Blackham, Alysia --- "Why Do Employment Age Discrimination Cases Fail? An Analysis of Australian Case Law" [2020] SydLawRw 1; (2020) 42(1) Sydney Law Review 1

Why Do Employment Age Discrimination Cases Fail? An Analysis of Australian Case Law

Alysia Blackham^[1]

Abstract

Employment age discrimination cases are notoriously unsuccessful in Australia. While it is arguable that most strong cases are settled through conciliation, serious questions remain: are those cases that proceed to the courts particularly weak? Or are there procedural or substantive legal hurdles that operate as barriers to the success of claims? As the first rigorous study of age discrimination case law across all Australian jurisdictions, this article evaluates these two questions, drawing on employment age discrimination case law at federal, state and territory level up to 2017. This article interrogates and maps, both qualitatively and quantitatively, potential legal barriers to age discrimination claims. It offers original and innovative insights into the cases that proceed to court, and why they fail. It argues that while some cases may be weak, there appear to be procedural and substantive hurdles that limit the success of age discrimination cases. This article offers compelling evidence of the need for legal reform, or for a more sympathetic interpretation of existing statutes by the courts, if individual enforcement is to be used as a means of addressing age discrimination.

I Introduction

Age discrimination remains prevalent in Australian society. In 2014, the Australian Human Rights Commission (‘AHRC’) interviewed 2109 Australians aged 50 years and over.^[1] This National Prevalence Survey found that 27% of respondents reported experiencing age discrimination in employment in the previous two years.^[2] Further, 32% of respondents were aware of other people experiencing discrimination in the workplace because of their age in the last two years.^[3] Age discrimination is a fundamental barrier to older workers participating meaningfully in the workforce,^[4] which is of increasing importance in the face of demographic ageing.

Despite the prevalence of age discrimination in Australia, employment age discrimination cases are notoriously unsuccessful. There has never been a successful case brought under the Age Discrimination Act 2004 (Cth) (‘Age Discrimination Act’). There are also noticeably few cases at state and territory level, and those that are brought are generally unsuccessful. It is likely that most strong cases are settled through conciliation, reducing the number of successful cases that proceed to hearing. At the same time, there is limited publicly available information about the relative strength of cases that are resolved through conciliation or settlement, as compared to those that proceed to court. Therefore, it is worth analysing why those cases that do proceed to the courts are unsuccessful: are the cases particularly weak? Or are there procedural or substantive legal hurdles that are operating as barriers to the success of claims?

This article evaluates these two questions, drawing on employment age discrimination case law at federal, state and territory level, and comparing cases under discrimination statutes to those under industrial relations legislation. It argues that while some cases may be weak, there appear to be procedural and substantive hurdles that limit the success of age discrimination cases. Thus, there is a need for legal reform or a more sympathetic interpretation of existing statutes by the courts if individual enforcement is to be used as a means of addressing age discrimination.

As has been mapped elsewhere, there are substantial differences between age discrimination laws in the Australian states and territories and federally.^[5] Thus, in drawing on age discrimination case law from all Australian jurisdictions, this article spans diverse legal frameworks and substantial doctrinal differences. This can make analysis and synthesis of case law challenging. For this reason, most scholars (sensibly) do not attempt to consider state, territory and federal law simultaneously.^[6] The risk of this approach, however, is that federal nations like Australia do not obtain an overall picture of how case law is developing — or, indeed, failing to develop — collectively across the various jurisdictions. It is difficult to accurately map trends or identify gaps in legal enforcement by studying one jurisdiction alone. Thus, there is a need for scholarship that attempts to bring together case law emerging from different jurisdictions.

It is beyond the feasible scope of one article to examine all relevant jurisdictional differences in detail,^[7] and the detailed analysis of specific doctrinal provisions and differences is not the aim here. Rather, this article interrogates and maps, both qualitatively and quantitatively, potential legal barriers to age discrimination claims, noting jurisdictional differences as they emerge. This article, then, departs from traditional doctrinal scholarship in both its aims and methods, as set out in Part II. With this method, it is difficult to generalise across and between different jurisdictions. However, it does offer interesting description and rich insights into the case law as it stands.

In Part II, I outline the method of the article, and its use of quantitative and qualitative content analysis. In Part III, I consider existing scholarship and theories as to why employment age discrimination claims in Australia fail. In Part IV, I outline the results of this study, identifying trends in age discrimination cases from quantitative and qualitative content analysis, and examining points of failure. In Part V, drawing on these findings, I consider the implications of this study and, in Part VI, offer suggestions for reform.

II Method

In this study, qualitative and quantitative content analysis was employed to analyse age discrimination cases handed down in Australian jurisdictions.^[8] Content analysis, while similar to doctrinal research,^[9] differs from traditional legal methods in two key ways. First, it can be used to analyse a broader range of texts than doctrinal analysis, which typically focuses on legal texts like cases and legislation. Second, and more relevantly here, content analysis analyses themes in texts, whereas doctrinal analysis focuses on harmonising, rationalising or systematising legal texts.^[10]

According to Hall and Wright, content analysis is preferable for analysing judicial decisions where there are multiple decisions of similar weight (ie rather than a few decisions of significant precedential weight), and what matters is the pattern across cases rather than a deep understanding of one case.^[11] Thus, what matters is collective — not individual — insights from cases: ‘[c]ontent analysts ... assemble a chorus, listening to the sound the cases make together’.^[12] This is particularly important in discrimination law, where most decisions are of low precedential weight (being heard and decided in lower courts, and with few decisions establishing weighty legal precedent), and what is of interest is trends in how cases are decided.

Content analysis involves identifying, coding and categorising documents.^[13] In this study, research was undertaken in three parts. First, the ‘universe of documents’ to be analysed was identified. This included judgments relating to age discrimination, including those under discrimination legislation and industrial relations legislation (including unfair dismissal law where age was alleged to be a factor in the dismissal) handed down in all Australian jurisdictions, from the commencement of age discrimination legislation in Australia until 31 December 2017.^[14]

Second, the sample of texts for analysis was determined. Documents were identified via keyword searches of case databases and annotated Acts.^[15] This search identified 139 cases. To refine the sample, cases were excluded where: age discrimination was not significant to the decision (as where age discrimination was only mentioned incidentally in the judgment);^[16] the case did not relate to work or employment; or where the case did not include an individual claim (as where the claim related to discriminatory terms in enterprise bargaining). Cases that were reported multiple times (for example, in preliminary hearing and/or on appeal) were only considered once, aggregating the issues for consideration across the various hearings and judgments. This refined the cases to a sample of 108 decisions, made up of 64 substantive and 44 procedural judgments.^[17]

Third, the texts were coded using themes derived from the literature and the documents themselves,^[18] and analysed using qualitative and quantitative methods.^[19] Quantitative analysis can overlook nuances in judicial reasoning, and obscure important trends in judicial analysis. Thus, it is important to pair quantitative techniques with qualitative analysis, including traditional doctrinal analysis. As Hall and Wright argue, qualitative and quantitative analysis reveal different, complementary insights into case law: the two methods combined are more powerful than each taken alone.^[20]

In adopting this method, this study joins a strong tradition of quantitative analysis of case law in the United State (‘US’).^[21] In relation to discrimination cases particularly, scholars such as Schuster and Miller,^[22] and Feild and Holley^[23] have used quantitative analysis to retrospectively study judicial reasoning. In their quantitative study of 153 US federal age discrimination cases using content analysis, Schuster and Miller found that a majority (81%) of age discrimination claimants were men, and most were in professional or managerial positions (57%).^[24] The authors therefore concluded that age discrimination law had ‘become the primary device used by white male professionals and managers to attack arbitrary personnel decisions’, particularly in the areas of dismissal and involuntary retirement.^[25] As part of her PhD research, Irving conducted qualitative and quantitative content analysis of Employment Tribunal decisions in age discrimination cases handed down in England and Wales between 1 October 2006 and 1 April 2010.^[26] This dual approach allowed Irving to map the nature of complaints and trace trends in judicial reasoning.

III Why do Cases Fail? Existing Hypotheses

Writing in 2013, MacDermott noted that few age discrimination complaints under the Age Discrimination Act proceeded to court, and none that did had ever succeeded.^[27] The situation is no different writing in 2020. However, MacDermott did not analyse cases at state and territory level, and was writing before the successful use of the Fair Work Act 2009 (Cth) (‘FWA’) adverse action provisions to address age discrimination in Fair Work Ombudsman v Theravanish Investments Pty Ltd.^[28] Thus, it is timely to review and extend MacDermott’s findings, to consider whether additional jurisdictions and different research methods offer new insights into the field.

Drawing on existing scholarship (which, admittedly, has largely focused on discrimination law in the federal jurisdiction), we can identify a range of hypotheses for why age discrimination complaints do not proceed to court, and why those that do might fail. Age discrimination cases can ‘fail’ or be unsuccessful at multiple points: pre-claim, at conciliation, pre-hearing, at hearing or judgment. While the focus of this article is specifically on why reported cases fail in court, court proceedings are part of a larger process of enforcement that may affect a claim’s success. Thus, the hypotheses presented below address a range of factors, at all stages of a claim.

First, the strong reliance on alternative dispute resolution (‘ADR’) in Australian discrimination law might redirect a substantial number of complaints away from formal legal processes.^[29] In all jurisdictions except Victoria, it is compulsory to first lodge a complaint with the relevant equality commission or agency before proceeding to court, and to participate in compulsory conciliation.^[30] A substantial number of age discrimination complaints are resolved via conciliation. The AHRC’s Willing to Work report, for example, provides broad information about the outcome of age discrimination claims filed with statutory agencies. Of complaints by older workers relating to age discrimination in employment finalised by the AHRC between 2012 and 2015, nearly half were conciliated (45.4%, or 103 of 227 complaints).^[31] The remainder were terminated,^[32] not pursued or withdrawn. This is a lower conciliation statistic than for age discrimination claims generally at the AHRC: 51% of all age discrimination complaints were conciliated in 2015–16.^[33] Thus, while many claims are being conciliated, this does not account for all age discrimination complaints: a substantial number are not being resolved through ADR. In 2014–15, for example, 26.6% of age discrimination complaints by older workers finalised by the AHRC were terminated due to no reasonable prospect of conciliation (representing 17 complaints in total):^[34] these would appear particularly likely to proceed to litigation.

While many age discrimination complaints are not successfully conciliated each year, few progress to a court or tribunal hearing. The Federal Circuit Court of Australia, which receives applications in relation to federal discrimination statutes in its ‘Human Rights’ jurisdiction,^[35] received only 75 applications in that jurisdiction in 2015–16, compared with 2,013 complaints about discrimination and breaches of human rights received by the AHRC across all areas in that same year. As a ballpark figure, then, around 3.7% of AHRC complaints progress to the court application stage; substantially fewer would progress to hearing or judgment. This is broadly consistent with the AHRC’s estimate of 3% of complaints proceeding to court.^[36]

Second, then, it is necessary to consider substantive and procedural legal barriers that might prevent legitimate age discrimination complaints from proceeding to court, or undermine their success at hearing. MacDermott found in her study of federal case law that most cases related to questions of practice or procedure, rather than substantive discussions of whether age discrimination had occurred.^[37] Around 25% of the cases at that stage were applications for summary dismissal,^[38] leading MacDermott to hypothesise that cases were either particularly weak, or that it was particularly difficult to prove age discrimination.^[39] Time limits may cause problems for some complainants, particularly for unfair dismissal applications under the FWA, where claims must be filed within 21 days.^[40]

In relation to the substantive legal limitations of age discrimination law in Australia, I have mapped the extensive exceptions to the prohibition of age discrimination in the states, territories and federally, which might exclude many claims from legal protection.^[41] I have also argued that Australian courts adopt a limited, narrow and non-purposive approach to age discrimination cases, which might deter claimants from pursuing legal avenues.^[42] Federally, cases might also struggle due to: a lack of evidence;^[43] difficulties in proving causation;^[44] the comparator requirement;^[45] an underdeveloped jurisprudence leading to legal uncertainty;^[46] the need for legal representation;^[47] no shifting of the burden of proof and difficulties proving that discrimination has occurred;^[48] the cost of litigation, including the risk of having to pay the other side’s costs;^[49] the emotional toll of pursuing a complaint;^[50] fear of victimisation;^[51] and intersectionality.^[52]

To address these issues, the AHRC’s Willing to Work report recommended a number of areas for review in relation to existing federal age discrimination laws. These included: approaches to standing, including by potentially allowing actions by representative organisations;^[53] considering the use of positive duties;^[54] removing the comparator test, and replacing it with a detriment test;^[55] moving to a system where parties bear their own costs, with the court retaining a discretion to award costs if it is just to do so;^[56] amending discrimination laws to encompass discrimination based on a combination of attributes;^[57] and reviewing the 21-day time limit under the FWA.^[58] Not all of these issues are relevant at state and territory level. For example, there is arguably no comparator test in Victoria or the Australian Capital Territory (‘ACT’).

Similar issues appear to emerge in England and Wales. Irving’s analysis of Employment Tribunal age discrimination cases discovered a number of points of failure that impeded age discrimination claims, including: jurisdictional limits, such as employee status, time limits, and territorial issues;^[59] difficulties shifting the burden of proof (particularly in relation to recruitment discrimination);^[60] issues establishing the presence of ‘discriminatory treatment’;^[61] exceptions;^[62] the presence of ‘justified’ discrimination;^[63] and cases that were underprepared or lacked evidence.^[64] Thus, these failure points are well-established in England and Wales, and at the federal level in Australia. It is still unclear, however, whether these failure points are affecting potentially meritorious complaints. This study therefore sought to inquire further into these issues.

IV Points of Failure in Age Discrimination Cases

A An Overview of the Sample

Of the 108 cases in the sample, 33 related to recruitment and 63 to dismissal (including 12 relating to retirement and 10 relating to redundancy).^[65] This differs dramatically from the results of the AHRC National Prevalence Survey, which found that the most common episodes of discrimination experienced by older workers were: limiting employment, promotion or training because of age (27%); jokes or derogatory comments based on age (21%); and perceptions that individuals had outdated skills, were too slow to learn new things or delivered an unsatisfactory job because of their age (21%).^[66] The least common instances of discrimination — being threatened with redundancy or dismissal, or being asked to retire; and poor treatment resulting in leaving the job, being made redundant or dismissed because of age — were each reported by only 3% of respondents.^[67] Thus, the cases in the sample are likely to be outliers, and the most extreme instances of age discrimination experienced. This makes sense, given the potential emotional, financial and reputational costs of pursuing an age discrimination claim: only the most serious cases are likely to be pursued.

Eighty-one cases featured a male claimant, and 28 featured a female claimant.^[68] Over 90% of claims (99 cases) related to claimants who alleged they were too old; only six related to claimants who were too young. Ethnicity was raised in 21 cases, and disability in 19 cases. The majority of claimants (68%, or 73 claimants) were men bringing a claim on the basis that they were too old.

Where specified or possible to discern, claimants were represented in 53 cases (49% of the sample); 50 cases had litigants in person (three cases were unclear; in two the claimant did not appear). The vast majority of representatives were solicitors or barristers; unions acted as representatives in only two cases. The majority of cases (69) were brought against employers in the private sector, rather than the public sector (39). Where identifiable, more claims (54) were brought by blue collar or unskilled workers than white collar or professional workers (45).^[69]

Where discussed, most claims alleged direct discrimination (58 cases), with only 17 cases discussing indirect discrimination. Cases were unevenly distributed across the jurisdictions: 34 cases were brought under the FWA, Workplace Relations Act 1996 (Cth) or Industrial Relations Act 1988 (Cth); 15 cases related to the Age Discrimination Act or Human Rights and Equal Opportunity Commission Act 1986 (Cth). Table 1 (below), lists the number of age discrimination cases in each jurisdiction and their prevalence per capita.

As Table 1 illustrates, cases are far more likely per capita in Tasmania than any other jurisdiction. This is distorted, however, by one repeat litigant, who brought cases against three separate employers.^[70] Even with these cases counted as one data point, Tasmania is still the most litigated jurisdiction, with one case per 104,940 residents. Victoria is the jurisdiction with the least age discrimination litigation per capita, despite the ability to apply directly to the Victorian Civil and Administrative Tribunal since 1 August 2011.^[71]

These jurisdictional trends may be explained partly by the demographics of each state and territory. For example, in 2017, Tasmania had the oldest median age in Australia at 42 years (compared with 37 years across the country).^[72] As illustrated by Table 2 (below), Tasmania also has the largest proportion of older residents in the country, with nearly 40% of the state’s population aged 50 or over. That said, Victoria does not have a particularly youthful population, but still has few age discrimination cases; and SA has a comparatively elderly population, but relatively few age discrimination cases. Thus, demographic trends do not fully explain the presence or absence of case law in each jurisdiction.

Table 1: Age discrimination cases for the Australian population by jurisdiction^[73]

	Population in 2017 (‘000)	Age discrimination cases (number)	Population per case (‘000)
Commonwealth (‘Cth’) / Australian population	24770.7	49	505.52
New South Wales (‘NSW’)	7915.1	30	263.84
Victoria (‘Vic’)	6385.8	4	1596.45
Queensland (‘Qld’)	4965	11	451.36
Western Australia (‘WA’)	2584.8	3	861.60
South Australia (‘SA’)	1728.1	2	864.05
Tasmania	524.7	7	74.96
Australian Capital Territory (‘ACT’)	415.9	1	415.90
Northern Territory (‘NT’)	246.7	1	246.70

B Successful Claims

Across all 108 (substantive and procedural) cases in the sample, claimants were successful in only 23 cases on the basis of age.^[74] Claimants were successful in 12 substantive decisions, as detailed in Table 3 (below). This represents a claimant success rate in substantive cases of 18.75%.^[75] The key common factor across the successful substantive decisions was compelling evidence of discrimination: age was expressly cited as a reason for dismissing^[76] or not recruiting^[77] the claimant in a number of cases, or was explicitly asked for in recruitment material.^[78] These claims were often supported by written evidence (such as letters citing age as a reason for dismissal); or testimony of others who witnessed or were later told about the discrimination.^[79] The cases were fairly evenly spaced across time: one was decided before 1999; five between 2000–9; and six from 2010–17. The rate of legal representation in these successful substantive cases (67%) was similar to that for unsuccessful substantive cases (61%).

The largest number of successful cases (four) was in Queensland. Only two cases were successful at the federal level: one under the FWA^[80] and one under the Workplace Relations Act 1996 (Cth).^[81]

Beyond these cases, there were also a number of successful unfair dismissal applications in the sample: in these cases, age might have been involved in the dismissal, but was not a basis for the decision.^[82] In some cases, complainants were indirectly successful, for example, where the Commonwealth unsuccessfully sought judicial review of a Human Rights and Equal Opportunity Commission (‘HREOC’) decision.^[83] Claimants were also successful in some procedural decisions, including in relation to: the grant of an extension of time;^[84] successfully arguing that they had filed their application in time;^[85] fending off an application for summary dismissal^[86] or to strike out their claim;^[87] seeking leave to proceed to Tribunal;^[88] or establishing that the Commissioner’s decision to dismiss the application was incorrect.^[89] Overall, though, the vast majority (78%) of cases in the sample had unsuccessful claimants: success was the exception, not the norm.

Table 2: Population of Australia (‘000), by age, states and territories at 30 June 2017^[90]

Table 3: Successful substantive cases in the sample

D Points of Failure

Some failure points were jurisdiction-specific. For example, 12 cases were brought under the FWA, a further 18 in the FWA and Workplace Relations Act 1996 (Cth) unfair dismissal jurisdiction, and another four under the Industrial Relations Act 1988 (Cth). This constituted a sizeable proportion of the sample (31% of all cases). In the vast majority of these cases, the claimant failed. A key procedural failure point under the FWA was time limits. The FWA requires that unfair dismissal or general protections applications related to termination be filed with the Fair Work Commission (‘FWC’) within 21 days after the dismissal took effect.^[104] The FWC may grant an extension of time if satisfied that there are ‘exceptional circumstances’ that warrant the extension. Only one application for an extension of time was granted;^[105] seven were refused;^[106] and one claim was held to be within time.^[107] Causation was also a recurring issue in the cases, as discussed below in Part IVE. While there were 10 cases in the sample relating to the Age Discrimination Act, only two were substantive; four cases were successful applications for summary dismissal. At the federal level, some claims had deficient pleadings: failing to allege that the employer acted for a prohibited reason;^[108] alleging behaviour that occurred before the commencement of the statute;^[109] pursuing a party who was not a respondent to the HREOC complaint;^[110] or failing to allege acts that amounted to discrimination, as opposed to bad practice.^[111]

Thirty cases were brought in NSW. Claimants were successful on the basis of age in three substantive decisions^[112] and five procedural decisions.^[113] In NSW, nine cases were applications for leave for the appeal to be the subject of proceedings before the NSW Civil and Administrative Tribunal or Administrative Decisions Tribunal. Under s 96 of the Anti-Discrimination Act 1977 (NSW), leave of the Tribunal is required before proceedings can commence if the matter is referred to the Tribunal at the requirement of the complainant. Seven cases failed at this hurdle.

Beyond these jurisdiction-specific issues, some factors affected cases across jurisdictions. For example, a number of claims fell beyond the scope of the statute: relating to a job ‘interview’ that was actually a career counselling session, and therefore did not count as ‘recruitment’;^[114] or where no job was on offer;^[115] claiming against another employee, not an employer;^[116] making a claim after retirement, when the claimant was no longer a ‘worker’;^[117] or making a claim relating to a management agreement, which was not ‘work’.^[118]

A thread running through many unsuccessful cases was a failure to produce sufficient evidence to support the claim.^[119] Indeed, a lack of evidence was raised in 32 of the 108 cases. A lack of evidence could relate to the alleged detriment or adverse treatment; causation; or the relevant groups for an indirect discrimination claim. These substantive failure points are developed in more detail below.

E Differential Treatment

In relation to differential treatment, some cases failed because there was no evidence of detriment^[120] or unfavourable treatment^[121] or different treatment.^[122] In cases where the alleged adverse treatment was dismissal, courts and tribunals sometimes found there to be no ‘dismissal’ — the claimant resigned,^[123] or casual employment came to an end,^[124] or there was a genuine redundancy.^[125]

The need for a comparator was flagged in 16 cases. The High Court’s decision in Purvis v New South Wales (Department of Education and Training),^[126] which related to the Disability Discrimination Act 1992 (Cth), loomed large in these cases. In Purvis, the Court was asked to identify the appropriate comparator for determining whether there had been less favourable treatment of a student with disabilities who was suspended and excluded from school due to his behaviour, which included kicking, punching and verbal abuse.^[127] The Disability Discrimination Act 1992 (Cth) defines direct disability discrimination as occurring ‘if, because of the disability, the discriminator treats, or proposes to treat, the aggrieved person less favourably than the discriminator would treat a person without the disability in circumstances that are not materially different’.^[128] The question for the Court was whether the student’s violent behaviour — which was attributable to his disability — should be also imputed to the comparator. The majority held that it should.^[129] For Gleeson CJ, ‘[t]he required comparison is with a pupil without the disability; not a pupil without the violence. The circumstances are relevantly the same, in terms of treatment, when that pupil engages in violent behaviour.’^[130] Purvis exemplifies a highly restrictive approach to the comparator requirement. This may well be an example of ‘hard cases making bad law’^[131] (or, perhaps, bad legislation):^[132] the absence of an unjustifiable hardship defence (which has now been extended to such cases) may have led the Court to distort the comparator requirement.^[133] Regardless, the highly restrictive approach in Purvis had a substantial influence on cases in this sample, meaning the comparator selected tended to be very narrow.^[134]

That said, narrow comparators were also evident in state case law, and in claims that preceded Purvis. In the NSW case of Duncan v Chief Executive, NSW Office of Environment and Heritage (No 2), for example, an actual comparator (which Mr Duncan could not identify) would be

any person not of [Mr Duncan’s] race or age who, like him, was afforded a priority interview in similar circumstances and was appointed to the position, or at any rate treated more favourably, despite being considered unable ‘to demonstrate the capacity to competently undertake the position within six months with the support of appropriate training and management’.^[135]

In Kennedy, a NSW claim by a call centre worker against the NSW Department of Industrial Relations as his employer, the comparator was identified as ‘a disgruntled employee who holds the strong and genuine view that they were not appointed to a position for reasons other than merit’.^[136] These are extraordinarily narrow comparators; it would be surprising (to say the least) if they actually existed. Unsurprisingly, then, in most cases, the courts noted that no actual comparator was available,^[137] meaning the courts had to rely on a hypothetical comparator.

In only two cases did the claimant successfully identify an actual comparator. In Virgin Blue Airlines,^[138] the Court rejected the argument that older and younger job applicants were not in circumstances that were the same or not materially different:^[139] all participated in the same recruitment process.^[140] Using this comparator, less favourable treatment could be established on the basis of statistical evidence, which showed that 90% of applicants were aged 35 or under, but 99.9% of those appointed were aged 36 or under.^[141] The claimants’ case was obviously assisted by a large pool of job applicants, which meant a comparator was available, and helped to demonstrate the discriminatory impact of the recruitment process.

Similarly, in Talbot,^[142] the claimant could identify other drivers with performance issues, and could illustrate that he received different, harsher treatment for driving infractions.^[143] There was no documentary evidence that Mr Talbot had received counselling or warnings, as had other drivers with performance issues.^[144] However, there was documentary evidence that explicitly cited the claimant’s age as a reason for not engaging him.^[145]

In other cases, when relying on a hypothetical comparator, as noted in Duncan, it is difficult to separate issues of proof relating to a hypothetical comparator from the question of causation; thus, ‘[i]t follows that the questions of less favourable treatment and causation should be approached ... as part of the same reasoning exercise.’^[146] This means that the comparator requirement was of limited usefulness in most decisions in the sample; the question of the comparator was conflated with issues of causation. The question, then, is whether the comparator requirement adds any value to the legal reasoning process, if a hypothetical comparator is relied on in many cases, and thereafter conflated with causation. A better approach may be to allow claimants to rely on an actual comparator, if one exists, but otherwise to remove the comparator requirement.

It is noteworthy that a comparator may not be required in relation to FWA adverse action claims. In EP2,^[147] the respondent’s argument that the claimant needed to show that there was less favourable treatment by reference to a real or hypothetical comparator was held to be ‘misconceived’:^[148] the claim was not based on discrimination, but that there was adverse action ‘because of’ protected attributes (here, race and age).^[149] Therefore, there was no need ‘to establish the complex matters required by the anti-discrimination laws’,^[150] including the comparator requirement. However, the claim still failed, as the claimant could not establish that he was dismissed: he resigned, so there was no adverse action.^[151] Thus, the claim was summarily dismissed.^[152]

There is arguably also no comparator requirement in Victoria or the ACT. Instead, for example, s 8 of the Equal Opportunity Act 2010 (Vic) defines direct discrimination as occurring ‘if a person treats, or proposes to treat, a person with an attribute unfavourably because of that attribute’.^[153] In Aitken v State of Victoria,^[154] the Victorian Court of Appeal held that it was ‘an unresolved question of law in Victoria’ as to whether the unfavourable treatment requirement still necessitated the use of a comparator.^[155] This may be compared with the decision in Re Prezzi and Discrimination Commissioner and Quest Group,^[156] where the ACT Administrative Appeals Tribunal held that the unfavourable treatment test in s 8(1)(a) of the Discrimination Act 1991 (ACT) did not require a comparator: ‘[a]ll that is required is an examination of the treatment accorded the aggrieved person or the conditions upon which the aggrieved person is or is proposed to be dealt with.’^[157] Campbell and Smith have argued that it is possible to interpret the ‘unfavourable treatment’ test in a way that overcomes the comparator requirement, but doing so means it becomes ‘over-inclusive in ways that Parliament would not have intended’, as the test potentially encompasses a wide and ill-defined range of interests.^[158] Thus, moving to an unfavourable treatment test is not necessarily a simple solution to the limitations of the comparator requirement.

F Causation

Relatedly then, causation was a recurring issue in the case law: causation was discussed in 24 of the 108 cases. While the tests for causation vary from jurisdiction to jurisdiction, many cases fell at this hurdle.

Under the FWA, for example, employers must not take adverse action ‘because of’ age.^[159] If an employer can show that age is not a ‘substantial and operative factor’ in the adverse treatment, the claim will fail.^[160] This affected a number of claims under the FWA.^[161] Drawing on Barclay,^[162] which related to industrial activity, an employer’s evidence as to the reason for their actions can be enough to show that age is not a reason for the treatment.^[163] For French CJ and Crennan J in Barclay,

[t]here is no warrant to be derived from the text of the relevant provisions of the Fair Work Act for treating the statutory expression ‘because’ in s 346, or the statutory presumption in s 361, as requiring only an objective enquiry into a defendant employer's reason, including any unconscious reason, for taking adverse action. The imposition of the statutory presumption in s 361, and the correlative onus on employers, naturally and ordinarily mean that direct evidence of a decision-maker as to state of mind, intent or purpose will bear upon the question of why adverse action was taken, although the central question remains ‘why was the adverse action taken?’.

This question is one of fact, which must be answered in the light of all the facts established in the proceeding. Generally, it will be extremely difficult to displace the statutory presumption in s 361 if no direct testimony is given by the decision-maker acting on behalf of the employer. Direct evidence of the reason why a decision-maker took adverse action, which may include positive evidence that the action was not taken for a prohibited reason, may be unreliable because of other contradictory evidence given by the decision-maker or because other objective facts are proven which contradict the decision-maker’s evidence. However, direct testimony from the decision-maker which is accepted as reliable is capable of discharging the burden upon an employer ...^[164]

While this implies that courts look to the subjective reason for the treatment, Gummow and Hayne JJ expressly rejected the distinction between subjective and objective reasons in Barclay:

to engage upon an inquiry contrasting ‘objective’ and ‘subjective’ reasons is to adopt an illusory frame of reference. Such an inquiry into the “objective” reasons risks the substitution by the court of its view of the matter for the finding it must make upon an issue of fact.^[165]

However, where an employer gives ‘direct evidence ... as to [their] state of mind, intent or purpose’,^[166] that may well be decisive. Indeed, an employer’s reason for acting will likely be accepted if the evidence is reliable, even if the reason itself is irrational. This makes causation a substantial and possibly determinative hurdle for claimants, which is nearly impossible to surmount, even with the reverse onus of proof under s 361 of the FWA.

In mapping the different approaches to the onus of proof and causation under the FWA and preceding legislation, Chapman, Love and Gaze compare the ‘Barclay approach’^[167] with a ‘broader approach’, where courts apply a ‘wider lens’ to ‘independently consider the extent to which the stated innocent reason of the employer, and sometimes more broadly the decision of the employer, is linked to the alleged prescribed ground’.^[168] These two approaches represent ‘two end points on a continuum or spectrum of judicial approaches to the reverse onus of proof and the causal link’.^[169] Understandably, where the broader approach is adopted, claimants are far more likely to be successful.^[170] Unfortunately for claimants, however, the Barclay approach appears to remain dominant.

In the sample of cases analysed in this article, the impact of Barclay is illustrated by Vink^[171]. In that case, the employer ‘categorically denied that age was a factor in his decision to dismiss the applicant’.^[172] The Federal Magistrate’s Court and, on appeal, the Federal Court of Australia, accepted that denial: ‘[i]t seems to me that, rightly or wrongly, Mr Ottobre believed that the applicant was incompetent when Mr Ottobre decided on 9 November 2011 to dismiss the applicant.’^[173] Even if that belief was wrong, it would not affect the quality of the denial. In Vink, there was some evidence to support the employer’s belief of incompetence — even if that evidence was later proven to be false.^[174] This

may well have weighed heavily in an unfair dismissal claim. ... In a case such as the present, however, where what was alleged was a contravention of s 351 of the FW Act, the focus was on the substantial and operative reasons which motivated Mr Ottobre to dismiss Mr Vink. Provided that those reasons did not include Mr Vink’s age, it mattered not that they were based on a mistaken assessment or were not supported by the weight of the evidence.^[175]

This was the case even though Mr Vink was likely told that the employer wanted a ‘youthful and vibrant’ culture. The issue was what the employer actually (subjectively) intended:

[I]t is probable that Mr Clerk [the general manager] did tell the applicant that Mr Ottobre wanted a vibrant and youthful culture. However, that is not to say that Mr Ottobre did want a vibrant and youthful culture or that that was his reason for dismissing the applicant.^[176]

Thus, the Court’s approach indicates that any plausible explanation for dismissal — which does not include a protected ground — will be sufficient to exclude a claim under the FWA, and that courts are willing to second-guess employer statements to look for their ‘true’ motivation. This will operate as a substantial obstacle to successful claims under the FWA.

This outcome may be compared with that in the earlier (pre-Barclay) case of Carr.^[177] In that case, the claimant gave evidence that he was told he was dismissed because the third party who was engaging his services did not ‘want young blokes working on the wind farms anymore’.^[178] The employer denied making this statement and claimed that Mr Carr’s dismissal was due to a lack of work, his relative lack of experience, and the relative (lack of) quality of his work. The Court could not determine which account to prefer, and could not ‘determine with sufficient certainty whether or not Mr Carr’s age was a material and operative factor when Mr Van Kempen decided to terminate Mr Carr’s employment’.^[179] The employer had therefore ‘failed to establish a defence that Mr Carr’s termination “was for a reason or reasons that do not include a proscribed reason” for the purposes of s 664 of the Act’, and Mr Carr’s claim succeeded on this ground.^[180] This may reflect the particular wording of s 664 of the Workplace Relations Act 1996 (Cth), which explicitly did not require the claimant to prove causation, and re-framed causation as a defence, to be proven by the employer:

(a) it is not necessary for the employee to prove that the termination was for a proscribed reason; but

(b) it is a defence in the proceedings if the employer proves that the termination was for a reason or reasons that do not include a proscribed reason ...

The framing of this section is arguably clearer than that in s 361 of the FWA, which says:

(1) If:

(a) ... it is alleged that a person took, or is taking, action for a particular reason or with a particular intent; and

(b) taking that action for that reason or with that intent would constitute a contravention of this Part;

it is presumed that the action was, or is being, taken for that reason or with that intent, unless the person proves otherwise.

Regardless of the similarities or differences between these sections, it is unlikely that the comparatively benevolent interpretation in Carr would be adopted in other cases post-Barclay.

These issues with causation under the FWA were echoed in other jurisdictions. In Travers,^[181] decided under the Age Discrimination Act, the reason for the treatment was ‘because the Board of Studies was dissatisfied with the manner in which Ms Travers supervised the examinations’, not age.^[182] In other cases, the employer gave a plausible ‘innocent’ explanation for the claimant’s treatment:^[183] age was therefore not the ‘real reason’ for their treatment.^[184] As under the FWA, an employer’s denial of age discrimination could be sufficient to exclude a claim, particularly in the absence of other supporting material.^[185] This reflects the fact that the onus of proof is generally on the claimant in non-FWA jurisdictions.^[186]

In only one case did the court reject an employer’s ‘innocent’ explanation. In Lightning Bolt,^[187] the Court rejected the argument that the claimants were dismissed due to a desire for more ambitious employees, as distinct from age discrimination:

Ambition is not necessarily a characteristic of the young, but neither can it be said that they are necessarily devoid of it. There is no inconsistency between desiring to employ people who are ambitious to advance beyond store work and who could be part of a trained pool who could be promoted to other areas as the need arose; and desiring to employ young people.^[188]

Thus, the reason was not ‘innocent’. Further, the Court noted that even if there were other reasons for the treatment, this did not invalidate that age was a substantial reason.^[189]

Other claimants provided insufficient evidence that the treatment was due to age,^[190] meaning the case as to causation was ‘largely based upon speculation’.^[191] Difficulties that claimants might encounter with proving causation, particularly in claims relating to recruitment, were acknowledged by the Tribunal in Kennedy:

A complainant seeking to establish that age ... played a role in a decision of an interview panel faces an onerous task. The notes (if any) taken by panel members, even where they survive, are not made available to applicants. The deliberations that led to the panel’s final decision are not held in public or recorded. The recollection of panel members is often blurred where a significant time has elapsed since the interview. This is exacerbated where the interviewer sits on many panels, as is often the case in the Public Service. Even where a ‘textbook’ interview has been conducted, factors other than merit may consciously or otherwise creep into the decision. In short, if a panel has allowed improper factors to contaminate their decision it is extremely difficult for a complainant to establish that this was so. While we acknowledge these practical difficulties, the complainant is not excused from discharging their evidentiary onus.^[192]

Thus, this claim failed due to a lack of evidence that age affected the recruitment decisions.^[193]

In McEvoy,^[194] a successful claim, causation was established as the employer failed to counter the claimant’s evidence regarding why he was dismissed. This was not due to a lack of opportunity — as the Tribunal noted,

[t]he Board invited Acorn to comment on Mr McEvoy’s account of his meeting with Ms Kelly. In a letter dated 22 May 2015, Solicitors for Acorn wrote that Acorn was unable to comment on Mr McEvoy’s account as Ms Kelly was ‘on leave of absence due to unfitness for work’. The Solicitors wrote that Mr McEvoy was dismissed from his employment due to ‘on-going problems with his performance’ but did not elaborate. In addition, the Solicitors wrote that Acorn denied ‘the allegation that rude jokes, swearing and filthy language were commonplace in its workplace’.^[195]

Ms Kelly, who dismissed the claimant, could not be found to give evidence in the proceedings — her employment had also ended, ‘subject to the terms of a “confidential settlement”’^[196] and she could not be found by a process server to serve a summons to appear.^[197] This left a gaping hole in the employer’s evidence:

Only Mr McEvoy and Ms Kelly were parties to their alleged conversation on 28 Februray [sic] 2014. In that sense, this is a case of word against word. Mr McEvoy gave sworn evidence and was cross-examined about his account. Ms Kelly, on the other hand, was not called by Acorn due to her unavailability. Instead, an unsworn statement was tendered and admitted in evidence. It is obviously relevant but, given the circumstances, especially the fact that it has not been directly tested, the weight it should receive is difficult to assess. On the other hand, on the face of it, Mr McEvoy presented as a plausible and truthful witness.^[198]

It is unclear whether Mr McEvoy’s claim would have succeeded had the employer given (any) evidence as to why he was dismissed. Thus, where an employer offers evidence of their motivation (which, of course, would rarely be acknowledged to be discriminatory), it appears that causation will be a difficult — if not insurmountable — hurdle for claimants.

G Exceptions

While it has been hypothesised elsewhere that the wide-ranging exceptions to age discrimination law will preclude many claims in practice,^[199] exceptions were raised in only 10 cases in the sample:

• Three cases related to exceptions for statutory provisions;^[200]

• One case related to the setting of ‘reasonable and appropriate’ minimum ages by qualifying bodies under s 49ZYG(2) of the Anti-Discrimination Act 1977 (NSW);^[201]

• One case related to exceptions for superannuation fund conditions;^[202]

• Five cases related to the inherent requirements of the position;^[203] and

• One case related to positive action.^[204]

This represents only a small subset of the sample; though, where an exception was relevant, it proved fatal to the claimant’s action in six cases.^[205]

H Indirect Discrimination

Indirect discrimination was raised in only a minority of cases (17). Establishing indirect discrimination caused serious issues for claimants, particularly in identifying groups for comparison;^[206] and/or their ability to comply;^[207] and/or whether the requirement was (un)reasonable.^[208] Only one of the 12 successful substantive decisions established indirect age discrimination.^[209]

I Jurisdictional Issues

Jurisdictional issues beset some claimants. Some attempted to bring both Age Discrimination Act and FWA claims.^[210] Others sought to bring claims at both state and federal levels,^[211] or to challenge federal legislation under state discrimination law.^[212] The correct jurisdiction for bringing a claim is often far from clear. For example, in B v Naval Reserve Cadets,^[213] part of the alleged conduct occurred on Federal Government property and part of it did not. Further, the Tribunal had to consider whether the Naval Reserve Cadets were an emanation of the Commonwealth, or whether the Commonwealth might be covered by state discrimination legislation. Resolving these questions necessarily entailed further submissions and an additional hearing:

The jurisdictional issue is a complex one and requires consideration of cases which have not been the subject of submissions. It would not be appropriate to determine the jurisdictional issue without the advantage of further submissions. The Tribunal is of the view that the appropriate course is to consider the jurisdictional issue at a later stage if the Commissioner’s decision is overturned. The advantage of this course is that the issue may be determined during a preliminary hearing when evidence may be called.^[214]

This reflects the complexity of identifying the best jurisdiction in which to bring a claim.

V Discussion

In sum, then, the cases in this sample demonstrate a wide range of failure points, in both procedural and substantive areas. Based on this sample, it is nearly impossible to establish a claim of age discrimination in court. Of course, the vast majority of age discrimination complaints are resolved through conciliation and mediation before reaching court.^[215] Thus, any attempt at quantitative analysis of judicial decisions only offers half (or, more accurately, much less than half) of the full picture of claims.^[216] This analysis would therefore benefit from corroboration and triangulation with conciliation statistics. Given the limited information that is released by Australian equality bodies in their annual reports on these detailed issues, this is difficult to do in practice.^[217] From this (albeit limited) sample of claims, some thematic observations emerge, related to the demographics of age discrimination claimants, the acceptance of age discrimination in the workplace, the prevalence of poor human resources practices in many workplaces, and the challenges facing young workers.

A Claimant Demographics

The prevalence of older, white male claimants in the sample reveals the potential for tensions between age discrimination law and other protected characteristics:^[218] white men are not normally those who are seen as needing protection from discrimination. This potentially raises a fundamental challenge to age discrimination law’s perceived legitimacy.

More generally, though, it is debatable why older women are so underrepresented in the sample of cases. This is unlikely to be because older women experience less age discrimination than men: in the AHRC National Prevalence Survey, there was no statistically significant difference in the rates at which men (28%) and women (26%) experienced age discrimination in employment.^[219] Women were also more likely than men to report that their experience of discrimination affected their self-esteem, mental health or stress levels.^[220] Thus, women may experience age discrimination at least as frequently as men, and may experience more severe personal repercussions. Thus, the absence of claims by older women is particularly concerning.

A few possible explanations may be put forward to explain these demographic trends. Older white men may have greater resources than women (both financial and otherwise), increasing their capacity to pursue a claim to court. By contrast, there is increasing recognition that older women are at risk of poverty and poor pension savings,^[221] reducing their financial capacity to pursue their legal rights. This is particularly the case for older women in precarious or insecure work arrangements, who are likely to have limited financial resources.

Alternatively, older women may be less likely to identify and rail against discrimination, having experienced discriminatory treatment throughout their working lives. In the UK, Grant’s study of women aged over 50 mapped the prevalence of discrimination on the basis of both gender and age against women across the life course, often spanning 30 or 40 years: ‘discriminatory practices were part of [women’s] everyday [work] experiences’.^[222] Older women in the study felt that the progress made by equality law was too little, too late to affect their own experiences in work.^[223] Drawing on these results, Grant argues that older women internalise discriminatory sentiment and a lack of belief in their own abilities. Thus, there is a ‘legacy of negativity based on the dubiety of legislation, past personal experiences and current perceptions’.^[224] It is possible that older women in Australia have experienced similar levels of discrimination across their own working lives, meaning they are less likely to complain about old age discrimination should they encounter it.

It may also be that there are simply more older men than women in employment in Australia, meaning men are also more likely to experience discrimination in employment. According to Australian Bureau of Statistics data, and as depicted in Table 4 (below), men had higher workforce participation rates than women for all age groups after the age of 50 in 2016–17 (when this study concluded).

Table 4: Australian labour force participation rate by age and gender, 2016–17^[225]

However, the difference in gender workforce participation rates is insufficient to account for the substantial disparity in claiming rates. In this sample, 81 cases (or 75%) featured a male claimant, and only 28 (26%) featured a female claimant.^[226] Thus, cases featuring male claimants were three times more likely than those featuring female claimants. The AHRC National Prevalence Survey found that those aged between 60 and 64 were most likely to report experiencing age discrimination in employment in the last two years (at 32% of respondents).^[227] In this age group, men aged 60 to 64 are only 1.27 times more likely to be in the labour force than women. The disparity in claiming therefore cannot be explained by workforce participation rates alone, though this may be part of the picture.

Finally, older women may have alternative grounds on which to pursue a discrimination claim, including on the basis of gender or caring responsibilities. Given the woeful success rates for age discrimination claims, it is possible that older women are strategically pursuing claims in other areas of discrimination law. This is a possibility that needs to be examined through further research.

B ‘Casual’ Ageism

A number of cases illustrate the normative acceptance of age discrimination in the workplace. For example, some cases opined that discriminatory language might be used to ‘save face’ for older workers. In Vink (FMCA), the Court suggested that: ‘[i]t may be that Mr Clerk told the applicant that Mr Ottobre (the managing director) wanted a vibrant and youthful culture because he thought that would be less hurtful than telling the applicant that Mr Ottobre thought he was incompetent.’^[228] Similarly, in Farah,^[229] the Court held that a reference to age in dismissal might have been a ‘face saving’ excuse to terminate the claimant’s employment:

At most, I am prepared to accept that when Mr Ahn first mentioned the possibility that he might decide in the following week to dispense with Mr Farah’s full-time services, at least temporarily, he might have made a comment that some staff might ‘relate better’ to a different manager whom he referred to as a ‘younger manager’. However, I doubt whether this would have been said by Mr Ahn with any belief that Mr Farah’s age was itself a real impediment to the continuance of his employment. I think it more likely that, if it was said, Mr Ahn made the statement as a ‘face saving’ excuse, which avoided the need to suggest any more personal failing by Mr Farah, or the need to admit that his doubts about employing Mr Farah went to the foundations of their previous mutual plans to take the café ‘up market’.^[230]

Other cases held that one-off comments did not mean that age was an operative factor in the claimant’s treatment^[231] and were not evidence of age discrimination.^[232] ‘Passing comment’ about retirement did not lead to an inference that treatment was due to age.^[233] In Pomplun,^[234] one age-based comment in private, with no evidence of detriment, was held to be unlikely to constitute harassment.

Beyond age-based comments, some decisions accepted that what might be age-based discrimination was actually just good business practice. In Martin v Donut King,^[235] it was held that a manager reducing a worker’s hours, to employ younger (cheaper) staff on junior rates, and avoid paying a higher rate of pay to an older worker, was not necessarily age discrimination:

It is also relevant to observe that the employer has the responsibility to manage the business in the most efficient and cost effective means and the lawful use of junior rates does not give rise to discrimination. There was no other allegation which went to the issue of age.^[236]

This may be compared with the successful unfair dismissal claim in Gateway Investments,^[237] where the employer’s argument that there was no work for the claimant was rejected, due to evidence that the claimant was dismissed at the same time that younger (cheaper) staff were being recruited to work at junior rates of pay,^[238] at the point where the claimant became eligible for senior rates of pay.^[239] This implied that age was a reason for the dismissal.

These cases, then, illustrate a degree of acceptance of age discrimination in the workplace. It appears that age discrimination law is unsuccessful in practice for addressing this sort of ‘casual’ ageism: a different approach is required. This is particularly important given the prevalence of ageist jokes and stereotypes in employment.^[240]

C Poor Human Resources Practices

The cases illustrate the presence of poor human resources practices in some workplaces. While courts might acknowledge that claimants have been subject to poor practice, it is not within the scope of legislation to address such issues. For example, in Vink (FMCA), the Court noted that:

It seems that Mr Clerk supported the applicant fixing the database, but Mr Ottobre simply wanted the applicant to fulfil his basic bookkeeping function. It seems that the applicant was the victim of a difference of opinion between Mr Clerk, as general manager, and Mr Ottobre, as managing director.^[241]

As a result, Mr Vink was dismissed, but bad practice alone was not sufficient to support a claim of age discrimination.

Similarly, in Arkley,^[242] the Tribunal noted that unfair practices were not necessarily discriminatory:

It is not uncommon for an employer or supervisor to treat an employee unfairly for any number of reasons, none of which have anything to do with discrimination. For example, the employer and/or supervisor may not give proper consideration to an employee’s excuses, which he or she considers reasonable, in response to allegations put to that employee. In this case Ms Witt may not have acted properly when disciplining or dismissing Mr Arkley because of a desire to save money or because she did not wish to trouble herself with Mr Arkley’s excuses. Alternatively she might not have considered his excuses to be reasonable. This all might be unfair, but it does not mean Mr Arkley has been discriminated against. A full hearing of this matter might demonstrate that the reasons given for Mr Arkley’s change in roster or the termination of his employment were unfair but that in itself is not enough to make out a case of discrimination or prohibited conduct on the basis of age.^[243]

Thus, the Commissioner was correct in dismissing the complaint.

Finally, in McKeown,^[244] the Tribunal was clear that it was not commenting on whether the dismissals concerned were inaccurate or unfair — just whether they were based on age.^[245] The employer’s argument that the employees who were let go lacked ‘flexibility’, and that junior cashiers tended to be more flexible, was accepted.^[246] This flags issues of intersectionality: while younger women might be ‘flexible’, older women with caring responsibilities tend to be less flexible. This issue was not pursued in this case, even though all those dismissed were women, and many had caring responsibilities. Age discrimination law therefore presently appears ill-suited for addressing poor human resource practices, though many claims in the sample arose from deficient organisational practices.^[247] As McKeown shows, there is a need for courts and advocates to interrogate organisational practices more deeply, to consider issues of intersectionality and assess whether poor practices might have disparate impact on particular age groups.

D Young Workers

The sample reveals a noticeable absence of claims by younger workers: only six cases were brought on the basis of being too young. Those claims that are brought, however, often involve allegations of appalling behaviour and abuse of power. In Gabryelczyk,^[248] for example, the claimant was told his treatment was ‘character building’;^[249] this included being struck in the face with car keys, trying to touch his scrotum and crotch, being hit on the knee with pliers, having a lit cigarette stubbed out on his arm, having a cigarette lighter held to his neck, being chased through a car park and hogtied, having a lit cigarette flicked into his eye, having his lunch regularly stolen, and being called offensive names.^[250] Despite the severity of the behaviour alleged, the claim in Gabryelczyk failed, as it was against the employee concerned, not the employer (the claim against the employer had already been conciliated successfully).^[251] This case suggests that young workers are in a precarious position, and illustrates the extreme abuses of power that are possible in the workplace. It also reveals the limitations of age discrimination law for correcting these power imbalances.

VI Conclusion

The study reported here provides renewed support for existing scholarly critiques of the process and substance of Australian discrimination law, particularly as it relates to time limits under the FWA,^[252] the comparator requirement at the federal level,^[253] causation,^[254] and the onus of proof.^[255] More particularly, limiting High Court of Australia decisions — such as Purvis and Barclay — are having a substantial impact in practice on individual claims.

Some of these issues could be addressed by legislative reform. For example, as noted in Part III, the AHRC’s Willing to Work report recommended a number of areas for review, with a view to amending existing age discrimination laws at the federal level.^[256] To the AHRC’s recommendations, we could add shifting the onus of proof, though this does not appear to have had much impact in the context of the FWA.

Of course, legislative reform will not cure all points of failure: even if, for example, we reviewed the comparator requirement, and replaced it with a test of ‘detriment’^[257] or unfavourable treatment, some meritorious claimants may still struggle to establish detriment or unfavourable treatment.^[258] Thus, even if the comparator requirement was removed, some claims might still fail: this may mean that the claims are of low merit; or may reflect the limits of age discrimination law for addressing ‘casual’ ageism.

The best way to examine the possible impact of reform in this area is to review claims brought in Victoria or the ACT, where the statute requires unfavourable treatment rather than a comparator. Within this sample, the four claims in Victoria all failed — not due to the comparator requirement, but due to: a lack of discriminatory behaviour and detriment;^[259] lack of evidence;^[260] being beyond the scope of state legislation;^[261] or lack of causation.^[262] The case in the ACT,^[263] while successful, did not include any detailed discussion of unfavourable treatment or the nature of the discrimination. Thus, at most it can be said that adopting an unfavourable treatment test will address one barrier to bringing a claim of age discrimination.

Even in the absence of legislative reform, more sympathetic and purposive judicial interpretation of existing statutes could do much to promote claimants’ prospects of success.^[264] This is illustrated, for example, by Chapman, Love and Gaze’s comparison between the ‘Barclay approach’ and a ‘broader approach’: where the broader approach is adopted, claimants are far more likely to be successful.^[265] Thus, legislative change is not necessarily required in this area, but a more contextual and critical judicial approach to interpreting the law and facts is required.

Similarly, a highly restrictive approach to the comparator requirement, as exemplified by Purvis and subsequent case law, does not require legislative reform to address. The artificiality and narrowness of Australian cases on this point can be compared with the Supreme Court of New Zealand age discrimination case of McAlister v Air New Zealand Ltd,^[266] which related to the demotion of pilots after the age of 60, in keeping with age-based rules in some airspaces like the US. The question for the Supreme Court was whether the comparator should be a pilot under the age of 60; or a pilot under the age of 60 who was unable to fly to destinations like the US (due to visa requirements or other conditions). The Court held that the latter, Purvis-style comparator was ‘too much’^[267] and would ‘appear to lead to an obvious result’^[268] by artificially ruling out discrimination at an early stage of the inquiry.^[269] Thus, a Purvis-style comparator would be artificial and failed to reflect a purposive interpretation of the statute.^[270] This decision throws the Australian case law into sharp relief: New Zealand courts adopt a dramatically different approach to Australian courts in the selection of a comparator in age discrimination claims. It would be open to the High Court of Australia to adopt a similar approach, without legislative intervention.

More generally, this study highlights the implications of a federal structure, and the non-uniform nature of Australian discrimination law. While particular hurdles (such as highly restrictive time limits) can be avoided by choosing a different jurisdiction, no legal framework appears ideal for claimants: levels of success were uniformly low. Further, choice of jurisdiction can lead to complex decision-making for claimants,^[271] which can generate confusion and incorrect choices.

Returning to the original question, it is unclear whether the employment age discrimination claims that progress to court in Australia are inherently weak, explaining the limited prospect of success. Instead, it is possible that a lack of legal representation, difficulty obtaining evidence and the importance of an employer’s subjective intent mean that age discrimination claims are set up to fail.

Overall, the cases reveal the difficulties of addressing age discrimination through individual claims. Even when claims are successful, damages for loss of a chance to be employed are likely to be low.^[272] While one successful claim achieved a high damages payout, this reflected the loss of a career,^[273] recognising that few older workers will be re-employed after experiencing age discrimination. To achieve meaningful change for older workers requires a different approach. This also reflects the prevalence of poor human resource practices in the sample, including unfair dismissals that could not be challenged in the unfair dismissal jurisdiction, changes to working time and shift allocations, and the growing use of casual contracts. While age discrimination law was often used to address employee grievances — perhaps as a jurisdiction of last resort — it proved a limited tool for addressing poor organisational behaviour. Rather than relying on individuals to address these organisational failings, a more proactive and preventative approach is required. It is time for positive duties to be imposed on Australian employers to address inequality.

^&#6[1] Associate Professor and Discovery Early Career Research Fellow, Melbourne Law School, University of Melbourne, Victoria, Australia. Email: alysia.blackham@unimelb.edu.au; ORCID iD: 0000-0002-8149-4887. This research was 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. I gratefully acknowledge research assistance provided by the Melbourne Law School Academic Research Service, and the insightful comments of the anonymous reviewers and participants at the Australian Labour Law Association 9^th Biennial Conference, where this research was first presented.

^[1] Australian Human Rights Commission (‘AHRC’), National Prevalence Survey of Age Discrimination in the Workplace: The Prevalence, Nature and Impact of Workplace Age Discrimination amongst the Australian Population Aged 50 Years and Older (2015) (‘National Prevalence Survey’).

^[2] Ibid 18.

^[3] Ibid 23.

^[4] AHRC, Willing to Work: National Inquiry into Employment Discrimination against Older Australians and Australians with Disability (2016) 38 (‘Willing to Work’).

^[5] For detailed discussion of the statutory frameworks in each jurisdiction, see Neil Rees, Simon Rice and Dominique Allen, Australian Anti-Discrimination and Equal Opportunity Law (Federation Press, 3^rd ed, 2018).

^[6] See, eg, Therese MacDermott, ‘Resolving Federal Age Discrimination Complaints: Where Have All the Complainants Gone?’ (2013) 24(2) Australasian Dispute Resolution Journal 102 (‘Resolving Federal Age Discrimination Complaints’). MacDermott considered only federal cases, not those at state and territory level.

^[7] On age discrimination law and its exceptions, see Alysia Blackham, ‘A Compromised Balance? A Comparative Examination of Exceptions to Age Discrimination Law in Australia and the UK’ [2018] MelbULawRw 4; (2018) 41(3) Melbourne University Law Review 1085 (‘A Compromised Balance?’).

^[8] Lisa Webley, ‘Qualitative Approaches to Empirical Legal Research’ in Peter Cane and Herbert M Kritzer (eds), The Oxford Handbook of Empirical Legal Research (Oxford University Press, 2010) 926, 941.

^[9] See Mark Van Hoecke, ‘Legal Doctrine: Which Method(s) for What Kind of Discipline?’ in Mark van Hoecke (ed), Methodologies of Legal Research: What Kind of Method for What Kind of Discipline? (Hart Publishing, 2011) 1, 4.

^[10] See ibid 11–17.

^[11] Mark A Hall and Ronald F Wright, ‘Systematic Content Analysis of Judicial Opinions’ (2008) 96(1) California Law Review 63, 66.

^[12] Ibid 76.

^[13] Michael Quinn Patton, Qualitative Evaluation and Research Methods (SAGE, 2^nd ed, 1990) 381.

^[14] The earliest cases in the sample were from 1996 (under the Industrial Relations Act 1988 (Cth) and in the Northern Territory under the Anti-Discrimination Act 1992 (NT)), though age discrimination was first added to the grounds of discrimination in the Human Rights and Equal Opportunity Commission Act 1986 (Cth) from 1 January 1990 by the Human Rights and Equal Opportunity Commission Regulations 1989 (Cth) reg 4(a)(i); and prohibited in South Australia from 1 June 1991. Compulsory retirement was progressively abolished in New South Wales from 1 January 1991: Anti-Discrimination (Compulsory Retirement) Amendment Act 1990 (NSW).

^[15] Including Barnet JADE <https://jade.io/>, individual Tribunal websites, LexisNexis AU cases <https://www.lexisnexis.com.au/aus/products/lexisnexisau/>, Westlaw AU cases <https://legal.thomsonreuters.com.au/products/westlaw/>, CCH Online (Discrimination Cases and Industrial Relations Cases) <http://www.wolterskluwer.cch.com.au/our-products/content-solutions/intelliconnect> , and the Fair Work Commission website <https://www.fwc.gov.au/>).

^[16] Some cases were identified that related to the administration of retirement policies, but these were not challenged as a form of age discrimination, so were excluded from the sample: see Stephen v Miller [1913] NSWStRp 9; (1913) 13 SR (NSW) 44; St George Building Society Ltd v Federated Clerks’ Union of Australia (NSW Branch) (1986) 15 IR 110; Australian Federation of Air Pilots v Bristow Helicopters Australia Pty Ltd (1993) 52 IR 450; Public Service Association (NSW) v Catholic College of Education Sydney Ltd (1987) 23 IR 235.

^[17] For this study, procedural decisions were retained in the sample, given that Australian interlocutory decisions reveal interesting insights into points of failure that might occur in procedural steps in the court process.

^[18] Gery W Ryan and H Russell Bernard, ‘Data Management and Analysis Methods’ in Norman K Denzin and Yvonna S Lincoln (eds), Collecting and Interpreting Qualitative Materials (SAGE, 2^nd ed, 2003) 259, 275–6.

^[19] Further details of these methods and results are on file with the author and available on request.

^[20] Hall and Wright (n 11).

^[21] See, eg, Fred Kort, ‘Predicting Supreme Court Decisions Mathematically: A Quantitative Analysis of the “Right to Counsel” Cases’ (1957) 51(1) American Political Science Review 1. Kort used quantitative analysis to predict claim outcomes (‘jurimetrics’).

^[22] 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.

^[23] Hubert S Feild and William H Holley, ‘The Relationship of Performance Appraisal System Characteristics to Verdicts in Selected Employment Discrimination Cases’ (1982) 25(2) Academy of Management Journal 392.

^[24] Schuster and Miller (n 22) 68.

^[25] Ibid 74.

^[26] LD Irving, ‘Challenging Ageism in Employment: An Analysis of the Implementation of Age Discrimination Legislation in England and Wales’ (PhD Thesis, Coventry University, 2012) <http://curve.coventry.ac.uk/open/items/ffc88163-6994-4400-bead-121298f52bd1/1/> .

^[27] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 102.

^[28] [2014] FCCA 1170 (‘Theravanish’).

^[29] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 108. See also Therese MacDermott, ‘Giving a Voice to Age Discrimination Complainants in Federal Proceedings’ (2017) 19(2) Flinders Law Journal 233. On the use of conciliation in discrimination law generally, see 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; Dominique Allen, ‘In Defence of Settlement: Resolving Discrimination Complaints by Agreement’ (2014) 14(4) International Journal of Discrimination and the Law 199.

^[30] In Victoria, direct access to the Victorian Civil and Administrative Tribunal was introduced in 2011: Equal Opportunity Act 2010 (Vic) s 122.

^[31] AHRC, Willing to Work (n 4) 417 (Table 4).

^[32] Complaints are terminated under s 46PH of the Australian Human Rights Commission Act 1986 (Cth), including on the basis that ‘the President is satisfied that ... there is no reasonable prospect of the matter being settled by conciliation’: s 46PH(1B)(b). Complaints terminated on this particular basis would appear to be those most likely to progress to court.

^[33] AHRC, 2015–16 Complaint Statistics (2016) 26 (Chart 8) <http://www.humanrights.gov.au/our-work/commission-general/publications/annual-report-2015-2016> (‘2015–16 Complaint Statistics’).

^[34] AHRC, Willing to Work (n 4) 417 (Table 4).

^[35] The ‘Human Rights’ jurisdiction includes federal unlawful discrimination matters under the Australian Human Rights Commission Act 1986 (Cth), and relates to complaints under the Age Discrimination Act, Disability Discrimination Act 1992 (Cth), Racial Discrimination Act 1975 (Cth), and Sex Discrimination Act 1984 (Cth). These statistics do not include applications under the Fair Work Act 2009 (Cth) (‘FWA’) or industrial legislation: see Federal Circuit Court of Australia, Annual Report 2015–16 (Commonwealth of Australia, 2016) 6, 28–9.

^[36] AHRC, 2015–16 Complaint Statistics (n 33) 2.

^[37] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 105.

^[38] Ibid 106.

^[39] Ibid.

^[40] FWA (n 35) s 774.

^[41] Blackham, ‘A Compromised Balance?’ (n 7).

^[42] Alysia Blackham, ‘Defining “Discrimination” in UK and Australian Age Discrimination Law’ [2017] MonashULawRw 23; (2017) 43(3) Monash University Law Review 760 (‘Defining “Discrimination”’).

^[43] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 107.

^[44] Ibid.

^[45] AHRC, Willing to Work (n 4) 324–5.

^[46] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 107.

^[47] Ibid 108–9, 110–111; AHRC, Willing to Work (n 4) 321–2.

^[48] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 107; AHRC, Willing to Work (n 4) 323–4.

^[49] MacDermott, ‘Resolving Federal Age Discrimination Complaints’ (n 6) 110; AHRC, Willing to Work (n 4) 322–3.

^[50] AHRC, Willing to Work (n 4) 323–4.

^[51] Ibid 325–6.

^[52] Ibid 326.

^[53] Ibid 332–3 (Recommendation 48).

^[54] Ibid 333–4 (Recommendation 49).

^[55] Ibid 336 (Recommendation 50).

^[56] Ibid 337–8 (Recommendation 51).

^[57] Ibid 338 (Recommendation 52).

^[58] Ibid 342–3 (Recommendation 56).

^[59] Irving (n 26) 160–69.

^[60] Ibid 171–80.

^[61] Ibid 169–71.

^[62] Ibid 189–94.

^[63] Ibid 201–28.

^[64] Ibid 195–6.

^[65] Other areas included: training (6), promotion (5), demotion (4), working time (8), harassment (10), benefits (5), performance (2) and other (5). As cases could relate to multiple areas, these numbers do not add up to 108.

^[66] AHRC, National Prevalence Survey (n 1) 38.

^[67] Ibid 39.

^[68] One case included both a male and female claimant, meaning these numbers add to 109, not 108.

^[69] Schuster and Miller found that in the US, most claimants were professionals or managers: Schuster and Miller (n 22) 74. However, some occupations are difficult to categorise in this binary way: call centre workers, for example, are probably better categorised as unskilled workers than professionals, though they are more likely to be seen as white collar than blue collar workers.

^[70] Von Stalheim v Davey Accounting Plus [2007] TASADT 7 (‘Davey Accounting’); Von Stalheim v KPMG [2003] TASADT 12 (‘KPMG’); Von Stalheim v Garrotts [2003] TASADT 9 (‘Garrotts’).

^[71] Equal Opportunity Act 2010 (Vic) s 122; see Dominique Allen, Addressing Discrimination Through Individual Enforcement: A Case Study of Victoria (Monash Business School, 2019) 10–11. It is possible that some potential state claims have been displaced to the federal jurisdiction.

^[72] ABS, Australian Demographic Statistics, Jun 2017 (Catalogue No 3101.0, 14 December 2017).

^[73] Source: Australian Bureau of Statistics (‘ABS’), Australian Demographic Statistics, Dec 2017 (Catalogue No 3101.0, 21 June 2018) and author’s own calculations.

^[74] But some were successful on a different basis.

^[75] This is lower than the comparable figure for the US, where Schuster and Miller found that employers won nearly two-thirds of cases: Schuster and Miller (n 22) 70 (Table 3).

^[76] Theravanish (n 28); McEvoy v Acorn Stairlifts Pty Ltd [2017] NSWCATAD 273 (‘McEvoy’); Talbot v Sperling Tourism & Investments Pty Ltd (2011) 211 IR 419 (‘Talbot’); Lightning Bolt Co Pty Ltd v Skinner (2003) EOC 93-260 (‘Lightning Bolt’); Webforge Australia Pty Ltd v Richards 2004 WAIRComm 13207; (2005) EOC 93-360 (‘Webforge’).

^[77] Bloomfield v Westco Jeans Pty Ltd (2001) EOC 93-161 (‘Bloomfield’).

^[78] Hosking v Faser (1996) EOC 92-859 (‘Hosking’); Willmott v Woolworths Ltd [2014] QCAT 601 (‘Willmott’).

^[79] Cf Carr v Blade Repairs Australia Pty Ltd (No 2) [2010] FCA 688; (2010) 197 IR 307 (‘Carr’), discussed below in text accompanying nn 177–80.

^[80] Theravanish (n 28).

^[81] Carr (n 79).

^[82] See Borg v NSW Greyhound Breeders, Owners & Trainers’ Association [2012] FWA 10013; Martin v Donut King Chirnside Park [2012] FWA 2905 (‘Martin v Donut King’); Stewart v Kalari Pty Ltd [2004] AIRC 27. See also Anderson v Thiess Pty Ltd, which illustrates that age may play a role in showing that a dismissal is unfair: [2015] FWCFB 478. Age could be relevant to the decision: see Chakalakis v M T Sullivan & Co Pty Ltd [2008] AIRC 425, [63].

^[83] Commonwealth v Human Rights & Equal Opportunity Commission [2000] FCA 1854; (2000) 108 FCR 378 (‘Commonwealth v HREOC (Hamilton)’); Commonwealth v Human Rights & Equal Opportunity Commission [2000] FCA 1150; (2000) 104 FCR 464; Commonwealth v Human Rights & Equal Opportunity Commission [1999] FCA 1524; (1999) 95 FCR 218 (‘Commonwealth v HREOC (Bradley)’). In these cases, the Commonwealth brought an application for review under the Administrative Decisions (Judicial Review) Act 1977 (Cth) of a Human Rights & Equal Opportunity Commission decision (that is, a finding and a recommendation under the Human Rights and Equal Opportunity Commission Act 1986 (Cth)).

^[84] Edwards v Tiger Airways Australia Pty Ltd [2017] FWC 4021 (‘Edwards’).

^[85] Forrester v Growers Market Express Pty Ltd [2015] FWC 8874 (‘Forrester’).

^[86] Rochas v University of Sydney [2004] NSWADT 14 (‘Rochas’); French v Gosford City Council [2003] NSWADT 273 (‘French’).

^[87] Vanden Driesen v Edith Cowan University [2012] FMCA 735; (2012) 269 FLR 422.

^[88] Wells v Glen Innes Severn Council [2010] NSWADT 281 (‘Wells’); Donohoe v Trustees of the Christian Brothers [2008] NSWADT 51 (‘Donohoe’); Goyal v National Electricity Market Management Co Ltd [2008] NSWADT 181 (‘Goyal’).

^[89] B v Naval Reserve Cadets [2004] TASADT 11.

^[90] Source: ABS, Australian Demographic Statistics, Dec 2017 (n 73) and author’s own calculations.

^[91] Jurisdiction key: Cth = Federal; ACT = Australian Capital Territory; NSW = New South Wales; NT = Northern Territory; Qld = Queensland; SA = South Australia; Tas = Tasmania; Vic = Victoria; WA = Western Australia.

^[92] McEvoy (n 76).

^[93] Theravanish (n 28).

^[94] Willmott (n 78).

^[95] [2013] QCAT 243.

^[96] Talbot (n 76).

^[97] Carr (n 79).

^[98] (2007) EOC 93-457 (‘Virgin Blue Airlines’).

^[99] Webforge (n 76).

^[100] [2005] NSWIRComm 209 (‘Gateway Investments’).

^[101] Lightning Bolt (n 76).

^[102] Bloomfield (n 77).

^[103] Hosking (n 78).

^[104] FWA (n 35) s 774.

^[105] Edwards (n 84).

^[106] Sternberg v Gables Reception Pty Ltd [2016] FWC 7892; Robb v Bond University Ltd [2016] FWC 1552; Armstrong v Police Citizens Youth Clubs [2016] FWC 766; Downie v Naladra Pty Ltd [2015] FWC 8905. Under the Workplace Relations Act 1996 (Cth), see Lin v University of Melbourne [2009] AIRC 61 (‘Lin’); Luppi v Spanish Club Ltd [2008] AIRC 465; Boes v Universal Rocks Water Garden Centre Pty Ltd [2007] AIRC 552.

^[107] Forrester (n 85).

^[108] Cavar v Green Gate Management Services Pty Ltd [2017] FCA 471 (‘Cavar’).

^[109] Boyn v Schering Pty Ltd [2008] FCA 961; (2008) 103 ALD 108.

^[110] O’Brien v Crouch [2007] FMCA 1976.

^[111] Ibid [29], [30].

^[112] McEvoy (n 76); Talbot (n 76); Gateway Investments (n 100).

^[113] Rochas (n 86); French (n 86); Donohoe (n 88); Goyal (n 88); Wells (n 88).

^[114] Retallick v Nestlé Australia Ltd [2006] NSWADT 343, [14].

^[115] McIntyre v Tully [2000] QCA 115; [2001] 2 Qd R 338, 344 [15] (‘McIntyre’).

^[116] Gabryelczyk v Hundt [2005] NSWADT 94 (‘Gabryelczyk’).

^[117] Gladdish v Queensland [2012] QCAT 721, [30].

^[118] Lidis v Top Dog Minders Pty Ltd [2011] QCAT 232, [21].

^[119] See, eg, Keys v Sydney Night Patrol and Inquiry Co Pty Ltd, where there was no evidence older workers were less able to comply with a recruitment test to found a claim of indirect discrimination: [2015] FCCA 776, [22] (‘Keys’); Martin v Donut King (n 82) [29]; Vye v Secretary, Department of Finance, Services and Innovation (NSW) [2016] NSWCATAD 117, [33]–[39] (‘Vye’); Coady v Sutherland Shire Council [2016] NSWCATAD 95, [18]–[19]; Neeson v Director-General, Department of Education and Training [2009] NSWADT 292, [46]; Shirley v Director-General, Department of Education and Training (No 2) [2009] NSWADT 235, [45], [48] (‘Shirley’); Heber v Glen Henney & Son Pty Ltd (No 3) [2008] NSWADT 168, [24] (‘Heber’); Pignat v Richmond Valley Council [2005] NSWADT 162, [28], [35] (‘Pignat’); Kennedy v Director-General, NSW Department of Industrial Relations [2002] NSWADT 186, [119] (‘Kennedy’); Serewko v Queensland [2009] QADT 5, [32]; Arkley v CatholicCare Tasmania [2017] TASADT 3, [15], [41], [45] (‘Arkley’); Davey Accounting (n 70) [57]; KPMG (n 70) [24]–[26]; Garrotts (n 70) [6]–[7]; Udugampala v Essential Services Commission [2016] VCAT 2130, [108] (‘Udugampala’).

^[120] Pomplun v Shoalhaven City Council [2010] NSWADT 113 (‘Pomplun’); Tanevski v Fluor Australia Pty Ltd [2008] NSWADT 217; Perera v Warehouse Solutions Pty Ltd [2017] VCAT 1267, [243]–[244] (‘Warehouse Solutions’). See also Navaratnam v Queensland, which related to a proposed process, which meant a previous process was no longer relevant: [2013] QCAT 131.

^[121] Naidu v Causeway Inn Pty Ltd [2015] VCAT 929, [31] (‘Naidu’).

^[122] Sidhu v Sydney Local Health District [2015] NSWCATAD 70 (‘Sidhu’); Plancke v Director General, Department of Education & Training [2001] NSWADT 137 (‘Plancke’); Allen v Newlands Coal Pty Ltd (No 2) [2014] QCAT 522, [37]; Perera v Director-General, Department of Education and Communities (Office of Communities) [2012] NSWADT 108 (‘Perera’).

^[123] Sun v EP2 Management Pty Ltd [2016] FCCA 1381, [37] (‘EP2’); Heber (n 119) [20]; Ng v Citigroup Pty Ltd [2008] AIRC 233, [49].

^[124] Bradford v Toll Personnel Pty Ltd [2013] FWC 1062; Thompson v Big Bert Pty Ltd [2007] FCA 1978; (2007) 168 IR 309 (‘Thompson’).

^[125] Maurer v S.U.M.M.S [2013] FWC 1661; Lin (n 106); Yaxley v Trust Bank of Tasmania [1996] IRCA 132. And, in cases under the Industrial Relations Act 1988 (Cth), claims failed where the dismissal was not at the ‘initiative’ of the employer: Adams v Australia Post (1996) 64 IR 309; Griffin v Australian Postal Corporation [1998] IRCA 15.

^[126] [2003] HCA 62; (2003) 217 CLR 92 (‘Purvis’).

^[127] Ibid 107–8 [34]–[38], 108 [40] (McHugh and Kirby JJ), 148 [183] (Gummow, Hayne and Heydon JJ).

^[128] Disability Act (n 35) s 5(1) (emphasis added).

^[129] Cf Purvis (n 126) 105–6 [27], 134 [129]–[130] (McHugh and Kirby JJ).

^[130] Ibid 100 [11] (Gleeson CJ). See also at 160–2 [222]–[232] (Gummow, Hayne and Heydon JJ), 175 [273] (Callinan J).

^[131] Colin D Campbell, ‘A Hard Case Making Bad Law: Purvis v New South Wales and the Role of the Comparator under the Disability Discrimination Act 1992 (Cth)’ [2007] FedLawRw 4; (2007) 35(1) Federal Law Review 111.

^[132] Susan Roberts, ‘The Inequality of Treating Unequals Equally: The Future of Direct Discrimination under the Disability Discrimination Act 1992 (Cth)?’ [2005] AIAdminLawF 8; (2005) 45 AIAL Forum 20, 30–1.

^[133] Ibid 30; Campbell (n 131).

^[134] Travers v State of New South Wales (Board of Studies Teaching and Educational Standards NSW) [2016] FCCA 905, [49] (‘Travers’).

^[135] [2013] NSWADT 78 (‘Duncan’) [64].

^[136] Kennedy (n 119) [129].

^[137] McEvoy (n 76) [71]–[73]; Hayne v YMCA NSW [2016] NSWCATAD 14, [17] (‘YMCA’); Duncan (n 135) [64]; Shirley (n 119) [40]; Heber (n 119) [7]–[8]; Pignat (n 119) [25]; Kennedy (n 119) [70]; Peacock v Human Rights and Equal Opportunity Commission [2005] FCAFC 45, [54]; Donohoe (n 88) [17]; Goyal (n 88) [18].

^[138] Virgin Blue Airlines (n 98).

^[139] Ibid 74628 [155].

^[140] Ibid 74627–8 [154].

^[141] Ibid 74628 [156], 74629 [167].

^[142] Talbot (n 76).

^[143] Ibid 434 [72].

^[144] Ibid.

^[145] Ibid 435 [75].

^[146] Duncan (n 135) [65] (emphasis in original).

^[147] EP2 (n 123).

^[148] Ibid [17].

^[149] Ibid [2].

^[150] Ibid [18].

^[151] Ibid [37].

^[152] Ibid [38].

^[153] See also Discrimination Act 1991 (ACT) s 8(2).

^[154] [2013] VSCA 28; [2013] 46 VR 676.

^[155] Ibid 687 [45]–[46]. This was cited favourably in Kuyken v Chief Commissioner of Police (2015) 249 IR 327, 355–6 [93] (‘Kuyken’). In Kuyken, it was not suggested that the Victorian Civil and Administrative Tribunal was wrong in failing to require a comparator: at 356 [95].

^[156] (1996) 39 ALD 729.

^[157] Ibid 736.

^[158] Colin Campbell and Dale Smith, ‘Direct Discrimination without a Comparator?: Moving to a Test of Unfavourable Treatment’ (2015) 43(1) Federal Law Review 91, 93, 101–2.

^[159] FWA (n 35) s 351(1).

^[160] Board of Bendigo Regional Institute of Technical and Further Education v Barclay [2012] HCA 32; (2012) 248 CLR 500, 522–3 [57]–[59] (French CJ and Crennan J), 542 [127] (Gummow and Hayne JJ), 546 [146] (Heydon J) (‘Barclay’). See also General Motors Holden Pty Ltd v Bowling (1976) 12 ALR 605, 619 (Mason J).

^[161] Silver v Rogers & Rogers (2012) 224 IR 439, 446 [32] (‘Silver’); Vink v LED Technologies Pty Ltd [2013] FCA 443 (‘Vink (FCA)’); Farah v Ahn [2012] FMCA 44 (‘Farah’).

^[162] Barclay (n 160).

^[163] Ibid 517 [44]–[45] (French CJ and Crennan J). See also at 542 [127] (Gummow and Hayne JJ).

^[164] Ibid 517 [44]–[45] (French CJ and Crennan J) (citations omitted). See also at 542 [127] (Gummow and Hayne JJ).

^[165] Ibid 540–1 [121] (Gummow and Hayne JJ). See also at 541 [126] (Gummow and Hayne JJ).

^[166] Ibid 517 [44]–[45] (French CJ and Crennan J).

^[167] Where, ‘if the decision-maker gives evidence that they did not take adverse action “because” of a prescribed ground, and that evidence is accepted, there will not be a breach. Evidence of surrounding circumstances may be relevant, but only to test the veracity of the evidence of the employer’: Anna Chapman, Kathleen Love and Beth Gaze, ‘The Reverse Onus of Proof Then and Now: The Barclay Case and the History of the Fair Work Act’s Union Victimisation and Freedom of Association’ [2014] UNSWLawJl 18; (2014) 37(2) University of New South Wales Law Journal 471, 488–9.

^[168] Ibid 498.

^[169] Ibid 489.

^[170] Ibid 498.

^[171] Vink (FCA) (n 161) on appeal from Vink v LED Technologies Pty Ltd [2012] FMCA 917 (‘Vink (FMCA)’).

^[172] Vink (FCA) (n 161) [26].

^[173] Ibid [26] quoting Vink (FMCA) (n 171) [48]. See also Vink (FCA) (n 161) [36]–[38], [42]–[44], [46]–[49].

^[174] Vink (FCA) (n 161) [45].

^[175] Ibid.

^[176] Vink (FMCA) (n 171) [35]. See also Farah (n 161) [94].

^[177] Carr (n 79).

^[178] Ibid 312 [13].

^[179] Ibid 315 [29].

^[180] Ibid.

^[181] Travers (n 134).

^[182] Ibid [51]. See also cases decided under the previous s 16 of the Age Discrimination Act, which required that age be the ‘dominant’ reason: Fernandez v University of Technology, Sydney [2015] FCCA 3432; Gardem v Etheridge Shire Council [2013] FCCA 1324; Thompson (n 124) 320 [44]–[45].

^[183] YMCA (n 137) [25]; Sidhu (n 122) [17]; Pignat (n 119) [35]; Plancke (n 122) [28]–[29]; McKeown v Carcione Nominees Pty Ltd [1998] WAEOT, [209], [213] (‘McKeown’).

^[184] Duncan (n 135) [70]–[76].

^[185] North v City of Sydney Council [2001] NSWADT 75, [51] (‘North’); Choong v Bridgestone Australia Ltd [2009] SAEOT 8, [28]–[29]. See also Gunn v Electroboard Pty Ltd [2003] TASADT 7, where the claimant’s evidence was rejected: [50].

^[186] North (n 185) [51]. That said, the shifting of the onus of proof under the FWA (Cth) is not sufficient to address this hurdle: see above n 167 and accompanying text.

^[187] Lightning Bolt (n 76).

^[188] Ibid 77165 [12].

^[189] Ibid 77166 [15]. The finding in this case was likely assisted by the fact some of the employer’s evidence was ‘completely untrue’: at 77166 [17].

^[190] Mooney v Commissioner of Police, New South Wales Police Service (No 2) [2003] NSWADT 107 (‘Mooney’); Arkley (n 119) [15], [41], [45].

^[191] Vye (n 119) [33]. See similarly Arkley (n 119) [15], [41], [45]; Perera (n 122) [89].

^[192] Kennedy (n 119) [72].

^[193] Ibid [119].

^[194] McEvoy (n 76).

^[195] Ibid [15].

^[196] Ibid [29].

^[197] Ibid [30].

^[198] Ibid [48].

^[199] Blackham, ‘A Compromised Balance?’ (n 7).

^[200] Keech v Metropolitan Health Service (WA) [2010] FCA 1332; (2010) 215 FCR 393 (‘Keech’) (see the analysis in Blackham, ‘Defining “Discrimination”’ (n 42) 786–7); Harley v Commonwealth [2011] FMCA 197 (‘Harley’); Commonwealth v Human Rights & Equal Opportunity Commission [1999] FCA 616; (1999) 57 ALD 623, 626 [15] (‘Commonwealth v HREOC (Manning)’).

^[201] Moore v TAB Ltd (2004) EOC 93-348 (‘Moore’).

^[202] Martin v Powerlink Qld [2004] QADT 36; (2005) EOC 93-363 (‘Martin v Powerlink’).

^[203] Setchell v Alkira Centre Box Hill Inc [2009] FMCA 288 (‘Setchell’) (though this might be more closely relevant to the claimant’s disability); Commonwealth v HREOC (Hamilton) (n 83) 395 [62], 396 [67]–[68]; Commonwealth v HREOC (Bradley) (n 83) 233 [29], 234 [32]; Qantas Airways Ltd v Christie (1998) 193 CLR 280 (‘Qantas Airways’); Khan v South Australia [2000] SADC 1, [11] (‘Khan’).

^[204] Khan (n 203) [13].

^[205] Keech (n 200); Commonwealth v HREOC (Manning) (n 200); Moore (n 201); Martin v Powerlink (n 202); Qantas Airways (n 203); ibid. The application of the inherent requirements exception was rejected in Commonwealth v HREOC (Hamilton) (n 83); Commonwealth v HREOC (Bradley) (n 83). Two cases were procedural, meaning the exception was not fatal at that particular stage: Setchell (n 203); Harley (n 200).

^[206] Shirley (n 119) [58]–[59]; Kennedy (n 119) [159]–[160]; Dewan v Main Roads WA [2004] WAEOT 7, [13]–[21], [30] (‘Dewan’).

^[207] Kennedy (n 119) [159]–[160]; Keys (n 119) [22].

^[208] Kennedy (n 119) [159]–[160]; Dewan (n 206) [51], [60]–[63].

^[209] Virgin Blue Airlines (n 98).

^[210] Cavar (n 108).

^[211] Birch v Wesco Electrics (1966) Pty Ltd [2012] FMCA 5; (2012) 257 FLR 237; Lin (n 106).

^[212] Shore v Max Employment Solutions [2016] VCAT 2200, [24] (‘Shore’).

^[213] B v Naval Reserve Cadets (n 103).

^[214] Ibid [26].

^[215] Blackham and Allen (n 29). In the US, see Schuster and Miller (n 22) 74.

^[216] Schuster and Miller (n 22) 74.

^[217] See the detailed discussion in Dominique Allen and Alysia Blackham, ‘Under Wraps: Secrecy, Confidentiality and the Enforcement of Equality Law in Australia and the UK’ (2019) 43(2) Melbourne University Law Review (advance).

^[218] Schuster and Miller (n 22) 74.

^[219] AHRC, National Prevalence Survey (n 1) 19.

^[220] Ibid 53 (Figure 32).

^[221] Nicola Heath, ‘Aged over 60 and Female? Here’s Why You Might Be at Risk of Poverty’, SBS News (online, 5 December 2017) <https://www.sbs.com.au/topics/life/culture/article/2017/11/24/aged-over-60-and-female-heres-why-you-might-be-risk-poverty>.

^[222] Diane Grant, ‘Older Women, Work and the Impact of Discrimination’ in Malcolm Sargeant (ed), Age Discrimination and Diversity: Multiple Discrimination from An Age Perspective (Cambridge University Press, 2011) 41, 43.

^[223] Ibid 62.

^[224] Ibid.

^[225] Source: ABS, Gender Indicators, Australia, Sep 2017 (Catalogue No 4125.0, 19 September 2017).

^[226] One case included both a male and female claimant, meaning these numbers do not add to 108.

^[227] AHRC, National Prevalence Survey (n 1) 18. But cf (n 1) Figure 8.

^[228] Vink (FMCA) (n 171) [35].

^[229] Farah (n 161).

^[230] Ibid [94].

^[231] Silver (n 161) 446 [32].

^[232] Coote v Commissioner of Police, NSW Police [2007] NSWADT 128, [11]. See also Imielska v Director-General, NSW Health [2012] NSWADT 25.

^[233] Mooney (n 190) [50].

^[234] Pomplun (n 120) [7].

^[235] Martin v Donut King (n 82).

^[236] Ibid [23] (citations omitted). Though Ms Martin’s dismissal was still unfair, and her claim was successful on that basis.

^[237] Gateway Investments (n 100).

^[238] Ibid [46].

^[239] Ibid [47].

^[240] See above n 66 and accompanying text.

^[241] Vink (FMCA) (n 171) [52].

^[242] Arkley (n 119).

^[243] Ibid [41].

^[244] McKeown (n 183).

^[245] Ibid [210]–[211].

^[246] Ibid [17], [132], [152]–[153].

^[247] In the US, see similarly Schuster and Miller (n 22) 74.

^[248] Gabryelczyk (n 116).

^[249] Ibid [27].

^[250] Ibid [11]–[20]. ‘Hogtied’ means having limbs tied together, rendering the person immobile.

^[251] Ibid [50]–[51].

^[252] AHRC, Willing to Work (n 4) 342–3.

^[253] See Campbell (n 131); Kate Rattigan, ‘Case Note: Purvis v New South Wales (Department of Education and Training): A Case for Amending the Disability Discrimination Act 1992 (Cth)’ [2004] MelbULawRw 17; (2004) 28(2) Melbourne University Law Review 532; Roberts (n 132).

^[254] Chapman, Love and Gaze (n 167).

^[255] Dominique Allen, ‘Reducing the Burden of Proving Discrimination in Australia’ [2009] SydLawRw 24; (2009) 31(4) Sydney Law Review 579; Jonathan Hunyor, ‘Skin-Deep: Proof and Inferences of Racial Discrimination in Employment’ [2003] SydLawRw 24; (2003) 25(4) Sydney Law Review 535.

^[256] See above nn 53–8 and accompanying text.

^[257] AHRC, Willing to Work (n 4) 336 (Recommendation 50).

^[258] Cf Campbell and Smith, who have argued that the unfavourable treatment test could be over-inclusive, meaning claimants may not encounter difficulties on that basis: Campbell and Smith (n 158) 101–2.

^[259] Warehouse Solutions (n 120) [243]–[244].

^[260] Udugampala (n 119).

^[261] Shore (n 212).

^[262] Naidu (n 121).

^[263] Bloomfield (n 77).

^[264] Indeed, as the FWA example shows, a sympathetic approach is required even where apparently generous legislative provisions are in place (such as the shifting of the onus of proof).

^[265] Chapman, Love and Gaze (n 167).

^[266] [2009] NZSC 78; [2010] 1 NZLR 153.

^[267] Ibid 171 [37] (Elias CJ, Blanchard and Wilson JJ).

^[268] Ibid.

^[269] Ibid 174 [51] (Tipping J).

^[270] Ibid.

^[271] Elizabeth Raper, ‘Pick Me: Choosing the Right Discrimination Law Forum’ (2010) 1 Workplace Review 30.

^[272] See Virgin Blue Airlines (n 98); McIntyre (n 115).

^[273] Lightning Bolt (n 76).