9. Women and affirmative action in the U.S.:the Johnson case


Another interesting parallel between the American and the European situation can be drawn from the only case that has been so far decided by the U.S. Supreme Court concerning the use of affirmative actions for improving the situation of women. As already mentioned, the instrument of affirmative action was first developed in the United States in order to remedy racial discrimination by providing better opportunities for blacks and other ethnic groups. But when the feminist movement arose in American society during the 1970s, it appeared that this mechanism could be fruitfully used to enhance the position of women and their chances (1). The extension of its use to this new sector was not uncontroversial, and it has given rise to a certain amount of litigation. The first, and hitherto only, case to come before the Supreme Court dates to 1986: in Johnson the Court had to decide on the lawfulness of a voluntarily adopted affirmative action plan for hiring and promoting minorities and women (2). In order to attain a work force whose composition would have reflected the proportion of minorities and women in the area's labour force, Santa Clara County Transportation Agency established yearly short-range goals in hiring and promoting employees, in which sex could be considered a relevant factor. The category of road dispatcher comprised several hundred employees, but no women; therefore the yearly plan reserved six percent of promotions for women. When a vacancy was announced, several people applied. The board decided to promote a female applicant even though her score was slightly lower than that of a male applicant, expressly stating that sex was one among several criteria to have been taken into account. On review, the Court decided that the selection was lawful on several grounds: first, the plan took a moderate, flexible, case-by-case approach in order to improve women's representation in the work force; secondly, the plan did not fix rigid quotas reserved for women, instead it set goals that should be achieved with flexibility, and therefore the encroachment on the legitimate expectations of other employees was limited and acceptable (3). It is important to note that in deciding this case the Supreme Court relied on a precedent, Weber, where the standards for judging the admissibility of affirmative action plans was the existence of a conspicuous imbalance in traditionally segregated areas (4). Other important decisions by the Court, starting with Justice Powell's opinion in Bakke, make use of a much stricter standard, the so-called "strict scrutiny test", under which affirmative actions are justified only if they redress an actual prior discriminatory practice in favour of people actually discriminated against (i.e. if they are strictly compensatory), if they correspond to a compelling public interest, and if the remedy is narrowly tailored to furtherance of that interest. As we shall see, in recent times the Court has moved back to the use of a strict scrutiny standard, rendering proof of a legitimate use of affirmative action much harder to provide.

If we compare the Johnson decision to that by the European Court of Justice in Kalanke, the most striking difference seems to derive from the standards used to evaluate the lawfulness of affirmative actions. Both systems recognize that affirmative actions constitute a departure from the fundamental principle of formal equality, and therefore both require some further justifying elements. The American system, through the case-law of the Supreme Court, has developed different (and sometimes conflicting) standards, which range from the strict requirement of actual discrimination by the creator of the affirmative plan against the people favoured by it, to a looser criterion whereby statistical evidence is used to prove the existence of conspicuous imbalances among different groups which justify action. Community law has devoted less attention to the characteristics of discrimination that requires the use of affirmative action; this might partly be due to the fact that tackling discrimination against several racial groups, with widely different problems and needs, requires greater sophistication and complexity of analysis than when dealing with just one "minority", as in the case of women. However, this is an ambiguous answer: since women are only improperly called a minority, not just because they form an extremely important part of the population, but even more because they are dispersed throughout all other social groups (economic, social, religious, ethnic, and so on), the use of sophisticated techniques for evaluating the degree of discrimination, and consequently the type of action required is as useful as it is in the American legal system.

In fact, Community law has developed similar techniques in a crucial field of discrimination law: indirect discrimination. Mainly through developments of the case-law of the ECJ, Community law has developed a body of rules that deals with behaviours that are formally neutral, but have factual consequences that are discriminatory (a good example is the worse economic treatment of part-time workers, which amounts to indirect discrimination if the majority of those workers happen to be women). To do so, the Court has relied on several kinds of evidence, among which statistical evidence is crucial, using it flexibly in order to evaluate the economic and social dimension of the phenomenon. Oddly, these methods have not been applied in the field of affirmative action. If we look again at the Kalanke decision, we note that even in the extensive considerations by Advocate General Tesauro little attention is paid to this issue: he started by distinguishing among different ways in which affirmative actions can tackle the problem of discrimination (par. 9), but he failed to analyse the way in which the German statute defines the standards for taking action. He could have pointed out that the criterion defining under-representation whenever women constitute less than half the work force is unfair, and that it should be compared against the statistical data on the working population of the area, or he could have raised similar objections. Instead, the Advocate General chose to focus only on the purpose of the affirmative action plan. By classifying it as aimed at equality of results instead of equality of opportunities, he qualified it as unlawful. But, as we have seen, this distinction seems to be unsound, whereas objections against the characteristics of the plan, and the proportionality of the means used, would probably have been more firmly grounded. The outcome is that whereas the Supreme Court in Johnson considered as lawful a decision that promoted a woman that was less qualified than her male counterpart, since preference due to sex was counterbalanced by several other elements (like flexibility, temporal limits, and so on), the European Court of Justice has rejected as unlawful the choice of an equally qualified woman because this amounted to imposing equal results and infringing the principle of equal formality.


(1) S. N. Mills, Debating Affirmative Action: Race, Gender, Ethnicity and the politics of inclusion, (New York, 1994); S.D. Clayton, F.J. Crosby, Justice, Gender and Affirmative Action, (Ann Arbor, Mi., 1992).
(2) Johnson v. Transportation Agency, Santa Clara County, (1987) 480 U.S. 616; 107 S. Ct. 1442.
(3) In a strong dissenting opinion to the majority opinion written by Justice Brennan, Justice Scalia points out that women under-representation in certain jobs stems from reasons completely different from these that determine the under-representation of ethnic and racial minorities: while in the latter case it may easily be related to discriminatory practices, in the former social attitudes may be the most important reason, and in this case affirmative action is not justified: "It is absurd to think that the nationwide failure of road maintenance crews, for example, to achieve the Agency's ambition of 36.4% female representation is attributable primarily, even if substantially, to systematic exclusion of women eager to shoulder pick and shovel. It is a "traditionally segregated job category" not in the Weber sense, but in the sense that, because of longstanding social attitudes it has not been regarded by women themselves as desirable work" (p. 688). Although it must be recognized that social practice may sometimes be more pernicious than open discrimination in segregating groups from certain activities, and it therefore forcefully requires contrasting action, there is a strong argument for being suspicious of any kind of paternalistic legal intervention that aims at imposing on people what is best for them, even if they do not want it.
(4) Note that in Johnson the Court fails to distinguish between affirmative actions that favour racial minorities and those that favour women; it is therefore unclear whether all the rules developed in the race discrimination field are applicable by analogy to sex discrimination: see. M. Rosenfeld, Affirmative Action and Justice, cit. note 5, pp. 198 and 204.