The Problem of Privilege Discrimination and The Solution of “Evidence Based Discrimination”
I am a professor at a university in the American South. I am in my sixth year there. In that time, I have served on numerous hiring committees. For readers outside of academia, hiring a new faculty member is an extensive vetting process that can take several months. You are hiring someone who may be there for decades. After being a member of so many committees, I have gained a good understanding of the hiring process.
I am also a black faculty member. I have observed with keen interest the often-ineffective efforts at my university to recruit black faculty. I have been an integral part of the discussions around this issue. These discussions have only increased in recent years. I have become a tenured member of the faculty and expected to participate in university governance. I am also an officer in my university’s black faculty association.
My experiences, observations, and discussions with others about improving diversity in hiring lead me to this conclusion: The anti-bias procedures we have in place that are meant to protect people from employment discrimination skew hiring in favor of privilege. This privilege is primarily white. In the context of diversity hiring, we must recognize this privilege discrimination and replace it with a clearly articulated, evidence-based discrimination in favor of diversity.
Although my focus here will be on academia, the problems and solutions presented can apply to hiring in many other white-collar occupations.
Anti-Bias Procedures in Hiring
My university employs a set of procedures to protect against implicit bias in hiring. I suspect other universities have also used similar procedures. Here is an example:
(1) Advertisements must be made public and listed for a certain amount of time. This is a way to address how most jobs are gained through personal connections.
(2) Evaluators do not know the race, gender, or sexual orientation of the person submitting materials for consideration. This is for the obvious reason that knowing these characteristics will bias the evaluators.
(3) Evaluators are not to view material from a candidate that was not provided in their application packet. This is to prevent someone from searching online and identifying information that would bias the evaluation.
(4) During phone interviews, questions are standardized so that each candidate gets the same series of questions. This is meant to prevent one candidate from getting softball questions and another candidate from being heavily scrutinized.
(5) The measures used to evaluate a candidate must come from the advertisement. This fourth practice is the one that gets the most attention because it is the most consequential for hiring. Candidates are evaluated according to what is called a review matrix. This is a series of measures that a committee member uses to evaluate the suitability of a candidate for the advertised position. The matrix is a way of connecting what is in the advertisement to the actual person.
Consider the advertisement for an Assistant Professor of Business below:
State University invites qualified applicants to join our team as an Assistant Professor of Business. We are looking for candidates who:
- Have an ability to teach Introduction to Business Technology
- Have an interest in interdisciplinary research collaborations with our growing Cybersecurity Department
- The potential for securing external research grants
- History of publishing in research journals
- We are an equal opportunity employer — women, sexual minorities, and racial minorities are encouraged to apply
A review matrix will take each of those bullet points and attempt to quantify them. For example, one can evaluate the ability of a candidate to secure research grants on a scale of “1” = no evidence, “2” — worked alongside others on a grant, “3” — has a record of securing grant funding, and “4” — has an exceptional record of securing grant funding.
The committee will read the potential resume and other materials and score that person.
An example of a review matrix is below, where three candidates have been evaluated by one member of a hiring committee.
Hiring committees are usually composed of five or six people, and scores for potential hires are averaged across the group.
And what about the “women, sexual minorities, and racial minorities are encouraged to apply” bit? Well, since the whole idea is to avoid bias in favor of merit, and someone does not earn their race, sex, or gender we cannot take those factors into account.
According to this imaginary committee member, John Turner is the best candidate. He is a tad bit weaker than Abigail and Denise in teaching. But his grant and research work is more substantial. Abigail’s overall resume is very good, but she is not as worthy as John. Finally, according to this matrix, Denise falls short with research and grant funding and is the least meritorious candidate.
The committee would then invite John Turner for an on-campus interview.
A lot of investment goes into hiring in academia, and I suspect that other places of employment may be more or less intensive. However, they share with the scenario I painted here the underlying logic of installing anti-bias practices that aim to produce the most meritorious candidate.
But there is a problem. People are not hired solely on merit. Because we maintain this fantasy, we end up discriminating in favor of privilege.
The most meritorious people — those who earned their place through hard work, are not always the ones who have the most external achievements.
Consider someone who comes from a wealthy home and has attended elite private schools from pre-K through university. That person will likely have “achieved” quite a bit. They will have participated in many extracurricular activities, can play an instrument, and may have learned a second language. At their well-funded universities, they would have access to many learning and research opportunities, as well as resource-rich social networks.
Now consider the working-class person who went to community college first, then to a local, public university to complete their education. They may have worked several jobs, taken care of family members, sacrificed time with family and friends in study hall to overcome a deficient public high school education. They will have certainly earned the qualification needed to apply for the current job. Given where they started, an argument could be made thay they have earned to a greater extent than the privileged candidate.
But they will not have the achievements of the privileged person. So, the question becomes, who exactly is more meritorious?
When my university uses anti-bias practices, they are assuming they are removing bias to uncover the candidate with the most merit. But they are, in reality, revealing the candidate who has achieved the most by leveraging privilege. This same achievement granting process likely unfolds in other organizations that invest a comparable amount of time in vetting and hiring someone.
This narrative I weave is not merely theoretical. It is an open secret that America is quickly becoming a country of haves and have-nots. As Yale law professor Daniel Markovits wrote in the Atlantic, “Legacy preferences, nepotism, and outright fraud continue to give rich applicants corrupt advantages…On average, children whose parents make more than $200,000 a year score about 250 points higher on the SAT than children whose parents make $40,000 to $60,000”.
This is especially damaging in the context of diversity searches, where most of those searches are focused on hiring black or Hispanic applicants. Those candidates are more often the products of a working-class academic trajectory, which means they do not have the same achievements (on average) as their white counterparts.
I saw this repeatedly on the hiring committees I served. We started many of those searches hoping that a black candidate would rise to the top — at least into the top three or four. But because our searches discriminated in favor of privilege, and black candidates are rarely privileged, hiring committees consistently turned up white candidates.
How Diversity in Hiring Should Actually Work
The point here is not to wholly refute the idea of hiring on merit but to show that anti-bias procedures are still biased. When we use anti-bias models assuming we are erasing bias, we end up discriminating for privilege, which disadvantages people of color, women, and working-class applicants.
What happens, then, is an organization will set out to increase diversity but do a series of hires that are not diverse. They then spout variations of statements like: “There seems to be a pipeline problem” or “We had some good candidates, but not good enough.”
When we think about diversity in hiring, we should first acknowledge that we are always discriminating in hiring. Then we can move towards making localized discriminatory decisions based on the values of the organization and the customers it serves.
This is a middle ground between two untenable decisions — hiring any “different” person simply because they are different, or posting the “women, sexual minorities, and racial minorities are encouraged to apply” and then shrugging when mechanisms of privilege prevent the organization from extending an offer to a minority.
I can use my university and city as an example of how an organization can find this middle ground.
My university is in Norfolk, Virginia. Norfolk has a population of 240,000 and is about 40% black. The city is redolent with African American history. The first black people were brought to America’s shores not far from Norfolk in nearby Jamestown, Virginia. We are a few hours’ drive from Richmond, the capital of the Confederacy. We have a relatively large population of black students for a non-Historically black college at 27%.
These demographic and historical facts set up several reasons why my school should focus on hiring black faculty:
- The tax-paying population of black parents (and their children who may be future students) would probably like to see more black scholars.
- Like many others, our university value community engagement, so a black scholar who is willing to apply their teaching and research to the black community would be highly valued.
- As some research has shown, black students who see themselves reflected in the faculty have better educational outcomes.
Identifying these factors would then justify diversifying our faculty and discriminating in favor of a black candidate. However, this discrimination would be balanced with the prior attempts to identify merit through anti-bias practices. We would not want any black candidate — but one who is qualified.
One way of doing this would be to add a racial variable to the matrix. By law, asking explicitly for a black candidate is illegal. Also, just because someone is black does not mean they will fit the diversity goals we have (many people reject group affiliation). Instead, we identified the abilities and competencies that would best suit the black population on campus and off.
A sample matrix, updated from the previous is below:
The evaluators on the committee would not know it, but Denise is black. In this new matrix, she would benefit. She did not have the privilege of being in an environment that allowed her to publish or work on grants. However, her research is focused on the black experience, and she has demonstrated in graduate school her willingness to work in the black community applying her research. She receives a “4”.
Meanwhile, Abigail is white, but that does not mean she is not interested in working with black students or the black community. She has published some research on racial issues and could, in a pinch, teach a course within the department that might focus on race or diversity. She gets a “2”. John, who was the most qualified person before the new measure, is now the least qualified. He gained no points for reflecting black student and community interest, and he receives a “0”.
So, has John been wronged? He went from the most qualified to the least qualified according to the new matrix. Well, let’s think it through:
- Denise’s hire is a unique circumstance. Not every organization will develop compelling reasons to target a black person. John can likely apply to many other places where there is not racial discrimination but instead privilege discrimination.
- If the university has clearly articulated a need for black faculty, then John is less qualified anyway. Parents and students would prefer to see black faculty on campus, and maintaining a strong core of black scholars focusing on the area would keep the historical connection strong.
- John could argue that even though the university is not discriminating by explicitly stating they want to hire a black person, they discriminate on race in practice by rigging the game — a kind of de facto discrimination instead of de jure discrimination. John would be right. But the alternative would be to discriminate against Denise for her lack of privilege.
This approach is flexible. An organization could just as easily have articulated a need to discriminate in other directions. Imagine you are a corporation that just moved its headquarters to a rural area of the country. Instead of recruiting managers globally, this corporation may decide to discriminate towards people living in the area. The rationale would be that they want to develop strong ties with the community, and hiring local people to their management team may help facilitate those ties. Arguments can be constructed for all populations that have been historically on the outside looking in, including transgender people, differentially-abled, and women.
The example I gave above was not real. But I contend that the approach outlined above — what I will call evidence-based discrimination — is more realistic given the sociological realities of privilege in the United States. It will also assuage many people’s concerns about how helpful diversity hiring is to the organization, its employees, and the individual being hired. I think three things must take place to put this approach to diversity in place.
Act on Diversity
This seems simple, but you would be surprised. Organizations often get by with stating they have a “commitment to diversity.” But this is very much different than saying, “we will have a diverse workforce.” The first statement is stagnant virtue-signaling that helps the organization look good but does nothing for people who have been traditional excluded from those jobs. It is selfish. The second phrase puts the organization in a position where they must do something or have reneged on a promise.
I have found that these throwaway statements are a significant obstacle to moving forward on diversity. I even suggest a move against making such a stagnant statement. Instead of having a “commitment” to diversity and plastering it all over social media and promotional materials, just go ahead and start the process of hiring somebody. Ultimately, it is better, as the young ones say, to “show receipts.”
Articulate the Discrimination
Action requires deciding in what direction you want your diversity hiring to go — in other words, how do you want to discriminate. Just hiring any “different” person makes sense for public relations and is technically creating diversity by bringing different people together in the same space. But that diversity is more meaningful when it is focused and directed towards hiring someone who brings diversity and also helps the organization.
The historical and demographic facts of Norfolk supported hiring more African American faculty for our public university. But if I were thinking about hiring at a public university in North Dakota, the emphasis now shifts to hiring someone of Native descent who is focused on Native issues.
Private organizations may not have a compelling civic reason to diversify their workforce, but they will have a compelling economic interest. Given American society’s culturally diverse makeup, certain products may need to be tailored to certain demographics. Having someone of that demographic on the team can help tap into that group’s interests or concerns. While I am not in private industry, this dynamic plays out repeatedly in social science, with racial, religious, or sexual minorities leading the charge in generating knowledge about those groups.
Businesses will also have a public relations interest. Consider a tech business that relocates from the West coast to a rural area in Appalachia. To become a part of that community — which may help them with future tax breaks — they may want to discriminate in favor of hiring people in that community. Instead of importing a recent computer science graduate from Cal Tech, they may explicitly articulate a need for graduates from the local college.
Thinking in terms of the type of diversity allows the organization to develop a clear rationale and justify to employees and potential hires what you are attempting to accomplish. Also, it prevents what I believe are rather misguided diversity hiring efforts, where organizations will take any random collection of “different” people and see that as diversity.
There is a common counterargument that diversity hires are less qualified and are just token hires for public relations. Clearly articulating why the given type of diversity is necessary goes a long way to addressing this admittedly weak argument. Moreover, if a metric is used that includes the type of variables modeled in the matrix above, that diversity hire is likely more qualified.
Know When You are Diverse Enough
Along with attempting to act on diversity and knowing what type of discrimination is needed, a clearly stated end goal is also needed. The organization, its employees, and the citizens it serves need to know when the goal of diversity has been met.
I think organizations struggle with this because they do not want to look as if they are explicitly hiring a minority — so they use the tired “commitment to diversity” line and only haphazardly if ever, diversify their workforce. I think this is a mistake.
Like any organization, precise goals lead to precise results. Goals can be articulated like:
- Hire two women of recent Hispanic heritage who can help create advertisements for our Hispanic customer base
- Hire six people who identify as queer, with at least two who are transgendered, to work in our non-profit organization dedicated to helping queer youth
One common goal is to aim for proportionality. This is the standard justification of “we have X amount of people as customers, so we want X number of employees.” In the past, I have advocated for proportionality. I have since come to realize that this was a mistake. The problem with proportionality is that it is not tied to the employee’s competencies or the organization’s deliverables. It is not the proportionality or the raw number of black employees that matter, but their likely impact and what that impact will be.
Having these types of precise goals lets you, the organization, and a sometimes-cynical public know when this diversity goal has been reached. Then, another goal can be articulated if need be.
Evidence-Based Discrimination and the Future
At my university, we still struggle with the low number of black faculty. I believe this is because we still cling to the false belief in meritocracy. But if we want to address the lack of minorities in organizations, we need to discard the notion that people are hired purely on merit. Instead, we need a more mature understanding of the hiring process that acknowledges discrimination. We can then channel that discrimination in socially acceptable ways.
My belief is that many white readers of this will recoil from this idea of open discrimination. They will see it, as they have other types of policies that are not oriented towards them, as being racist. I imagine they may see what I am advocating for as reverse discrimination. I am sympathetic to those claims. However, there is never no discrimination. So let’s choose the right one.
Moreover, given the growing income inequality in the United States I suspect that privilege will have an even greater impact on hiring in the decades to come. The same evidence-based discrimination practices that would discriminate in favor of Hispanics, women, transgender, and the differentially-abled, will also be applied to poor or low-income job applicants. In the same way that white women were biggest beneficiaries of affirmative action policies, white folks will be a major if not the major beneficiaries of evidence-based discrimination.