Individual v. Collective, Live and Learn, Trust

Media Matters Criticizes Me (sort of) and Unknowingly Makes a Major Concession in the Process

So Media Matters, a truly unbiased organization if there ever was one, ran an article criticizing conservatives for denying the alleged wage gap between men and women. The article particularly criticized Alex Castellanos, who got into a bit of tiff with Rachel Maddow about said wage gap.

Well it just so happens that Alex Castellanos used my article on the wage gap as a major source for his conclusions. (Yep, I got cited by name in The Daily Caller, what what).  I’ll get to the article by Media Matters in a second, but first I need to respond to the commenter, one MidnightWriter, who noticed where Castellanos got much of his information:

It wasn’t all that difficult to eviscerate.

Castallenos admitted to getting his facts from this blog which, in turn, borrowed heavily from John Stossel, James Bennett, and Thomas Sowell.

Oh boy, I’m about to be eviscerated! It is true, however, that my entry was mostly a summation of research by James Bennett, Thomas Sowell, Warren Farrell and Denise Venable.

This was the strongest source ol’ Alex could find to back his arguments? We’re asked to accept some rather broad ground rules centered on what can be called same pay for the same work (such as the idea that history professors are not paid as well as business professors — something offered with no numbers to support that thought, nor are we given any numbers that tell us the averages of what male and female professors in those fields are paid). We’re offered two rather misleading charts; one that shows that more men receive college degrees incomputer science (which has nothing to do with equal pay for equal work), and a chart that offers weekly median earnings (the issue is averages).

Average full time business professor: $111,621, average full time history professor: $82,202. The average associate and assistant business professor make $93,767 and $87,248 respectively. The average associate and assistant history professor make $63,228 and $52,626 respectively. (Source: here). Of full time, associate and assistant professors in business, women make up 17.9%, 29.1% and 37.3% respectively. In history, women comprise about 18%, 36% and 44% respectively.  Google searches no están difícil. Furthermore, this was one small point that was supposed to be so obvious it didn’t require a citation (the sky is blue by the way) to illustrate why inequality doesn’t necessarily equal discrimination.

Both graphs were supplemental and not even referenced in the text. A degree in computer science is well known to come with a higher wage than say, sociology. (Are you really going to ask for another citation here?) And medians are actually better than averages typically because they eliminate outliers that would otherwise skew the data. I explained this in a later entry of the Lies, Damned Lies series and you can enjoy this guy’s take on it.

And, of course, there was that classic bit of cherry picking, college educated, never married women between the ages of 40 – 64 actually earn more than college educated, never married men between the ages of 40 – 64. Seriously? That ever so slender slice of the pie is supposed to prove just what now? Here we have one select demographic group where the women earn, roughly 20% more than the men — well, geez Louise, how much worse does that make it in all the other categories for women to end up earning 23% less overall?

The argument is so exceptionally and epicly bad it will require a list of bullet points:

> MidNightWriter forgets to mention that I listed four statistics, not just that one, illustrating the wage gap was fallacious

> What exactly is MidNightWriter’s explanation for the gap I mentioned? Do employers discriminate in favor of never married women without children over the age of 40 but just lay the hammer down to all other women? Why would they do this? Oh wait, one of the other statistics MidNightWriter forgot to mention that I included stated unmarried women ages 21-35 make the same as unmarried men of that age. Those poor 18-20 and 36-39 year old women!

> Never once does MidNightWriter mention the crux of my argument and the whole reason to segment off never married women without children: The marriage asymmetry hypothesis. This is like criticizing Darwinism without mentioning his theory of evolution.

> Nor does MidNightWriter mention all of the variables I discuss that contribute to gap. Nope, the only thing that matters is the raw comparison, a comparison which is, quite obviously, meaningless.

And finally:

I’ve got to say, though, I did love the title — “Lies, Damned Lies, and Statistics.” The author knows his Mark Twain, but apparently has no concept of irony.

Maybe MidnightWriter was discussing the fact that Mark Twain popularized the term, but I doubt it. Sorry MidnightWriter, Twain didn’t come up with it, he attributed the saying to  Benjamin Disraeli (although there is some controversy as to whether Disraeli came up with it either).

Anyways, enough with MidNightWriter, what about the Media Matters article. Well the article lists out a bunch of conservatives, including Castellanos, who have criticized the wage gap. Then it blows them out of the water with an AAUW report:

Overall, the regression analysis of earnings one year after graduation suggests that a 5 percent pay gap between women and men remains after accounting for all variables known to affect earnings. Women who choose male-dominated occupations appear to earn more than do other women. Undergraduate majors in business and management, engineering, health professions, or public affairs and social services enhance both women’s and men’s earnings. [AAUW, April, 2007]

Wait, what? … 5%?* That’s it? What happened to 77 cents on the dollar? Hold on, they’ve got more:

 Bush Labor Department: The “Adjusted Gender Wage Gap … Is Between 4.8 And 7.1 Percent.” In January 2009, the Bush administration’s Department of Labor published a report written for the department by CONSAD Research Corporation. While downplaying the existence of wage inequality, Deputy Assistant Secretary of Labor Charles E. James stated in a foreword to the CONSAD report that after controlling for several variables, there was “an adjusted gender wage gap that is between 4.8 and 7.1 percent.” [CONSAD Research Corp, 1/12/09]

Fascinating. This is the same report I mention in the expanded version of this article I put into my new book. I’ve never said that some sort of wage gap is impossible, just that it’s very much exaggerated if it does exist at all. Indeed, I think the CONSAD report misses some things, which I will get to shortly.  But either way,  I haven’t seen any “92.9 – 95.2 cent” pens in protest of the patriarchy. Nor have I heard Media Matters or any other liberal apologize for distorting the wage gap by some 300%!

And OK, I’m a man so I guess I’m not supposed to have an opinion on these things, but ladies, at what point does the wage gap become trivial? 1%? 3%? 5%? If guys still pay for dates most the time and the wedding rings we buy are a wee bit more expensive, maybe we have a date/wedding ring premium. Or maybe it’s this (warning, much profanity ensues):

Regardless, even saying it’s 5% or thereabout is questionable. The idea that you can control for every variable is a fallacy. Russ Roberts has a great discussion with Jim Manzi on this very subject I highly recommend. As Jim Manzi explains:

I’ve built thousands of regression models in my life, and they are not useless; they are useful for certain purposes. What I argue is that they are not capable of determining reliable, useful, and nonobvious effects of interventions…

…if I have a regression analysis that tries to predict a set of variables–say, I want to predict what unemployment will be as a size of the size of the employment, hypothetically, the size of the population, economic growth rate in a prior period, the education level of the population, and so on, and I say: I am trying to use this to measure the effect of changing education levels on unemployment. And all the variables other than education level are meant to be controls, or to hold constant these other effects we described. That if I neglected to include a variable in my model for, let’s say, the amount of immigration into the society, that turns out to be causally important, that what happens is by failing to include that, I modify or create instability in all the parameter estimates, including the estimate of the variable I care about. And therefore, if I’ve left out any significant variables from my equation, the estimate of the impact of the variable I care about is called into question.

And my argument is that all regression models, including… one that I built and take it apart in detail to show how this is true [referencing his book], all models like this are subject to omitted variable bias because we can’t get data on all the potential causes. The complexity of this phenomena overweighs our ability to build terms, to build interaction terms, and so on; and they are always subject to this problem of significant omitted variable bias. Such that we cannot rely on their results.

Yes, we can get close, but be very careful with phrases like “studies prove” or “a regression analysis showed.” There’s always some uncontrolled variable. Indeed, the CONSAD report doesn’t appear to take years at the same job into account (men 5 years, women 4.4 years) or total time spent out of the labor force. What about people paid on comissions? The CONSAD report, for its part, does mention:

If the salesperson’s wages includes commission on sales, discrimination by customers could result in a substantial gap in wages between male and female salespersons and service workers. (Pg. 55)

Amueod-Dorantes and the Mach find that piece rates and commissions are more prominent in increasing the wages of men relative to women…” (P. 93)

The word “commission” or “commissions” doesn’t appear in the AAUW report. Also, it’s not necessarily customer discrimination, for example, the CONSAD report also mentioned men paid with tips, i.e. waiters and waitresses, earned 11% less (Pg. 93). That could be an aberration, or discrimination on the customers part or something else entirely. Customers could discriminate slightly against male waiters and female cars salespeople or something along those lines that’s much more nuanced than Media Matters would have you believe.

With regards to the overall wage gap, it could just be that men dedicate themselves slightly more to their careers and women seek a better work/life balance. In other words, motivation is very important and not easily quantified and basically impossible to control for.  For example, a poll by MBACareers.comrevealed that men and women who obtain MBA’s have very different career aspirations:

Additionally, the survey revealed the long-term career goals for male and female MBAs differ as well. Men acquiring an MBA aspire to become president or CEO of both public and private companies or to start their own businesses. Women MBAs, however, ranked management consulting, executive level vice-president positions and non-profit executive management high among their career goals.

Indeed, I would go so far as to say women are wiser in this respect. But it’s a wise trade off that hurts the bottom line. And while young women have become a lot more career oriented, much of the data on the wage gap is from older men and women who, for better or worse, have and have had a more traditional mindset.

Finally, what about negotiating for raises? The evidence is a bit anecdotal but women appear less likely to do so. There are certainly things we could do to rectify this, but they have to do more with education and societal norms. Having the government negotiate or help negotiate on women’s behalf seems like the most patronizing thing imaginable.

Discrimination exists, without question, but it is not nearly as prevalent as Media Matters, Rachel Maddow or MidNightWriter would have you believe. And if the wage gap even does exist, it’s not anywhere near as big as they would have you believe either.

Photo Credit:


*While not mentioned by Media Matters, I should note that the AAUW report does say the gap widens to 12% later in life. If you average that out, it makes for about a 8-9% gap which is just above the high end of the CONSAD report.


For more Swift Economics, subscribe now to our RSS Feed
Follow Swift Economics on Twitter
LIKE Swift Economics on Facebook



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s