Has the wage-education locus for women been worsening?

That question is the focus of some recent research by Chen Huang.

Women’s labor force participation rate has moved from 61% in 2000, to 57% today.  It seems two-thirds of this change has been due to demographics, namely the aging of the adult female population.  What about the rest?  It seems that, relative to education levels, wages for women have not been rising since 2000:

I discover that the apparent increase in women’s real wages is more than accounted for by the large increase in women’s educational attainment. Once I condition on education, U.S. women’s real wages have not increased since 2000 and may even have decreased by a few percentage points. Thus, the locus of wage/education opportunities faced by U.S. women has not improved since 2000 and may have worsened. Viewed in that light, the LFPR decrease for women under age 55 becomes less surprising.

You can consider that another indicator of the Great Stagnation.


Is "education" simply years of education here? Or does it account for field of education? Because without the latter, I'm inclined to dismiss "education" for being a meaningless proxy of useful education.

Gender studies is valuable.

"the Great Stagnation", 4% GDP growth this year. Hmmm! Maybe you need to rethink that who Great Stagnation thing.

Krugman on growth : https://m.youtube.com/watch?v=cZ5dVBa3grw
Krugman on stock market: “Still, I guess people want an answer: If the question is when markets will recover, a first-pass answer is never.”
Krugman on health issues : “Cholera. In a US territory. Among US citizens. In the 21st century. Heckuva job, Trumpie."

When is the left correct about anything? 1st pass answer is never.

In May of 2017, the Trump Administration based its budget on growth of 3.3%. Alan Blinder said that was impossible as was 3.0% -- he said 2%. Blinder argued "to get to 3% you need productivity to increase 300 to 400 percent - does anyone think that's likely?" That was an exaggeration and not sure why he said that. To go from 1.4% productivity to 3.0% is only around a 100% increase.

So far... 3Q 2017 = 3.2%, 4Q 2017 = 2.9%, 1Q 2018 = 2.2%, 2Q 2018 = 4.0% (?) for an average of 3.1%.

So women have been choosing to study economically-irrelevant subjects?

Who would have guessed a degree in Women's Studies only takes you so far.

Wait are you telling me that academic majors that produced such classics like "why we should hate men" in the wapo might not be good investments for evil bourgeois companies?

It may actually be just the opposite. Women's educational attainment has been steadily rising, so there has been more competition among women for employment requiring economically-relevant higher education.


Not vouching for the theory, but supply has gone up.

so this is the 'education is signalling' version of "educational attainment".

Well when they borrow 100k to get a poetry or history degree, that's not going to raise their wages. They're just playing a zero sum status game for the jobs that they could have already had without a college degree. So it's actually worse because of the debt. The government needs to consider more carefully what degrees they subsidize for, or not subsidize post-secondary degrees at all.

Can you imagine the gnashing of teeth if women's studies were defunded?

My oh my ...

It looks like Fat Studies is being defunded across the US. So at least one field of Critical Studies has died a death without anyone caring.

"Once I condition on education, U.S. women’s real wages have not increased since 2000 and may even have decreased by a few percentage points."

Unsurprising for those of us who see education as being mostly a positional good.

I can't download the paper, but starting at year 2000 seems to be a concern, as well as no mention on how the women's numbers compare to men's.

US participation rate was 7.1% higher in 2000 for The US as a whole, and 7.0% higher for women. So nothing to see here.

Commenters on this blog often love to point to degrees of low expected wage value. There are fewer poetry and history majors now than in recent memory, and I just looked up women's studies degree production, which is fewer than 3000 per year in the entire country. Degrees and certifications in low-paying but job-generating fields, such as education and social work, are dominated by women, which is certainly part of the equation. But so too is accounting, physical and occupational therapy, pharmacy degrees and other higher paying fields.

What about all the women getting computer science and electrical engineering degrees?

Oh, wait ...

Agreed, the Blank Studies majors are a red herring. As is poetry and the infamous OWS protestor who majored in puppetry.

As mentioned upthread this is probably an artifact of the base year.

Plus it wouldn’t be Blank Studies.

Anyone who’s been on a campus recently knows it’s biology and psychology that are the gendered majors.

I've noticed over the past 10 years in my field that women have been more inclined to seek flexible work arrangement roles. In fact, I've never had a man request such an arrangement. This obviously is due to social and cultural norms that place women on the front line of child care. This increased desire for part-time roles may be because employers are more open to flexible work arrangements than the past due to changes in telecommuting and other computer driven changes in the workplace. In other words, I can more easily monitor workplace productivity than the past without the need to have a person on site.

In that light, it seems possible that changes in workplace behavior and priorities by women could be accounting for at least a portion of the decrease or "stagnation" of real wages. In fact, it could mean that women's real wages have increased on a per hour basis since the statistics cannot capture these subtle changes in pay. Quite often, the women remain on a "full-time" payroll with a downward adjustment to base pay to account for the decrease in hours worked which is never officially recorded anywhere. For example, I've agreed to a 30 hour work week for several women while they maintain full-time benefits, but with an adjustment to base pay at 75% of the full-time rate which their male counterparts also earned. Effectively, the women received a pay increase when the cost of benefits are included. A statistician would have counted these women as 40 hour employees and a lower base pay. The statistician would also have determined that these women were being "held back" based on years of service, etc. when, in fact, they were working fewer hours than their male counterparts and had no desire to progress to a management level requiring 40 hours or more. Bottom line is that the statistics you see cannot account for the subtle changes in workplace conditions for women that are occurring. The "stagnation" issue may be nothing more than bad data collected by unwary statisticians.

A 1.33 point difference is one third of that metric under discovery. Good work if the researcher spotted it, the null hypothesis looms.

FYI ...


This along with Tom T.'s insightful comment above:

"Women's educational attainment has been steadily rising, so there has been more competition among women for employment requiring economically-relevant higher education"

Are likely explanations.

This post was just Tyler signaling. Again.

The likely culprit here is the cost of daycare, since women are (not always rationally) traditionally saddled with more child care responsibilities. A daycare worker can only watch the same number of toddlers as 50 years ago. That gives daycare a bad case of Baumol's Disease, and on the margin pushes women out of the labor force (despite their rising education).

"Saddled with more child care responsibilities" is an interesting description of the problem.

Ten years ago in my suburb, most of the mothers were college-trained SAHMs. The prevailing theory was, once you are part of a family, you have to do what's best for everybody in the family.

Now, with higher housing and education costs, both parents work and kids are raised by grandparents (in Asian families) and nannies/au pairs/etc. among the rest. Mothers still have to get home earlier than fathers because tradition, and so while more of us are working our salaries have flattened because there are higher priorities than climbing to the top by working 60+-hour weeks.

It's only if you value higher wages above every other human goal that you see this as some kind of retrograde development.

Partial explanation: what were once high-school educated secretaries are now college-educated "executive assistants" who are doing the same work and are not surprisingly getting similar pay.

Seems likely that a lot of this can be explained by the fact that marginal students do less well than strong students, and that most of the growth in education is among the marginal students.

How can a "locus" worsen?

When there are too many of them, you get a plague of locus. Then the rivers turn red and all the first born die.

Imagine a graph with the x axis representing the amount of education (say number of years, assuming people get their degrees on time) and the y axis representing annual earnings from wages or salary. The line is what's called the "locus". If for every x y goes down, I'd certainly call that "worse".

No, it's gone down. Why drag in the value judgement?

Controlling for education creates problems due to ability bias and Simpson's paradox. As college graduation rates increase, the marginal college graduates are going to be more talented than the average high school graduate, but less talented than the average college graduate. As a result, this will make both college graduates and high school graduates less productive on average, and bias average and median wages downwards for both groups.

Comments for this post are closed