Favoritism: Playing Out at an Office Near You

Here’s a riddle: A company’s workforce is composed of 80% white men. Is there discrimination going on? Or is it favoritism?

That’s the question posed in a new paper by economists at Maastricht University and the University of Texas, who found that favoritism is a far more powerful force than outright bias.

Their conclusion might help explain the lack of diversity in the skyscrapers and industrial parks of corporate America, and assist companies and regulatory agencies in coming up with effective plans to create more heterogeneous workplaces.

“It’s not about reducing hatred but reducing excess liking, and these are not the same things,” said Dan Hamermesh, an economist at the University of Texas and one of the authors of the paper.

The paper gets deep into the dynamics of what we normally call discrimination, separating out hostility against an out-group (a group to which, say, a hiring manager does not belong) from preference for people similar to oneself. The authors refer to these attitudes as exophobia and endophilia.

The distinction is a subtle one – at first glance, they seem like two sides of the same coin – but the authors point out that they can produce quite different real-world outcomes.

For example, if a white boss is responsible for distributing $300 in bonus money to three workers, one white, one black and one Hispanic, each worker should get $100 presuming all other variables are equal. But an endophilic supervisor might give the white worker $110, leaving the other two $190, or $95 each. Meanwhile, an exophobic boss who is biased against African-Americans might give the black worker $90, leaving $210 for the others, or $105 each. “In both cases the difference in outcomes between the white and the black worker is $15, but white-Hispanic relative wages differ under each alternative,” the authors write.

In an experiment conducted at the School of Business and Economics at Maastricht University, in the Netherlands, the researchers randomly revealed or concealed the names on students’ final exams, allowing some graders to infer the nationality and gender of the students. Most of the individuals were German or Dutch, and their names generally reflected their origins. Researchers also had access to the graders’ nationality and gender.

In total, 1,495 exams scored by 42 graders were examined.

Among those graders who’d seen the names, the authors found “substantial” favoritism by nationality, but no evidence of discrimination. A student who matched the grader’s nationality received a higher score if her name was visible, equivalent to moving from the median score to the 57th percentile. But in unmatched pairs, where discrimination would presumably be evident, grades were about the same whether names were revealed or not.

On average, gender had no effect on scores.

The authors point out that while discrimination still exists in some cases, favoritism is more pervasive, and suggest that anti-discrimination policies or workplace-diversity programs, which tend to focus on revealing and fighting biases, should instead be oriented toward helping workers and managers understand their unconscious preferences.

“We’re so hung up on accusing people of … discriminating against others, but maybe it’s not the most efficient way to ameliorate the problem,” said Hamermesh.