Damaging Stereotypes: the Return of ‘Hoovering as a Hobby’

Damaging Stereotypes:
the Return of ‘Hoovering
as a Hobby’

Reg Graycar*


More than a quarter of a century ago, I first discovered that cases about personal injury damages assessment, a seemingly gender neutral area of law, graphically illustrate how what appear to be neutral or ‘black letter’ rules and legal principles can in fact operate in ways that significantly disadvantage women (Graycar 1985a; 1985b). Throughout the late 1980s and 1990s and into this century, I have explored the gendered nature of damages assessment in some detail (see Graycar 1992; 1993; 1995; 1997; 2002; 2003). Examining how those damages are calculated, and the many ways in which they tend to be reduced in cases involving women, casts light on some of the ways that unstated assumptions about gender, and about the differing roles of women and men, affect legal decision making more broadly.

This chapter looks back at that work and asks whether some of that critique has had the effect of ameliorating any of that identified gendered disadvantage. However, before that question can be addressed, it is necessary to explain the gendered nature of damages assessment in somewhat more detail.

There are three main ways in which women can be disadvantaged in the context of damages assessments, and I consider each of these in this chapter. First, where women are injured in accidents, their damages for loss of earning capacity (the loss of capacity to do paid work that is brought about by the accident) tend to be artificially depressed by gendered assumptions about women’s lack of attachment to the paid labour market and the assumption that women’s paid work is secondary to their role as mothers and carers. Secondly, a woman’s loss of capacity to work in the home, doing non-market or caring work, has historically been treated by the law as a loss to someone else, through the action for loss of consortium. Some jurisdictions have either abolished the action for loss of consortium, or extended it to women for loss of their husband’s consortium, but for reasons explained below, neither approach effectively addresses the gendered assumptions about the nature of the loss. Even in those cases where such a loss is treated as a woman’s own loss, rather than a loss to someone else, as the discussion below will show, it tends to be treated as a loss of amenity – a non-economic loss. This characterisation is significant because damages for non-economic loss have increasingly been targeted by legislatures for either reduction or abolition altogether.

Finally, assumptions about gender also affect the assessment of damages for the costs of caring for accident victims. Many accident victims need considerable amounts of post accident care, care that is often provided by close family members. But that formulation of the issue is, like many such propositions, deceptively gender-neutral: most of the carers of accident victims are women. Care for children, the aged, the sick, people with disabilities, and people otherwise unable to look after themselves, is considered quintessentially women’s work – even where, as will be illustrated, it is done by men – and valued (or perhaps, more accurately, devalued) accordingly.

The examples given in this chapter are mostly, though not exclusively, from Australia. Many of them are not particularly recent. But their continued relevance lies in the fact that an area of law generally seen as somewhat prosaic and certainly not usually seen as raising gender issues (in the manner of, say, sexual assault law, or family law) is, when closely examined, replete with unstated gendered assumptions. The purpose, therefore, of unmasking the ‘hidden gender’ of damages assessment is to remind us that any area of law is equally susceptible to a gendered analysis (compare Graycar and Morgan 2002).

Damages for loss of earning capacity: loss of
capacity to do paid work

When women are injured in an accident, assessment of their damages can be negatively affected by stereotypes and assumptions. Sometimes this happens explicitly, for example by the use of depressed ‘female’ wage figures or ‘female’ work life tables for women.1 Equally importantly, though less overtly, sometimes this happens implicitly through the use of stereotypes and gendered assumptions about women’s lack of attachment to the paid labour market. Each of these will be considered in turn.

The use of gendered (or racialised) actuarial data to
calculate loss of future earning capacity

There is a considerable literature in North America dealing with the use of explicitly gendered and sometimes race-specific actuarial tables in the context of damages assessment.2 This trend of using gendered wage tables does not seem to have captured widespread currency in Australia, although it is certainly not wholly absent from damages assessments. For example, in one case in NSW, the plaintiff was a young woman and the Court of Appeal had to choose between using average female earnings to estimate her loss of future earning capacity, or using median female earnings (Rosniak v Government Insurance Office [1997]). Notably, no question was raised about whether it was appropriate to use female wage rates, which have been and continue to be consistently lower than men’s.3

By contrast, the use of such data has been commented upon critically by courts in both the United States and Canada. One of the earliest such discussions was in a 1991 decision from British Columbia called Tucker v Asleson [1991]. There the female plaintiff4 argued that her loss of future earning capacity should be assessed on the basis of tables of average earnings for a university educated man (an amount of $947,000) while the defendant argued that female tables should be used (because of the history of sex discrimination in employment, this would have amounted to a mere $302,000). The trial judge accepted the plaintiff’s argument about the use of male wage figures, a decision that was not disturbed by the majority of the British Columbia Court of Appeal (Tucker v Asleson [1993]).

In 2005, the Ontario Court of Appeal affirmed a decision of the Ontario Superior Court of Justice which had decided to reject the use of female wage data in assessing damages for a 17-year-old girl catastrophically injured in an accident, who, at the time of her injury, had not yet completed high school (Walker v Ritchie [2005]).5 Both the trial court and the Court of Appeal reviewed the Canadian case law on this issue, starting with Tucker v Asleson. The trial judge had concluded:

In my view, the use of statistical figures which reflect the entire population, without division as to gender … avoids the problem of having two separate tables and then having to choose between them and apply what seem like appropriate adjustments. The choice of the general average figures for university graduates seems especially appropriate here where the court is attempting to make a forecast stretching many years into the future, and where a couple of the suggested future professions, barring the accident, are teaching in which pay equity has been achieved, and human kinetics, where a good deal of the employment is with government or government supported institutions, where pay equity is mandated.

(Walker v Ritchie [2004]: [135])

The Ontario Court of Appeal endorsed this approach, noting that while damages assessments must be based on the particular evidence in the individual case, a ‘court must be equally cognizant of the fact that gender-based earnings statistics are grounded in retrospective historical data that may no longer accurately project the income a person would achieve in the future’ (Walker v Ritchie [2004]: [45]).

In their recently published treatise, The Measure of Injury: Race, Gender, and Tort Law, Martha Chamallas and Jennifer Wriggins provide an overview of the status of gendered and racialised wage tables in assessing personal injury damages in the United States (Chamallas and Wriggins 2010: 158–70). They refer to a 2008 judgment where it was held that the use of raced life expectancy data to calculate personal injury damages in tort cases violates the equal protection and due process guarantees enshrined by the US Constitution (Chamallas and Wriggins 2010: 156; see also Wriggins 2008: 53–7). The court was hearing a case involving an African American victim of the 2003 Staten Island Ferry crash, and held that ‘by allowing the use of “race”-based life expectancy tables, which are based on historical data, courts are essentially reinforcing the underlying social inequalities of our society rather than describing a significant biological difference’ (McMillan v City of New York [2008]).

While the focus of this decision was raced life expectancy figures, Weinstein J also commented that ‘Courts are increasingly troubled by “race”- and gender-based figures for calculating loss of future income’ (253) before going on to review some of the cases that had applied raced earnings related data, to the detriment of plaintiffs. Those he considered included Wheeler Tarpeh-Doe v United States [1991].6 Although it would perhaps be premature to see the decision in McMillan as having the effect of reversing the trend of using raced or gendered wage tables, the focus on the dubious constitutionality of that practice, at least in relation to life expectancy, does suggest that it might soon come to be viewed as inappropriate (Chamallas and Wriggins 2010: 166).7

Chamallas and Wriggins also refer to the approach taken by Kenneth Feinberg, the Special Master of the federal September 11 Victim Compensation Fund, who ‘made a choice to ignore race and to reject the use of gender-based statistics that would have lowered awards for families of female victims’ on the basis of considerations of public policy and equity (Final Rule 2002; Chamallas 2003). They claim (perhaps optimistically) that what is most significant about these two recent US examples is:

that, after years of neglecting the issue, some courts are finally expressing doubts about the legality and fairness of gender- and raced-based assessments and are reaching to reform damage calculations in a manner consistent with constitutional principles and civil rights norms.

(Chamallas and Wriggins 2010: 166)

But while it is important, at least symbolically, to confront the inherent bias that flows from using gendered or raced wage tables, the decision as to which table to use may not in fact be anywhere near as important to the final outcome of a case as a trial judge’s choice of what figure to use as the appropriate reduction for vicissitudes (the issue of ‘vicissitudes’ is discussed below, in the context of ‘implicit’ rather than explicit ways in which damages are reduced). In Tucker v Asleson [1991] while the judge used male rather than female figures, he then applied a 60 to 65 per cent reduction for vicissitudes, leading to a much smaller award. So the result is a somewhat hollow victory for the plaintiff. What this demonstrates clearly is that these explicitly gendered rationales for reduction of damages are perhaps not so separate as they may superficially appear from the less explicit uses of gender stereotypes. The same factors that lead to women’s average earnings being significantly less than those of men (such as the unequal distribution of work in the home and discriminatory assumptions about women’s lack of attachment to paid work) are relied on by courts to reduce damages for vicissitudes. Moreover, the use of gendered wage tables and a reduction for vicissitudes would constitute double discounting. On this basis, I would endorse the cautionary view of Supreme Court of Canada Justice Dickson in Andrews v Grand and Toy Alberta [1978]:

The apparent reliability of assessments provided by modern actuarial practice is largely illusionary for actuarial science deals with probabilities, not actualities…. [A]ctuarial evidence speaks in terms of group experience. It cannot, and does not purport to, speak as to the individual sufferer.


Assumptions about women’s lack of attachment
to the paid labour market

While gendered wage data may not have generated the same debates in Australia as in North America, the case law in this country is replete with examples of the implicit rationales – the ‘damaging stereotypes’ – that are used to reduce awards for loss of earning capacity to women accident victims (compare Graycar 1995). A case that reached Australia’s High Court provides a clear example (Wynn v NSW Insurance Ministerial Corporation [1995a]). Maree Wynn was injured in a motor vehicle accident when she was 30 and a senior employee of American Express. She had been promoted several times, had a number of staff responsible to her, and worked long hours, often working at home until 1 or 2 am. The extensive computer work aggravated her whiplash injury, and she was forced to resign at age 32. The work she undertook subsequently was far less remunerative and only part time. By the time her damages claim came to trial, she had married her long-term partner (with whom she lived while employed by American Express) and had a child.

The main basis of the respondent’s challenge to the trial judge’s award was, as put by Handley JA in the NSW Court of Appeal, that the judge ‘had failed to make proper allowance for vicissitudes. He held that it was not probable that but for injury, the plaintiff would simply have retired to the laudable but limited role of housewife and mother and abandoned her business career’ (Wynn v NSW Insurance Ministerial Corp [1994]: 61,740). The trial judge had assessed damages on the basis that Ms Wynn would have worked for American Express until age 60 and, after considering all the factors raised (such as possible maternity leave etc), reduced her assessed damages by 5 per cent for vicissitudes.

In the NSW Court of Appeal, much was made of the stressful nature of her job and the onerous responsibility it carried, along with the very long working hours. With regard to child care, the Court said: ‘If the plaintiff … continued her demanding business career after marriage, and after the birth of her child or children, she and her husband would necessarily have been faced with the necessity of engaging a full time nanny for the children and substantial household help during the week’ (Wynn v NSW Insurance Ministerial Corporation [1994]: 61,741), and her damages were reduced to take this cost into account.

The Court also expressed a number of reservations about the trial judge’s finding that she was likely to continue working at a senior executive level. First, they did not accept that she would be further promoted because that would have required another overseas posting:

The Court held that the trial judge had erred in allowing only 5 per cent for vicissitudes. Not only was the possibility of ‘burn out’ not taken into account, but the Court did not agree with the trial judge that it was ‘quite probable’ that she would have been further promoted (though there is no mention of any evidence to the contrary that was before the court). After adding ‘a fair allowance’ for domestic help, a figure the Court expressly acknowledged was based on no direct evidence (‘the Court must do the best it can’), the Court of Appeal summed up as follows:

The allowance for vicissitudes … should include two years’ absence from work to have two children (8 per cent of the 23.75 years [the estimated period of working life]) together with an allowance for the prospect that the plaintiff would be unable or unwilling to remain in her job which placed such heavy demands on her time, energy and health and the love and patience of her husband. The plaintiff, of course, could have worked until sixty or later in a less demanding job but would then have earned substantially reduced salary and benefits…. A fair allowance for such vicissitudes in my opinion would be 20 per cent and this with the 8 per cent allowance for having two children gives a total deduction for vicissitudes of 28 per cent which I would adopt.

(Wynn v NSW Insurance Ministerial Corporation [1994]: 61,742)

Applying this reasoning, Ms Wynn’s damages for loss of future earning capacity were reduced from over $700,000 to $411,350. She appealed and during the special leave application (the hearing in which the High Court considers whether to hear the case) one member of the High Court, McHugh J asked: ‘Well, supposing the applicant had been a male, could you imagine a judge making a finding like this’? (Wynn v NSW Insurance Ministerial Corporation [1995b]).8 Or is it more likely that in that event, the court might instead have described the plaintiff as another court described a similar, though not chromosomally challenged, plaintiff, as ‘a young man with bright prospects, who has been deprived of the ability to choose to continue his career’? (Tucker v Westfield Design and Construction Pty Ltd [1993]: [28]). 9

Wynn’s case resonates with many of the cases I came across when researching damages cases in the 1990s:10 what they have in common is a tendency to treat women’s paid work as marginal, as worthy of comment, as requiring an explanation, rather than as something that adult gender neutral people just do. The judgments often provide an explanation for why a woman works, coupled with an underlying assumption that should the particular reason given for her employment disappear she would no longer engage in paid work (see Graycar 1995). For example, a woman works because her husband left her and she’s a single parent (Harper v Bangalow Motors Pty Ltd [1990]: 9); or she might become one (Wallenv Hird [1993]). One woman works to escape her husband who is violent (Stekovic v City Group Pty Ltd [1994]: [84]). Another works because her husband is unemployed and therefore cannot support her and the children (Angelopolous v Rubenhold, [1991]: 8), while yet another wants to help her daughter to attend university (Randall v Dul [1994]). One woman’s religious beliefs were said by a judge to lie behind her view that ‘her role was to provide financial support to her maximum capacity for her husband and children’ (Kelson v Transport Accident Commission [1994]). A young South Australian woman had her damages reduced on appeal because it was held that she was unlikely to take over her mother’s role in the management of a family business as she had three brothers and the business may have to support their families.11 Perhaps the best explanation comes from Lord Denning in 1974:

Many a married woman seeks work. She does so when the children grow up and leave the house. She does it, not solely to earn money, helpful as it is, but to fill her time with useful occupation, rather than sit idly at home waiting for her husband to return. The devil tempts those who have nothing to do.

(Langston v AEUW [1974]: [987])

A common assumption in the Australian case law is that sole parents are more likely than women in two parent households to be in paid work: in fact, the opposite is true, both in Canada and in Australia, according to data from both the Australian Bureau of Statistics and Statistics Canada.12 I mention this to draw attention to a phenomenon that seems common in these cases: that is, basing fact-finding and judicial decision making on completely erroneous assumptions, a theme considered more fully below.

Paid work for women, particularly married women, is often seen to be in direct competition with other aspects of their lives – with other roles they fulfil or are expected to fulfil. In one Australian case, the court decided that a woman would not be successful running her own business because ‘she may have succumbed to competing family demands’ (Becin v GEC Australia and Ors, [1993]). Women’s capacity to bear children is also used, in a number of different (and often contrasting) ways, to disadvantage women. For young women, damages are discounted because they may in the future have time out of the workforce to have children, irrespective of whether they indicate that they did not want to do so, or planned to have no more children (compare Handley JA in Bondin v Lamaro [1994]: [5]; Dunford J in Partridge v GIO[1993]: [8]). In 1996, the British Columbia Court of Appeal reduced a trial judge’s award, deciding that the plaintiff would not have spent her working life at her pre-accident employment: ‘She hopes to raise a family when her spouse is suitably employed’ (Lee v Swan [1996]: [55]). But an older woman might have her damages reduced when she no longer has children to care for (Reece v Reece [1994]: [3]) or because she is considered unemployable after a history of time out of the paid workforce for family responsibilities (Kirby P in NSW Insurance Ministerial Corporation v Rayner [1993]: [6]), or because, in the words of one judge, she ‘may well have taken breaks from her employment, for example, when her children married and had families to visit and to assist them with their children …’ (Tully v G J Coles [1993]).

Just like in Wynn’s case, a court may consider that the workplace might prove too demanding for a woman who could not be expected to keep up such a pace (see Priestley JA in Rasmus v GIO [1992]: [4]) or, that a woman’s husband might not want her to undertake full-time paid work because, as she and he ‘aged and became financially secure, her husband’s attitude might have induced [her] to retire early or to reduce her working hours’ (Park v Hobart Public Hospitals [1991]: [12]). And, while the New South Wales Court of Appeal treated the difficulties that would confront Ms Wynn in travelling overseas to secure her promotion as almost insurmountable, another female plaintiff’s award was reduced since she ‘may well have taken breaks from her employment … during any transfers in his work by her husband’ (Tully v G J Coles [1993]). For yet another woman, damages for future economic loss were reduced because of her husband’s peripatetic employment since, according to the court, ‘there must also be taken into account … the consequences of being married to a serviceman’ (Isabella Smith v Michael Smith [1991]: [69]).

When the High Court reviewed the decision in Wynn, they allowed the appeal in part. The Court decided that a more appropriate reduction for vicissitudes was 12.5 per cent and refused to discount the award to allow for the costs of child care pointing out that such costs may be incurred by men or women whether or not the child’s mother is in the paid workforce. The Court also said that there was ‘nothing in the evidence to suggest that the appellant was any less able than any other career oriented person, whether male or female, to successfully combine a demanding career and family responsibilities’ (Wynn v NSW Ministerial Corporation [1995a]: [494]).

Gendered assumptions about the economic consequences of marriage for women have also historically informed awards of damages to surviving female spouses in wrongful death actions. In De Sales v Ingrilli [2002b], the High Court reconsidered the issue of taking into account the prospects of a widow’s remarriage as a factor in reducing such an award. The case received some media attention as it involved assessing the widow’s prospects of remarriage by reference to her likely ‘attractiveness’ to a new partner. During the special leave hearing, the following exchange ensued between Hayne J and Gaudron J (the latter was at that time the only woman on the High Court):

HAYNE J: But here, the leading judgment in the Full Court, seems to deal with the subject entirely after some introductory remarks by saying, ‘Only for my part, I would think that a woman of the appellant’s age and credentials, a 20 per cent deduction would be appropriate’. What does his Honour mean?

GAUDRON J: What his Honour means is she is relatively good looking.

(De Sales v Ingrilli [2002a])