AustLII Home | Databases | WorldLII | Search | Feedback

University of New South Wales Law Journal

Faculty of Law, UNSW
You are here:  AustLII >> Databases >> University of New South Wales Law Journal >> 2020 >> [2020] UNSWLawJl 28

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Blackham, Alysia; Temple, Jeromey --- "Intersectional Discrimination in Australia: An Empirical Critique of the Legal Framework" [2020] UNSWLawJl 28; (2020) 43(3) UNSW Law Journal 773


INTERSECTIONAL DISCRIMINATION IN AUSTRALIA: AN EMPIRICAL CRITIQUE OF THE LEGAL FRAMEWORK

ALYSIA BLACKHAM[* ]AND JEROMEY TEMPLE[**]

Australian equality law is still largely dependent on individual enforcement to achieve systemic change. The degree to which discrimination law acknowledges and accommodates intersectional discrimination is a question of growing pertinence. This article bridges theoretical scholarship on intersectionality and empirical statistical evidence of how people experience discrimination in Australia, drawing on data from the 2014 General Social Survey, to critically evaluate the extent to which Australian discrimination law is able to accommodate intersectional experiences of discrimination. We argue that there is a fundamental disconnect between the legal framework, which focuses on separate and distinct ‘grounds’ of discrimination, and how people actually experience discrimination in practice, which is multiple and overlapping. This article offers concrete suggestions for how the legal framework and data collection could be improved to better integrate intersectionality in Australian discrimination law.

I INTRODUCTION

There is growing recognition of the complexity and cultural embeddedness of discrimination in Australian society.[1] While laws have been introduced at the federal, state and territory level to provide for individual actions to obtain redress for discrimination, this individual complaints model has failed to achieve meaningful systemic change.[2] To remedy this, there is increasing focus on the role of equality agencies and other enforcement bodies, and the strategic actions that can be taken to address discrimination.[3] For now, however, equality law is still largely dependent on individual enforcement to achieve systemic change. Thus, it is important to recognise and seek to address key limitations in the individual enforcement model, while recognising and acknowledging the potential for broader and more revolutionary change.[4]

While there are many doctrinal challenges facing equality law, the degree to which it acknowledges and accommodates intersectional discrimination is a question of growing pertinence. Scholars such as Atrey have undertaken extensive theorising of the extent to which intersectionality might be embedded in discrimination law.[5] This scholarship has not extended to considering the Australian legal framework. Further, there is a potential disconnect between theoretical accounts of discrimination law and intersectionality, and empirical scholarship on how people actually experience discrimination.[6] Indeed, despite the long history of intersectionality scholarship, there is a fundamental lacuna in empirical legal evidence of intersectionality as it relates to discrimination law.

In this article, we bridge theoretical scholarship on intersectionality and empirical statistical evidence of how people experience discrimination in Australia, drawing on data from the 2014 General Social Survey (‘2014 GSS’) collected by Australia’s official statistical agency, the Australian Bureau of Statistics (‘ABS’). Building on a theoretical account of intersectionality and how it relates to discrimination law (Part II), we critically evaluate the extent to which Australian discrimination law is able to accommodate intersectional experiences of discrimination, including by comparison to international experiences in the United Kingdom (‘UK’) and Canada (Part III). Drawing on statistical evidence of how discrimination is experienced in Australia, using data from the 2014 GSS, we argue that there is a fundamental disconnect between the legal framework, which focuses on separate and distinct ‘grounds’ of discrimination, and how people actually experience discrimination in practice, which is multiple and overlapping. We show that disadvantage is compounded by having multiple protected characteristics, and argue that multiple discrimination, across multiple contexts, affects a substantial proportion of the population. We therefore conclude in Part V by evaluating how the legal framework and data collection could be improved to better integrate intersectionality in Australian discrimination law.

II INTERSECTIONALITY AND DISCRIMINATION LAW

Intersectionality poses a fundamental challenge to ideas of ‘discrimination’ and the existing framing of discrimination law. Discrimination law in Australia is generally structured around protected ‘grounds’ or characteristics, prohibiting disadvantageous treatment on the basis of fixed categories.[7] In Victoria, for example, the Equal Opportunity Act 2010 (Vic) prohibits discrimination on the basis of the attributes of:

• age;

• breastfeeding;

• employment activity;

• gender identity;

• disability;

• industrial activity;

• lawful sexual activity;

• marital status;

• parental status or status as a carer;

• physical features;

• political belief or activity;

• pregnancy;

• race;

• religious belief or activity;

• sex;

• sexual orientation;

• an expunged homosexual conviction; and

• personal association (whether as a relative or otherwise) with a person who is identified by reference to any of the above attributes.[8]

Focusing on particular grounds helps to disambiguate acceptable and unacceptable distinctions between individuals, meaning that the ‘grounds’ of discrimination law are socially and historically contingent.[9] Indeed, recent reforms to discrimination law in Australia have seen the addition of multiple new attributes or grounds.[10]

However, there is increasing doubt as to whether people do belong, or should be seen as belonging, to fixed and unchanging ‘groups’ of this nature. Young has argued that groups are better identified relationally, instead of by focusing on attributes and categories.[11] Groups are defined in relation to each other, rather than any one privileged group occupying a ‘universal’ position.[12] This changes our view of difference from one of ‘other’ and deviation, to one of ‘specificity, variation, [and] heterogeneity’.[13] Group difference is about relations between groups, and interactions with social structures; it is therefore more appropriately seen as contextual, socially constructed, and contingent.[14] Groups themselves are unified by a sense of affinity and social interactions, not fixed characteristics, and are therefore fluid and capable of self-definition.[15] Thus, grounding anti-discrimination law in fixed categories risks embedding a view of difference as exclusion, antagonism and otherness.[16] Discrimination law’s reliance on protected grounds might therefore serve to indirectly reinforce societal norms.

Further, everyone has multiple, intersecting identities.[17] In contrast to this multiplicity, reliance on fixed grounds to anchor anti-discrimination law creates ‘single-dimension “silos”’ that undermine the flexibility and efficacy of the legal framework.[18] For Crenshaw, established ideas of discrimination encourage us to think about disadvantage ‘along a single categorical axis’,[19] limiting inquiry to the most privileged members of that group, and (in the case of sex discrimination, for example) ignoring class and race-based disadvantage.[20] By contrast, intersectional discrimination is where disadvantage is compounded, and the sum of disadvantage is greater than its individual components.[21] Protected characteristics interact to produce disadvantage which is unique and distinct from discrimination based on any one individual ground.[22] This is qualitatively different and more challenging than the idea of multiple discrimination, where disadvantage is additive.[23] Ignoring the insights of intersectionality risks marginalising those who have multiple protected characteristics, and concealing or obfuscating discrimination that is not easily contained or explained by a single silo or axis.[24]

Intersectionality therefore challenges discrimination law to ‘embrace the complexities of compoundedness’[25] by reorienting itself at the intersection of disadvantage.[26] For Atrey, intersectionality is a radical demand, focusing on both sameness and difference and understanding people’s multiple identities holistically, with the ultimate aim of transforming group disadvantage.[27] A fulsome understanding of intersectionality cannot be addressed by simply adding marginalised groups into the dominant legal framework or ways of thinking; properly addressing intersectionality requires a fundamental rethinking of existing structures[28] and, indeed, the entire framework of discrimination law.[29]

While theories of intersectionality pose fundamental challenges to the existing legal framework, and undermine essentialist understandings of identity,[30] they are beset by practical and theoretical problems. First, it is unclear how we should conceive of the ‘intersection’, especially when group identities are fluid and changing, and people may be advantaged in some areas and disadvantaged in others.[31] The notion of an ‘intersection’ may be insufficiently dynamic to accurately capture the complexities of identity.[32] Second, intersectionality may be too focused on the individual, downplaying the importance of social processes.[33] Third, and integrating the two previous concerns, it is unclear how a meaningful understanding of intersectionality should be operationalised, including in law and the legal framework. Intersectional thinking risks devolving into the ‘infinite elaboration of inequality subgroups’, fragmenting and confounding conceptual ideas of equality and disadvantage.[34]

Atrey responds to these criticisms by focusing on the contextual and transformative nature of intersectionality. Given intersectionality is grounded in context, it is concerned with identity categories exactly because of the inequalities and power relations that attend them.[35] Thus, intersectionality is deeply embedded in social processes and power dynamics, using the individual as a lens for focusing on systemic power disparities. Indeed, intersectionality’s focus on transformation emphasises the importance of social processes and looking beyond the individual.[36] Intersectionality avoids the infinite regress issue in much the same way, by focusing on identity categories purposefully and in a grounded way, with the aim of redressing and transforming disadvantage.[37] In this way, too, intersectionality can recognise the fluidity of identity, through a critical reclamation of identity categories.[38]

The question remains, of course, whether this focus on transformation, social processes and power dynamics has sufficient utility to overcome the abiding practical and theoretical problems inherent in intersectionality. While intersectionality offers an improved lens for understanding disadvantage and its practical manifestations, particularly when compared to a siloed approach to protected characteristics, it remains a theory which is difficult to comprehend, communicate and implement in a statutory framework. It is to these practical questions of implementation that we turn in the next section.

III INTERSECTIONALITY AND THE LEGAL FRAMEWORK

Despite the importance of intersectionality for understanding disadvantage, the discrimination law framework in Australia is fundamentally ill-equipped for dealing with instances of intersectional discrimination.

A Federal Discrimination Law

At the federal level, discrimination law is contained within five separate statutes, four of which are based on different protected grounds – sex, race, age, and disability[39] – with noticeable inconsistency between the statutes. There is minimal attempt to accommodate or integrate intersectional disadvantage or even multiple discrimination. The only concession to individuals experiencing intersectional discrimination lies in the causation provisions in the statutes: for example, section 16 of the Age Discrimination Act 2004 (Cth) provides that if discrimination is ‘done for 2 or more reasons; and one of the reasons (whether or not it is the dominant or a substantial reason) is [age]’, then the act ‘is taken to be done because of the age of the person’. This means that a claim of multiple or intersectional discrimination is at least not explicitly prohibited by the existing statutes,[40] though there is no explicit recognition of how discrimination may intersect or compound across protected characteristics.

That said, the current structure of federal anti-discrimination law in Australia creates three key problems for intersectional claims. First, requiring claimants to prove each ground of discrimination separately under different (potentially inconsistent) statutes substantially increases the burden of proving multiple or intersectional discrimination. Second, the structure of discrimination law – and the disaggregating of different identities – is inconsistent with how identity is lived and discrimination is experienced in practice.[41] Third, and more specifically, if a different actual or hypothetical comparator is required to prove discrimination on the basis of each protected characteristic, this is likely to seriously jeopardise the success of intersectional claims.[42] The law is unclear on this point: how to choose a comparator for intersectional claims is far from self-evident.[43] Atrey argues that using a single comparator in claims of intersectional discrimination is unsatisfactory, as it is unguided by principle and complicated in practice to choose the appropriate comparator, and the approach fails to acknowledge and reflect the nature of intersectional discrimination.[44] As it is currently used, comparison in discrimination law is ‘highly resistant to intersectionality’, undermining the prospects of success for an intersectional claim.[45]

Recognising some of the difficulties facing intersectional claims in Australia, the Australian Human Rights Commission’s (‘AHRC’) own priorities for law reform therefore include ensuring that that the federal legislative framework becomes:

• consistent, including across key definitions in discrimination statutes, unless differentiation is required to accommodate something distinct about the particular ground; and

• intersectional, such that protections for different grounds ‘work together easily’.[46]

Similarly, the Willing to Work: National Inquiry into Employment Discrimination against Older Australians and Australians with Disability report on age discrimination recommended ‘[t]hat the Australian Government amend federal discrimination laws to apply to discrimination based on a combination of attributes protected under federal discrimination laws’.[47]

This was taken up in the proposed (but ultimately abandoned) Human Rights and Anti-Discrimination Bill 2012 (Cth) (‘Bill’). Had it been pursued, the Bill would have consolidated all federal discrimination statutes, and included recognition of multiple (though not intersectional) discrimination in section 19(1): ‘A person (the first person) discriminates against another person if the first person treats, or proposes to treat, the other person unfavourably because the other person has a particular protected attribute, or a particular combination of 2 or more protected attributes’.

When the Bill was abandoned, so too was the chance to recognise multiple discrimination in a unified federal statute, let alone intersectionality.

B State and Territory Law

Australian States and Territories have adopted unified anti-discrimination statutes, with each jurisdiction having one piece of legislation that covers all protected grounds.[48] However, there is no explicit recognition of intersectionality in any state or territory statute, and only minimal accommodation of multiple discrimination. While state and territory legislation does not require that one attribute be the only reason for particular treatment, there is minimal recognition of intersectionality. The Equal Opportunity Act 2010 (Vic) illustrates the general approach taken in most jurisdictions: section 7(1)(a) defines discrimination as beingdirect or indirect discrimination on the basis of an attribute’ (emphasis added), though section 8(2)(b) provides that ‘[i]n determining whether a person directly discriminates it is irrelevant ... whether or not the attribute is the only or dominant reason for the treatment, provided that it is a substantial reason’.[49]

The Australian Capital Territory (‘ACT’) is the only jurisdiction with explicit recognition of multiple discrimination, following amendment by section 6 of the Discrimination Amendment Act 2016 (ACT). Section 8 of the Discrimination Act 1991 (ACT) now says:

(2) For this section, a person directly discriminates against someone else if the person treats, or proposes to treat, another person unfavourably because the other person has 1 or more protected attributes.

(3) For this section, a person indirectly discriminates against someone else if the person imposes, or proposes to impose, a condition or requirement that has, or is likely to have, the effect of disadvantaging the other person because the other person has 1 or more protected attributes (emphasis added).

This section, while the most progressive in Australia, reflects an additive view of discrimination, rather than the complex ‘compoundedness’ of intersectionality. It is also unclear what effect (if any) this ACT provision is having in practice. The revised section 8 commenced on 24 August 2016, meaning there has been little time for relevant cases to progress to a hearing (and, indeed, few discrimination cases in Australia proceed to a court or tribunal).[50] In a survey of ACT Civil and Administrative Tribunal discrimination decisions reported between 2017 and April 2020, two potentially intersectional cases related to events prior to 24 August 2016,[51] and two other potentially intersectional cases were dismissed as being frivolous and vexatious and lacking in substance.[52] Only one decision addressed the amended section 8 in any detail: in Complainant 201808 v Transport Canberra and City Services[53] the claimant alleged age and race discrimination and sexual harassment, as a young person of Croatian heritage. Despite the more lenient test in section 8 of the Discrimination Act 1991 (ACT), the Tribunal appeared to consider these grounds separately, and the claim failed on causation:

even assuming that these acts – individually or collectively – constitute unfavourable treatment, the Discrimination Act is breached only where the ‘true basis or real reason’ for that conduct is the existence of a protected attribute (in this case, the applicant’s age or her race or a characteristic thereof). There must be a proven causative relationship between the discrimination and the protected attribute. ...

There is simply nothing in the evidence before the Tribunal that could found an inference that any of the incidents of alleged unfavourable treatment were due to age or race or any characteristic thereof, or that any of the unfavourable events were due to the complainant’s age or race.[54]

Rather than being due to age or race, the Tribunal inferred that any unfavourable treatment was ‘a direct result of the complainant’s own conduct and behaviour and the nature of her interactions with her colleagues’.[55] This was not characteristic of her age or race. The approach of the Tribunal in this case does not engage with the complexity of identity or the ‘compoundedness’ of intersectionality. Rather, it reflects the additive approach adopted in the legislation itself.

C International Perspectives

Australia is not alone in neglecting to accommodate intersectional and multiple discrimination in its legislative framework. Few European Union (‘EU’) countries have embedded intersectionality in their statutory framework, and it is a concept that rarely appears in EU case law.[56] This may be because, as in Australia, intersectionality is increasingly conflated with multiple or additive discrimination. That said, even multiple discrimination is only included in a handful of national legislative frameworks.[57] Confining understandings of intersectionality to additive or multiple discrimination ignores the radical implications of intersectionality as a theory of inequality, and may effectively negate any need for legal change.[58]

The Equality Act 2010 (UK), for example, includes a legislative section relating to combined or additive discrimination, but makes no mention of intersectionality; even this more limited section has never become operative. Section 14 relevantly says:

A person (A) discriminates against another (B) if, because of a combination of two relevant protected characteristics, A treats B less favourably than A treats or would treat a person who does not share either of those characteristics. ... B need not show that A’s treatment of B is direct discrimination because of each of the characteristics in the combination (taken separately).

Section 14 acknowledges that combined discrimination is different to just ‘additive’ discrimination, and that it may not be possible to prove discrimination on the basis of each characteristic separately. That said, Fredman has critiqued the framing of section 14 on the basis that it is confined to only two protected characteristics, meaning that those who differ the most from the ‘norm’ are least likely to receive legal protection.[59] Further, Solanke has argued that section 14 does not challenge the ‘silos’ of protected grounds in the Equality Act 2010 (UK); instead, ‘the intersectionality provision has been designed to accommodate the silos’.[60]

The limited scope of section 14 of the Equality Act 2010 (UK) may be contrasted with the broader provision in section 3.1 of the Canadian Human Rights Act, RSC 1985, c H-6, which says: ‘For greater certainty, a discriminatory practice includes a practice based on one or more prohibited grounds of discrimination or on the effect of a combination of prohibited grounds’.

Hepple has argued that the failure to explicitly recognise intersectional or multiple discrimination in the Equality Act 2010 (UK) does not need to result in a gap in legal protection: courts can still consider two or more grounds, and cases in the UK have been successful on this basis.[61] However, this is unlikely to address the very real challenges faced by claimants pursing multiple or intersectional claims;[62] indeed, United States (‘US’) studies have found that multiple claims are less likely to succeed.[63] Further, these legislative provisions – across multiple jurisdictions – appear to be only a pale reflection of the complexity and radical potential of intersectionality. This may reflect the difficulty of practically implementing such radical ideals; or may just indicate that more concerted progress and reform is required.

D Intersectionality Case Law

The challenges of pursuing an intersectional claim in the Australian context are reflected in the very small number of cases that explicitly address intersectional issues.[64] While there are some historical cases that present intersectional issues,[65] these are far from the norm. Mansour argues that this might reflect overlapping factors, such as the small number of discrimination cases in Australia,[66] and that claimants may not present an intersectional claim as such, instead focusing their claim on one particular characteristic or ground.[67] (Or, indeed, multiple grounds may be abandoned as a claim progresses).[68]

Those who experience intersectional discrimination may also be less likely to engage with the legal system.[69] In her study of age discrimination case law in Australia, for example, Blackham found that the majority of claimants were older men (68% of the sample, or 73 claimants).[70] This is consistent with Schuster and Miller’s quantitative study of US federal age discrimination cases, which found that a majority (57%) of 153 age discrimination claims were brought by white men in professional or managerial positions.[71] Thus, intersectional claimants may be less likely to utilise legal mechanisms to address their rights, or at least to progress to a public court hearing.

It is difficult to assess the extent to which intersectional complainants are also under-represented at complaint or conciliation level. The AHRC and similar state and territory bodies do not publish sufficiently disaggregated data to determine who is making claims and on what basis,[72] and complaint files are otherwise confidential.[73] While the bodies publish some limited claimant demographic data in their annual reports, and selected conciliation case studies are available on the AHRC and New South Wales conciliation registers, this is not sufficiently disaggregated data to assess questions of intersectionality. This may be because the number of claims in each group or category in such publication would be so small that it would risk identifying certain claimants.[74]

In the public realm, few discrimination decisions explicitly engage with the idea of ‘intersectionality’. One exception to this is the Queensland decision of Till v Sunshine Coast Regional Council,[75] (‘Till’) where intersectional discrimination was integral to a challenge to access arrangements to a patrolled beach on the Sunshine Coast. Mr Till claimed that a patrolled beach was not accessible because one route had steps, and another route had a dog off-leash area. Mr Till claimed that this indirectly discriminated against him on the basis of (1) his parental status, (2) family responsibilities, and (3) association with persons of particular attributes (that is, the age and impairment of his children); and that his children had been indirectly discriminated against because of their age and impairment.

The Tribunal accepted this intersectional argument, noting:

Mr Till presents his claim with the intersecting attributes of parental status, family responsibilities and association with persons of a young age. The Tribunal broadly accepts that accompanying children or very young children is a characteristic that a parent generally has, or is imputed to have, and that taking children out as either part of a parent’s exercise regime or the children’s exposure to the outdoors, is part of Mr Till’s family responsibilities as he cares for and supports two dependant children. ... The Tribunal is therefore satisfied Mr Till is unable to comply with the terms imposed because of a characteristic of a parent (accompanying young children) and because of his family responsibilities to care for and support two dependant children, based on his concerns about safety.[76]

The Tribunal was also not particularly exacting when establishing whether people without these attributes would be better able to comply with the access path to the beach:

in keeping with the broad intention of the law to promote equal opportunity, and in light of no submission on this point from the Respondent, the Tribunal accepts the general premise that a higher proportion of people who are not accompanied by very young children can comply with the terms regarding [access].[77]

Thus, potential difficulties with identifying a comparator did not beset Mr Till’s claim. Ultimately, however, the Tribunal found that the access requirements were reasonable, meaning Mr Till’s claim did not succeed. Till indicates that courts and tribunals may accommodate intersectional claims, even where the statute does not make explicit provision for intersectional discrimination.

However, other cases that potentially illustrate intersectional discrimination are run with limited recognition of potential overlap between grounds. In Travers v New South Wales,[78] for example, the claimant alleged that she had been discriminated against on the basis of age and disability in her role as an exam supervisor, allegedly being denied breaks to go to the toilet and eat to manage her diabetes, dismissed, and told she was ‘too old to work’, ‘forgetful’ and ‘to fuck off and don’t come back’.[79] Ms Travers’s age and disability claims were considered essentially in tandem in the Federal Circuit Court; the age complaints were summarily dismissed, and the claim of disability discrimination for a failure to provide reasonable adjustments was allowed to proceed.

Even assuming Ms Travers could establish these alleged facts, the Court held that she could not show that she was treated less favourably than her comparator. In this case, the comparator was held to be a person who did not have her disabilities or was of a different age;[80] that is, someone who had also ‘failed in her task of properly supervising the examinations’.[81] Further, Ms Travers would not be able to prove that her treatment was because of her age or disability: rather, it was because her employer ‘was dissatisfied with the manner in which Ms Travers supervised the examinations’.[82] This ignores the potential link between Ms Travers’s disabilities and age, and her inability to supervise exams effectively without reasonable breaks (though these issues could be pursued in the reasonable adjustments claim). Thus, the construction of the comparator requirement and causation proved fatal, at least to the age discrimination claim.

Similarly, Thompson v Big Bert Pty Ltd[83] raised issues of indirect age and sex discrimination when a bar attendant had her casual shifts reduced. Ms Thompson alleged that the owner of the hotel wanted to replace older staff with ‘young glamours’, and that her shifts were varied to a pattern inconsistent with her caring responsibilities as a single parent, to force her out of employment.[84] The Court held that the changes to Ms Thompson’s working arrangements ‘were initially prompted by [the manager’s] response to the need to reduce the bar wages bill and the overall number of working hours’, then driven by ‘Ms Thompson’s persistent and unwelcome complaints about the change to her working arrangements’, which the manager found ‘tiresome and unacceptable’.[85] Of course, an unexpected change to a single parent’s working hours is likely to raise particular challenges; the Court held, however, that there was no basis for any claim of indirect sex discrimination, as Ms Thompson’s position was no different to that of other single parents.[86] While the Sex Discrimination Act 1984 (Cth) prohibits direct discrimination on the basis of family responsibilities, it does not make indirect discrimination on this ground unlawful.[87] This case reflects the challenges of pursuing intersectional claims within an incomplete and inconsistent legislative framework.

Overall, as Gaze and Smith argue, ‘law fails to acknowledge [claimants’] actual experience, in which the attributes that affect them are a whole and may not be separable, and the disadvantage that each attracts may actually be compounded by their combination’.[88] What this Part has shown, however, is that there is very limited data from the courts and equality agencies about those experiencing intersectional discrimination, and whether the legal system is able to support their needs. Data, case law, and the legislation itself are largely silent on intersectional discrimination; it is likely that this represents a significant black hole in legal protection.

In the Parts that follow, we seek to address this lack of information using data from the 2014 GSS to provide an empirical illustration of individuals’ actual experiences of discrimination. We show how disadvantage is compounded by multiple characteristics, and argue that multiple discrimination, across multiple contexts, affects a substantial proportion of the population.

IV EMPIRICALLY MAPPING DISCRIMINATION IN AUSTRALIA

The 2014 GSS offers one of the most substantial datasets on individual experiences of discrimination in Australia. The 2014 GSS was conducted by the ABS between March and June 2014, and represents a comprehensive dataset with nationally representative estimates and a large sample size. The 2014 GSS is uniquely positioned as a survey run by the government that captures experiences of discrimination in Australia, being designed to understand the ‘multi-dimensional nature of relative advantage and disadvantage across the population, and to facilitate reporting on and monitoring of people’s opportunities to participate fully in society’.[89]

Data for the 2014 GSS was collected using face-to-face interviews along with prompt cards and a Computer Assisted Interviewing (‘CAI’) questionnaire.[90] The 2014 GSS included an initial sample of 18,574 private dwellings, of which 16,145 dwellings were used due to issues of scope or uninhabited dwellings. In total, 80% of households fully responded, yielding a sample of 12,932 people aged 15 years and over.

The 2014 GSS collects demographic data from respondents, which broadly maps onto the ‘grounds’ protected under discrimination law. For the purposes of this study, we mapped this data onto the following variables:

• Gender (Male, Female);

• Age (18–34, 35–59, 60+);

• Country of Birth (Australia, Mainly English-Speaking Background (‘MESB’) countries, Non-English Speaking Background (‘NESB’) countries);

• Disability Status (No, Yes);

• Heterosexual Status (No, Yes);

• Parent of a Dependent Child (No, Yes); and

• Carer (No, Yes).[91]

The 2014 GSS also measures individuals’ experiences of discrimination. Respondents were asked: ‘In the past 12 months, that is since this time last year, do you feel that you have experienced discrimination or have been treated unfairly by others?’ Those responding ‘yes’ were further prompted: ‘In the past 12 months, in which places or situations do you feel that you have experienced discrimination or have been treated unfairly?’ A prompt card was then shown to the respondent listing:

• at home;

• at work;

• at the shop;

• on public transport;

• at school or university;

• in a restaurant or bar;

• on the street or in a public place;

• online;

• applying for work/jobs;

• applying for or keeping a flat/apartment or housing of any kind;

• dealing with police;

• dealing with the courts;

• dealing with government officials;

• dealing with people involved in healthcare; and/or

• other.

The grounds or characteristic to which the discrimination was attributed was measured by asking: ‘Thinking about your most recent experience of discrimination in Australia, do you think it was because of any of the following?’ A prompt card was then displayed listing:

• your skin colour;

• your nationality, race or ethnic group;

• the language you speak;

• the way you dress or your appearance;

• your gender;

• your age;

• a disability or health issue;

• your marital status;

• your family status;

• your sexual orientation;

• your occupation;

• your religious beliefs;

• your political position; and/or

• other.

Even at a first glance, it is clear that these questions are imperfect for considering the implications of individual lived experiences for an intersectional legal framework. First, what individuals regard as ‘discrimination’ may be far removed from what the legal system would hold to be ‘discrimination’; there is a potential disjunction between legal and popular understandings of this notion.[92] Second, the question about protected grounds does not fully capture intersectional discrimination as such; rather, it focuses on individual protected characteristics, gathering data on additive or multiple discrimination. While we can therefore measure multiple discrimination based on these self-reports, we cannot obtain a full picture of intersectional discrimination using this data. This partly reflects the difficulty of operationalising, describing and measuring intersectionality, at least with this style of questioning. Third, the grounds or attribute question was only directed to individuals’ most recent experience of discrimination; other experiences could have related to other or different grounds. Overall, then, this dataset is an incomplete addition to our understanding of intersectionality, though it flags the potential importance of intersectionality for the legal framework.

Recognising these limitations, using these questions we developed two additional measures. First, multiple discrimination, where an individual reports more than one type or ground of discrimination – for example, race discrimination alongside age discrimination. Second, multiple context discrimination, where an individual reports exposure to discrimination in multiple contexts – for example, in the workplace and in public settings.

In addition to the specific issues raised by this data for measuring intersectional discrimination, it is also important to bear in mind the more general limitations of the dataset and the analyses that results from it. While the 2014 GSS represents high quality data on experiences of discrimination in Australia, it is limited in five additional ways. First, data from the 2014 GSS are cross-sectional and represent one point in time. We cannot (and do not) draw causal relationships between discrimination and certain demographic characteristics. Second, our measures of discrimination are self-reported over a period of 12 months. Thus, recall bias over this timeframe may be an issue for some respondents. Third, some respondents may feel uncomfortable disclosing instances of discrimination, again biasing prevalence rates downwards. Fourth, the 2014 GSS sampling frame was limited to people living in private dwellings, such as private family homes. Those living in high care nursing homes, hospitals or other non-private dwellings were not included in the survey. Further data collection would be necessary to generalise the findings to these populations. Fifth, our measures only relate to interpersonal discrimination. Structural or systemic discrimination is an important component of the overall level of unfair treatment experienced by vulnerable groups and will not be captured in these measures.

While recognising these limitations, the 2014 GSS data still reveal telling findings about the prevalence of discrimination and multiple discrimination among Australian respondents. Table 1 displays the prevalence of any form of discrimination, disaggregated by the above demographic groups. It is clear that there is considerable heterogeneity in the levels of exposure to discrimination across these measures. For example, over one in three respondents within the following groups report discrimination: middle age female non-heterosexuals (49.6%), young female and male carers (38.5% and 36.8% respectively), young male non-heterosexuals (36.7%), and young and middle age people living with disabilities. In contrast, reported discrimination is lower among older Australians in general.

2020_2800.jpg

However, these broad prevalence rates of discrimination omit any understanding of the grounds and context of discrimination experienced by different demographic groups. Table 2 displays the type or ground of discrimination reported for those reporting exposure to any type of discrimination in the last year.[93] Respondents could report multiple types of discrimination, and this is referred to as ‘multiple discrimination’ in the table. As would be expected, the type or ground of discrimination reported is strongly associated with an individual’s demographic characteristics. For example, older Australians are more likely to report age discrimination (38.4%) relative to middle age Australians (16.9%); women are more likely to report gender discrimination (22.4%) relative to men (8.7%); and people from NESB countries are more likely to report race discrimination (82.2%) when compared to those born in Australia (17.4%).

These simple comparisons gloss over the total level of discrimination reported by impacted groups – in total, 38.9% of those reporting any form of discrimination, report experiencing multiple types of discrimination. As a specific example, around 22% of those experiencing discrimination who have a disability attribute it to their underlying health condition. However, this population also strongly reports age discrimination (24%), race discrimination (27%) and other forms of discrimination (46%). Therefore, focusing only on disability discrimination considerably underestimates not only the prevalence but also the range of discrimination experienced by Australians living with disabilities.

Table 2: Type of Discrimination Recorded, by Demographic Characteristics (%), 2014



Type of Discrimination Reported


Race or
Gender
Age
Disability
Marital
Sexuality
Other
Multiple
Ethnicity
Status
Age










18-34
36.9
19.7
21.1
5.4
10.1
4.3
53
44.4

35-59
34.8
15.5
16.9
14
10.1
2.6
46.9
38.5

60+
27.9
11.5
38.4
10.3
5.7
<1
40.1
29.4
Sex










Male
38.8
8.7
19.7
7
5.5
4.2
53.1
37.5

Female
30.7
22.4
22.7
12.9
12.4
2.5
44.9
40.1
Country of Birth









Australia
17.4
19.8
24.7
12.9
11.6
4
55.5
35.1

MESB
32.7
16.8
22.2
8.9
5.5
5.3
52.3
38.3

NESB
82.2
5.6
11.7
3.2
4
<1
28.2
49.7
Disability










No
38.9
16.4
19.9
3.1
8.8
2.8
50
38.7

Yes
27.1
15.6
23.7
21.8
9.9
4.2
46.4
39.2
Heterosexual









No
24.1
19.2
20.4
13.7
10
31.7
51.1
42.2

Yes
35.5
15.8
21.5
9.8
9.1
<1
48.4
38.6
Parent










No
32
17.3
22.6
11
7.4
3.8
49.6
38.1

Yes
37.5
14.6
19.7
9.2
11.6
2.6
47.4
39.9










All
34.4
16.1
21.3
10.2
9.2
3.3
48.6
38.9

Notes: Race or ethnicity discrimination includes respondents reporting any of ‘your skin colour’, ‘your nationality, race or ethnic group’, or ‘the language you speak’. Other discrimination includes respondents reporting any of ‘the way you dress or your appearance’, ‘your religious beliefs’, ‘your political position’, or ‘other’.

Table 3 places discrimination within the context of the place of exposure – for example, in the workplace or in an educational setting. Among all respondents reporting discrimination, the key contexts of exposure include the workplace or job search (53.4%), public (41.1%) and government or health related environments (16.3%). Almost half of respondents indicated multiple context discrimination in the previous 12 months (47.7%). Not surprisingly, about one quarter of people living with disabilities, or persons caring for those with a disability or long-term health condition, cite government services and health care as an important point of exposure to discrimination. For non-heterosexual Australians, educational settings are a key point of exposure, and about 55% report multiple context discrimination.

Table 3: Context of Discrimination Reported, by Demographic Characteristics (%), 2014



Context of Discrimination Reported


Home
Work
Legal
Public
Education
Gov’t
Other
Multiple
Health
Age










18-34
6.7
58.9
13.1
49
10
9
19.7
54.9

35-59
9.7
56.1
11.9
38.1
5.8
20.3
13
45.7

60+
10.6
35
8.3
33.9
<1
24.3
17.1
32.8
Sex










Male
7.6
57.2
15.1
41.7
7.1
15.5
17
48.8

Female
9.5
50.1
8.4
40.5
11.9
16.3
15.7
46.9
Country of Birth









Australia
10.1
51.5
12.5
36.6
10.5
16.8
17.8
45.8

MESB
8.6
60.8
8
32
9.9
14.6
14.7
39.3

NESB
4.6
55.7
10.1
56.9
7.3
14.3
12.6
56.2
Disability










No
6.4
57.3
10.5
41.9
9.3
11.2
15.2
47.3

Yes
12.3
46.8
13.1
39.7
10.3
23.7
18.1
48.4
Heterosexual









No
7.3
41.2
8.6
43.7
35.3
12.7
23.8
54.3

Yes
8.8
54.7
11.8
40.8
6.8
16.3
15.4
47
Parent










No
9.9
54
11.5
40
3.8
16.4
17.8
46.2

Yes
6.9
52.5
11.5
42.4
17.2
15.4
14.3
49.8
Carer










No
8.2
56.6
11.9
39.5
9.4
13.1
15.3
47.4

Yes
10
43
10
46.2
10.7
25.1
19.2
49










All
8.6
53.4
11.5
41.1
9.7
16
16.3
47.7

Notes: Public includes ‘at the shop’, ‘on public transport’, ‘in a restaurant or bar’, ‘on the street or in a public place’.

Finally, types or grounds of discrimination also interact with the context in which it is experienced, as depicted in Table 4. Respondents reporting race, sex, age and/or sexuality discrimination were all more likely to report discrimination in public, when compared to those not reporting these forms of discrimination. For example, 60% of those exposed to sexuality discrimination report that it occurred in a public setting. Sex and age discrimination are also particularly pronounced in the workforce (reported by around 63% of respondents). Around 43% of Australians exposed to disability or health-based discrimination note a government service or healthcare-based context of discrimination. For those reporting multiple discrimination, the majority report work (61%) and public (56%) contexts of discrimination. A further one in five report government services or healthcare (19.5%) or other contexts (19.3%).

Table 4: Type of Discrimination by Context of Discrimination Reported (%), 2014



Context of Discrimination Reported


Home
Work
Legal
Public
Education
Gov’t
Other
Multiple
Health
Type of Discrimination







Race or Ethnicity









No
10.3
54
11.6
32.8
9.3
16.8
16.5
42.4

Yes
5.5
52.1
11.3
56.8
10.5
14.4
15.8
57.8
Gender










No
8.1
51.5
11
39.4
9.2
16.2
16.7
45.3

Yes
11.1
62.8
14.1
49.8
12.4
15
13.8
60.4
Age










No
8.8
50.5
11.2
39.5
9.6
15.7
16.2
45.2

Yes
7.8
63.8
12.5
47
9.8
16.9
16.6
57
Disability or Health









No
8.4
54.2
11.3
41.5
9.8
12.9
15.5
46.6

Yes
10.5
46
12.8
37.5
9
42.8
22.7
58
Marital or Family Status








No
7.8
52.9
11.1
40.6
10
15.1
16
46

Yes
16.8
58.3
15.4
45.4
6.7
24.4
19.3
64.8
Sexuality










No
8.7
53.5
11.6
40.4
9.3
15.9
15.9
47.4

Yes
5.3
49.1
9
60.2
21.7
17.2
27.1
58.4
Other










No
7.8
51
10.4
44.1
9.1
15.1
11.7
47.1

Yes
9.5
55.8
12.7
37.8
10.2
16.9
21.1
48.4
Multiple










No
9
48.8
10.2
31.8
9
13.7
14.4
35.4
Yes
8
60.6
13.5
55.7
10.8
19.5
19.3
67.1

Notes: Public includes ‘at the shop’, ‘on public transport’, ‘in a restaurant or bar’, ‘on the street or in a public place’. Race or ethnicity discrimination includes respondents reporting any of ‘your skin colour’, ‘your nationality, race or ethnic group’, or ‘the language you speak’. Other discrimination includes respondents reporting any of ‘the way you dress or your appearance’, ‘your religious beliefs’, ‘your political position’, or ‘other’.

V DISCUSSION

These empirical findings provide a statistical illustration of how discrimination is potentially compounded for impacted groups. Those who belong to multiple impacted populations are far more likely to report experiencing discrimination. Further, this discrimination is likely to align with relevant protected characteristics: migrants from a non-English speaking country are more likely to report race discrimination compared to those born in Australia; women are more likely to report sex discrimination relative to men; and older Australians are more likely to report age discrimination relative to middle age Australians. However, focusing on only one protected characteristic tends to significantly understate individuals’ lived experiences of discrimination: about 40% of those exposed to discrimination report experiencing multiple discrimination. Moreover, almost half of those exposed to discrimination report experiencing multiple context discrimination. While the workplace and public life are key contexts of discrimination across all grounds, for people with disabilities, healthcare and government services are important areas of exposure.

Overall, then, our findings underscore the importance of contextualising individual experiences of discrimination. Discrimination is not experienced in a simple or straightforward way: in practice it is multiple, overlapping and complex. In many cases, it is not confined to one ground or one context. It is pervasive across areas of public and private life.[94]

One thing is plainly evident: the existing Australian legal framework does not accommodate this complexity and overlap. At the very least, there is a need for a legal framework with more coherent and consistent statutes across different grounds or protected characteristics. More fundamentally, however, there is a need to build recognition of intersectional discrimination into the legal framework. While our data can only directly illustrate the prevalence of multiple or additive discrimination, it clearly implies that discrimination is compounding, complex, and overlapping. There is therefore an urgent need for recognition of intersectionality in the legal framework. This can be compared with the sheer scarcity of intersectional case law in Australia. As Atrey argues: ‘The rarity of successful claims of intersectional discrimination should give us a sobering idea of the urgency of intersectionality as a juridical project’.[95] The practical question, though, is how we should best accommodate or operationalise intersectionality, especially given the complexity of the idea itself.

Mansour argues that there are two ‘trends’ in international legal reform to accommodate intersectionality: first, adopting one equality Act for all protected grounds, with standardised tests and definitions; and, second, making explicit provision for combined claims, though jurisdictions differ as to whether this is on the basis of ‘multiple’, ‘intersecting’ or a ‘combination’ of attributes.[96] The drafting of this provision in Canada is preferable to that in the UK, as discussed in Part III(C).[97] Alternatively, Victoria Legal Aid has recommended ‘that the test for discrimination ... explicitly state that reference to a protected attribute is a reference to one or more protected attributes’.[98] Adopting such a provision is important at both a technical legal level, and at a symbolic level, in directing the attention of practitioners, policy-makers and equality agencies to the challenges facing those experiencing intersectional discrimination.[99] This could be complemented by the removal of the comparator test, including by moving to a test of unfair disadvantage or treatment.[100] The Equal Opportunity Act 2010 (Vic), for example, defines direct discrimination as occurring ‘if a person treats, or proposes to treat, a person with an attribute unfavourably because of that attribute’.[101]

Effectively embedding intersectionality in discrimination law is a gargantuan project – as Atrey demonstrates in the context of other jurisdictions, reframing discrimination law to enable intersectional claims involves re-examining every part or aspect of discrimination law, to ensure it operates seamlessly as a whole for intersectional claims. For Atrey, this involves a reconsideration of: the scope, test and text of the prohibition of discrimination; grounds; use of comparators; defences; the burden of proof; and remedies. As Atrey argues: ‘It is only when each of these independently and simultaneously responds to intersectionality favourably that an intersectional claim may succeed’.[102] While it is beyond the feasible scope of this article to consider all of these areas in detail, it is clear that making explicit provision for intersectional claims in discrimination law is just the first step of this project of legal reform.

Even without legal change, there is still scope for a more purposive and sympathetic approach to the interpretation of existing legal frameworks, in a way that would assist intersectional claimants, as is evident in Till. Till demonstrates that the strategic interpretation of the law could remedy some of these legal gaps: indeed, Atrey concludes that the wording of statutes is ultimately less important than how they are interpreted.[103] Sympathetic – or ‘intersectionality-friendly’[104] – judicial interpretation is therefore important both practically and symbolically for those from impacted groups.[105]

Solanke argues that the concept of stigma can help to embed intersectionality in the legal framework.[106] For Solanke, structural stigma is a contextually embedded approach to identifying what sets individuals apart from others. It can be informed by protected grounds, but is not limited to them

by thinking about discrimination as stigma, we disrupt existing categories – we are no longer thinking about identity per se but about arbitrary social meaning attached to certain attributes, statuses and conditions in a way that strips away the right to equal regard. It may be that there is just one attribute, status or condition that stigmatises or it may be that there are many which intersect. Structural stigma can thus provide a framework for a methodological approach to intersectional discrimination.[107]

For Solanke, an intersectional approach would sit alongside ground-based discrimination law; while disruptive to existing legal frameworks, intersectionality is not necessarily destructive of established legal rules and mechanisms.[108] An intersectional approach could be adopted where law would otherwise deny a remedy and/or where discrimination is particularly blameworthy or heinous.[109] Judges interpreting these sorts of provisions may need to take social science and empirical evidence into account as part of a social framework analysis.[110] For them to do so, however, there needs to be robust empirical evidence available about how intersectional discrimination may affect individuals in practice.[111] This article offers one piece of this empirical puzzle in the Australian context.

Beyond legal reform, there is also a need for more targeted support of those experiencing intersectional discrimination, to assist them to bring a claim.[112] Further, there is a practical need for better data collection around multiple and intersectional experiences of discrimination. The 2014 GSS data are obviously limited: the survey is confined to self-reports of discrimination, focuses on discrete ‘grounds’ of discrimination (rather than how grounds intersect), and does not consider how groups experiencing discrimination engage (or fail to engage) with equality agencies and other enforcement bodies. More fundamentally, though, the 2014 GSS data are, at present, only available for one point in time. Although the 2014 GSS survey is conducted as a repeated cross-sectional survey, the discrimination module has only been included once. The ABS should include the discrimination module in future cross-sections to gauge levels of discrimination across time and how discrimination is experienced by different demographic groups in the population. This provides an important litmus test of the efficacy of legal and policy interventions, and societal change over time. The ABS is in a unique position to inform the Australian community on the prevalence of discrimination, through the 2014 GSS as well as other microdata collections. The ABS data is of very high quality, sampling a wide segment of the Australian population and (with its data collection enforced by the Census and Statistics Act 1905 (Cth)) achieves very high response rates. The ABS is therefore in a unique position to collect quality data of this nature, and there is a clear need for such questions to be included in future iterations of the 2014 GSS survey.

Apart from ABS collections, there is also a need for more focused attention and data collection on intersectionality by equality agencies in particular, to illuminate whether particular groups are more or less likely to assert their legal rights to be protected from discrimination. Without this data, it is difficult to assess the effectiveness or gaps in the legal framework, particularly given the evident lacuna in publicly accessible case law.

Finally, there is a need for more detailed consideration of how intersectional discrimination is experienced at a personal level, and its potential impacts and consequences for those from impacted groups. For example, there is now considerable international evidence that race discrimination has a detrimental impact on a range of mental and physical health outcomes,[113] behaviours and life outcomes.[114] Research has demonstrated that this deleterious impact occurs through several key pathways:

(1) cognitive, emotional and physical strain, stress or damage impacting upon mental, physical, spiritual or social wellbeing;

(2) reduced engagement in adaptive behaviours (such as physical activity);

(3) maladaptive behaviours (including alcohol and drug use);

(4) compromised access to key health-promoting settings (like employment and education);

(5) attenuated benefit from everyday routine activities (like sleep); and

(6) heightened contact with health-damaging exposures (including toxic substances).[115]

The consequences of intersectional discrimination must be investigated via the type of rigorous, empirical analyses that have been demonstrated within the field of race discrimination research. At present, our empirical understanding of intersectional discrimination and its practical experience is severely limited: future research needs to engage with this empirical lacuna, particularly in the Australian context.

VI CONCLUSION

As the Australian population becomes more diverse, and we become more attuned to the culturally embedded nature of discrimination, it is essential that the legal framework accommodates and reflects intersectional experiences of discrimination. Intersectionality poses a fundamental challenge to the existing framing of Australian discrimination law, with its focus on discrete ‘grounds’ and, at the federal level, division into discrete discrimination statutes for each ground. The empirical data presented in this article provide a clear illustration of the need to amend the legal framework to better reflect our lived experiences of discrimination, which are overlapping, multiple and complex. The time has come to adopt a more contextually-driven framework for addressing discrimination, which is consistent with the lived reality of impacted groups in society.


[* ] Associate Professor and Discovery Early Career Research Fellow, Melbourne Law School, The University of Melbourne. Alysia’s contributions to this research were funded by the Australian Government through the Australian Research Council’s Discovery Projects funding scheme (Project DE170100228). The views expressed herein are those of the author and are not necessarily those of the Australian Government or Australian Research Council.

[**] Associate Professor, Demography and Ageing Unit, Melbourne School of Population and Global Health, The University of Melbourne.

[1] Beth Goldblatt and Linda Steele, ‘Bloody Unfair: Inequality Related to Menstruation’ (2019) 41(3)

Sydney Law Review 293, 295–6.

[2] See, eg, Belinda Smith, ‘A Regulatory Analysis of the Sex Discrimination Act 1984 (Cth): Can It Effect Equality or Only Redress Harm?’ in Christopher Arup et al (eds), Labour Law and Labour Market Regulation: Essays on the Construction, Constitution and Regulation of Labour Markets (Federation Press, 2006) 105; Belinda Smith, ‘From Wardley to Purvis: How Far Has Australian Anti-discrimination Law Come in 30 Years?’ (2008) 21(1) Australian Journal of Labour Law 3; Belinda Smith, ‘It’s About Time: For a New Regulatory Approach to Equality’ [2008] FedLawRw 5; (2008) 36(2) Federal Law Review 117; Dominique Allen, ‘Remedying Discrimination: The Limits of the Law and the Need for a Systemic Approach’ [2010] UTasLawRw 5; (2010) 29(2) University of Tasmania Law Review 83; Dominique Allen, ‘Reducing the Burden of Proving Discrimination in Australia’ [2009] SydLawRw 24; (2009) 31(4) Sydney Law Review 579.

[3] See, eg, Dominique Allen, ‘Wielding the Big Stick: Lessons for Enforcing Anti-discrimination Law from the Fair Work Ombudsman’ (2015) 21(1) Australian Journal of Human Rights 119; Dominique Allen, ‘Strategic Enforcement of Anti-discrimination Law: A New Role for Australia’s Equality Commissions’ [2010] MonashULawRw 28; (2010) 36(3) Monash University Law Review 103; Dominique Allen, ‘Barking and Biting: The Equal Opportunity Commission as an Enforcement Agency’ (2016) 44(2) Federal Law Review 311. See also Rosalind Croucher, ‘Free and Equal: An Australian Conversation on Human Rights (2019)’, Australian Human Rights Commission, (Web Page, 19 February 2019) <https://www.humanrights.gov.au/our-work/rights-and-freedoms/publications/free-and-equal-australian-conversation-human-rights-2019>; Australian Human Rights Commission, Respect@Work: National Inquiry into Sexual Harassment in Australian Workplaces (Report, 2020) 441–616.

[4] See also Shreya Atrey, Intersectional Discrimination (Oxford University Press, 1st ed, 2019) 27.

[5] Ibid.

[6] Cf Atrey, who seeks to bridge intersectionality theory and discrimination law doctrine: ibid 25.

[7] This is consistent with the approach in the UK and European Union: Sandra Fredman, Discrimination Law (Oxford University Press, 2nd ed, 2011) 112.

[8] Equal Opportunity Act 2010 (Vic) s 6.

[9] Fredman (n 7) 110.

[10] See, eg, Sentencing Amendment (Historical Homosexual Convictions Expungement) Act 2014 (Vic).

[11] Iris Marion Young, Justice and the Politics of Difference (Princeton University Press, 1990) 171.

[12] Ibid.

[13] Ibid.

[14] Ibid.

[15] Ibid 172.

[16] Ibid 170.

[17] Fredman (n 7) 139.

[18] Iyiola Solanke, ‘Infusing the Silos in the Equality Act 2010 with Synergy’ (2011) 40(4) Industrial Law Journal 336, 336.

[19] Kimberle Crenshaw, ‘Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics’ [1989] (1) University of Chicago Legal Forum 139, 140.

[20] Ibid.

[21] Beth Gaze and Belinda Smith, Equality and Discrimination Law in Australia: An Introduction (Cambridge University Press, 2017) 84. See also Sarah Hannett, ‘Equality at the Intersections: The Legislative and Judicial Failure to Tackle Multiple Discrimination’ (2003) 23(1) Oxford Journal of Legal Studies 65, 68.

[22] Mary Eaton, ‘Patently Confused: Complex Inequality and Canada v. Mossop(1994) 1(2) Review of Constitutional Studies 203, 229.

[23] Gaze and Smith (n 21) 84. See also Hannett (n 21) 68. Atrey describes these ideas as lying on a continuum, with single-axis discrimination at one end, and intersectionality at the other: Atrey, Intersectional Discrimination (n 4) 78.

[24] Crenshaw (n 19) 140.

[25] Ibid 166.

[26] Ibid 167.

[27] Atrey, Intersectional Discrimination (n 4) 2.

[28] Crenshaw (n 19) 140.

[29] Atrey, Intersectional Discrimination (n 4) 2–3.

[30] Emily Grabham, ‘Intersectionality: Traumatic Impressions’ in Emily Grabham et al (eds), Intersectionality and Beyond: Law, Power and the Politics of Location (Routledge-Cavendish, 2009) 183, 184.

[31] Ibid; Davina Cooper, Challenging Diversity: Rethinking Equality and the Value of Difference (Cambridge University Press, 2004) 48.

[32] Grabham (n 30) 185.

[33] Joanne Conaghan, ‘Intersectionality and the Feminist Project in Law’ in Emily Grabham et al (eds), Intersectionality and Beyond: Law, Power and the Politics of Location (Routledge-Cavendish, 2009) 21, 29. But see Beth Goldblatt, ‘Intersectionality in International Anti-discrimination Law: Addressing Poverty in Its Complexity’ (2015) 21(1) Australian Journal of Human Rights 47.

[34] Conaghan (n 33) 31.

[35] Atrey, Intersectional Discrimination (n 4) 57.

[36] Ibid 62.

[37] Ibid 59.

[38] Ibid 58–9.

[39] The fifth – Australian Human Rights Commission Act 1986 (Cth) – includes a different enforcement structure for some additional grounds, including: race, colour, sex, religion, political opinion, national extraction, social origin, age, medical record, criminal record, marital or relationship status, impairment, mental, intellectual or psychiatric disability, physical disability, nationality, sexual orientation, and trade union activity.

[40] Cf the previous wording of the Age Discrimination Act 2004 (Cth) s 16, as at 10 December 2008. Until the Act was amended in 2009, this provision was even less generous, providing that age needed to be the ‘dominant’ reason for the Act: see also Australian Human Rights Commission, ‘Federal Discrimination Law’ (Handbook, 2016) 14–15 [2.2.4].

[41] Andrew Thackrah, ‘From Neutral to Drive: Australian Anti-discrimination Law and Identity’ [2008] AltLawJl 7; (2008) 33(1) Alternative Law Journal 31, 32. See also Hannett (n 21) 66; Jennifer Nielsen, ‘How Mainstream Law Makes Aboriginal Women “Disappear”’ [2004] IndigLawB 28; (2004) 6(1) Indigenous Law Bulletin 23, 23.

[42] On comparators and intersectional discrimination, see Hannett (n 21) 81–5. On comparators generally, see Colin Campbell and Dale Smith, ‘Direct Discrimination without a Comparator? Moving to a Test of Unfavourable Treatment’ (2015) 43(1) Federal Law Review 91. This could be avoided by moving to a test of disadvantage, as is the case in Victoria and the ACT.

[43] Shreya Atrey, ‘Comparison in Intersectional Discrimination’ (2018) 38(3) Legal Studies 379, 380–1.

[44] Ibid 382.

[45] Ibid. This is consistent with Blackham’s empirical study of age discrimination claims in Australia, which found that the comparator requirement was a common point of failure for all claims, not just intersectional claims: Alysia Blackham, ‘Why Do Employment Age Discrimination Cases Fail? An Analysis of Australian Case Law’ [2020] SydLawRw 1; (2020) 42(1) Sydney Law Review 1, 17–20 (‘Why Do Employment Age Discrimination Cases Fail?’).

[46] Australian Human Rights Commission, ‘Priorities for Federal Discrimination Law Reform’ (Discussion Paper, October 2019) 6.

[47] Australian Human Rights Commission, Willing to Work: National Inquiry into Employment Discrimination against Older Australians and Australians with Disability (Report, 2016) 338.

[48] However, while contained within one Act, the statutes in NSW, SA and WA contain separate sections with the tests for each protected ground: see Anti-Discrimination Act 1977 (NSW); Equal Opportunity Act 1984 (SA); Equal Opportunity Act 1984 (WA). Other statutes are more unified.

[49] See similarly Discrimination Act 1991 (ACT) s 4A(2); Anti-Discrimination Act 1977 (NSW) ss 4A, 7; Anti-Discrimination Act 1992 (NT) s 20; Anti-Discrimination Act 1991 (Qld) s 10(4); Equal Opportunity Act 1984 (SA) s 6(2); Anti-Discrimination Act 1998 (Tas) s 14; Equal Opportunity Act 1984 (WA) ss 5, 8.

[50] Alysia Blackham and Dominique Allen, ‘Resolving Discrimination Claims outside the Courts: Alternative Dispute Resolution in Australia and the United Kingdom’ (2019) 31(3) Australian Journal of Labour Law 253.

[51] Ezekiel-Hart v Reis [2017] ACAT 76 (on appeal from Ezekiel-Hart v Reis [2017] ACAT 3); Australian Capital Territory v Wang [2019] ACAT 65 (on appeal from Wang v Australian Capital Territory [2015] ACAT 5 and Wang v Australian Capital Territory [2016] ACAT 71).

[52] Cheluvappa v University of Canberra [2018] ACAT 108; Mewett v University of Canberra [2018] ACAT 61 (many aspects of this claim also related to conduct in 2014, before the amendments to Discrimination Act 1991 (ACT) s 8).

[53] [2018] ACAT 132.

[54] Ibid [97], [101] (Senior Member Robinson).

[55] Ibid [104] (Senior Member Robinson).

[56] Iyiola Solanke, ‘A Method for Intersectional Discrimination in EU Labour Law’ in Alan Bogg, Cathryn Costello and ACL Davies (eds), Research Handbook on EU Labour Law (Edward Elgar Publishing, 2016) 446. See, eg, Shreya Atrey, ‘Illuminating the CJEU’s Blind Spot of Intersectional Discrimination in Parris v Trinity College Dublin(2018) 47(2) Industrial Law Journal 278.

[57] Solanke, ‘A Method for Intersectional Discrimination in EU Labour Law’ (n 56) 455.

[58] Ibid 454.

[59] Fredman (n 7) 143.

[60] Solanke, ‘Infusing the Silos in the Equality Act 2010 with Synergy’ (n 18) 348.

[61] BA Hepple, Equality: The Legal Framework (Hart Publishing, 2nd ed, 2014) 78.

[62] See the discussion of Bahl v The Law Society [2004] EWCA Civ 1070 in Solanke, ‘Infusing the Silos in the Equality Act 2010 with Synergy’ (n 18) 340–1.

[63] Minna J Kotkin, ‘Diversity and Discrimination: A Look at Complex Bias’ (2009) 50(5) William and Mary Law Review 1439; Rachel Kahn Best et al, ‘Multiple Disadvantages: An Empirical Test of Intersectionality Theory in EEO Litigation’ (2011) 45(4) Law and Society Review 991.

[64] This trend is repeated internationally: see Atrey, Intersectional Discrimination (n 4) 29.

[65] See, eg, Dao v Australian Postal Commission [1987] HCA 13; (1987) 162 CLR 317; Fares v Box Hill College of TAFE (1992) EOC 92-391; Wiggins v Department of Defence [2006] FMCA 800; (2006) 200 FLR 438; Wu v Cohen [2000] HREOCA 36. These cases are discussed in Julia Mansour, ‘Consolidation of Australian Anti-discrimination Laws: An Intersectional Perspective’ (2012) 21(2) Griffith Law Review 533, 542.

[66] This may reflect the importance of alternative dispute resolution in redirecting claims: see Blackham and Allen (n 50).

[67] Mansour (n 65) 542. This is understandable, as US studies have found that multiple claims are less likely to succeed: Kotkin (n 63).

[68] Wang v Australian Capital Territory [2015] ACAT 5, [10] (Senior Member Anforth); Herbert v Star Aviation Pty Ltd [2019] FCAFC 136, [6] (Katzmann, Colvin and Thawley JJ).

[69] Mansour (n 65) 542.

[70] Blackham, ‘Why Do Employment Age Discrimination Cases Fail?’ (n 45) 9.

[71] Michael Schuster and Christopher S Miller, ‘An Empirical Assessment of the Age Discrimination in Employment Act(1984) 38(1) Industrial and Labor Relations Review 64.

[72] Mansour (n 65) 542.

[73] Dominique Allen and Alysia Blackham, ‘Under Wraps: Secrecy, Confidentiality and the Enforcement of Equality Law in Australia and the United Kingdom’ [2020] MelbULawRw 1; (2019) 43(2) Melbourne University Law Review 384.

[74] On these issues of transparency and reporting, see also ibid.

[75] [2016] QCAT 530 (‘Till’).

[76] Ibid [20] (Member Clifford).

[77] Ibid [23] (Member Clifford).

[78] [2016] FCCA 905.

[79] Ibid [16] (Judge Manousaridis).

[80] Ibid [47], [49] (Judge Manousaridis).

[81] Ibid [49] (Judge Manousaridis).

[82] Ibid [51] (Judge Manousaridis).

[83] [2007] FCA 1978; (2007) 168 IR 309.

[84] Ibid 310 [1]–[2] (Buchanan J).

[85] Ibid 320 [44] (Buchanan J).

[86] Ibid 321 [50] (Buchanan J).

[87] Sex Discrimination Act 1984 (Cth) s 7A.

[88] Gaze and Smith (n 21) 90.

[89] Australian Bureau of Statistics, General Social Survey: Summary Results, Australia, 2014 (Catalogue No 4159.0, 29 June 2015).

[90] Data for the 2014 GSS were collected by the ABS under the provisions of the Census and Statistics Act 1905 (Cth). Confidentialised data and access to the Remote Access Data Laboratory (‘RADL’) were made available to the authors for this study through the ABS and Universities Australia Agreement. Ethics approval for this project was granted by the University of Melbourne – Ethics ID: 1953686.1.

[91] Carer, as defined by the ABS, includes persons caring for those with a disability or long-term health condition. It does not include parents caring for children who may be temporarily ill.

[92] This may actually lead to under-reporting: see Australian Human Rights Commission, Everyone’s Business: Fourth National Survey on Sexual Harassment in Australian Workplaces (Report, 12 September 2018) 23–4 (‘Everyone’s Business’).

[93] In Table 2, we include as race or ethnicity discrimination instances where respondents report any of ‘your skin colour’, ‘your nationality, race or ethnic group’, or ‘the language you speak’. This is an imperfect measure of race or ethnicity discrimination, but seeks to capture the range or responses and prompts used in the 2014 GSS.

[94] See Everyone’s Business (n 92) 21–3.

[95] Atrey, Intersectional Discrimination (n 4) 212.

[96] Mansour (n 65) 545.

[97] Ibid 551.

[98] Victoria Legal Aid, Submission to Attorney-General and Minister for Finance and Deregulation, Parliament of Australia, Consolidation of Commonwealth Anti-Discrimination Laws (1 February 2012) 24.

[99] Mansour (n 65) 551.

[100] Blackham, ‘Why Do Employment Age Discrimination Cases Fail?’ (n 45) 20.

[101] Equal Opportunity Act 2010 (Vic) s 8. See also Discrimination Act 1991 (ACT) s 8(2). It is unclear whether a comparator is still required in Victoria: see Aitken v Victoria [2013] VSCA 28; (2013) 46 VR 676, 687–8 (Neave and Priest JJA). A comparator is not required in the ACT: see Re Prezzi and Discrimination Commissioner and Quest Group (1996) 39 ALD 729, 736 [22] (President Curtis, Members Attwood and Corkery). See also Campbell and Smith (n 42) 93, 101–2.

[102] Atrey, Intersectional Discrimination (n 4) 25.

[103] Ibid 140.

[104] Ibid.

[105] Ibid 213–14.

[106] Solanke, ‘A Method for Intersectional Discrimination in EU Labour Law’ (n 56) 465–71. On stigma generally, see Iyiola Solanke, Discrimination as Stigma: A Theory of Anti-discrimination Law (Hart Publishing, 2017).

[107] Solanke, ‘A Method for Intersectional Discrimination in EU Labour Law’ (n 56) 466; Iyiola Solanke, ‘Putting Race and Gender Together: A New Approach to Intersectionality’ (2009) 72(5) Modern Law Review 723, 739 (‘Putting Race and Gender Together’).

[108] Solanke, ‘A Method for Intersectional Discrimination in EU Labour Law’ (n 56) 469.

[109] Ibid 470–1.

[110] Ibid 471. See also Solanke, ‘Putting Race and Gender Together’ (n 107).

[111] On the issues arising with courts using empirical evidence, see Alysia Blackham, ‘Legitimacy and Empirical Evidence in the UK Courts’ (2016) 25(3) Griffith Law Review 414.

[112] Mansour (n 65) 551; Nielsen (n 41).

[113] Yin Paradies et al, ‘Racism as a Determinant of Health: A Systematic Review and Meta-analysis’ (2015) 10(9) PLoS ONE e0138511: 1–48.

[114] YC Paradies and DR Williams, ‘Racism and Health’ in Stella R Quah and William Cockerham (eds), International Encyclopedia of Public Health (Elsevier, 2nd ed, 2017) 474.

[115] Yin Paradies et al, ‘Racism as a Determinant of Health: A Protocol for Conducting a Systematic Review and Meta-analysis’ (2013) 2 Systematic Reviews 85: 1–7.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2020/28.html