Wal-Mart, The Supreme Court, and Discrimination II

This is a response the Emily's post on Wal-Mart, The Supreme Court, and Discrimination.


With all due respect to Professor Tanya Hernandez and Emily, I find myself disagreeing with a number of the assertions in both of their arguments.

Professor Tanya Hernadez makes the case that because 1.5 million women felt strongly enough to sue Walmart, one can draw the conclusion that, at the very least, there was something fishy going on at Walmart. If 1.5 million would have gone through the trouble of actually suing Walmart, then indeed there would be a strong intuitive argument to be made that Walmart was generally misbehaving in some way. Of course, I think Prof. Hernadez used some pretty impressive legal shorthand to finesse this point into existence. In a class action lawsuit one must opt out if one falls in the included class. One and a half million women did not actively pursue a case against Walmart, rather the majority of these women simply did not take the proactive steps to opt out. And people wonder why everyone dislikes lawyers.

Emily argues,

I hope I do not need to convince anyone that gender bias still exists today. Women are still generally seen as lacking the characteristics warranting promotion into managerial and higher positions. Women also make $0.77 to the $1 that men make nationwide for full-time, year round work. Women are still required to take on the majority of child rearing and household responsibilities, making it more difficult to work outside the home and/or limit availability. Unless specific criteria are spelled out for those in charge of making pay and promotion decisions, it is likely that unconscious (or even conscious) stereotyping and gender bias will influence those decisions. This is especially the case when the managers who are given this subjective power themselves might not be the most educated and/or unbiased people.

I disagree strongly with the leading premise of the argument. I think it is incredibly important for people making the claim of bias to argue it, and argue it well. We as a culture tend to make the argument that any difference between groups of people has to be attributed to either a)institutional bias or b) societal bias. This leaves out other reasonable options. If we are unwilling to consider inherent differences between the groups and different choices made by those in different groups, we have limited our ability to approach difficult questions with an open mind.

Emily goes on to say that women make 77 percent of what men make. I do not deny this. I would ask her to explain the reasons behind this gap. Here are some reasons I would present to explain this bias: 1) women tend to negotiate less when offered a salary; and starting salaries heavily influence wages at a later date in a career, 2) women are indeed the primary caregivers in our society, meaning that they often take years away from their career, losing momentum and raises; 3) there is a societal expectation for women to work less so as to devote more time to their home lives; and 4) there is gender discrimination.

I will briefly present some criticisms of the statistics behind the Walmart case specifically, and then move on to the more general question of how our society ought to approach wage differentials generally (note: Jeremy being the data-head provided the following information):

1. Richard Drogin, the statistician retained by the Respondents, did not conduct statistical power analysis to determine the minimum detectible effect given the extremely large sample size. With a sample of nearly four million people, it would be difficult not to find significance, even with relatively insubstantial differences between men and women.
2. The “unexplained” wage differentials between hourly male and female Walmart employees generally fell within the typical range of adjusted national gender wage differentials (4.8 to 7.1 percent). Salaried employees were a notable exception. However, when Dr. Drogin included a variable for “job position,” the differential for salaried employees fell between 5.46 and 7.07 percent – well within the adjusted national range. Dr. Drogin argues that this variable should not be included, as it may be tainted due to prior discrimination. While this may be true, failing to include this variable biases the regression in favor of the Respondents, with whom the burden of proof rests.
3. As many of the amicus briefs for the Petitioner noted, regression models using aggregate (rather than store-specific) data assume rather than prove commonality in terms of managerial decision-making. The regression models used by Dr. Drogin also assume additivity, when there could be a number of important interactions among different variables in the equation.. This possible misspecification could upwardly bias the gender coefficients and generate larger “unexplained” wage differential estimates. 

I’m back, and glad Jeremy is so smart. What adds to all of this is that women who actually applied for management positions were hired at a higher rate than their male counterparts. So the question becomes:Is this indicative of a culture of discrimination? I would argue no, and that for this reason a class action suit just doesn’t make sense. If you would like some really enjoyable evidence of why Scalia might be a bastard but really sharp read his back-and-forth with the lawyers representing the class action suit about how Walmart both had an institutional policy of discrimination about promotions in place and how it had no institutional policy regarding promotion at all.

The reason this case makes me cringe is because 20 years ago Anita Hill wasn’t the exception she was the rule. When talking to a woman over 40 about their experiences in the workforce, it makes me cringe to be a man and at all associated with what they went through. I’m not arguing we are in a post-gender bias society. But I am arguing that this is a terrible lawsuit and avoids addressing the questions corporations and society must deal with in our current world.

Which brings us to the real question, what is a corporation like Walmart responsible for when making hiring decisions? I have two examples to clarify this question. If a bright, hard-working, woman were to have religious convictions that her main job was to be a mother and wife and she only wanted to work 20 hours a week, should Walmart be responsible for pushing her to apply for a promotion? I think they clearly should not. If an equally bright, hard-working, woman was unsure whether she should apply for a promotion because she sees mostly men in managerial roles and isn’t sure how she should fit in, does Walmart have a responsibility to have some procedures in place to help her and women like her? I think absolutely they should, for two reasons. First,it behooves Walmart to hire the best people, and this is an untapped resource. Second, we know that in societyas awhole, and corporations individually, we have leftover biases and institutional structures that reflect a not-too-distant discriminatory past. Correcting that ought to be a priority of both society and companies.

Women and men are inherently different. If we attribute all differences to discrimination, we end up making the word lose all meaning. I would ask Emily specifically and feminists generally if they think that women’s much higher college entrance and graduation rates are discrimination based? Do they think it is discrimination that this economic downturn has disproportionately affected men, specifically younger men? I think these are not examples of discrimination, but if we use the criteria that all differences have to be attributed to discrimination, we’re trapped in a pretty terrible loop.

So here’s my challenge going forward. Examine Walmart on the merits of its case. Look at the numbers behind wage differential, and let’s understand why this differential exists. Understanding this, we can have a better idea of what we ought to attempt to change. Let’s not write off arguments like Lawrence Summers’s speech on academia, science, and gender because it presents uncomfortable conclusions. When we choose to willfully ignore that the world is a complex place, we end up at conclusions that are quite simply untrue. There are real questions to be answered here. Should corporations be a part of helping to change societal norms via the roles women play in the corporations? How should we address not only discrimination where we find it (forcefully I hope) but also what to do about the really difficult questions like how to deal with the fact that our culture seemingly socializes girls in a way that inhibits them from achieving the same monetary success as their male counterparts. I’m not arguing against addressing discrimination, I’m arguing that the Walmart case is not perhaps as clear as Emily believes and that the belief that our gender differences are a result of discrimination is counterproductive.

-Bo