How to design anti-discrimination policies that actually work

Editor’s note: Twelve months on from the killing of George Floyd, The Economist is publishing a series of articles, films, podcasts, data visualisations and guest contributions on the theme of race in America. To see them visit our hub

IN 1969 JULIUS LESTER published a book about the black-power movement called, “Look Out, Whitey! Black Power’s Gon’ Get Your Mama!” Since then, approaches to managing friction between different racial groups have become less confrontational and more professional. For many American institutions this takes the form of diversity training.

Like most city forces, the Minneapolis police department puts its officers through diversity training. Tanya Gladney, an ex-cop who has trained officers in nearby St Paul, says such training became standard in police departments 15 or 20 years ago. It has not stopped forces from shooting unarmed citizens, a disproportionate share of them black men. The curriculum has been tweaked and extended to incorporate the latest research into “unconscious bias”. Since then, in Minneapolis alone, Jamar Clark was shot dead by a white officer while pinned on the pavement in 2015; Philando Castile was shot dead by a Latino officer while inside his own car in 2016; and Justine Damond, an Australian woman who called the police about a possible assault, was shot dead by the Somali-American officer who arrived to investigate in 2017.

Diversity training for police officers has been updated again. The latest curriculum was developed in New Orleans, home to a police department that became notorious after Hurricane Katrina for its violence. The training, which is called EPIC (Ethical Policing is Courageous), emphasises the responsibility of officers who stand by while colleagues use excessive force. What started in New Orleans has been rolled out across the country with the help of Georgetown University, under the brand name ABLE (Active Bystandership for Law Enforcement). Focusing on a violent cop’s silent buddies, says Ms Gladney, aims to change “what we mean by a bad apple”. Perhaps it will work. Yet in constantly tweaking their diversity training in the hope of finding something that does, police departments struggle with a higher-stakes version of a problem confronting many American institutions.

When an organisation faces internal claims of racism, or decides to change hiring or promotion in response to external prods, the impulse is strong to show that this is a priority. Over the past year, many more bodies have offered or mandated training, backed up with reinforcement sessions. Lots have hired chief diversity officers. Russell Reynolds, a headhunter, estimated before George Floyd’s death that half of Fortune 500 companies had such a post, and says demand has since leapt. But average tenure in such positions is short, at about three years. And the people who take on these jobs often feel that they are being asked to do something impossible.

That sense of impotence is reflected in academic studies. Frank Dobbin, of Harvard, and Alexandra Kalev, of Tel Aviv University, who reviewed studies of the effectiveness of diversity training, found that by making staff think more about racial attitudes, companies may actually reinforce negative stereotypes. Diversity training can also be divisive. White employees respond better to messages that are race-neutral, something Martin Luther King understood when he talked about judging people “not by the colour of their skin but by the content of their character”. For many non-whites, that approach is off-putting and suggests a desire to skirt around the problem, which is why, despite its masterly inoffensiveness, this King quote has often been controversial.

It is a fair bet that in 2021 most Americans do not wish to indulge in racist stereotyping. There are powerful internal motives (people like to believe they are acting justly) and external ones (negative consequences for careers) to avoid doing so. And yet it happens all the time. Jennifer Eberhardt, a black social psychologist at Stanford, tells of boarding a plane with her (black) son when he was five. On spotting the only black adult male on the aircraft, Ms Eberhardt’s son said “that man looks like Daddy!” And then he added, “I hope he doesn’t rob the plane!” As Ms Eberhardt talked to her son about negative stereotypes, her academic work on bias and her unpaid labour as a parent briefly merged.

So-called unconscious bias may show up in people who think they are committed opponents of racism. It is controversial because some see it as an excuse (it was my unconscious that did it, not myself) while others see it as a sin that must be confessed before redemption is possible. Yet an awareness of implicit bias is useful in helping people to reflect about how they think, even if training them to be aware of it does not eliminate racism, whether in police departments or in Wall Street banks.

Conscious unconscious

Some evidence comes from policing. In the shooter test, subjects are shown images of people in a short space of time. Some are armed and some not. Some are black and some not. The person taking the test has to judge in a split-second whether the use of force would be justified. Katherine Spencer, Amanda Charbonneau and Jack Glaser, of the University of California, Berkeley, summarise their findings. “People do not want to shoot unarmed suspects at all, let alone in a racially discriminatory way, and most would be particularly motivated to avoid exhibiting this kind of discrimination. Nevertheless, most do.”

A cocktail of ancestral bias, media stories and a hangover after the crime wave from the 1960s to the 1990s, all mixed in unknown proportions, may account for this. But bias is malleable, meaning it can be reduced. Ms Eberhardt’s research suggests that slowing people down, giving them time to think what they are doing, can help. That fits with the shooter-test finding that people are more likely to exhibit stronger shooter bias if they have first been set a hard set of anagrams to solve and so are tired of thinking. Ms Eberhardt has worked with Nextdoor.com, a social network that had an epidemic of users reporting sightings of suspicious-looking black men, who on closer inspection turned out to be just black men. Getting users to slow down and be more specific about what they saw reduced the problem. She has also worked with Oakland’s police department, which found that when officers first had to explain why they were stopping people in the street, they stopped far fewer black men.

Mr Dobbin and Ms Kalev have recommendations for what works instead. It includes recruitment programmes to identify talented women and minorities, mentoring programmes, diversity task-forces and anything that puts “existing higher-ups in touch with people from different race/ethnic/gender groups who hope to move up”. Such initiatives nominally help minorities. But their real beneficiaries may be managers. Someone whose closest encounters with black males involve watching American football on TV may have different mental shortcuts to someone whose African-American colleague helps him untangle his budget.

When those suspicious of each other are brought together, hostile feelings can fade. That can lessen the sense of outrage at whatever divided them in the first place. But it may also dilute the impulse usually required to make further progress. Sociologists who study interactions between groups talk of “the paradox of harmony”. As a commitment to diversity has become core to capitalist America, activists and academics who work on race find that they too are struggling with the paradox of harmony.■

Full contents of this special report

Race in America: What it means to be an American
African-Americans: The evidence of things not seen
White Americans: The souls of white folk
* Anti-discrimination policies: Black power’s not gonna get your mama
Racial categories: The all-American skin game
Reparations: The freedman’s bureau
The future: The price of the ticket