Carpe Diem

Team Obama struggles to explain, defend the 12% gender pay gap at the White House, first reported here seven months ago

whitehouseFor about the last seven months starting last September, I’ve been blogging about the 12% gender pay gap at the White House illustrated above, see my original post here, see a December 2013 post here, a January 2014 post here, another January 2014 post here, a February 2014 post about February 20 being “Equal Pay Day” at the White House for 2014, and a March 2014 post here. Not much happened until this week, when the story percolated long enough I guess, that it finally got some media traction and now it’s everywhere: Drudge Report, CNN, Bloomberg TV, Washington Post, New York Times, CBS news, talk radio, etc. And the result is that Jay Carney and Team Obama are now frantically trying to explain their own 12% gender pay gap, just as Obama issued two executive orders today concerning fair pay for women employed by federal contractors.

Here are some of the media reports over the last few days highlighting the White House’s hypocrisy on the gender wage gap:

1. Video (note that the opening graph features a variation of the chart above) and report below via Newsbusters:

On Tuesday’s CBS This Morning, White House correspondent Major Garrett completely dismantled President Obama’s left-wing talking points on the supposed gender pay gap of women making 77 cents on the dollar compared to men, reporting: “The White House is getting…roughed up by hits own pay equity rhetoric.”

Garrett used the administration’s hypocrisy on the issue to fact check the false claims: “An analysis of White House salaries, which nobody here disputes, shows that the median income of female staffers is 88% of that of male staffers….Now, the White House said its gender pay gap is tied to job experience, education, and hours worked, among other factors. This matters because those explanations, according to the Labor Department, explain a good deal of the gender pay gap nationally.

The big difference in these stories, when President Obama discusses this issue nationally he doesn’t mention those other work variables, only the broad figure that 77 cents per dollar is what women earn compared to men in median wages.”

2. Watch a video below of Jay Carney squirming as he unsuccessfully tries to defend the 12% gender pay gap at the White House, courtesy of the Washington Post, and see its report here titled “The White House’s own wage gender gap.

3. The New York Times reported yesterday that “As Obama Spotlights Gender Gap in Wages, His Own Payroll Draws Scrutiny.”

Even as Mr. Obama seeks to make an issue of the gender gap in compensation across the country, however, his own hiring is facing some scrutiny. The recent study, by the conservative American Enterprise Institute, showed that the median annual salary for women in the White House last year was $65,000, while the median annual salary for men was $73,729. The study was based on White House salary data.

4. In this video segment, CNN’s John King called the White House’s push for equal pay “A textbook case … of do as I say, not as I do.”

5. Fred Lucas reporting for The Blaze gives me a little credit in his report “Economics Professor: White House Holds Itself to a Different Standard on Gender Pay Equity.”

6. Here’s from another CNN report “Inside Politics: Equal Pay gap reaches White House“:

For the past several elections, Democrats have adopted the equal pay issue and made the equality of paychecks a huge priority. But the issue of equal pay plagues President Obama’s White House. An analysis by the conservative American Enterprise Institute found that women staffers made about 88 cents on the dollar, compared with male staffers.

Bottom Line: Along with my WSJ op-ed today with Andrew Biggs (“The ’77 Cents on the Dollar’ Myth About Women’s Pay“), it was a pretty good day for a dose of some overdue statistical and economic sanity, and a good day for exposing the ’77 cents on the dollar’ myth that has gradually morphed from a faulty or incomplete analysis of gender wage differentials into more of a political lie.

26 thoughts on “Team Obama struggles to explain, defend the 12% gender pay gap at the White House, first reported here seven months ago

  1. Heh, well at least they cite this data now, even if only as coming from a “study” from the “conservative American Enterprise Institute,” which is obviously a dog whistle for liberals that they can ignore what those conservative meanies are saying. :) Was there an actual full “study” put out by the AEI or are they just referring to your blog posts?

    I also liked this video put up by the guy who put up the John King video, particularly the “stunned” reaction by the anchor at the end to Carney’s explanation for the WH wage gap. :D

  2. Professor Perry,

    Could you please include some additional information on the methodology you used, specifically with regards to how you coded each employee. I am looking at the original .csv file of 2013-Report-White-House_Staff which contains 460 unique entries of individuals coded as Status=Employee or Status=Detailee.

    I am interested in running a more detailed regression analysis, as I am sure you know that is how one can draw inference from data, and I would like to replicate your method as close as possible. In your descriptive statistics from your 12/8/13 post you list 228 females and 231 males, with the former earning $65,000 and the latter earning $73,729. Total employees for that analysis would be 459. In your 4/9/14 post you list 229 female employees and 232 male employees, but your earnings totals for each group are the same. The total employees in your that analysis would be 461. However, the dataset provided by the White House, which you cite as your data source, contains 460 total individuals. What is the correct number of individuals you selected from the original 460 member dataset? If the total is 461, where did you obtain the additional individual in light of the original data’s 460?

    Additionally, how did you determine the gender of each individual? This variable is not included in the original data and some of the first names seem ambiguous. For example, two Senior Writers are named Casey and Jordan. What outside source did you use to determine whether these individuals are male or female? Also, did you control for the individuals that have their status listed as “Detailee” as opposed to those listed as “Employee?” It would appear that detailed employees are those that worked for another agency/federal organization and maintained their pay at the federal scale applicable at that former agency. So, for example, Seth Wheeler the highest paid at $225,000 of the 460 individuals was detailed from his position with the Federal Reserve, where the pay scale is higher. This difference in pay scales between the Federal Reserve and the White House likely leads Seth to bias the figures towards men, when his pay may be better explained by the difference in pay scales. Surely someone with your experience and knowledge in economics and advanced econometrics can understand how these ‘detailed’ individuals may have an influence on pay rate and therefore bias any generalized inference based on descriptive statistics.

    Lastly, how did you control for Position Title and the individuals status within the current federal pay scale? The Position Title is included in the dataset, so (if you did control for this) I would imagine you just coded each position title as an ordinal variable. However, I am unsure what you used as your source in controlling for pay scale status. Again, I am sure you recognize how an individual’s pay scale status will likely have a strong influence on pay rate.

    I have no doubt that you have already done your due diligence in conducting a robust analysis of this data to confirm the existence of a pay gap that is correlated with gender. In the nature of academic transparency and to further the body of knowledge on this issue I am, as stated previously, interested in replicating your findings and building off them using more advanced methods. I appreciate your time and look forward to a reply.

    • Sean: I will send you my most recent updated Excel spreadsheet of White House employees. Here are some details:

      1. Out of 460 WH employees listed in the original database, there are 229 female WH staffers with positive salaries, and 229 male staffers with positive salaries. Two employees with $0 salaries were deleted. I have updated my blog posts to reflect 229 female and 229 male employees.

      2. To determine gender (not provided by the WH), I went through a pretty time-consuming process to determine the gender of each individual by his/her first name. That process included Internet searches, Facebook searches, media reports, searches of SS name records, etc.

      3. Reporters from the Daily Caller and Washington Post have also gone through the same process and I have been in contact with both of them, and my finalized database agrees with both of theirs.

      4. In some of my original blog posts, I had several very minor errors in the salary database that have now been corrected. In fact, the finalized databases shows a 13.3% gender pay gap at the WH, with a median female salary of $65,000 (as I reported before) and a median male salary of $74,958. Therefore, we could say that WH women make LESS than 87 cents for every $1 men earn.

      5. Note that Jay Carney, nor Team Obama or the WH, have challenged the 12% pay gap; possibly because it’s actually greater than that.

      • Professor Perry,

        I received your email and I appreciate you sharing the fruits of your data collection so openly. All else aside, I appreciate your prompt reply and your openness. I downloaded a more detailed dataset from FedScope that consists of individual employee data with variables such as education, years of experience, gender, pay rate, government pay scale, etc., but it does not include names and the White House Position Title seems to be a generic name that does not match any official listing.

        I believe that a cursory summation of median pay between genders in any capacity fails to truly explain whether a systematic gender gap exists as it does not control for factors that are known to cause pay discrepancies. A cursory review of the White House employees data showed that those with the same Title, excluding “detailed” employees, were paid the same or only slightly differently. The “detailed” employees appear to be biasing the results although I have not further reviewed in which direction. With regards to some of the more modest differences, surely one would agree that, ceteris paribus, a federal employee with 5+ years of service would be paid more than a federal employee with 1 year of experience. This same claim goes for any analysis of the private sector as well. The fact that the complex issue of gender and pay is being debated, at least in the media, using basic descriptive statistics is disappointing in light of the multitude of academic studies published in peer reviewed journals. Such studies spend time developing advanced methodological models employing multivariate regression analysis, which provide a more accurate analysis of the issue. A quick search on Jstor resulted in about 20,000 results for “gender pay gap,” and a look at the literature review in some of the more recent articles demonstrates a variety of theories on why such a gap exists. The gap does not appear to be ideological, and it is unfortunate that an important issue like equal pay between the sexes has turned into a talking point for ideological advancement.

        With that said, I again thank you for you emailing me with the information that you used.

        • Sean, “an important issue like equal pay between the sexes has turned into a talking point for ideological advancement” Why is equal pay between the sexes, races, etc. an important issue? In the US it was government through separate but equal that perpetuated state sanctioned discrimination. The important issue should be to get government as far away as possible from private contractual matters.

          • Why is equal pay between the sexes, races, etc. an important issue?

            It’s not, but people like to make assertions with no real oomph behind them, hoping nobody will call them on it.

        • it is unfortunate that an important issue like equal pay between the sexes has turned into a talking point for ideological advancement.

          Stupid is as stupid does. If you want politicians to “do something” about some perceived wrong, them by necessity that issue will be come a “talking point for ideological advancement”.

          Grow up.

        • Sean

          All that detailed examination of tree bark may have cauesd you to miss the forest.

          I believe the point of this post is to point out the hypocrisy of a President who would take up the fight against a “gender pay gap” when his own White House suffers from that very disease.

          As you and Prof. Perry have pointed out through careful analysis of the data, there is no gender pay gap at the White house that isn’t explained by factors other than gender. I other words, there is no evidence of gender discrimination in pay at the White house. The same can be said (and has been said) about the private sector.

          Yes, it is extremely disappointing that someone would dishonestly use a non-issue like gender discrimination to make political capital, especially when that person is the President of the United States.

    • If you are truly interested “In the nature of academic transparency and to further the body of knowledge on this issue” as you say you are, then do the work yourself. You have all the data and you have listed all the details that need to be considered so do your own study and then post the results.

      You could have easily send Dr. Perry an email requesting his data but instread you posted this very long and detailed post in a veiled attempt to discredit his findings. Unfortunately for you, Dr. Perry is a real economist who is concerned only with truth and facts. He study will withstand any analysis you attempt.

      Nice try though Sean.

    • Sean, I think you’re missing the point of Mark’s chart. The whole point is to lump together detailees and those of varying experience and accomplishment, as that is the standard Obama has chosen for his misleading claims that women garner 77 cents for every dollar a man earns in the broader economy. Mark has pointed out many times how that is not a fair comparison, yet Obama and his administration continued to trumpet that misleading stat.

      Well, turnabout is fair play, such misleading methods have now been turned around on them. They can have fun explaining these other factors or eat crow based on their own methods. Great fun for those of us watching! :D

  3. @Mark Perry …

    It took about six months for your blogging and the study exposing the ‘equal pay for equal work’ Democrat talking point to begin the ‘melt down’ process, of that fallacious Democratic talking point, with its arrival at the White House.

    When will the Federal –ONe siZe doESn”’T fiT aLL– minimum wage discussion arrive at the White House?

    [what have you done for me lately ;-) ]

  4. The WH gender pay gap is less than the overall gender pay gap. 88% to 77% (or 82% with current data). What about the residual of the ceteris paribus condition?

    Okay, I notice in Perry’s chart that he refers to median annual salaries. What’s the mean?

  5. ” In this video segment, CNN’s John King called the White House’s push for equal pay “A textbook case … of do as I say, not as I do.””

    i would even take this a bit further and say it is a case of projection and prejudice.

    “I have good reasons for the things I do and I do not discriminate, but THEY certainly must be discriminating, look at the results.”

    it’s this faith in their own rightness and righteousness while assuming others lack the same incentives and virtues that is so galling.

    “MY reasons are good reasons, but YOURS, even though i do not even know you, must be prejudice.”

    • Krugman wrote about it this week. Don’t you get it? Liberals are just darned smarter than conservatives and they do not have any influence over the press….

      “People want to believe what suits their preconceptions, so why the big difference between left and right on the extent to which this desire trumps facts?

      ” One possible answer would be that liberals and conservatives are very different kinds of people — that liberalism goes along with a skeptical, doubting — even self-doubting — frame of mind; “a liberal is someone who won’t take his own side in an argument.”

      Another possible answer is that it’s institutional, that liberals don’t have the same kind of monolithic, oligarch-financed network of media organizations and think tanks as the right.

      Whatever it is, I think it’s important: people are people, but politics doesn’t seem to have the same stupiditizing effect on left and right.”

        • I think the first sentence is William’s and is sarcastic, the rest is Krugman’s. William’s quoting isn’t done well so it’s hard to tell what’s his, but the rest of it sounds like Krugnuts. Also, I don’t read that tripe at the NYT, so I can’t verify the source.

    • m: “it’s this faith in their own rightness and righteousness while assuming others lack the same incentives and virtues that is so galling.

      Precisely. There’s this extremely annoying attitude some people have that they know better how to manage my life than I do.

  6. I am just a lowly Engineer not an Economist, but I do have a many years of observing how businesses are run. I can say with confidence that if a company could hire women at 77% of what they pay men, there would be no unemployed women.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>