In reply, Ms Ingram's submissions pointed to the additional data available for purchase from the ABS, noting that it was not taken into account because it was not provided to QBE's experts.[107]
6. As set out above, Associate Professor Pitt agreed that prevalence rates do not indicate what sort of assistance a person might need including in the range between counselling and hospitalisation. In respect to the additional ABS data he said that, had they been undertaking an incidence analysis, then that data would have been useful.[108] When asked in evidence in chief why he had used prevalence rather than incidence rates, he said that in his view the former are more appropriate because:

. . . in this context, and the reason is that a pre-existing medical condition for mental illness, while it exists in the current policy, it is never triggered, so if you were to move to a situation where the blanket claims exclusion for mental illnesses on anxiety and affective disorders was removed, you would need to revisit the existing pre-existing medical condition as it relates to those pair - or that pair of mental illness forms

Right? Right. In order to underwrite a travel insurance policy, many of which are purchased at the click of a mouse on the Qantas site, if you were to underwrite that with a statement about pre-existing medical condition relating to substance abuse, with such a high prevalence for substance - for mental illness, there's such a high prevalence for mental illness, you would need to do some significant underwriting associated with that. It's true that many people who are suffering a mental illness are not aware of it. They don't acknowledge it. They don't get it treated.[109]

1. I understand that Associate Professor Pitt was suggesting that it would be too difficult to underwrite insurance given that many people with mental illness do not acknowledge those conditions and so will not disclose a pre-existing condition when applying for insurance. Noting that the policy allows claims to be refused when an insured fails to make proper disclosure,[110] this does not help me understand why prevalence rates assist in determining how many insured will develop mental illness between being issued with insurance and undertaking travel or whether those people are more or less likely to claim.
2. Associate Professor Pitt later suggested that the overall method he and Dr Kyng used was close to an incidence rates analysis. That explanation was not given when he was invited to comment on Ms Bateup's report, nor was it clear to me.[111]
3. Mr De Luca also gave evidence in chief that it would be difficult for QBE to verify claims arising from mental illness. He referred to mental illness claims being difficult to predict and having a range of triggers. In cross- examination he conceded that in Ms Ingram's case she had been asked for and was able to supply evidence about the nature of her illness and inability to travel. He agreed that many insurance claims are difficult to predict. Mr De Luca referred to the risk of a mental illness episode overseas and suggested that, being in an unfamiliar environment, might be a factor increasing that risk. As discussed earlier, Mr De Luca's impressions about these matters arose from his observations from media and other generally available information rather than on the statistical material in the reports annexed to his affidavit or on any training he had received. [112] That evidence did not assist QBE's case.
4. In circumstances where Associate Professor Pitt acknowledged that prevalence data has limits to what it can show and where he agreed incidence data could assist in analysing relevant matters, a methodology which relied only on the former seems to me to be less reliable, particularly in conjunction with the other uncertainties discussed above.

Financial impact on QBE's business and alternatives

1. I noted earlier that Ms Bateup made suggestions about how QBE could minimise the financial impact of being unable to rely on the mental illness exclusion.
2. QBE was critical of her suggestions on the basis that these were not matters particularised in the claim and because there was no modelling or data to show they would have had the suggested consequences. I understood the suggestions were made as possible ways to offset the financial consequences of the removal of the mental illness exclusion. That would be relevant to whether there was unjustifiable hardship. The matters raised by Ms Bateup addressed that question having regard to her professional experience.
3. At the hearing, QBE's counsel suggested that the concept of offsetting losses in the travel insurance business against other profit-making products was contrary to QBE's regulatory obligations. No evidence about those obligations was called at the hearing. In the closing written submissions, the basis for the suggestion was set out in detail. Reference was made to section 32 of the Insurance Act 1973 (Cth) and prudential standards issued under that Act by the regulatory authority, the Australian Prudential Regulation Authority (APRA). The submission set out in detail material relating to the history and content of APRA's Prudential Standard GPS115 which addresses capital adequacy requirements. The submissions concluded that, if the mental illness exclusion was removed from the policy, QBE would suffer a loss and that would require it to hold an additional amount of capital to secure its travel insurance business under GPS 115. The submissions asserted that it would not be appropriate for QBE to rely on other parts of its business to offset the losses which would be incurred in the travel insurance business. It further contended that it would not be appropriate to have regard to the overall financial position of QBE when considering this matter as the proper question was to consider the effect of the removal of the mental illness exclusion solely within the context of the travel insurance business.[113]
4. Not surprisingly, in the reply submissions objection was taken to QBE purporting to introduce new evidence of this kind through submissions.[114]
5. The applicant's written submissions noted that the APRA standard does not show how the minimum capital requirements are to be held across the business. They said that the highest QBE could put the effect of the standard was that it would make it inevitable that QBE would be required to hold an additional amount of capital. They commented that might simply be the cost of doing business in a non-discriminatory way. Ms Ingram's submissions also noted that the existence of these regulatory requirements did not appear to have prevented other insurers from offering policies which do not include general exclusions based on mental illness.
6. The applicant's submissions contended that, if this was an important point for QBE, it could have tendered evidence of fact and, if necessary, expert evidence on the issue, given it had a lengthy period of time to prepare its defence in this case. The submissions noted that Mr De Luca gave evidence that the travel insurance business was a small component of QBE's total business and that he thought QBE's travel insurance business would have a 'minor' level of influence on its overall performance. He was otherwise unable to give evidence about the regulatory requirements or QBE's compliance with those requirements.[115]
7. As none of these matters regarding the APRA requirements were able to be put to Ms Bateup, because they were only raised in the closing written submissions, it was submitted for Ms Ingram that the rule in Brown v Dunne applies. Ms Ingram contended that is not open to the tribunal to now have regard to matters relating to the prudential standards for the purposes of considering whether QBE has proven that section 29A of the DDA applies.
8. I accept those submissions regarding the failure to produce evidence to support the submissions and the application of the rule in Brown v Dunne. I take the same approach to other assertions of fact unsupported by evidence which were included in the written submissions about, for example, the impossibility of QBE maintaining insurance at the levels it does without being able to rely on the mental illness exclusion or the impact of those matters on QBE's Australian financial services licence.[116]
9. Strangely, having criticised Ms Bateup for not providing modelling in respect of her suggestions including that there could be an increase in the premium, QBE contended that I should take into account the likelihood that an increased number of claims in respect of mental illness would require an increase in not only the travel insurance policy but also potentially in other policies.[117]
10. Even if I was entitled to take judicial notice of the APRA standard and associated material, given the reservations I have about the calculation of the claimed losses relied on by QBE, I would not be able to be satisfied about how that standard might affect QBE overall or in respect of its travel insurance business.

1. On the evidence and for the reasons discussed above, I am not satisfied that the conclusion contained in the Pitt/Kyng report is well founded. Accordingly, I am not satisfied that the report demonstrates that 'it is reasonable for insurers [such as QBE] to exclude mental health illness related claims from their travel insurance policies' or that 'the insurer would have to increase the price of the insurance otherwise or bear losses for offering the insurance at the current premium rates.'[118]
2. I will now turn to the express matters raised by section 11A in the context of the DDA section 29A defence.

1. When determining whether a hardship that would be imposed would be 'unjustifiable', it is necessary to take into account all relevant circumstances of the particular case, including those set out in section 11(1) as follows:

(a) the nature of the benefit or detriment likely to accrue to, or to be suffered by, any person concerned;

(b) the effect of the disability of any person concerned;

(c) the financial circumstances, and the estimated amount of expenditure required to be made, by the first person;

(d) the availability of financial and other assistance to the first person;

(e) any relevant action plans given to the Commission under section 64.

Example: One of the circumstances covered by paragraph (1)(a) is the nature of the benefit or detriment likely to accrue to, or to be suffered by, the community.

1. QBE called no evidence directed towards these matters other than the Pitt/Kyng report and Associate Professor Pitt's evidence.
2. The general question raised is whether the removal of the discriminatory treatment would impose an unjustifiable hardship on the relevant party.
3. Subsections (1)(d) and (e) do not arise here.

The nature of the benefit or detriment likely to accrue to or be suffered by any person concerned

1. If the mental illness exclusion had not been included in the policy, Ms Ingram would have benefitted because her claim for cancellation costs would have been paid. Had the mental illness exclusion not been included, Ms Ingram would not have suffered hurt and felt stigmatised. Ms Ingram submitted that the mental illness exclusion has an impact on the wider community, in particular the stigmatising of mental illness.[119]
2. QBE said that there is no evidence that there would be any benefit to it if the mental illness exclusion was removed, to the contrary it would suffer a detriment having regard to the Pitt/Kyng report. As discussed above, QBE submitted that as a consequence of the increased losses, premiums for travel insurance and possibly other policies might have to be increased leading to detriment for other policy holders.[120] No evidence in support of that assertion was produced.
3. QBE submitted that benefits might accrue to other policy holders who would have had their claims met at the relevant time and into the future. Those claims may have extended beyond cancellation claims to claims regarding mental health expenses incurred overseas.[121] I accept those are likely benefits.
4. QBE submitted that it is conceivable that it may benefit from not having the mental illness exclusion because it may attract a greater share of the travel insurance market. It contended that it would be speculative to make such a finding and that, given the actuarial evidence, it was likely there would be a net loss to QBE.[122] No evidence was produced to support the market share submission.

1. This factor requires consideration of the actual disability of the persons concerned, rather assumptions or generalisations that may be made about its effect.[123] In King, Robertson J took into account the effects on Mrs King and also on Jetstar's operations and other passengers, including those who may be delayed.[124]
2. Ms Ingram contended that those affected would be Ms Ingram and also other persons who have or may develop a mental illness who hold insurance under the terms of the same policy.
3. QBE accepted that, had Ms Ingram been indemnified for her cancellation expenses, she would not have experienced any further effect because of her disability. It contended that the effect on other policy holders and shareholders would be significant because of the forecast losses and risk of a need to increase premiums. The effect on the community was suggested to be increased insurance premiums or fewer travel insurance providers.[125]

Financial circumstances and estimated amount of expenditure

1. It is understood that this factor cannot be given greater weight than the others. It, like the other factors, must be considered within the context of the legislation and the circumstances of the case.[126]
2. I refer to the earlier discussion about the Pitt/Kyng report and associated matters.
3. I accept that in King, the fact that Jetstar's profit was to be significantly impacted was regarded as sufficient to conclude that the change to the policy sought by Mrs King would impose an unjustifiable hardship.[127]
4. QBE contends that, even on the most favourable calculations, the net loss it would incur if the mental illness exclusion was removed would be at best 2% and more likely 7%. Accordingly, it said that this factor would weigh significantly in its favour given the approach taken in King. QBE noted that the circumstances it faces are dramatically different to those considered in Scott & Anor v Telstra Corporation Ltd, where the cost to Telstra of taking non-discriminatory action was the equivalent of 30 cents per annum per subscriber. [128]

1. I am satisfied that, if the mental health exclusion were removed from the QBE policy, the effect on Ms Ingram and other policy holders subject to the same terms, would be that they might have had their claims met (section 11(1)(a) and (b)). I say 'might' in respect of other policy holders because any claim would be subject to proof of hospitalisation or confinement (or other prerequisites) and subject to the impact of the pre-existing condition clause. In Ms Ingram's case there was no dispute she met the prerequisites to claim if the exclusion was removed.
2. I accept that the community would benefit from an action which would lessen the stigmatising effect of negative attitudes towards mental illness (section 11(1)(a)).
3. Clearly the burden of any detriment arising from the removal of the mental illness exclusion would fall on QBE, its policy holders and shareholders because the impact would be primarily financial (section 11(1)(a) and (c)).
4. As discussed above, I have found that the content of the Pitt/Kyng report and the evidence of Associate Professor Pitt and Mr Dong did not prove that QBE would suffer an absolute loss in the range of 2% to 12% (or 13%) if the mental health exclusion was removed or more generally that it is reasonable for insurers (such as QBE) to exclude mental health illness related claims from their travel insurance policies or that insurers would have to increase the price of the insurance otherwise or bear losses for offering the insurance at the current premium rates.
5. I have not been persuaded that I should have regard to the regulatory requirements said to apply in the absence of evidence as to what they are and how they would affect QBE. While QBE has asserted that it would have to increase premiums in respect of the travel insurance policies, and possibly others, no evidence about what the additional cost would be has been produced. I cannot weigh up whether an unjustifiable hardship would arise from a premium increase because I do not know if an increase on the average $92 policy would be modest or substantial.
6. QBE submitted that the removal of the exclusion would cause it an overall loss on its travel insurance business. I am mindful that, accordingly to the approach taken in King, a reduction in profits might be sufficient to amount to an unjustifiable hardship. Due to the significant reservations I have about the Pitt/Kyng report and the evidence produced in this case by QBE, I cannot safely find that there would be a reduction in profits. I find that, in so far as QBE relied on the financial factors to demonstrate that it falls within section 29A of the DDA, it has not met its burden of proof.
7. The consequence is that, when undertaking the balancing task required by section 11, there is an absence of sufficient material for me to determine that it would be an unjustifiable hardship for QBE to be unable to rely on the mental illness exclusion. The scales weigh in favour of people like Ms Ingram being able to be properly assessed on their policy claims in the same way people with physical disabilities are assessed.
8. In those circumstances, I find that section 29A of the DDA does not apply and so cannot be relied upon under section 27(1)(a) of the EOA in respect of either claim of discrimination.
9. The consequence of this conclusion is that I have found none of the exceptions relied on by QBE apply and so the direct discrimination engaged in by it was unlawful.

1. In her particulars of claim, Ms Ingram sought:
   • Declarations that QBE unlawfully discriminated against her in respect of the two claims made;
   • Compensation for economic loss;
   • Compensation for non-economic loss; and
   • Costs.
2. Ms Ingram originally also sought a written apology but withdrew that application.

The tribunal's powers to make findings and declarations

1. The tribunal has the powers granted it by the Victorian Civil and Administrative Tribunal Act 1998 (Vic) (VCAT Act) and by the various enabling enactments, which include the EOA.
2. The tribunal must act fairly and according to the substantial merits of the case. It is bound by the rules of natural justice. It is not bound by any practices or procedures applicable to courts of record, such as rules regarding pleadings. The tribunal must conduct each proceeding with as little formality and technicality and determine each proceeding with as much speed, as the requirements of the VCAT Act and the enabling enactment and a proper consideration of the matters before it permit.[129]
3. Section 125 of the EOA says that, after hearing the evidence and representations of the parties, the tribunal may find that a person has or has not contravened a provision. If it finds there has been a contravention, the tribunal may make a range of orders including for a person to pay to the applicant an amount the tribunal thinks fit to compensate for loss, damage or injury suffered in consequence of a contravention. The tribunal may also make an order that the person do anything with a view to redressing any loss, damage or injury suffered as a result of the contravention.
4. Under section 124 of the VCAT Act, the tribunal has the power to make a declaration concerning any matter in a proceeding instead of or in addition to any other orders. A declaration may be made by a presidential member or a member, such as myself, who falls within the definition of an 'Australian lawyer' contained in the Legal Profession Uniform Law (Victoria). A declaration is a discretionary remedy which should not be made if the issue is hypothetical in nature and would have no real or practical utility.[130]
5. Ms Ingram sought a clear public statement that QBE discriminated against her. She contended that a declaration in this case would be important in part because the exclusion is used as a basis on which all claims made on the basis of mental illness are denied. She said that this practice is ongoing as the exclusion remains in QBE's policies and it affects many consumers.[131]
6. In its written submissions, QBE contended that:
   • As Ms Ingram had only sought a declaration under section 124 of the VCAT Act and not a finding under section 125(a) of the EOA, she has no entitlement to orders for further action under the EOA or otherwise, including for compensation for economic or non-economic loss;
   • The tribunal ought not to make a declaration as no finding had been sought and a declaration ought not to be made as an alternative;
   • As no finding under the EOA had been sought, no consequential orders may be made and so a declaration would serve no purpose;
   • The tribunal ought not to allow Ms Ingram to amend her application to seek a finding under section 125 of the EOA; and
   • The tribunal ought not to amend the application on its own initiative under section 217 of the VCAT Act but rather should hold Ms Ingram to the case she has pleaded and the remedies she sought in her particulars of claim.[132]
7. It is correct that Ms Ingram did not seek a finding under section 125 of the EOA in her particulars of claim.
8. If QBE were correct in respect of its proposed reading of section 125, then I would not be able to make any order in this proceeding under the EOA or otherwise. As noted in QBE's submissions, the tribunal is bound to make findings based on the evidence and informed by the relevant law. Section 125 of the EOA is the means by which the tribunal brings finality to proceedings brought under the EOA. In my view, whether or not an express application has been made for a finding, it is the tribunal's task to make relevant findings of fact and law to enable it to make orders.
9. I am mindful that the tribunal is not a court of pleadings and is required to conduct each proceeding with as little formality and technicality as required. In circumstances where Ms Ingram brought a claim that alleges that QBE's conduct amounts to direct discrimination in breach of a prohibition in the EOA and where QBE concedes that the case was fought on the question whether it had contravened section 44,[133] in my view, it goes without saying that she has sought a finding under section 125 of the EOA. There is no need for Ms Ingram to apply to amend her particulars of claim to be entitled to seek a finding under section 125.
10. As for the application for a declaration, I accept the parties' submissions as to the circumstances in which a declaration is appropriately made. Ms Ingram's submissions drew attention to decisions which discussed the power of a declaration, particularly in human rights cases. The power has been described as a form of 'final vindication'[134] and having an 'educative purpose'.[135]
11. I accept that a declaration may have the important function described above. However, given that my findings on the question of the application of the exceptions relied upon have turned on the way QBE prepared for and ran this case, I have decided not to make a declaration to ensure that an impression is not given that my decision automatically extends beyond the dispute between these parties and, in particular, to avoid an impression that it applies to all insurers.

1. Ms Ingram claimed the sum of $4,292.48 being the costs associated with her cancelled trip after taking account of a refund from the credit provider. In her closing written submissions, Ms Ingram also sought interest on this sum in the closing written submissions. I assume that application was made under section 125 of the EOA which potentially empowers the tribunal to make an order for interest. No interest rate was specified in the submission. As interest was not sought in the particulars of claim and the basis for it has not been particularised, no order for interest will be made.
2. QBE does not dispute that sum has been established as the claimable loss on the evidence.
3. Having regard to my findings of discrimination, I will make an order for that amount to be paid.

1. In her closing written submissions, Ms Ingram sought the sum of $20,000 in respect of non-economic loss. Reliance was placed on Ms Ingram's evidence that the refusal of indemnity caused her to feel upset, angry, frustrated and like she had been treated as a second rate citizen. In her witness statement, Ms Ingram stated that she felt stigmatised about her depression and thought her mental illness was being less valued than a physical disability. She stated she did not plan on becoming unwell and would not have chosen it. She could not understand why she was being treated differently to someone who accidentally or unintentionally breaks a leg or gets food poisoning. Ms Ingram's evidence was that the ongoing refusal left her with feelings of hopelessness because she thought the experience with QBE might be an example of what she could expect for the rest of her life. She questioned whether she should disclose her mental illness again in the future to an employer or other insurer for fear of being discriminated against. The refusal of indemnity continued to remind her of the time when she was very ill and suicidal which was and continued to be distressing to her. Her feelings of anger, upset and frustration continue. The ongoing refusal reminded her of the worst time of her life. She told the tribunal she felt unimportant. [136]
2. Ms Ingram's mother, Mrs Linda Darragh, gave evidence that, when she informed her daughter that indemnity had been denied, Ms Ingram cried, was angry and frustrated and could not understand why that decision had been made.[137]
3. QBE submitted that any award of compensation for non-economic loss should be for a sum of no more than $5,000. It noted that Ms Ingram's evidence was that the decision to refuse indemnity did not cause Ms Ingram to feel scared, unsafe or suicidal. It noted there was no medical evidence as to psychological injury or any evidence that showed Ms Ingram's depression had been aggravated or exacerbated.[138]
4. QBE distinguished a number of cases in which large sums have been awarded, including in respect of sexual harassment and circumstances where the applicant had suffered very significant and long term consequences, including mental illness or diagnosed psychological damage.[139] As I understood it, QBE relied on the 2003 decision in Bassanelli when proposing the sum of $5,000. In that case, the Federal Court confirmed that it was appropriate to exclude any element for damages for Ms Bassanelli's campaign for fair treatment of cancer sufferers.
5. I was referred to the decision of Deputy President Nihill in Slattery v Manningham City Council,[140] in which the sum of $14,000 was awarded in respect of Mr Slattery's exclusion from all buildings owned, occupied or managed by the council for a period of some three years.
6. The parties each referred to my decision in Dziurbas v Mondelez Australian Pty Ltd.[141] QBE contended that, without wishing to trivialise the hurt felt by Ms Ingram, the hurt and humiliation suffered by Mr Dziurbas is not comparable to that she experienced.[142]
7. In Dziurbas, I discussed what I considered to be an appropriate approach to the assessment of non-economic loss in discrimination cases.[143] Like QBE, I distinguished cases of sexual harassment. I have had regard to the following matters:
   • Under section 125 of the EOA, compensation can be awarded for psychological injury, humiliation and hurt feelings provided that there is a causal link between the breach of the EOA and the claimed loss, damage or injury;
   • Non-economic loss will often be hard to quantify in monetary terms. Each case must turn on its own circumstances. Reference to other awards might provide some limited guidance, but cannot be decisive;
   • A discriminator takes his or her victim as he or she finds that victim;
   • The breach of the EOA does not have to be the sole cause of the claimed loss or damage. It is enough to demonstrate that contravention of a relevant provision of the EOA was a cause of the loss or damage sustained; [144]
   • Community standards now accord a higher value to compensation for pain and suffering and loss of enjoyment of life than before. As compared to the past, modern society places a higher value on the loss of enjoyment of life and the compensation of pain and suffering;[145]and
   • The purpose of the award of damages is to attempt to measure, in monetary terms, the hurt that has been done to the applicant by the conduct or act in breach of the EOA and the assessment should mirror the approach that would be taken if this case were to be heard in a court. [146]
8. As to the application of the 'egg skull rule' here, the applicant's submissions said that, QBE must take Ms Ingram as it finds her, even if a denial of indemnity would cause her greater anguish than the average citizen. It was submitted that Ms Ingram's reaction was neither unusual nor irrational.[147]
9. I accept QBE's contention that any award of compensation should not include an element which reflects Ms Ingram's very public efforts to raise the profile of her claim.
10. As I understand it, the hurt and humiliation experienced by Ms Ingram did not centre on disappointment at not receiving a refund of the costs of the cancelled trip. The upset was directly linked to the fact that QBE had, by way of its exclusion, refused to refund those costs because she had suffered a mental rather than physical illness. I accept Ms Ingram's evidence about her reaction to that decision and in particular the fact that she found it to be stigmatising. I accept and place significance on Ms Ingram's evidence that the unfavourable treatment by QBE has raised in her mind concerns about disclosing her mental health history in future for fear of further discrimination.
11. I consider the broader community would share that reaction given the efforts being made by governments, health practitioners, non-government organisations and individuals to have open discussion about mental illness and to encourage members of our community to seek help and treatment for mental health problems.
12. The circumstances of this case include my finding that at the relevant points in time the decision to include the exclusion clause in the policy and to refuse Ms Ingram indemnity was not based on proper actuarial data or statistics or evidence which demonstrated it was justified.
13. This proceeding has had a lengthy history and one consequence has been that the Ms Ingram's feelings as described have persisted. While it is difficult to separate that matter from any other ongoing symptoms she might have experienced or the effects of the publicity surrounding this case, I am mindful that the time taken for the matter to be completed has been an added burden. More than three years have passed since indemnity was first refused by QBE.
14. Having regard to the matters listed above, I consider an award of no more than $5,000 to be out of step with recent decisions, including of the Federal Court of Australia and out of step with community standards.
15. In all of the circumstances of this case, I will order that QBE pay to Ms Ingram the sum of $15,000 for non-economic loss.
16. For the sake of certainty, I note that award is made in respect of the overall claim for non-economic loss and, had I found against Ms Ingram in respect of her first claim of discrimination, the award would have been the same.

1. Ms Ingram applied for costs thrown away in respect of an application to tender new evidence from Mr Dong and additional evidence from Associate Professor Pitt. She also seeks costs of the proceeding, less those already awarded.
2. QBE contends that no costs order ought to be made.

(1) Subject to this Division, each party is to bear their own costs in the proceeding.

(2) At any time, the Tribunal may order that a party pay all or a specified part of the costs of another party in a proceeding.

(3) The Tribunal may make an order under subsection (2) only if satisfied that it is fair to do so, having regard to -

(a) whether a party has conducted the proceeding in a way that unnecessarily disadvantaged another party to the proceeding by conduct such as -

(i) failing to comply with an order or direction of the Tribunal without reasonable excuse;

(ii) failing to comply with this Act, the regulations, the rules or an enabling enactment;

(iii) asking for an adjournment as a result of (i) or (ii);

(v) attempting to deceive another party or the Tribunal;

(b) whether a party has been responsible for prolonging unreasonably the time taken to complete the proceeding;

(c) the relative strengths of the claims made by each of the parties, including whether a party has made a claim that has no tenable basis in fact or law;

(e) any other matter the Tribunal considers relevant.

1. The usual approach to section 109 may be summarised as follows:
   • The presumption in section 109 is that costs lie where they fall;
   • The tribunal has a wide ranging discretionary power to award costs if the tribunal is satisfied that it is fair to do so;
   • The tribunal may have regard to the specific matters set out in section 109(3)(a) to (d) and, under (e), any other relevant matter; and
   • An order ought to only be made where it is fair to do so.
2. Under section 109(3)(e), the tribunal may take into account the nature of the claim. It is long established that the Human Rights List (and its predecessor, the Anti-Discrimination List), operates broadly as a no-costs jurisdiction and that is particularly relevant to EOA claims.

Ms Ingram's applications for costs[148]

1. Ms Ingram's submissions contended that several applications by QBE to introduce further lay and expert evidence on the second hearing day, caused her expert to be required to be at the tribunal for two days rather than one. The new or further evidence included the additional data sheets prepared by Mr Dong, which had not been served prior to the hearing. In addition, time was taken for Mr Dong to give his evidence and additional evidence was adduced from Associate Professor Pitt in respect of the manner in which he used the statistical data received from Mr Dong. Ms Ingram said that there was no reasonable excuse for QBE to produce that evidence so late given it had more than a year available to it to brief experts, prepare its reports and ensure they were served in time for Ms Ingram's advisers and expert to review them and prepare their response. In addition to relying on the many procedural orders made by the tribunal for the filing of expert material, Ms Ingram's solicitors had written to QBE's solicitors on 2 October 2015 seeking confirmation that all documents and materials relevant to QBE's expert evidence had been produced. On 13 October 2015, QBE's solicitors wrote saying that all such material had been produced.
2. Ms Ingram also sought the costs associated with the second hearing day as she said that day was occupied by new evidence (in the form of the new data sheets and Mr Dong's evidence) and applications to produce new evidence. The submissions said that it had been anticipated that the expert evidence would be completed on day two of the scheduled hearing and so the third hearing day would not have been required.
3. Ms Ingram sought the costs of the whole proceeding, other than those already ordered and those associated with an interlocutory hearing held in November 2014.
4. On behalf of Ms Ingram it was contended that it would be fair for those costs to be paid by QBE because QBE conducted the proceeding in a manner which disadvantaged her by:
   • Failing to file the Pitt/Kyng report in accordance with the tribunal's practice note until the first day of the hearing;
   • Failing to serve all briefing and other materials relevant to the expert evidence until the second hearing day;
   • Failing to serve a witness statement made by Mr Dong; and
   • Asking for an adjournment on the second hearing day to present new evidence. That application was refused but the application took up a considerable amount of hearing time.
5. Ms Ingram said that, as QBE had more than a year to prepare for the hearing, it could have no reasonable excuse for the above conduct.
6. In respect to the relative strengths of the parties' cases, Ms Ingram noted that QBE denied that it had engaged in any discriminatory conduct and then also relied on its defences. It specifically denied that providing indemnity under contracts of insurance was a 'service' under the EOA but then, at the hearing, conceded it was and also conceded that there was no issue about Ms Ingram proving that she had been treated unfavourably on account of her mental illness from the time indemnity was declined. As a consequence of these late concessions, Ms Ingram had to prove these aspects of her claim. Further, Ms Ingram said that QBE's reliance on the section 47(1)(b) EOA and section 46(2)(f) exceptions was untenable.
7. Finally, Ms Ingram said that there was significant disparity between her power as a young woman and that of a large, publicly listed company. She contended that the public interest element of this proceeding was also relevant. She contended that, the absence of a costs order would be likely to discourage other complainants from bringing claims of discrimination of public importance to the tribunal.

1. QBE contended that no further costs order ought to be made. Referring to section 109, it said that the usual rule that the parties bear their own costs should apply. In particular, it said:
   • There was no significant factual dispute between the parties in respect of the lay evidence. The focus of the context concerned the expert evidence;
   • QBE was not responsible for unreasonably prolonging the hearing - three days were allotted for the hearing and that is the time it took;
   • Neither party advanced a claim that was untenable in fact or law. Not only were the positions taken by the parties reasonably arguable but they were strongly asserted for good and proper reasons; and
   • An aspect of the proceeding involved a degree of complexity. The evidence regarding actuarial and statistical data had to be considered and none was adduced by either party that unnecessarily added to the technical evidence necessary to consider the statutory defences.

1. I start by noting that it is not unusual in a hotly contested matter for unexpected evidence to be adduced or for there to be mid-hearing applications in respect of new or additional evidence or for adjournments. The tribunal's approach to ruling on those matters is guided by its obligation to act fairly and in accordance with the rules of natural justice. The mere fact that there were some unexpected applications and a need for time to consider matters and obtain expert comment is not a matter on its own to attract an award of legal costs. Something more is required.
2. In not ensuring that its expert evidence complied with the tribunal practice note, QBE failed to comply with the tribunal's four orders dealing with that matter. It is correct that the absence of the required material in addition to the substantive expert report led to additional questions of Associate Professor Pitt, but as compared to his almost entire day in the witness box, that was a minor matter. The necessity for Associate Professor Pitt to give some evidence in chief, contrary to the tribunal's directions, occupied approximately an hour and a half. While unfortunate and apparently caused to some extent by a failure of preparation, such an event is, unhappily, reasonably commonplace in litigation at the tribunal and elsewhere.
3. Noting that Mr Dong's evidence took around two hours in total, I am not satisfied that the legal costs associated with that evidence warrants a costs order. As a matter of fairness to QBE, it was allowed to call the evidence and ultimately it assisted the tribunal to understand the material which underpinned the Pitt/Kyng report.
4. The adjournment application in order to seek new evidence took around three hours in total on day two of the hearing. Given the length of time Associate Professor Pitt was required to give evidence, I consider it unlikely that the hearing would have been completed in two days had that time not been spent on the adjournment application. At worst the three hours was lost. In my view, that was not an unexpected or unusual consequence of the conduct of litigation and does not entitle Ms Ingram to an order for any of the second day's costs
5. On the question of Mr Dong's evidence, I am satisfied that the lateness with which he was called and the late tender of different data sheets caused disadvantage to Ms Ingram as her expert was required to be present at the hearing longer than might have otherwise been necessary and to be asked to consider that material 'on the run'. While it was always intended that the expert witnesses be present for each other's evidence and have an opportunity to provide further expert evidence in response to each other, had Ms Bateup had the full data sheets relied upon and a witness statement from Mr Dong earlier, her report would have been more complete and it is likely she would have only been required at the tribunal for one day. In these circumstances, I am satisfied that QBE is liable to pay the costs of Ms Bateup's attendance for one day and I shall so order.
6. I will now turn to the general application for costs in the proceeding.
7. I have dealt with the matters relating to the failure to comply with the tribunal's practice note, the late service of materials, calling Mr Dong late, and the application for an adjournment above.
8. Given the conclusions reached on Ms Ingram's substantive application, I consider that the section 47(1)(b) of the EOA and section 46(2)(f) of the DDA defences relied on by QBE were weak. It was clear from the evidence QBE's solicitor gave in relation to the adjournment application that the requirements of those provisions and how they might apply to this proceeding had been misunderstood.[150] Arguably, the evidence of Mr De Luca was of limited use as much of it was directed to those defences. However, as ought also to be apparent from these reasons, Mr De Luca's evidence was relevant to my understanding of the background to the defences relied on and to some extent to my assessment of the section 29A of the DDA defence.
9. The ultimate consequence of QBE's approach to the case was to lead to it fail to prove its defence and be refused an opportunity to call entirely new evidence. Those matters assisted Ms Ingram to succeed but not in such a manner as to allow me to conclude that the defence was hopeless from the outset and ought never to have been relied upon. Further, the evidence called in respect of the three exceptions was essentially all the same and was required to consider whether the exceptions could be proven.
10. While I have noted the power disparity mentioned, I do not consider that to be a basis on which a costs order should be made. Ms Ingram has been represented by Victoria Legal Aid, as I understand it a body empowered and funded to provide legal advice and representation in cases of individual and community wide merit. While I am aware that the majority of the legal costs of VLA and Ms Harris of counsel will likely be paid from the public purse, I do not consider that to be a reason to order costs. This has not been a test case but rather is a case of a kind commonly run in the tribunal. My findings have turned on the evidence placed before me in circumstances where each party has very properly pursued its interests vigorously.
11. I am not persuaded that fairness calls for a costs order to be made in the face of the usual position set out in section 109 of the VCAT Act requiring parties to bear their own costs. Accordingly, the applications for costs are refused other than in respect of Ms Bateup's extra attendance.

[1] Particulars of defence dated 18 September 2014, paragraph 22.

[2] See section 6 of the EOA which sets out the attributes protected by the Act.

[3] Aitken & Ors v State of Victoria - Department of Education & Early Childhood Development [2012] VCAT 1547.

[4] Closing written submissions dated 20 November 2015, paragraph 98.

[5] Closing written submissions, paragraph 96.

[6] Section 13 of the EOA has the effect that discrimination is not prohibited if an exception applies to the conduct.

[7] Exhibit MD-1 to the affidavit of Mr De Luca National Manager - Sales, Distribution and Strategy with QBE (Exhibit R1).

[8] Evidence of Mr M De Luca, transcript, day 1, page 80.

[9] Policy, page 31. A claim may also be made if a business partner or employee made the cancellation necessary but that aspect of the clause is not relevant here. The term 'confinement' is not defined in the policy and was not the subject of submissions. I proceed on the basis that it does not extend to compulsory treatment of the kind regulated under the Mental Health Act 2014 (Vic) or its predecessors and so inpatient treatment for mental illness is captured by the term 'hospitalisation'.

[14] Policy page 28 and definitions pages 24 to 25.

[16] Particulars of complaint dated 28 August 2014, paragraph 20.

[17] Particulars of complaint dated 28 August 2014, paragraph 22.

[18] J Gardner, An Equality Act for a Fairer Victoria: Equal Opportunity Review Final Report, June 2008.

[20] Closing written submissions, paragraphs 19 and 20.

[23] Closing written submissions, paragraphs 68 to 70 and paragraphs 92 to 95.

[24] [1998] HCA 28.

[25] Reply submissions, paragraphs 10 to 14 - Slaveski v Smith [2012] VSCA 25 at paragraphs 24 and 45; Nigro v Secretary to the Department of Justice [2013] VSCA 213 at paragraph 85; and Victorian Toll & Anor v Taha [2013] VSCA 37 at paragraph 190. See also Momcilovic v The Queen [2011] HCA 34; Bare v Small [2013] VSC 129; and Kuyken v Chief Commissioner of Police [2015] VSC 204 at paragraph 78.

[26] See Project Blue Sky v ABA [1998] HCA 28; and Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation [1981] HCA 26 per Gibbs CJ at paragraph 6.

[27] Closing written submissions, paragraph 33.

[28] Closing written submissions, paragraphs 21 to 36.

[30] [2001] FMC 15 at paragraph 15.

[31] Closing written submissions, paragraphs 96 to 100.

[32] [1986] HCA 24.

[33] Closing written submissions, paragraphs 88 to 90.

[34] Reply submissions, paragraphs 3 to 5.

[36] Closing written submissions, paragraphs 46 to 56.

[37] Reply submissions, paragraphs 8 to 15.

[38] [2001] FMC 15.

[39] See QBE Travel Insurance v Bassanelli [2004] FCA 396.

[40] Bassanelli at paragraphs 31 and 79.

[41] Closing written submissions, paragraph 112.

[44] During the hearing and in the written submissions there was argument about whether QBE was confined to the above statistics or could rely on all of the material contained in the documents annexed to Mr De Luca's affidavit. Given my findings, it has not been necessary to determine this question finally.

[45] [1982] HCA 70 - closing written submissions, paragraph 107.

[49] Transcript, day 1, pages 64, 74, 85 and 108 to 109. Also see Associate Professor Pitt's evidence regarding reliance on the blanket exclusion at transcript, day 3, page 20.

[50] Closing written submissions, paragraph 82 relying on Jones v Dunkel (1959) HCA 8.

1. [51] Reply submissions, paragraphs 27 to 34.

[52] Closing written submissions, paragraph 116.

[53] Orders for expert evidence were made in the proceeding on 2 February 2015, 24 April 2015, 27 July 2015 and 26 August 2015. On each of the last three occasions, orders for costs thrown away were made against QBE.

[54] Closing written submissions, paragraphs 114 to 115.

[56] This summary is based on the parties closing written submissions which describe the case law on section 29A.

[57] Closing written submissions, paragraph 85.

[58] Closing written submissions, paragraphs 119 to 127.

[59] [2012] FCA 8.

[63] King v Jetstar Airways Pty Limited [2012] FCAFC 115, at paragraph 29.

[64] Reply submissions, paragraphs 16 to 34.

[65] Closing written submissions, paragraph 132.

[67] This material is taken from Exhibit R4.

[71] Closing written submissions, paragraph 117.

[72] Letter from Dr Kyng dated 26 November 2015 and letter from Ms Ingram's solicitors dated 27 November 2015.

[73] As I understand that QBE has no data for mental illness based cancellation claims, the monetary amounts used were based on the quantum of physical medical cancellation claims.

[74] Mr Dong - transcript, day 2, page 94 to 95; and Associate Professor Pitt - transcript, day 3, page 8.

[75] Transcript, day 2, pages 95 to 97.

[77] Transcript, day 2, page 76. I note that in the closing written submissions that ratio was stated as being 95% but that appears to be an error.

[80] Introductory section of report under heading 'Instructions'.

[85] Transcript, day 3, pages 27 to 28.

[89] For example, the evidence about other insurance policies was of little assistance. While some insurers had a similar exclusion, others did not. There was no evidence before me as to why the various insurers took different approaches or as to the premiums payable or the comparative benefits or terms of each. Many of the policies post-dated the relevant events. In these circumstances, I have not considered those policies further.

[90] Transcript, day 3, pages 67 to 69.

[91] Transcript, day 3, page 103 to 104.

[92] Closing written submissions, paragraph 111 to 113.

[99] Closing written submissions, paragraphs 114 to 115, referring in part to Associate Professor Pitt's evidence at transcript, day 3, pages 22 to 24.

[100] Closing written submissions, paragraphs 105 to 107.

[101] Closing written submissions, paragraph 136(d).

[102] Transcript, day 3, page 77 to 78.

[103] Transcript, day 3, pages 38 to 39.

[104] Transcript, day 2, pages 24 to 25.

[105] Closing written submissions, paragraphs 98 and 110 to 114.

[106] Closing written submissions, paragraph 136(b).

[107] Reply submissions, paragraphs 58 to 59.

[109] Transcript, day 3, pages 20 to 21.

[110] Pages 5 to 6 regarding pre-existing conditions and page 8 regarding the insured's duty of disclosure.

[112] Transcript, day 1, pages 97 to 98, 108 to 109

[113] Closing written submissions paragraphs 140 to 146.

[114] Reply submissions, paragraphs 37 to 51.

[116] Closing written submissions, paragraphs 131(b).

[117] Closing written submissions, paragraph 131(f) and 132(b).

[118] Final paragraph of the Pitt/Kyng report.

[119] Closing written submissions, paragraph 121 to 122.

[120] Closing submissions, paragraph 131(b), (e) and (f).

[121] Closing submissions, paragraph 131(d) and (f).

[122] Closing written submissions, paragraph 131(c).

[123] Francey v Hilton Hotels of Australia Pty Ltd [1997] HREOCA 56.

[125] Closing written submissions, paragraph 132.

[128] (1995) EOC 92-717.

[129] Sections 97 and 98 of the VCAT Act.

[130] Ainsworth v Criminal Justice Commission [1992] HCA 10.

[131] Closing written submissions, paragraph 132.

[132] Closing written submissions, paragraphs 156 to 170.

[133] Closing written submissions, paragraph 169(c).

[134] Per Bell J in Kracke v Mental Health Review Board [2009] VCAT 646 at paragraph 820.

[135] Per Hollingworth J in Alinta Asset Management v Essential Services Commission [2007] VSC 353 at paragraph 13.

[136] Dated 23 March 2015, paragraphs 23 to 24, 29 to 30 (Exhibit A1) and transcript pages 51 to 52 (cross-examination) and pages 52 to 53 (re-examination).

[137] Witness statement dated 23 March 2015, paragraph 19 (Exhibit A2) and transcript, page 59.

[138] Transcript, pages 51 to 52; Closing written submissions, paragraphs 172 to 175.

[139] I note that arguably QBE incorrectly distinguished the tribunal's decision in Bevilacqua v Telco Business Solutions (Watergardens) PL No 2 [2015] VCAT 693 - see paragraph 50 of the decision and paragraph 59 of the reply submissions.

[140] [2013] VCAT 1869 and [2014] VCAT 1442.

[141] [2015] VCAT 1432.

[142] Reply submissions, paragraph 65.

[144] GLS v PLP [2013] VCAT 221 at paragraph 275, per Garde P referring to I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd [2002] HCA 41 per Gaudron, Gummow and Hayne JJ at paragraphs 56 to 57.

[145] Richardson v Oracle [2014] FCAFC 82, per Kenny J at paragraphs 95 to 97, including reference to the Victorian Court of Appeal's decision in Amaca Pty Ltd v King [2011] VSCA 447, an asbestos related negligence case.

[146] Tan v Xenos (No 3) [2008] VCAT 584 at paragraph 556.

[147] Closing written submissions, paragraph 147 referring to DP McKenzie's decision in Styles v Murray Meats Pty Ltd [2005] VCAT 914 at paragraph 100.

[148] Reply submissions, paragraphs 68 to 80.

[149] Closing written submissions, paragraphs 192 to 193.

[150] Transcript, day 2, pages 46 to 47.