Policies in Practice

Abstract

Bill Bielby reflects on his role as an expert witness for women seeking class action status for a discrimination suit against Wal-Mart. He argues that sociologists must take the stand to point out the difference between policies and practices. Without expert testimony, courts can too easily believe that discrimination ends when well-intentioned policies are put in place.

In oral arguments in March 2011, Chief Justice John Roberts posed a hypothetical question: “What if you had a situation where you had a company with a very clear policy in favor of equal treatment of men and women… Yet you still have the same subjective delegation system. Could you have a class of women who were harmed by this subjective policy, even though it was clear that the policy of the corporation favored equal employment opportunity?”

One thing organizational sociologists know—at least, those of us who study workplace inequality—is that the answer is “Yes, and in circumstances that have been found in many large organizations.” Policies on the books don’t guarantee anything.

Without our evidence, courts can conclude that the existence of well-intentioned anti-discrimination policies is equivalent to the practice of workplace parity.

Here’s the back-story and why I think this basic sociological point is worth thinking over. Eleven years ago, Betty Dukes and five other women went to Federal Court. They sought class action status for their lawsuit alleging that Wal-Mart, where they’d all worked, discriminated against women in pay and promotion. At Wal-Mart, the plaintiffs argued, women made up about two-thirds of the hourly workforce, but less than a third of salaried managers. In this highly sex-segregated work setting, a “tap on the shoulder” system of promotion permitted local managers to make staffing decisions with hardly any company oversight. The subjective system created pay bias and severely constrained women’s opportunities at Wal-Mart.

It took four years, but a federal judge in San Francisco agreed to let these six named plaintiffs represent the claims of up to 1.6 million other women, and his decision was upheld in two rounds of appeals. But when the case got to the U.S. Supreme Court, conservative judges like Roberts were skeptical about the group’s claims.

I have served as an expert witness in this case testifying on behalf of the plaintiffs, and the experience has reminded me once again of how difficult it can be for lawyers, judges, and the media to understand and accept sociological knowledge.

In the media, this issue has typically been framed as one of “unconscious” or “hidden” bias. In fact, a few years ago, Fortune described Dukes v. Wal-Mart as “the war over unconscious bias,” and Business Week summed up expert testimony with the glib line “white men can’t help it.” Days before oral arguments were to begin, in fact, my hometown newspaper referenced my participation as an expert witness in the trial, writing “a University of Illinois professor’s theory that white men have unconscious bias against women and minorities is likely to be center stage.” But I have no such theory, and the answer to Justice Roberts’s question has little to do with what’s in people’s heads. Instead, it’s about whether a company adopts practices that have been shown to create and sustain bias or whether they implement those that have been shown to minimize bias.

Almost all large companies have written policies mandating equal treatment of men and women. Some even hold their managers accountable for following guidelines that minimize how gender (and race) stereotypes can influence decisions about hiring, assignments, and promotion, and they systematically monitor those guidelines’ effectiveness. Others, though, have written policies that end up disconnected from actual practices, let alone the organizational units charged with addressing diversity and equal opportunity. When sociologists like myself are asked to testify in class action Title VII discrimination cases like Dukes v. Wal-Mart, this is often what we’re being asked to do: determine whether policy and practice amount to nothing more than symbolic management.

While I faced some harsh (or worse: dismissive) responses to my testimony, I’m confident that sociologists must continue to present our data and analysis in cases like these. Without our evidence, gathered with the conceptual and methodological tools of organizational sociology, courts can conclude that the existence of well-intentioned anti-discrimination policies is equivalent to the practice of workplace parity. Workers and managers understand that management handbooks are just paper; the policies that employers enforce are the ones that they take seriously.

One thing I know is that organizational policies are not the same as organizational practices. Another thing I know is that ending discrimination requires conscious effort, not empty promises.