Quotas, it is said, inevitably lower quality. Quotas lead to hiring a person who otherwise wouldn’t get hired. It’s obvious, isn’t it, that if she were good enough, she would get the job on her own? Quotas take situations that are fair — tough, sure, but fair nonetheless — and make them unfair. Surely you don’t think sex makes a difference in how well you do your work!
Those are the attitudes I bump into when I talk about quotas — the most extreme form of intervention in hiring processes. Those attitudes, however, fly in the face of recent research. It turns out that there are good reasons to think quotas actually do the opposite: they raise quality.
The impediments to fair and quality-based hiring processes are many. An increasing literature shows that we have implicit biases which keep us from making objective evaluations of the work done by men and women; we inevitably consider sex when we evaluate individuals. In the essay Quotas, microaggression and meritocracy, I put it like this:
Academics hold tightly to the view that progress in our system is meritocratic. Hiring, decisions about article publication, citation of the work of our peers, the awarding of research funds, raises, promotions and more are determined rationally, we believe, as a result of the objective evaluation of clearly stated requirements for advancement. An increasing body of research, however, makes it clear that equally qualified men and women are viewed differently when hiring, that women have less access than men to positions of prominence in article authorship, that citation patterns reflect the sex of the author, that prestigious funding agencies have systems which set the bar lower for men than for women, and that the CVs of men and women are evaluated differently for promotion (Vernos, 2012; Donald, 2013; European Research Council, 2012; Maliniak, Powers and Walter, 2013; West, Jacquet, King, Correll and Bergstrom, 2013; Ministry of Science and Innovation, 2011; Wenneras and Wold, 1997). (reference list)
Can you really recognize quality?
A newly published article, Elite male faculty in the life sciences employ fewer women, shows that part of the problem with increasing the number of women faculty in the natural sciences is found in the employment patterns in the very most prestigious labs.
This article shows a difference between the behavior of male and female heads of labs. Earlier simulation studies could not distinguish between the behavior of men and women; the new paper is a study of actual data gathered after the fact and as such is messier with more variables that cannot be controlled for. Its results are an invitation to more research: The absence of a difference between men and women when it comes to implicit bias is clearly a topic that we are not yet done studying.
Independent of that, what is clear from any of these studies is our commitment to quality: We intensely want to believe that we are capable of recognizing quality when we see it. But study after study shows us that it’s not so simple. In many situations, we are simply not up to the task. We need to acknowledge this and to imagine processes which in fact do not leave us on our own. We need processes that support our quest for selecting the best. This is part of a conceptualization of human capital and many such processes are available to us.
I’d like to take the most controversial one, however, and put it on the table with what I hope will be a fresh perspective.
How quotas raise quality
Quotas can take the form of insisting that a particular position go to a member of an under-represented group, or that a particular percentage of a group must be made up of members of particular sub-groups.
One common objection to quotas is that they are unfair. But quotas do not get introduced in situations that already are fair. They are a tool to pursue fairness — to correct unfairness.
The other common objection is that quotas necessarily lead to a reduction in quality. If you use a quota, it is said, you will hire someone who isn’t otherwise good enough. If you insist that companies put women on their boards, we used to hear in Norway, you’ll lower the overall quality. That’s the stereotype.
But research suggests that this is exactly wrong. Research shows us that quotas can raise quality. Everyone who is a friend of gender balance work needs to learn these arguments.
When anecdotes become anecdata
The stereotype is a classic case of basing conclusions on something other than research. It bases conclusions on logic — bad logic, it turns out — and stories. These stories — “we had a woman president at our university once, but it didn’t really work out” — get repeated so often.
In fact, they are anecdotes that with sufficient repetition start to get treated as data. We might call them anecdata.
Anecdata are not useful as we try to build our knowledge. Three studies on quotas show something different.
A simulation published in Science in 2012 shows that affirmative action attracts more highly qualified women. The use of affirmative action measures in this study — including, but not limited to quotas — leads to a change in the applicant pool. In fact, the inclusion of some affirmative action measure in the announcement of conditions had the effect of attracting more highly qualified women to the extent that it was almost never necessary to actually use the affirmative action measure to get the desired gender balance.
Up with zippers
In 2013, Sweden’s Research Institute of Industrial Economics published a study called Gender quotas and the crisis of the mediocre man: Theory and evidence from Sweden.
This paper examines the effects of quotas on the lists of political parties. Lists of candidates for elections alternative between men and women, so that there is a “zipper quota” on the group giving 50% men and 50% women.
The authors evaluate group competency and demonstrate that the zipper quota raises the overall competency of the groups and, in particular, it raises the competency of the group of men on the list.
Think about that.
Adding a zipper quota raises the competency of the group overall and of the group of men. What does that mean?
It means that mediocre men are being replaced by more highly qualified women. The overall compentency is raised because more highly qualified people replace lower qualified people.
And for the group of men, lower qualified men are no longer present.
In this case, research shows that a quota on the composition of a group increases the quality of that group. This is a realization of the classic argument that we need to make use of all of societies resources. And it shows that exactly the opposite of the stereotype about quotas is what actually happens.
Watch the European Gender Summit speech related to this blog entry!
Getting the most out of human capital
Now in 2014, a new study called Socially gainful gender quotas argues from a human capital perspective the following:
We study the impact of gender quotas on the acquisition of human capital. In the absence of quotas, women consider their chances of getting top positions to be lower than men’s. The lure of top positions induces even men of relatively low ability to engage in human capital formation, whereas women of relatively high ability do not expect to get top positions and do not therefore engage in human capital formation. Gender quotas discourage men who are less efficient in forming human capital, and encourage women who are more efficient in forming human capital. We provide a condition under which the net result of the institution of gender quotas is an increase in human capital in the economy as a whole.
We know that we stumble in our attempts to select the best, and the obvious response to stumbling is to carry a crutch. We need help. We need interventions. And those can take many forms. But no matter what interventions we consider, it’s important to identify them based on research, and not on anecdotes, and not even on anecdata. Research is the route to knowledge.
And when we look at the research on quotas, it turns out that what we thought we knew — that they necessarily lower quality — is exactly wrong.
Both in terms of attracting stronger applicants and in terms of their impacts on groups, quotas increase quality. Period. That’s what the research tells us. What should we do with this knowledge? Quotas are the most extreme measure. And they carry other challenges — not least of all stigma for the target group. Whatever we may decide to do, however, let’s set the stereotypes aside and see what research shows about the positive effects quotas can bring for everyone.
This blog entry is based on my comments at Gender Summit 4.
Share
12 Comments
6 Trackbacks
- 2 ways quotas for women raise quality: The Video!
- The two body problem – the biggest issue for advancement of women in science and academia? | Cate Macinnis-Ng
- Women top managers are a rare sight in Norway - Curt Rice
- FWSA Blog » 50 shades of sexism in the academy
- Bingo and Beyond | hypatia dot ca
- Where’s the evidence? A little science about bias and gender equality | Curt Rice
Republish
I encourage you to republish this article online and in print, under the following conditions.
- You have to credit the author.
- If you’re republishing online, you must use our page view counter and link to its appearance here (included in the bottom of the HTML code), and include links from the story. In short, this means you should grab the html code below the post and use all of it.
- Unless otherwise noted, all my pieces here have a Creative Commons Attribution licence -- CC BY 4.0 -- and you must follow the (extremely minimal) conditions of that license.
- Keeping all this in mind, please take this work and spread it wherever it suits you to do so!
“But quotas do not get introduced in situations that already are fair. They are a tool to pursue fairness — to correct unfairness.”
This is immensely false. When the government mandates that 40% of the boards of public Norwegian companies must be women, this happens regardless of any demonstrated unfair selection process.
When the education system gives women bonus points to apply for engineering college, this happens without any demonstrated unfairness being present.
Quotas and affirmative action are implemented without regard for specifics, they assume that unfairness is taking place, where none has been proven or indicated.
Simulations (!) may show that you get higher quality, but what do real world results show?
What about the problem with the quotas giving the appearance that women who are given the position wouldn’t otherwise have qualified, cementing (or creating!) sexist views on women in the workplace? Don’t they risk always being seen as having made the cut simply for having the “correct” reproductive organ?
Quotas seem to be a destructive shortcut to rectify problems that may not have actually existed in all the places they are being applied.
Instead of private, random discrimination of women, you get state-sponsored discrimination of men. How is that any better? Isn’t it in fact worse to have the government institutionalize discrimination?
Do quotas actually counter sexism – or merely run around it? If the goal is less sexism, it would seem the goal is not being accomplished.
Thanks for stopping by and leaving these thoughts, George. It won’t surprise you that I disagree with your position, but let me make a few specific comments anyway. The idea of quotas on boards in Norway (introduced by a Minister from Norway’s Conservative Party, Ansgar Gabrielsen) grew out of the observation that women were not being considered for board positions in anywhere the same numbers as men, as under 10% of board seats were filled by women. Now, one might argue that 90/10 was actually the portion of *qualified* men and women for those positions, but that turned out not to be true, at least not by any explicitly describable criteria. That is, there were plenty of women with the same kinds of backgrounds and experiences as the men on boards, they just weren’t being seen in the board-selection processes (which often are quite informal). The law has changed that and everyone in Norway is happy with it. In fact, the current conservative government is considering expanding it.
Regarding the engineering students, there is a massive shortage of engineers in Norway. When nearly half the population is ignoring that as a potential career, there are serious economic consequences, so many measures have been introduced to try to educate more engineers total. One of those (sets of) measures is to try to attract more women. The bonus points send a signal and the percentage of women in those programs is steadily increasing, although there is still significant gender segregation when we look at engineering sub-fields.
I would claim that it is a point about fairness to see qualified candidates in selecting boards and that it is a point of fairness to work to make a broader spectrum of careers appealing to everyone, giving greater choices in life to anyone. So, I stand by my claim in these two alleged counter-examples: in this cases, quotas are introduced in cases to enhance fairness.
Now, you do have an important point where I think you’re right, namely that quotas open the door to women being viewed as having gotten a job when they are underqualified. That is an important issue. But that’s why I wrote this post, i.e. to try to show that the facts don’t support that view.
The goal is less sexism, I agree. It’s a long-term goal and quotas feed it.
Using quotas for recruitment could well be illegal, at least in the UK, depending on the precise way you do it. Positive discrimination is just as illegal as any other form of discrimination, however for the past 3 years or so it has been legal to favour *equally qualified* candidates from under represented groups. IANAL but I am pretty sure that it would also be legal to do a zipper quota on interviews.
One of the best legal approaches for bigger companies has to be to get a recruitment agency to do it for you and treat equality as a performance measure of the agency – not the candidates. Tell the agency that you expect the demographics of the candidates they supply to be suitably diverse, it is totally legal to monitor the quality of your suppliers, and if diversity of good candidates is a quality metric then it is up to the agency to figure out how to send you a range of good candidates.
In my 20 years as an academic I have only positive experiences with the use of quotas (for both women and men) for selection of board members and in employment processes. One of the latest additions in gender balance procedures is so called “search committees”. This is a committee that have responsibility to make sure that qualified persons of both genders are asked to apply for a job, or are considered for a board. This makes sure that women consider jobs, are “seen”, and regards themselves attractive. In addition to quotas, which I regard instrumental in the development of a modern society, the search committees are one of the best organizational means in hope of a more just treatment of highly qualified women.
The search committees are an important example, indeed, and I completely agree. Having said that, I would like to do more with them. In particular, I would like to train both men and women who are on search committees about implicit bias, to heighten their awareness. I see too many cases in which the language used about women is too different from the language used about men, or when seemingly equally qualified people are treated too differently, even by female committee members. So, the balance in the committees is important, but we’re not done yet!
“A system of affirmative action in which employers are instructed to hire a women when a male and female applicant are found to be equally well qualified” is double sexist behavior. First of all you let gender be the final selection criterium, which is just…lame and uninspired problem solving. Second, you implicitly bias (pun intended) yourself to only apply the sexist measure in favour of women. Perhaps consider that in many industries it is actually men who are stigmatized and discriminated. it works both ways, but of course you don’t care about that, do you.
I’m glad that we agree, however, that affirmative action is one approach to solving the problem, even if we would characterize it differently.
I’d be curious to know which industries you have in mind in which men are stigmatized and discriminated against. There are some industries in which there are more women than men (nursing, teaching at some levels, etc), but that fact alone isn’t enough to support your claim. Indeed, even in those industries, the management positions are often dominated by men. So, I’m ready, if you have examples.
“It means that mediocre men are being replaced by more highly qualified women. The overall compentency is raised because more highly qualified people replace lower qualified people.”
Not sure what your statement is here, it’s exactly the opposite of a quota does. It replaces the more suitable person for the job for one that has been chosen because of some non-relevant characteristic, be it sex, race or nationality. To say that this person is more qualified than the person being replaced is an empty statement, there’s nothing that makes it true.
Thanks for taking the time to leave your thoughts here. But, having said that, I’m not quite sure to say to you except reiterating the argument in the piece. My claim is that your assertion about the effect of quotas is incorrect. Specifically, if I read you correctly, you are committed to the view that quotas necessarily lower quality. The point of my piece was to demonstrate on the basis of published research that this view is wrong. For example, quotas do not mean that someone is chosen “because of some non-relevant characteristic.” In the one research article I discussed, quotas changed the applicant pool, yielding stronger candidates and therefore the selection of someone who was best and who also happened to fulfill the desired profile. So, they were chosen “because” they were best, but it was the announcement that a quota could be used that attracted them to apply anyway.
Regarding the specific quote you open with in your comment, the reference is to the quality of a group, in this case elected officials. So, the overall competency which is being raised is the competency of the group, and the reason for that is that the quota lead to selection of women who were more qualified than the men they replaced. In other words, *without* a quota, people are being chosen for a non-relevant characteristic — in this case male-ness.
So, my goal is to challenge the position you default to, namely that quotas lower quality. I don’t understand what you mean when you assert that a statement is “empty” after having argued against it doesn’t make sense to me. It is incorrect that there is nothing that makes it true; it can be true, in this case, if quality can be measured. But you disagree with that. Why?
The statement is empty (or false, if you may) because, as you say in your comment, it can be true. It’s not always true, which makes it unusable. Is just as saying that a quota diminishes the quality of the group, it’s not necessarily true so I can’t make that statement either. I just state that the best suited candidate can be put aside because of external reasons. The problem I see in your argumentation is that you very quickly move from a “studies suggest” to ” The authors … demonstrate…”. There did not demonstrate it, the observed that this was the case in a very specific context, let’s say public politics in Sweden. That doesn’t meant that the “zippers” are for example applicable to hiring in private companies. I truly agree that a quota can result in stronger candidates applying, but that doesn’t mean the group as a whole is better off. You can end hiring some strong candidates that would have otherwise not applied, but at the same time having to hire a couple “mediocre” ones for which there was a more suitable match on the other group. And the process gets more complex if there’s several quotas ( native-american women, black males…), which is the case in some organizations.
I think part of what we’re discussing is perhaps terminology where we don’t necessarily disagree. For me, if a statement can be true under some conditions and false under others, then it isn’t empty. But, that’s not so important perhaps. I agree that it will not always be true that using a quota will lead to improvements in quality. What I intended to argue for in this piece is that the “stereotype” that quotas necessarily lower quality is false. That is not a necessary result of using quotas. I argued for this by suggesting that announced intention to possibly use a quota can affect the applicant pool, which in some cases leads to applications from qualitatively better members of the group being targetted. And I suggested that a quota on a group, in the form of a zipper, can raise the quality of the group by getting in members from the target who do better than members of the majority. So, I would certainly not see quotas as a fix-all for the issues. And I agree, if that’s what you’re saying, that finding the best qualified candidates is the goal. What’s clear to me, though, is that this is not happening, and more highly qualified women are being skipped over for less qualified men, and that requires interventions to fix. Quotas are the most extreme form of such interventions. For more on skipping over women, you might enjoy my “When women are good at math, they still don’t get hired.”
Thank you for writing this article.
I have never worked at a company who did any sort of affirmative action measure, although I have been the target for some hostility from coworkers who believed i was the recipient.
Reading the comments on your article makes me wonder what it will take to break down the inherent gender biases in our culture as so far no one commenting seems to understand even the basic premise of your article. If a well written and researched article by someone specializing in the field fails to be understood then I’m not sure what it will.
Maybe just more time, as each generation seems capable of only so much change. I know both my sister and I working in male dominated fields have had plenty of struggles being treated fairly and even though are Dad has been witness to this he continues to hold onto his old belief system.
I guess the best we can do is plant seeds and nurture them until they bear fruit.
Keep up the good work! I look forward to more articles. Here in Canada we still have a long way to go and it is a tremendous help to have multiple resources with which to combat gender bias.