Affirmative action beneficial, but also risky

Hey there, time traveller!
This article was published 22/10/2016 (1063 days ago), so information in it may no longer be current.

Affirmative action to revise outcomes is a bad idea. There… I have said the unthinkable. Now, before my friends and colleagues strike me off their invitation lists and my inbox groans under the weight of inbound indignation, I need to explain myself.

Let’s start at where the idea of affirmative action started. John F. Kennedy first referred to "affirmative action" in Executive Order 10925, which called for the employment of people "without regard to their race, creed, colour or national origin." Lyndon Johnson followed up with an even stronger statement as part of his "war on poverty."

The result of the ideals espoused by Kennedy and Johnson was to create a host of preferential hiring policies in government. Indeed, very quickly in the ’70s, the majority of black university graduates became employed in government. Large businesses climbed on board, as well. Black incomes shot up.

Throughout the economy, however, this success was not repeated. In small and medium-sized companies, it was and remains business as usual, since affirmative-action legislation often excludes them. Certainly black- and female-owned businesses are on the rise, but in the most dynamic part of the economy — high tech — the game is still played by white guys. When Steve Jobs and Steve Wozniak wanted to commercialize their ideas, they did not call in their black friends or their girlfriends (if they had any). They called on their network of white friends to offer pathways to finance and business connections.

The real "work" in an economy does not occur in the bureaucracies of government and its rules to make outcomes fairer. Rather, it is in the myriad of social and cultural practices that confer advantages in a multitude of ways, too. This is why parents scrimp to send their children to private school.

Affirmative-action programs would appear well-suited to reduce the economic costs of discrimination, but they can morph into a host of practices that transfer unfairness among disadvantaged groups like cooks handling a hot potato. University admissions has one of the most common approaches to affirmative action that reserve places for high achievers within the designated class.

In the United States, such quota systems have increased the numbers of black people attending universities, but the cost has been to deny entry to the children of Asian immigrants who tend to qualify for university in large numbers. Statistically, this was inevitable. If the total number of places at a university is fixed, reserving some for high achievers of a designated group will exclude the high achievers of other non-designated groups, some of which will have better scores than those in the designated group. Aside from unfairness, such decisions are economically inefficient.

Affirmative-action programs that designate employment positions or university slots are intended to attack the outcome of a long process of discrimination on the basis of any attribute not related to work or academic productivity.

George Stigler, the Nobel laureate from the University of Chicago, applied economic theory to clarify the economic cost of discrimination. When an employer chooses not to hire, or to underpay or award slow promotion to a particular group, Stigler argued not only the disadvantaged group feels the impacts, but when that disadvantaged group grows in proportion, such as is the case with women, we all bear the cost of discrimination as we fail to engage the best and the brightest.

Let’s consider something closer to home. The percentage of females attending the University of Manitoba has steadily risen over the last 50 years, starting from 15 per cent in 1960 to currently 54 per cent, a number that has held steady for the last decade.

Now consider the percentage of women in undergraduate nursing, the quintessential "female" occupation, is 86 per cent; some may be surprised 14 per cent of nursing undergrads are men, but this number has not changed much in 15 years. In engineering, traditionally a "male" occupation, 25 per cent of undergrads are women, a number that has also remained unchanged for 15 years. If we were to dial back to 1960, nursing and engineering would be overwhelmingly female and male, respectively.

But is a 50-50 gender parity the ideal? If, in the name of affirmative action, we were to mandate that these faculties maintain a 50-50 gender on admission, we would exclude better-qualified women from nursing and better-qualified men from engineering.

Designating an outcome is rather like the International Olympic Committee deciding we need diversity in the 100-metre sprint. It could start me on the 75-metre line and make Usain Bolt start at zero. It would be an exciting photo finish, watched only by my family, and completely destroy the Olympics.

Affirmative action is vital to a just society and efficient economy, but only if it means correcting the earliest disadvantages that create inequality of opportunity. Affirmative action that mandates end outcomes, such as designating spots at university, corporate board positions, or the roster of Supreme Court justices, creates its own unfairness and reduces our collective economic well-being.

Gregory Mason is an associate professor of economics at the University of Manitoba and a senior consultant at PRA Inc. His views are his own.