Social Justice, Current Events

Myths About My Views on the Myth of the Gender Wage Gap

Thanks to being linked on reddit a few months ago, my LearnLiberty video on the gender wage gap is now close to 300,000 views on YouTube.  The video was one of four designed to rebut commonly held economic myths.  In this case, the myth is that women earn 75% of what men do because of labor market discrimination. What’s been interesting about the reaction to the argument there is that both “left libertarians” and “right libertarians” have criticized it.  That reaction tends to be a sign that I probably got it pretty much correct!  However, I do want to clarify a few points I was trying to make there, and have probably made better in longer presentations of the argument at IHS seminars and other venues over the years, so that my basic points are clear. In other words, let’s tackle three myths about my argument that the claim that women earn 75% of what men do is a myth.

1.  Myth:  Horwitz was arguing that labor market discrimination plays no role at all in the gender wage gap.

The point of the video was not to argue that labor market discrimination has nothing to do with wage differentials, but rather to point out that there are a lot reasons and pretty good evidence to suggest that when we control for human capital, compensating differentials and the like, most of the gender wage gap disappears.  The most reliable studies I know put the “unexplained residual” at 5% or less.  No study I’m aware of, other than ones of very specific markets, concludes that human capital and compensating differentials and other economic factors explain the entire gap.  Folks are welcome to see my Economics of Gender syllabus where we tackle human capital, compensating differentials, and discrimination models in one chapter each.

Yes, other disciplines might provide arguments and evidence for a larger role for discrimination.  Great, let’s have at it.  Economists need to take them seriously and they need to look at the economic evidence.  Let’s see where it goes.  So for my left-leaning critics:

Reality:  Horwitz was arguing that only a small part of the gender wage gap remains unexplained by the factors discussed in the video, and that remainder may well be due to discrimination.

 2.  Myth:  Horwitz thinks sexism is not causally related to the gender wage gap.

It seems like some people didn’t watch the last minute of the video as they think I was denying the existence of sexism or that sexism plays any role in economic outcomes.  If you watch all the way through, I point out at the end that sexism in how children are socialized into gender roles can indeed affect economic outcomes.  If it’s true that girls really are discouraged from math, for example, or if they grow up believing that certain jobs are “women’s jobs,” then they will acquire their human capital accordingly, and may well earn less than the very men they went through school with whose human capital looks different.

It also matters that men don’t do as much as women do with respect to household production.  Equalize the second shift more and you’ll likely make wages more equal in the process.  Why men don’t contribute more is a matter of serious contention, but it’s certainly possible that misogyny is part of the story. So yes, sexism matters.

However, that is not the same thing as saying that the gender wage gap comes from labor market discrimination.  It is not labor markets that are the problem here, but the socialization of children that comes before they enter the labor market.  Change the socialization and labor markets outcomes will pattern differently by gender.  So yes, some portion of the gender wage gap might be due to sexism, but not necessarily or only in the labor market but elsewhere in society and the culture.  If we think that sexism is bad, or if we want to see more equality in wages, then attack the sexism in the places it has the most influence, but don’t imagine that labor market interventions such as anti-discrimination laws or worse, comparable worth, are going to solve the problem. So for my left-leaning friends again:

Reality:  Horwitz thinks sexism is causally related, but mostly via the socialization processes that influence the choices people make as they acquire their human capital before they enter the labor market, with labor market discrimination playing a (much) smaller role.

3. Myth:  By accepting the argument that misogyny in the culture indirectly explains the gender wage gap, Horwitz is tacitly endorsing interventionist solutions, not to mention making nice to the evil feminists.

From the right side of the libertarian spectrum comes the charge that all this talk of socialization and discouraging girls from math and science is all a bunch of lefty PC talking points that aren’t really true.  And from some quarters is the suggestion that if one believes that stuff, one is somehow tacitly endorsing government-enforced solutions to them.  For what it’s worth, I’m persuaded by the evidence I know of outside of economics (gasp!) that those problems are real, at least to some degree.  I would agree that they continue to lessen over the years, but they have not totally disappeared.  And there’s no doubt about the evidence on men’s contributions to household production being low.  Data from 2004 had women doing twice the hours as men.  Granted that’s down from 8 times the hours 40 years earlier, but if economic theory is right in suggesting that hours spent on household production should be in inverse proportion to the relative wages of the couple (think opportunity costs), then unless the average woman is making half of the average man, men are under-contributing in the household.  Again, there’s a number of possible reasons why, but the fact is there for all to see.

Accepting that sort of evidence hardly means one sees interventionist public policy as the way to deal with it, although changing some polices, such revamping the tax code to end the secondary earner bias by instituting a nice low, flat tax would sure help.  Sexism is a cultural problem and there are plenty of ways to try to work on the cultural side of matters to address it.  It may also cure itself as younger generations adopt different views from their parents, and the trends of the last 40 years indicate that this seems to be happening.  One can be hard-headed about the market generally not being a source of discrimination, while being soft-hearted about changing the culture in ways that might lead to different, and more equal, market outcomes.  Not every problem demands a political solution and persuading men that it’s a good thing to take more responsibility for raising their kids and washing the dishes is a non-policy way to address that sexism, as is more generally persuading recalcitrant men to look at women in the workplace as equals. So for my right-leaning friends:

Reality:  The feminists are not only not evil, they are right about a number of the problems, but wrong about the role markets play in causing them and that the state plays in potentially remedying them.  Acknowledging the existence of sexism doesn’t ipso facto commit you to any particular way of dealing with it.

Cross-posted at Coordination Problem.

  •  I want to both defend
    and criticize Steve’s positions. Let me start out by saying that I consider
    Steve one of the good guys in the libertarian movement. I want that to be very


    First the defense:

    Steve says “From the right side of the libertarian spectrum
    comes the charge that all this talk of socialization and discouraging girls
    from math and science is all a bunch of lefty PC talking points that aren’t
    really true.  And from some quarters is the suggestion that if one
    believes that stuff, one is somehow tacitly endorsing government-enforced
    solutions to them.”


    The ones who say this are know-nothings and lack any
    critical thinking faculties whatsoever on this topic.  They have a knee-jerk reaction to anything
    remotely perceived as PC, so their anti-PC stand is just as sheep-like as any
    left-winger’s might be.  I have taught
    Psychology of Women for over 20 years and I could write a whole book on how
    gender socialization affects both women and men, telling them in subtle and not
    so subtle ways what abilities they are capable of, what jobs and careers they
    should seek, how they should behave, and much more that affects what jobs and
    careers women and men seek. All this is based on peer-reviewed data done with
    scientific studies appearing in academic journals. I dare any of these
    know-nothings to come up with scientific data that says otherwise. They
    wouldn’t be able to because it doesn’t exist. I can run circles around them any
    day because their views are based on their own ideological bias free of any
    constraints of knowledge or critical thinking. Mine is based on hard data. They
    lose, I win. And so does Steve because he is entirely correct on this part of
    his video.


    As for the silly assumption that this position means we
    endorse government-enforced solutions: Steve’s libertarian credentials are
    impeccable and mine are even more impeccable. I have been a libertarian
    activist for nearly 50 years, long before some of Steve’s critics were
    undoubtedly even born. My credentials, which if I may immodestly say, are
    considerable, can easily be found on Facebook or Google. Nowhere have I ever
    advocated government solutions and I daresay that the same is true for Steve.
    So, critics, put up or shut up. It is your own fantasy, not any reality. You
    are pathetic. Go away.


    However, to move on to my criticism of Steve’s
    video…I  could also write a lengthy
    essay on discrimination against women and how it affects not just what jobs
    they apply for, but what jobs they get, how much money they make, how they are
    valued, and how likely they are to advance to higher-paying jobs. Since the
    studies on which my claims are based are done by psychologists, I can’t come up
    with a dollar-figure, but I can provide evidence that clearly shows that
    discrimination has an indirect impact on how much money women make. Steve
    didn’t discuss that in his video and I think he should. Any evaluation of such
    complex phenomena needs an inter-disciplinary approach that I rarely see among


    Here are just some of the references and links that support
    my contention that discrimination against women in the workplace plays a larger
    role in pay differences than Steve suggests in his video

    Even when women ask for raises, they don’t get as much money
    as men do.

    This research article points out how
    discrimination can subtly affect women’s chance to advance and be recognized as
    a leader. This in turn affects how much money she makes. Here’s one of the
    points from that pdf: “Even with equal qualifications and achievements to
    those of their male counterparts, women are perceived less favorably in terms
    of their abilities and accomplishments–this stereotyping tends to be reflected
    in evaluations and promotions and places women at a disadvantage for advancement.”  And less advancement
    means less pay!


    To continue from the article linked
    above: “Women are also more likely to be perceived as ‘family-focused’ and
    ‘unwilling to travel’ and therefore tend to be passed up for promotions.”  “Given
    equivalent qualifications, women are less likely to hired and promoted,
    especially for roles that are traditionally stereotypically masculine, such as
    engineering.” So Steve is saying women are less likely to be engineers.
    Well, sure if they are less likely to be hired as engineers, that will slow
    them down. BTW women who go into engineering often face discrimination in the
    college classroom and have a higher dropout rate.


     I also suggest the  Journal
    of Social Issues, vol 57, 2001: Gender, Status and Leadership. Many articles
    there which show that women have a tough time being perceived as leaders.
    Another reason why they are less likely to advance to better-paying positions
    than men.


    Here are just some of the other
    findings that can affect women’s advancement and therefore their pay.
    [Individual citations upon request. Immediate source = A New Psychology of Women by Hilary Lips]. 

    The quality of women’s work tend to be undervalued.  The same piece of work is frequently
    evaluated less favorably when it is attributed to women than when it is
    said to have been done by a man.Besides unfavorable evaluations of performance, women
    also often receive other negative assessments of their competence. Their
    success is often devalued by being explained as “luck,” and less likable
    than competent men.Women’s jobs are widely perceived as “secondary” and
    men’s job as “primary,” thus resulting in (among many other things) men
    having greater influence at work, and the continuation of the definition
    of equality at work as equality with men who do not have home
    responsibilities. This hinders many women’s chances for advancement
    regardless of their actual home circumstances.


    Here are some results reported in Half the Human Experience by Janet
    Shibley Hyde:

    Managers are more likely to select a man rather than a
    woman for important assignments (e.g., a “major sale”). Whose resume is
    likely to help them advance?Though there are no actual gender differences in the
    effectiveness of leaders, women are less likely to be perceived as
    effective leaders (though this interacts with leadership style).  Who is more likely to advance?


    These are only some of the ways that
    subtle and not so subtle discrimination against women and stereotypes about
    women in the workplace affects how much they can advance in their jobs and
    therefore how much they can be paid. So it is not merely a matter of “choices
    and preferences” or even gender stereotypes that affect which jobs women pick.  Economists can’t just look at differences in
    job preferences; they have to look at what goes on within companies before they have a true understanding of gender
    pay gaps. Only an interdisciplinary
    approach can give us an accurate picture. Steve, want to collaborate?

  • “but if economic theory is right in suggesting that hours spent on
    household production should be in inverse proportion to the relative
    wages of the couple (think opportunity costs), then unless the average
    woman is making half of the average man, men are under-contributing in
    the household”

    Perhaps you could explain how economic theory leads to that result? I would have said that economic theory implies that, ceteris paribus, the individual with lower wages should do more of the household production, but it does not imply, or even suggest, proportionally more.

    Consider the simple model where output in household production is proportional hours spent doing it,  market income similarly proportional, and the partners are equally good at household production. Isn’t it obvious that the partner with the lower market wage should do all of the household production?

  • Because I hate apps and refused to link from my Facebook account, I had to use a Yahoo account that has no information about me. HAHA.  But it left off my last name.  My name is Sharon Presley. I have a Ph.D. in social psychology; I am the co-founder of Laissez Faire Books, the Executive Director of the Association of Libertarian Feminist; co-editor of “Exquisite Rebel: The Essays of Voltairine de Cleyre” and a whole bunch more other stuff. You can find me or ALF easily on Facebook.

  • And my apologies for the formatting of the commentary I wrote. Not used to this blog and its formatting restrictions!

  • David: Only in a sexist universe where “fairness” is arrived at with with a calculator by an economist.  What are you saying? That the only criterion for “worth” is how much money you make? Do I misunderstand you?

    • blogster1

      you clearly have no understanding of economics.  Factors such as supply, demand, risk, competition, market regulation, externalities etc all affect salaries. 

      Quit using the shaming language and put a logical argument.  I noticed that in the studies you reference above you do not quote any statistics, nor the methodology or assumptions the studies were based on.  It’s all generic and qualitative.

       By and large the market decides salaries and wages.  Some industries are government regulated (e.g. education) and funded, while others are largely free market (e.g. professional sports).  They are subject to different forces and therefore different salary outcomes.

      If I work mining coal and demand for coal from China is high – guess what – like what is happening in Australia – pay is going to be high.  Simple.  Understanding economics might help you instead trying to ram the discrimination peg in the economics hole, despite it not fitting.

      You might say pay is not fair – maybe.  But it has nothing to do with sexism!

  • David: Only in a sexist universe where “fairness” is arrived at with with a calculator by an economist.  What are you saying? That the only criterion for “worth” is how much money you make? Do I misunderstand you?

  • I’m probably getting in over my head here, but I’d like to suggest that the problem of gender bias may go deeper than market forces, economics or taxation. It may be related to how money enters the economy, how money is spent into the economy, and how decisions are made as to what projects to favor when money is spent into the economy.

    In an attempt draw an analogy, if I’m not mistaken, the normal way a human body makes blood is via its bone marrow. But consider that a patient is not allowed or encouraged to allow blood to enter its body this way, but only through artificial means that are pre-determined by doctor/hospital protocol.

    Stated differently, to resuscitate an argument that was occurring directly after the Civil War regarding the labor of former slaves (and should apply to modern women), money is supposed to enter the economy directly through one’s labor/wages as “property,” not from central planners in New York or DC as “income” derived from a regulated currency. 

  • Anonymous

    Erm, while I have to agree with most of this, I have to point out that if you think ‘ If it’s true that girls really are discouraged from math, for example, or if they grow up believing that certain jobs are “women’s jobs,” then they will acquire their human capital accordingly, and may well earn less than the very men they went through school with whose human capital looks different.’, you then can’t assert that sexism _in the job market_ is not that relevant.

    Let us assume we are in a parallel universe, where instead of Janitor, there are two different jobs, called Kanitor and Lanitor. Let us postulate that what in our universe is done by Janitors is functionally divided between them: Kanitors empty wastebaskets, replace bathroom supplies, etc, and Lanitors clean floors and windows.

    Now, thanks to some weird gender stereotyping, Kanitors are 80% male, and Lanitors 80% female. This appears to be entirely voluntary, as far as everyone can tell. Little girls play with mops, little boys play with wastebaskets.

    Oh, and Kanitors get paid 30% more than Lanitors. …yeah, something seems a bit off here. Who decided what the value of their job is?

    In our universe, we don’t have Kanitors and Lanitors, but we do have exactly this sort of thing. It’s not enough to say ‘Women often choose to go into low paying jobs because of socialization’.

    At some point you have to ask, why, exactly, are jobs that women are socialized to go into low paying to start with? Why does teaching pay so badly? Why does nursing pay so badly? Hell, why do _all_ jobs that have to do with being sociable and interacting with other people (Which women are ‘good at’, and you can chalk that up to either society or actual gender differences, I don’t care), pay so shitty?

    It’s one thing to point out that the relative differences _inside_ a specific job are 5%. (Which, I must point out, is an average, and there are plenty of circumstances where it is much higher.)

    But that assumes that the average in an industry is somehow set by magical external forces and couldn’t possibly itself be sexist based on how much we, as society, value work we see a ‘men work’ or ‘women work’.

    • The Woman

      Yeah, this.  Isn’t there research showing that as fields change from female to male dominated, or vice versa, the rate of compensation changes?  Examples which leap to mind: teaching was originally a male-dominated profession, and in the early 20th cen became female dominated and the wages collapsed.  Likewise, psychotherapy was originally male-dominated and now is female-dominated, and the wages have been collapsing.  I have heard computer programming was originally female-dominated, but became male-dominated and wages surged.

      Now, it’s entirely possible that if there is such a correlation, it’s a product of the same gendered differences in the preference that Steve alludes to in his video (which I quite like): such as, women prefer part time work at a greater rate than men, so perhaps when a field is female-dominated, part-time positions are more normative, with the expected drop in hourly compensation  (I have reason to think this may be what happened in psychotherapy).


  • Why are women medical doctors more likely to become family practitioners and work in philanthropic hospitals while men medical doctors are more likely to become specialists and have private practices? This self-segregation certainly affects wage differences among doctors. As has been suggested, men and women tend to self-segregate into different fields, for the most part. No question much of this is social, but one does have to wonder how much is biologically-driven. And of course the biological gives rise to the social. I particularly recommend the work of Kenrick on evolutionary psychology and complex adaptive social systems along these lines.

    • The Woman

      Depends on what you mean by “biologically driven”.  If you mean “sugar and spice and everything nice, and that’s what girls are made of”, i.e. that women are naturally more caring than profit-oriented, in comparison to men, and thus gravitate to professions that feel more comfortable based on those presumed values… not so much.  If, however, you mean, “aware as they are of their vulnerability to pregnancy, they make steely-eyed decisions to pursue forms of work which may not be as lucrative, but are more reliable sources of income”, then maybe yeah.

      Look, I’m working on opening a private practice (not in medicine, but in a medical field) this year.  The only reason I could possibly do such a financially risky thing is that I do not and will never have children; I don’t have to worry that my availability to nurture my burgeoning practice will be suddenly compromised by doctor-ordered bed-rest to prevent miscarriage, or needing to take six weeks to recover from complications during an emergency caesarian, or my infant needing open heart surgery, or having to figure out breast feeding on a job I can’t possibly have a child at.  While it’s possible that something untoward could happen to my health — it could happen to any of us — I’m not planning on taking up sky diving or rugby, either.

      Starting a small business is risky.  So is pregnancy and childbirth.  I think a lot of women reasonably conclude that if you’re planning on doing one, maybe you shouldn’t be planning on doing the other at the same time.

  • Pingback: Friday Sex Links! « Sex with Timaree()

  • Pingback: Gender Gap Data | Economic Thought()

  • None

    Oh, feminists aren’t using the government to get what they want? Well, that’s a relief. It’s a good thing we pay these female professors to study how they’ve been oppressed. Surely, they’ll be good social engineers and try to help men stop discriminating against women. Oh, wait…

  • Pingback: Christina Hoff Sommers’ “Maternal Feminism”: Same Old Sexism, New Packaging()

  • Pingback: I’m Whatevs About Mansplaining, But Not About Culture()

  • Yes, this post was worth revisiting (H/T Boettke). 🙂

    Thank you Steve.

  • Pingback: I’m Steve Horwitz, Professor of Economics at St. Lawrence University. My article, “There is No Such Thing as Trickle-Down Economics” recently made the front page of Reddit. AMA! - MLG $MOKER$()