Populations M and F have different median weekly earnings, Mdn[sub]M[/sub] and Mdn[sub]F[/sub]. Thus the overall gap that you want to discuss, 77% (or whatever – looks like it’s narrowing since that reading in 2012), Mdn[sub]F[/sub]/Mdn[sub]M[/sub]. A subpopulation of M that works deadly jobs, d, which has a lower median weekly earnings than M (Mdn[sub]d[/sub] < Mdn[sub]M[/sub]), cannot make the gap between mM and mF larger. So not only does d not contribute to the gap, it *decreases *the gap. No regressions needed. Men choosing low-paying jobs that lower the overall pay of men cannot increase the overall pay gap. That’s true even *if *deadly jobs pay more than jobs that would otherwise have comparable duties. Raising my wage from $600/week to $700 doesn’t change the gap if Mdn[sub]M[/sub] and Mdn[sub]F[/sub] are $830 and $750, respectively.
It IS different because most daycare workers are women and they often get quite close with the kids. You see them cuddling them, reading to them, wiping their noses, and comforting them when they are hurt. I know many daycare children who will take their first steps to a daycare worker. Say their first words to a daycare worker. That daycare worker might pottytrain the kid.
So yes, they often are the only “mommy” that kid knows.
Males are few and far between in daycares and usually only with the older ones.
I find more women than men have “hobby” jobs.
For example, lets say its a woman who really likes sewing, quilting, pottery, painting, scrapbooking, or crafts. Such a woman might open a store catering to those. Such a store might not make much money (it might barely even pay the rent), but can be alot of fun.
Running a bed and breakfast might be similar. Also being an artist, musician, or writer.
They might have a spouse who has a good job, a parent who bankrolls them, or they might be single and living with several roommates.
Now the thing is such a “job” wouldnt make much money. But the person running such a business would get tons of job satisfaction.
Such a woman would also be a part of that wage gap statistic.
I want to point out, first, that while the gender wage gap is most commonly measured with the median, this does not exclude analysis of the mean.
The sub-discussion here has been compensating differentials – better pay for more dangerous work, at the same skill level – which are conventionally measured with normal regression techniques (i.e. condition mean analysis), so I’ve been thinking in means and using the specific word “average”, not “median”. But we can talk in the language of medians. It doesn’t change anything.
This last assertion is flatly mistaken.
This is a basic math error, a problem with medians. The only way for risky work to possibly “decrease” the median wage for men (compared to work they would otherwise do!) would be if men who earned more than the middle-earner suddenly decided to leave their relatively high-paying job to take on high-risk work that was lower-paying than the middle-earner. The way for the median to change is for the person in the middle to change, and risky work can only change the middle-earner to a lower-wage person (“decreasing” the median) if risky work somehow attracted high-earners into lower-paying and risky jobs, from above the median to below the median.
It is not reasonable to posit that high-earning men generally give up those good positions to earn less at jobs that are very risky.
The literal statements here are technically true, but miss the point entirely.
Just before, the assertion was that men earning more from dangerous work could “decrease” the median wage gap, if they happened to earn less than the median. That was not at all true. Here it’s noted that a shift into a higher-paying but more dangerous job might possibly leave the median unchanged. This is entirely true. It could possibly leave the median unchanged if the effect were exclusive to earners below the median, and disappeared by the time we reached the middle worker, so that the middle worker continued to earn the same amount.
___
| |
| |
^ v
xxxxxxxxxxOxxxxxxxxxx
But what this requires is that literally every risky job in the country is below the median. If even a single worker below the median happens to earn extra compensation, above the middle-earner, for taking on a riskier career despite no change in skills, then the middle-earner will change.
________
| |
| |
^ v
xxxxxxxxxxOxxxxxxxxxx
Just a single occurrence of this can change who is standing in the middle of the male wage distribution. It won’t be a big difference, just as one individual getting a better job will not measurably push the mean. But to make the point yet again, the vast majority of workers in the most dangerous jobs are men. This includes, for example, oil-rig workers. It happens to be the case that the least skilled workers on an oil-rig earn around the median wage. The last time I visited an oil rig, the grizzled foreman (huge ugly scar down his face, who said he was also missing some toes) informed us that there were workers not terribly long out of high-school who were quickly earning near six figures. I saw no women on the job site. Ordinarily, we should want to have a more representative sample than this. It would be reasonable to be skeptical of anecdotes like this… except we’re talking about the median now, not the normal average, which means one example can literally be all that is necessary. Those men would simply not have been earning the same amount if they’d chosen less dangerous work, with the skills and education that they had. Just one of them going from under the median to above it can change that median worker.
A lot of very dangerous jobs don’t pay well. But there are some that pay above the median, the workers are largely men, and the alternative jobs for at least some of those men would be below the median. And in fact, I believe I have personally met some of these men.
The statistical issue here is looking at potential changes in medians in this manner tends to be hard. The effect is not large, compared to other things that affect median wages. But the point is that the wage effect exists. The things that push the mean of a population around will also tend, in most cases, to push the median around as well.
This is why I’d prefer to talk about means when discussing compensating differentials, because the logic is slightly easier. It’s fairly straightforward to run a multiple regression to isolate the effect on wages from risky jobs for basic prediction (if not full inference). Talking about medians can be a little more tricky, for people without much experience, which is why it’s so easy to assert the mathematical mistake that taking on risky work for better pay might possibly “decrease” the median. It’s the trickiness of medians that leads to errors like that, where a discussion of the mean is more likely to avoid that flavor of error. It’s not a large effect, not “critical”, but it’s there, and I’ve been responding to the mistaken assertion that it doesn’t even exist. “Dangerous jobs actually don’t pay a premium and are irrelevant to the wage gap discussion.” This is simply false, and switching from means to medians and back doesn’t make it any less false.
The bottom line here hasn’t changed.
Dangerous work pays better than safer work, at equivalent skill levels.
Men do almost all of the most dangerous work. Those men get paid better than women who have equivalent skills who choose safer work.
This effect is a small part of the raw wage gap (median or mean). It’s not “critical”. There are much more important things going on.
But the effect still exists. The logic here is not difficult, despite the tricksome nature of medians. The weird defensiveness around this relatively simple idea is unnecessary, and counterproductive. We need to be able to discuss simple facts reasonably if we want to eventually discuss more complex facts.
Along with your “dangerous jobs=more money” thread, I wonder about the effects of the high pay by professional athletes, who are mostly men, comes into the picture?
Along with the players their are the coaches, trainers, scouts, etc… This goes all the way down to the high school level. Note - the NFL has just 1 female referee.
Even most womens sports like womens basketball and softball have male umpires and referees.
Dangerous work is not a big effect, but it visibly moves the needle a bit when looking at large national samples. It’s not critical. It’s small compared to the effect of “knowledge” (education, skills, etc.), but at least it can be seen.
I very much doubt professional sports would visibly move the needle at all, especially when considering those athletes in the minors who earn (relative) peanuts just on the hope of getting a major payday in future. Despite the flashy salaries, they’re a minuscule part of labor force. The effect is not literally nonexistent, but I just can’t believe it would be perceptible in any broad sample.
A quick back of the envelope calculation: there are 32 NFL teams with a salary cap of around $180 million each. Not every team ends up flush against the cap, but we can guesstimate that NFL players take home something like $5 billion annually. NBA, NHL, and MLB rosters are smaller, but certainly the total pay in the top level professional sports must be order of magnitude $10 billion. The minor leagues obviously pay quite a lot less, but those players often supplement income by working jobs in the off-season, out of sheer necessity. So the amount they make below the mean probably decreases that $10 billion figure somewhat, but I suspect that’s the right general order of magnitude, i.e. it’s probably closer to $10 billion than to $1 billion.
There are order of magnitude 100 million working men, so professional sports raises the mean pay for men by ~ $100 annually, give or take a relatively large percentage. That’s more than I would have expected, but I don’t see an obvious error – which could just mean I’m oblivious! – and I suspect I’m in the right ballpark, as it were.
And again, your only looking at players. There are also well paid coaches, trainers, and scouts.
Then there are some marginal sports - rodeo, pro soccer, boxing, MMA, car racing, etc… hardly a female around.
If you look at tennis and golf only a handful of women make any kind of real money compared to the men. And even then, most golf and tennis coaches are men. Here is alink to Venus Williams’s coach.
The gender pay gap is one of those oft trotted out “facts” that few ever question. President Obama said in a speech that women make $0.77 for ever $1 men make and people just accepted that is how it is (to be clear Obama was repeating a long claimed number with that…it was not new).
The number is arrived at by simply figuring out the median of how much men make and how much women make and divide the two you get the $0.77 figure. But that is a really blunt way to look at it that misses a lot of nuance.
Right on the face of it your bullshit meter should start going off. Why would a company pay a man more money if they can get a 23% price break on a woman doing the same work? Even if the company is run by a bunch of misogynist pricks the economics will force their hand eventually.
There is a wage gap but it is not nearly as big as that $0.77 claim and what does exist has little to nothing to do with discrimination (doubtless there are some anecdotes). More like $0.95 area. Most of the rest of the difference has an explanation.
If you want details on it I recommend listing to the Freakonomics podcast on this (or you can read the transcript). I will quote some and link it below:
Did you read Hellestal’s post? He says exactly this, but then he goes on to make the point that it’s still potentially problematic that women overwhelmingly make “choices” that leave them so economically disadvantaged. It’s true that people have the right to make choices and we don’t want to undercut that. But there’s also something disturbing about the fact that the outcome of those choices leads to such radical differences. Women remain responsible for so much of family life: not just taking care of kids, but also parents and in-laws at the other end of their career. That women are making the choice to live up to a responsibility that so many–male and female–see as uniquely theirs, and so taking a hit to their earnings power doesn’t mean that the 77% isn’t “real”–it means it has causes that are much more complex and harder to solve.
I also think that, given recent events, we have to consider the real possibility that a non-trivial number of women have made bad career moves as a consequence of sexual harassment–either they left a job for a less desirable one to escape an untenable situation or they had their career path disrupted in retaliation. Years later, that looks like they took a less rigorous career path and now are doing a job that pays less.
What I am saying is that it’s not enough to say “The wage gap is mostly the result of women choosing to take care of family and avoid toxic situations, not sexism” because the fact that women have to make those choices–and men don’t–is itself institutional sexism.
Finally, I don’t buy the argument that the invisible hand would fix wage disparities. People aren’t paid what they are worth, they are paid based on what someone thinks they are worth. IF employers decide that mothers are less productive/reliable, if they think women tend to be “catty and back-stabbing”, confirmation bias will find plenty of evidence, and that will be reflected in pay. It’s quite possible that Bob gets promoted over Lisa because they perceive him as adding more value, even when it’s not true.
I agree it is not fair that the decisions to prioritize family (be it to raise kids or care for an elderly parent) over work has fallen mainly on women’s shoulders. We could say that it is unfair and/or outmoded societal norms that leads to this but I am not sure that is true anymore. Maybe it once was but now women are able to join the workforce and yet are choosing to be homemakers.
It is not unfair for a man who has put in 20 years of work at a job to be paid more than a woman who put in 10 years and took 10 years to raise her kids. It would be unfair if we made women leave the workforce to do that but it seems more often than not women choose to do it of their own accord.
Also, I am not convinced sexual harassment plays a large enough role in this to be statistically significant. I am not saying sexual harassment does not exist. Far from it and certainly it plays a substantial role in some industries like entertainment. But even here it is not cut-and-dried. For instance female models are paid substantially better than male models despite doing the same work. Is this sexism or something else?
I think the invisible hand would work here even in the face of misogyny. A 23% wage difference is big. Unless men are 23% better at work than women it makes no sense to pay more for the same job. At the end of the day the bottom line reigns supreme in corporate America and some middle manager under pressure to cut costs will figure out that a woman can do the job as well as a man and cost 23% less is the way to go. Their competitor, Misogyny Inc. down the street, will soon either have to follow suit or go out of business.
Shareholders want maximized returns and rarely care one whit about anything else.
Women still have to choose between having children/having them raised in the way they think is in the best interest of those kids and having a career. Much more often, men can have both.
Many women don’t stay home/work a leas demanding job because they want to. They stay home because they are putting what they believe is best for their kids over what’s best for themselves. That’s a burden we still put disproportionately on women.
Everyone has choices to make. It is a burden if the choices are made for you or you have no choice. If a woman chooses to have children and chooses to stay at home I do not see how it is a burden. Why isn’t it a burden for the person who now has to work to support the family?
As that poll above shows a majority of women these days seem to prefer, as a choice, to be a homemaker. They don’t have to. They could choose to work instead but they don’t. Also, while still rare, stay-at-home dads are much more of a thing these days than they once were so there are options and the couple needs to work out what is best for them.
Here’s a short Pew piece on stay-at-home dads that I’ve only glanced through: Growing Number of Stay-at-Home Dads | Pew Research Center
They really shouldn't describe the people they are talking about as "stay at home dads", though,because that gives the impression they are talking about men who have chosen to stay at home specifically to care for their home and/or family. They aren't really. Of the 2 million "fathers who don't work outside the home" , only 21% say they are home mainly to care for their home or family. 23% are home mainly because they can't find a job, 35% are home due to illness or disability and 22% are home for other reasons, such as being in school or retired. Those home because of illness , disability or an inability to find work would not doubt be working if they could, and those who are mainly home because they are retired or in school almost certainly no doubt be retired or in school even if there were no children under 18 in the household.
21% up from 5%, so certainly movement. But still small, yes.
There’s a very important consideration that has been missing from this thread so far, including my own posts: one big reason the woman tends to be the partner to stay home is that she’s usually the one who earns less in the first place, so there’s less of a hit on the family income. Sometimes her job simply doesn’t offer flexible hours. And sometimes her pay wouldn’t cover the cost of childcare.
I respect women who stay home with their children. I respect women who can’t or don’t. But I get frustrated with people who try to guilt-trip women into staying home–“Your baby will call some caregiver ‘Mama’!” “You’ll miss his first steps!” “Strangers will raise your kids!”–and then blithely shrug off the wage gap with, “Hey, it was your choice to stay home!”
The poll you quote (and referred to again in a later post) asks women what they would rather do. It doesn’t ask why. I’d rather stay home with my son, but not because I like it–because I think he’d like it, and it would be better for him and our family. And I say that even though my husband is a SAHP, which makes me poignantly aware of how few women have this choice. In my experience, working moms who want to stay home say they want to stay home because they feel so guilty about leaving their kids, and stay at home moms are staying home largely because they think that’s what’s best for their kids and their family dynamic.
No one is suggesting that women get fast-fowarded to the salary they would have had if they hadn’t left the work force for ten years. But it’s a fact that a lot more women have to chose between being childless and taking a hit to their earnings. Many fewer men have to make that choice. They and their wives both still assume that it’s the woman’s inherent responsibility to do so. It’s incredibly naive to think that’s just gone.
There are also structural changes that would make this less of a problem: recasting good, professional jobs as part-time, allowing couples to both work half-time, or one person work half-time and at least keep a career trajectory. High-quality pre-K for all started at 3 or 4. Shifts in expectations around “emotional work”: having to be the events coordinator, social media intern, community liason etc., etc. is a real burden that limits women’s opportunities.
Pay differentials are not sexual harassment.
Let’s say 10% of women have left at least one job over their career that was otherwise a good career choice because of either direct sexual harassment or a hostile sexist environment. I don’t mean getting paid less–I mean being touch inappropriately, or feeling like they were in a situation that was headed in that direction. I would also include things like co-workers that engage in very crude sexual conversations around you, or try to engage you in inappropriate sexual conversations. I mean feeling unsafe or routinely humiliated. Would you consider that statistically significant? Because I would, and I think that’s a low estimate.
It doesn’t have to be 23% for it to be meaningful. No single factor is going to be responsible for it. But it’s foolish to trust the invisible hand to get it perfect. It responds to perception, not reality. Do you really think that there’s no effect from that sort of bias?
Making a choice is not sexual discrimination. For whatever reasons a woman chooses to stay at home the point is it is now a choice she can make.
Sexual harassment certainly needs to be addresses in society but I am not following how this relates to pay differences between men and women.
Please read/listen to the Freakonomics episode I linked to earlier. The professor studying this issue notes that pay differences have very little to do with sexual discrimination (as in paying William more than Mary just because he’s a guy). That is not to say you can’t find anecdotes where it happens but overall pay disparity has other causes.
Let’s look just at data for a moment. I read the Freakonomics transcript. Nothing new there. Goldin basically took a very simple look at a complex issue. That’s fine for his intended audience: people who mistakenly think that the gender wage gap means that McDonald’s has a separate and lower wage scale for female burger-flippers than male burger-flippers.
But there IS a gender-based pay differential in jobs with merit-based promotions and bonuses.
- And according to research presented in Fortune magazine, women’s performance reviews tend to be more critical
AND the criticism tends to be more personality/behavioral vs. the skills-based criticism of male employees:
This is the stuff that determines merit bonuses and promotions. Merit pay and promotions are one key factor that determine which parent earns more. The parent who earns less is more likely the one to stay home.
There’s a real issue here. It’s not merely one of imagined grievances.