Monday, August 03, 2009

Most or All of The Pay Gap Disappears After Controlling for Marriage and Having Children

The Department of Labor recently released its annual study Highlights of Women’s Earnings in 2008 (HT: Christina Sommers) and opens the report with the following statement:

In 2008, women who were full-time wage and salary workers had median weekly earnings of $638, or about 80% of the $798 median for their male counterparts. In 1979, the first year for which comparable earnings data are available, women earned about 62% as much as men. After a gradual rise in the 1980s and 1990s, the women’s-to-men’s earnings ratio peaked at 81% in 2005 and 2006.

MP: Doesn't the BLS' use of the term "male counterparts" (Webster definition: "one remarkably similar to another") imply an "apples to apples" comparison between male and females workers, as if all relevant explanatory factors have been controlled for, i.e. the ceteris paribus condition has been imposed?

For example, buried in the BLS report are these two interesting facts:

1. In Table 1, it is reported that for those workers who have never been married, women make 94.2% of their "male counterparts."

2. In Table 8, it is reported that for single workers with no children under 18 years old (marital status includes never married, divorced, separated and widowed), women make 95.6% of their "male counterparts."

In these two examples, I would argue that the term "male counterparts" is much more valid than the BLS' use of the term in the opening statement of its report when it is referring to all workers.

Since marriage and having children affect male and female earnings differently, men and women workers can't really be considered "counterparts" in a statistical sense, and any unadjusted comparisons would be comparing apples to oranges. In fact, some research has concluded that the factors of age, marriage and motherhood explain all of the male-female pay gap.

For example, in a 2005 NBER working paper "What Do Wage Differentials Tell Us about Labor Market Discrimination?" by June O'Neill (Professor of economics at Baruch College CUNY, and former Director of the Congressional Budget Office), she conducts an empirical investigation using Census data and concludes that:

"There is no gender gap in wages among men and women with similar family roles. Comparing the wage gap between women and men ages 35-43 who have never married and never had a child, we find a small observed gap in favor of women, which becomes insignificant after accounting for differences in skills and job and workplace characteristics.

This observation is an important one because it suggests that the factors underlying the gender gap in pay primarily reflect choices made by men and women given their different societal roles, rather than labor market discrimination against women due to their sex."


At 8/03/2009 9:23 AM, Anonymous gettingrational said...

The gender gap perception must be maintained so that the Democratic party maintains control. The perception that mean employers are taking wage advantage of women does two obvious things: 1) more likely to vote Democrat because it caters to victim classes; 2) easier to promote additional tax obligaions on the bully employers.

I am glad my wife was home during pre grade school days with our children. Her two undergraduate degrees and one post graduate degrees currently help provide our family with a good income. I think we contributed to statistical warping during her gratifying days at home with the kids and a lot less income.

At 8/03/2009 12:10 PM, Anonymous Anonymous said...

Ben Wattenberg made this point in his book 25 years ago, "The Good News is the Bad News is Wrong."

Even then, there was no gender gap or race gap. Just as there really weren't 3 million homeless (a made up number by Mitch Snyder). Neither were the USSR and the USA morally equivalent. And the air and water weren't getting dirtier with the evil bushhitler in office.

There really aren't 47 million Americans without health care. One in five children in America don't go to bed hungry. Men aren't more likely to beat their wives on Super Bowl Sunday. AIDS isn't a serious threat to heterosexuals. And global warming isn't going to cause the oceans to inundate cities around the world.

If liberals had to tell the truth, what would they be able to campaign on?


At 8/03/2009 2:59 PM, Blogger Angie said...

Can't help but notice that at no point do women counterparts earn more.

I also can't help but think that it's not OK for women to essentially take a cut in pay if they marry and/or have kids when the same effect doesn't happen to men.

Having done payroll for several years in several firms has also shaped my opinion.

What makes me un-liberal is that I don't think I need the Federal government to fix it for me.

At 8/03/2009 4:21 PM, Blogger juandos said...

"There really aren't 47 million Americans without health care"...

Over at Professor John Lott's site it was noted: Fox News poll indicates that 27 million without health insurance

At 8/03/2009 6:40 PM, Blogger Plamen said...

Angela, could you elaborate on why you think it's not okay... (your second paragraph). Why is equality of outcome required, if men and women, on average, put in a different amount of time at work and have different average uninterrupted career spans? Please, note, I have no problem with a claim along the lines of "men should be homemakers more often" - which every family can freely negotiate. If that's your claim (as I suspect because of your last paragraph), then we're on the same page.

At 8/03/2009 7:23 PM, Anonymous Anonymous said...

Not having health insurance is not the same thing as not having health care. Everyone in this country has health care.

At 8/03/2009 9:24 PM, Anonymous Mika said...

(1) 27 million without health care is still way too many - still a terrible travesty, not indicate of a country that claims to be - and likes to think it is - the best.

(2) Not having to pay at the ER is NOT having health "CARE"! Indigent die in their forties not because they can go to the ER but because they have no ongoing treatment and care.

(3) Orig post does not address the "glass ceilings" that still exist throughout corporate America, which keep the most talented and dedicated women from promotions, while hiring and promoting less able men above them. I know of what I speak. . . .By the way, I'm a male.

At 8/05/2009 11:04 PM, Blogger OBloodyHell said...

The whole pay-gap crap is, was, and ALWAYS has been a load of hooey:

"The U.S. Census Bureau found that as early as 1960, never-married women over 45 earned more in the workplace than never-married men over 45."
- Warren Farrell -

(Farrell is a former chairman of NOW, a PhD psychologist, and author of The Myth of Male Power and Why Men Earn More)

In short, it's a complete and total crock.

The way they pull these numbers out of their rectal cavity is to classify 25 hrs per week as "full time". I am fairly certain there isn't a single person who actually works full time who would concur that 55% of 40 hrs per week is "full time". As a matter of fact, if you actually work less than 40 hrs a week on a steady basis, then you ain't "full time", *period*.

And, for most of that, it would also make sense to create additional categories for >50hrs and >60hrs per week. Most of the people I know who actually work and earn near to or more than six figures do so by regularly working 60 hrs per week... so it's hardly rational to compare their efforts and compensation to someone working less than half that.

At 8/05/2009 11:05 PM, Blogger OBloodyHell said...


"While we acknowledged that glass ceilings that kept women out of the top, we [have] ignored the glass floors that kept [them out of the bottom]. Thus the 'Jobs Rated Almanac' reveals that the majority of the 25 worst jobs 'happen to be' male dominated."
- Warren Farrell -

At 7/07/2010 9:47 PM, Blogger echidne said...

There is a *large* and careful field of economics which looks at these questions.

The methods are multiple regression analyses controlling for all relevant variables that might account for variation in earnings.

Those studies, when properly carried out, do not show that the whole wage gap can be explained by women's child-rearing or marriage choices.

In fact, comparing only full-time workers hides the fact that part-time workers (who are the ones who have responded to that childcare obligation) in fact earn considerably less.

On the particular comments made in the post: Comparing single workers with no children will get you an over-representation of the very young. Initial earnings differences are always very small. It has additional difficulties, one of them being the fact that single men earn less than married men and single women earn more than married women. So it's still not comparing apples with apples.


Post a Comment

<< Home