Economists Do It With Models

Warning: “graphic” content…

Bookmark and Share
Correcting The State Of The Union, Gender Edition…

January 21st, 2015 · 15 Comments
Gender

I suppose that should technically read “Correcting The State Of The Union Address,” since I make no claims as to my ability to fix all of the nonsense that is currently going in the U.S. Anyway, I of course had a number of (usually nitpicky) objections regarding President Obama’s State of the Union address last night, but I know by this point that people have to choose their battles. So here’s mine…from the speech:

Today, women make up about half our workforce. But they still make 77 cents for every dollar a man earns. That is wrong, and in 2014, it’s an embarrassment. A woman deserves equal pay for equal work. She deserves to have a baby without sacrificing her job. A mother deserves a day off to care for a sick child or sick parent without running into hardship – and you know what, a father does, too. It’s time to do away with workplace policies that belong in a “Mad Men” episode. This year, let’s all come together – Congress, the White House, and businesses from Wall Street to Main Street – to give every woman the opportunity she deserves. Because I firmly believe when women succeed, America succeeds.

I am soooooo sick of this statistic, since it basically suggests that a woman shows up at a workplace and her boss is like hey, you look like you might have ovaries, here’s $0.77 rather than $1. And that’s not what is actually happening. Yes, it is true that, on average (actually, comparing medians if you want to be technical), a woman in the U.S. earns 77 percent of what a man in the U.S. earns, but that figure doesn’t control for any relevant determinants of income- schooling, industry, tenure, etc. Therefore, I cringe whenever the “equal pay for equal work” line is trotted out, since “equal work” would imply that whoever is handing out this 77 percent figure did in fact run some sort of regression that would control for enough to get to a point where the comparison was at least close to equal. In the spirit of actually wanting to understand the gender pay disparity issue and not just quote a meaningless number, let’s look at some actual research from Claudia Goldin. Some helpful excerpts:

Men and women begin their employment with earnings that are fairly similar, both for full-time year-round workers and for all workers with controls for hours and weeks. In the case of the latter group, relative earnings are in the 90 percent range for the most recent cohorts even without any other controls. But these ratios soon decline and in some cases plummet to below the 70 percent level.

Translation: We’re basically at a place now where young men and women don’t differ substantially in their levels of education (in fact, I think women are actually outperforming in terms of educational attainment according to a number of metrics), so when comparing the initial situations of these young people, the divide is 90-some-odd cents on the dollar, not 77. And this is without taking into account the fact men pay be sorting into higher-paying jobs. That said, there seems to be a shift in gender disparity as people move on their lives that should be examined.

The main takeaway is that what is going on within occupations—even when there are 469 of them as in the case of the Census and ACS—is far more important to the gender gap in earnings than is the distribution of men and women by occupations. That is an extremely useful clue to what must be in the last chapter. If earnings gaps within occupations are more important than the distribution of individuals by occupations then looking at specific occupations should provide further evidence on how to equalize earnings by gender. Furthermore, it means that changing the gender mix of occupations will not do the trick.

Translation: Convincing men and women to enter the same occupations wouldn’t make the gender disparity go away, so let’s perhaps stop focusing on that so much as a potential solution.

The clear finding is that the occupations grouped as Business have the largest negative coefficients and that occupations grouped as Technology and Science have the smallest ones. That is, given age and time worked residual differences for Business occupations are large and residual differences in Technology and Science are small. In fact, for the “young” group (less than 45 years old) some Technology and Science occupations have positive coefficients.

Translation: The female “penalty” differs a lot by occupation, and in some cases there is no penalty and even a benefit to being female.

As I will later demonstrate using data on occupations in business and law, the impact of hours on the gender gap is large and goes far to explain much of the gender earnings gap. Individuals who work long hours in these occupations receive a disproportionate increase in earnings. That is, the elasticity of earnings with respect to hours worked is greater than one.

Translation: Within an occupation (in some cases), being a high earner (even on a per-hour basis) requires long hours and, as is shown in another part of the paper, working a particular schedule. This feature explains a lot of the gender discrepancy and is a result of women and men selecting into these situations at different rates, especially as women start caring for families.

The gender gap in annual earnings for the JDs and MBAs, although large by year 15, is almost entirely explained by various factors, such as hours worked, time out of the labor force, and years spent in part-time employment.

Translation: This is not an ovaries penalty story, at least not directly.

What, then, is the cause of the remaining pay gap? Quite simply the gap exists because hours of work in many occupations are worth more when given at particular moments and when the hours are more continuous. That is, in many occupations earnings have a nonlinear relationship with respect to hours. A flexible schedule often comes at a high price, particularly in the corporate, financial, and legal worlds.

Hopefully there is no translation needed here. The overall point of presenting this is that, in order to craft an actual solution to a problem, it’s crucially important to identify what is causing the problem. As a society, we seem to have decided that a gender pay differential is a problem. However, the lack of understanding of the nature and cause of the problem is going to prevent the problem from being solved. The information provided above suggests that any legislation of the “equal pay for equal work” form, for example, will be mostly ineffective, since observed differences in pay are in fact largely explained by inequalities in either job tenure or work quantity. In order to solve the problem, then, policymakers must look one step behind the curtain and think about how to mitigate the effects of differences in worker hours or tenure rather than just keep trotting out a well-worn sound bite to overshadow the real issue.

Econgirl out. *mic drop*

Tags: Gender

15 responses so far ↓

  • 1 Steven // Jan 21, 2015 at 7:12 pm

    If you cross out the ‘equal pay for equal work’ line. Obama is saying exactly what you are. And the pay-gap disappears in countries (think Scandinavian countries) which have more progressive policies to address these types of differences. Incentives for men to take parental leave etc.

    The simple catch-phrase isn’t his policy, but it’s included because people like simple phrases.

  • 2 Reader // Jan 21, 2015 at 8:57 pm

    Without having fully read Goldin, I wonder if you are potentially mixing up your causal arrow.

    You say:
    (1) Women are less likely to select into long hours professions.
    (2) Long hours professions are likely to be high paying.
    (3) Therefore, women have fewer high paying positions in the same industry.

    This story can just as easily be told as:
    (1) Women are less likely to have high paying positions in the same industry.
    (2) High paying positions usually have long hours.
    (3) Therefore, women are less likely to have long hours positions.

    Both interpretations would be consistent with high hours predicting the wage gender gap.

    To draw a concrete example, compare:
    -Women are less likely to be on the partner track at a law firm due to bias factors, and thus do not work any amount of longer hours associated with being on that track.
    -Women are less likely to want to work the long hours associated with opting into the partner track, and thus do not end up among the high paid partners as often.

    These two would look the same, datawise. So it’s unclear whether women “select” into these situations in the way your translation asserts.

  • 3 Steven // Jan 21, 2015 at 9:11 pm

    sure women might ‘select’ into these hours, that’s the point. But that would still be a problem. Why do women ‘select’ into these? Perhaps because there is not enough support for women to take parental leave and re-enter the workforce. Perhaps there is a societal view that women are responsible for raising children and there is negative discrimination for any parent that leaves the workforce for parenting.

    From Obama’s speech:
    “A mother deserves a day off to care for a sick child or sick parent without running into hardship – and you know what, a father does, too.”

    As I said in my earlier post, some countries have done a lot to correct these types of things that are encouraging women to self select into lower hour roles and taking a career hit for it. And in fact, the pay gap disappears considerably as a result.

  • 4 Dave // Jan 21, 2015 at 9:19 pm

    These statistics match my real-world experience.

    But I’m not sure there is a way around this as long as women tend to be the primary care-givers of children. I’m in the technology sector (where, as your data indicates, women tend to make as much or more than men *if* they put in in equal work). The problem is that, if you’re the primary care-giver of a child, there is no way you could put in “equal work” at many (?most?) tech jobs. Hours and work environment in tech tend to be incredibly flexible; however, in return for offering you that flexibility, employers expect you to be flexible with them. Those who don’t treat flexibility as a two-way street tend to be passed for promotions and bonuses. And, if you’re the primary care-giver of a child, it’s hard to be flexible when the boss calls and says, “Can you be in Toronto tomorrow morning to meet with customer X?”, or “Can you work late tonight? — the system went haywire and the ops on call can’t figure it out. I need you to help them.”.

    As an example, a couple of years ago, I lead a team at a consulting firm. Our customer was in London — we were in Texas, so they were six hours ahead of us. We came in at 2:30am, our time, so we could work the same hours as our customer was working. As a result, our customer was super happy and signed up for additional work, and, thus, everyone on the team got a sizable raise or bonus. If one of the people on the team had been the primary care-giver of a child, there is no way we could have done that — there aren’t any day cares in the area that are open at 2:30am!

  • 5 Steven // Jan 21, 2015 at 9:27 pm

    perhaps more men should be the primary carer for children.

  • 6 Dave // Jan 21, 2015 at 9:31 pm

    I’m all for that, Steven, but it doesn’t seem to be a trend that is taking off.

  • 7 Steven // Jan 21, 2015 at 9:37 pm

    Hence Obama bringing it to our attention and the need for a policy response such as has happened in other countries that now have a (much) lower pag-gap.

  • 8 dullgeek // Jan 21, 2015 at 9:56 pm

    But Steven, isn’t the president crossing into the realm of legislating people’s choices? Meaning, it makes for bad statistics if men and women choose these types of roles, so we need to do something about that. If families are each choosing these roles, then why should the government intervene in those choices?

    It’s one thing if there’s some systemic problem where men and women, with the same experience, same education, and in the same job get different pay. If gender is the *ONLY* difference that explains pay differential then that’s one thing. But Jodi is arguing that isn’t happening, and yet you still think the government should do something about it.

  • 9 Steven // Jan 21, 2015 at 10:45 pm

    But the choices that are available to men and women are different as a result of cultural pressure, biology and a host of other factors. Obama is proposing ways to actually widen these choices for women and men, and not ways to make them for us. It’s not for the government to intervene in those choices I agree, but the government can do something about making those options available without being penalised for it. For example, policy intervention could make me being a stay at home dad a real option, and for my wife to have a career and still be a mum a real option for her.

    I’m pretty sure Jodi is also arguing the government do something about it. It’s just that the something that should be done is different to what would be done in the “equal pay for equal work” scenario. As I said, it’s more about incentives and choices being equal and available to both genders.

  • 10 doyle // Jan 22, 2015 at 5:13 pm

    In my experience in accounting, finance, and large corporations I have never seen any direct gender pay or title discrimination. Great article, simple to read and logical. #loveconomics

  • 11 Michael // Feb 7, 2015 at 1:57 pm

    “you look like you might have ovaries, here’s $0.77 rather than $1. And that’s not what is actually happening”

    …ok srsly I mean this in the nicest way but . . . yes? This is exactly what happens, on every damn level? The other things also happen, too.

  • 12 Michael // Feb 7, 2015 at 1:57 pm

    “perhaps more men should be the primary carer for children.”

    Given the incentives, why would we? The punishment is unrelenting and eternal.

  • 13 Owen // Feb 10, 2015 at 12:48 pm

    That is an empirical claim, Michael.
    Where is the evidence that this is what motivate the bosses? And do we have a list of pay checks of men and women in the same jobs with the same hours, in order that we can prove their is a difference in the way you are describing?

  • 14 Daren Zenner // Mar 16, 2015 at 1:58 am

    Daren Zenner…

    Correcting The State Of The Union, Gender Edition……

  • 15 purple and yellow wedding invitations // Jul 7, 2016 at 4:11 am

    The culture being different might also play into Nike’s I’m not sure, but the "coulda, woulda, shoulda" might take into account what a prudent company would do *at that This is getting well out of my wheelhouse I still predict Nike suffers zero consequences outside of some mild consumer discontent that has no provable effect on

Leave a Comment