The Supreme Court issued its decision in the Dukes v. Wal-Mart sex discrimination case yesterday, a frustrating ruling that doesn’t challenge the existence of bias, but that exempts the company from accountability. The case highlights the difficulty of addressing discrimination at a time when intentional bias is both illegal and socially unacceptable, and yet obvious gender and racial gaps remain. If much, perhaps even most, discrimination is unintentional on a personal level, what responsibility do employers (or our government, or each of us as individuals) have for addressing its institutional consequences? The court decided 5-4 that up to 1.5 million female employees cannot file suit together as a class. The court’s conservative majority raised questions not just about whether the women were discriminated against through the same mechanisms, but also about the validity of the plaintiffs’ central argument–that the combination of a highly centralized corporate culture and excessive discretion among managers systematically disadvantaged women. Wal-Mart’s numbers are not in question. Women comprise more than 65 percent of hourly employees, but only 34.5 percent of managers. This is significantly different from similar retail chains, in which women hold 56.5 percent of management jobs. It takes women on average 4.38 years to rise to a management post at Wal-Mart, but takes men only 2.86 years. Of 41 Wal-Mart regional vice presidents, only five are women, and only 9.8 percent of Wal-Mart’s district managers are women. Wal-Mart’s internal documents acknowledge that they are far behind the rest of their field. The plaintiffs in the case argued that Wal-Mart’s corporate culture invited managers to act on their own worst instincts. They cited the research of William Bielby, a sociologist who posits that people naturally hold stereotypes and biases, often unconsciously, and we act on them when we have the power to do so and nothing stops us. At Wal-Mart, male managers acted on their bias against equitable promotions and pay because the company’s centralized practices and policies give them huge amounts of discretion in personnel decisions. The discretion itself is the policy, and it stands out in a company whose corporate headquarters micromanages nearly everything, down to choosing the temperature and music in every store. Sophisticated computer systems and dozens of daily reports let headquarters know exactly what is happening on an hourly basis. But Wal-Mart’s top management chose to let store-level managers keep on discriminating. The gateway to promotion at Wal-Mart is its management training program; one has to go through this to become an assistant manager. Wal-Mart offered employees no information about how to get into the program until three months before this case was filed, and there was no system by which an employee could even express interest in it. (If there had been, there would be records of how many male and female employees had tried to become managers and been turned down.) Instead, managers identified potential future colleagues with a tap on the shoulder. Wal-Mart will argue that either coincidence or just a few sexist managers among many account for the glaring gap in the numbers of male and female management recruits, rather than a known set of central policies. Certainly, there has been some blatantly sexist behavior among Wal-Mart managers, such as management meetings in which men called their female colleagues “little Janie Qs.” But mostly, Wal-Mart’s system runs on silence. Silence about what exactly are the criteria for management positions; silence about the additional subjective criteria that individual managers apply for promotion; silence about the actual availability of management positions; silence about how you decide whether to give an employee a raise of 10 or 25 cents per hour. Male managers fill all that silence, the plaintiffs’ lawyers and expert witnesses said, with subjective decisions that are often influenced by stereotypes. Our laws do not require discrimination to be deliberate or even conscious before they require a remedy. So-called “disparate impact” is supposed to be enough for the law to step in. In this case, not only was there obvious disparate impact, but Wal-Mart’s knowledge of the gap supports the charge of knowing disparate treatment as well. Unfortunately, conservative members of the court ruled against the notion that the company is responsible, saying that different plaintiffs were discriminated against in too many different ways for the company to be systematically responsible. In the majority opinion, Justice Antonin Scalia writes that common elements tying all these employment decisions together were “entirely absent” in this case. Understanding Implicit Bias The numbers don’t mean that all male managers at Wal-Mart are intentionally sexist. Their biases are implicit rather than overt, and most of these managers are probably unaware of having them. In 1995, researchers at Harvard and Stanford universities developed the Implicit Association Test (IAT), which measures reaction time to examine subconscious bias. The researchers assert that human beings place information about the world into personal “schemas,” or world views. Schemas allow for implicit stereotyping and perceptions about in-groups (the group you belong to) and out-groups (groups you aren’t a member of), which can translate into behaviors that are discriminatory, or that produce inequitable outcomes. In the project’s online test, 75 percent to 80 percent of self-identified whites and Asians show an implicit racial preference for white relative to black. Everyday people, including the researchers who direct this project, are found to harbor negative associations in relation to various social groups (i.e., implicit biases) even while honestly reporting that they regard themselves as lacking these biases. When implicit bias is combined with the human reliance on first impressions, the result can be devastating. As Malcolm Gladwell noted in the New Yorker, the impression from the handshake that precedes a job interview colors impressions of the interview itself. “The first impression becomes a self-fulfilling prophecy: we hear what we expect to hear,” Gladwell writes. These patterns apply to other forms of discrimination as well. The Restaurant Opportunities Center of N.Y. (ROC-N.Y.) has conducted multiple studies of employment patterns in the nation’s fastest growing private sector industry. Restaurants, especially in the high-end market, are marked by a rigid racial hierarchy, and generally shut out women altogether. In ROC-N.Y.’s first study, interviews with employers revealed their rationale for raced and gendered decisions. They wanted “tall, beautiful” people in the front of the house jobs–these are the workers who make more money and actually speak to diners. For the more dangerous, low-wage jobs at the back of the house, they prefer “hard workers” who are immune to the poor wages and conditions. Simply counting the workers through observation makes it quite obvious that only white men, and the very occasional white woman, meet the beauty criteria in most such workplaces. Immigrants have accents too thick to explain the menu, employers say, while women can’t take the fast pace and informality (i.e. sexual harassment) of restaurant work. In a later study, ROC-N.Y. ran a matched-pair test, sending applicants with the exact same qualifications but of different races and genders to inquire about jobs. Without fail, white men got interviews at double the rate of others. This hierarchy is so ubiquitous as to become invisible unless you pay attention. Again, the bias is unconscious, but the result is the same. That’s why the remedies have to be intentional. Discrimination doesn’t just check itself. In my own reporting, I’ve met one after another immigrant busser who was repeatedly passed up for promotion–or didn’t even know about an opening–only to end up training a white actor or student who was hired instead. The front of the house jobs are never posted, and men of color never got tapped on the shoulder, no matter how well they understood the menu. A manager doesn’t have to hold conscious bias to replicate a workplace’s hierarchies in his own decision making. “If you want to succeed, you pick up what the company wants just by seeing it,” says Saru Jayaraman, the cofounder of ROC-N.Y. “A savvy manager doesn’t require the company to tell him to discriminate, although there’s plenty of that too. He understands its visual codes that are embedded in a segregated workplace.” Americans will be tempted to take this decision as proof that Wal-Mart is not guilty of gender discrimination, and employers will take heart from the Supreme Court turning a blind eye. If some bad managers make sexist decisions, companies will say, that can’t be helped; gender stereotyping is an intractable problem. But the real lesson is that Americans can’t rely on the courts alone to check all forms of bias, especially the kinds that don’t require explicit direction. Wal-Mart and other corporations need to hear from everyone–consumers, workers, and other employers who are building equitable workplaces. The message we send has to go beyond following the letter of the law to challenge the social norms that keep qualified people from getting the jobs they deserve.