7
$\begingroup$

Premise

In the United States, women have been replacing men in the workforce, or rather were from sometime prior to 1976 until the mid-90's. (The general upward trend is from the recent recession, sorry, I'm recycling a graph from another presentation).

People not working

Presumably, this occurred when a more qualified woman got a job over a male counterpart as women entered the workforce.

Given that, one would expect the economy to grow substantially as overall productivity rose.

Yet US GDP growth did not appear to outpace the world over that period, since not all the world liberalized their gender roles simultaneously.

GDP Growth Rate

Question

Why is that?

Is there a flaw in the reasoning, or were the effects simply not visible at the scale I've looked because the data was too volatile or confounded by external factors.

I'd be interested in any studies that dealt with this topic and offered potential causal mechanisms for breaking the presumed link between (again, the presumed) increased efficiency, and output growth.

Thanks!

Sources:

  • LNS10000001 - Civilian noninstitutional population, male
  • LNS11000001 - Civilian labor force, male
  • LNS10000002 - Civilian noninstitutional population, female
  • LNS11000002 - Civilian labor force, female
  • Google
Improve this question
$\endgroup$
7
  • $\begingroup$ This is a valid and important puzzle (though the answer might ultimately be as opaque as "a productivity slowdown in the US happened to coincide with the time when women were entering the workforce"). But I don't follow your logic for why women entering the workforce would necessarily increase productivity; surely it is not always true that expansion of the workforce increases productivity. $\endgroup$ – nominally rigid Dec 20 '14 at 7:37
  • $\begingroup$ My argument is that women entering the workforce without increasing the workforce participation rate is indicative that they were replacing men. If they were replacing men, it follows (I believe) that it is because they were more efficient workers than the men they replaced. Consequently overall output should increase. I hypothesize they were kept out of the workforce for cultural reasons, and that once that barrier was removed they had the same (or similar) talent distribution as men, so that the top 10% of women were free to replace the bottom 10% of men (roughly) every year until parity $\endgroup$ – Jason Nichols Dec 21 '14 at 5:39
  • $\begingroup$ ah. But is "women replacing men" an accurate description of the time series? The secular trends in employment/population are indeed in opposite directions, but the roughly 1965-1990 period of rapidly rising female employment did not coincide with a more rapid decrease for men. (And I don't know why we would expect it to, either; that hypothesis seems lump-of-labor-ish to me.) $\endgroup$ – nominally rigid Dec 21 '14 at 5:49
  • $\begingroup$ no, they have increased their percentage representation in the workforce: from same earlier presentation $\endgroup$ – Jason Nichols Dec 21 '14 at 5:51
  • 1
    $\begingroup$ I know - like I said, the very long-term trends are in opposite directions, but the period of heaviest female entry (1965-90) did not coincide with a period of unusually strong male exit. Male employment/population also declined at a similar pace both before and after the great female employment boom - see trends here. $\endgroup$ – nominally rigid Dec 21 '14 at 5:53
2
$\begingroup$

I'm eyeballing that the unemployment rate of man rose by 3% while the rate for woman fall by 13. Clearly not all of those newly working woman replaced man.

You choose a rather long period with incredible changes to the US economy; The one I think responsible for the effect you saw being the shift from manufacturing to service.

The Situation in 1977 can be seen on Page 10 (4 in the original numbering) of the 1977 Census. It shows 44% of all employees working in manufacturing and 39% in service (Retail + "Selected services industries".

Contrast that to the 1997 Economics Census (Page 20 or 14) where "Retail Trade" alone employed more people than manufacturing - and Services almost double compared to manufacturing.

That shift favored woman (and probably was only possible due to more woman choosing not be stay-at-home moms) which explains the diminishing employment gap.

(The methodology of those reports has changed considerably so the exact percentage rates can certainly not be compared -- but the shift is so big that the conclusion stands unchallenged)

Improve this answer
$\endgroup$
2
$\begingroup$

Answer:
From the research that I have done in order to answer this question, I believe that the following premise is flawed:

Presumably, this [women replacing men in the workforce] occurred when a more qualified woman got a job over a male counterpart as women entered the workforce.

With the hiring of women who were just as qualified (or, in some cases, less qualified) than men, we would not expect the economy to grow substantially, and we would not expect US GDP growth to outpace the world over that period.

My Reasoning:
In the early 1900s, women's rights became an important issue in the Western world. Women began entering the work-force during World War II in order to replace the men who had gone to the fronts. The wages of women, however, were much less than the wages paid to men (in some cases, it was only 60%).
It was not until the 1960s that the women's rights movement, now called "feminism," began to gather momentum, and President Kennedy signed the Equal Pay Act of 1963. This supposedly allowed for wages to be set by the job and not influenced by the sex of the worker. While this law reduced the wage gap, the gap continued to exist and still exists today.

Let us think about hiring employees with a wage gap from an employer standpoint. You have a vacancy in a particular position (with a wage of \$100 for men and \$80 for women) and have the option of hiring either a man or a woman. The man can do the job 90% well (meaning that his is only worth $100 * .9 = $90). The woman, on the other hand, can do the job 80% well (meaning that a man of similar skill is only worth $100 * .8 = $80). Therefore, the man is worth $90/$100 = 90% of his salary, and the woman is worth $80/$80 = 100% of her salary. Thus, the woman gets hired even though she is less qualified than the man (which is the contradiction to your premise).

Improve this answer
$\endgroup$
9
  • 1
    $\begingroup$ That's a contentious speculation. I like it. I don't know that's right, but it's the first time I've heard someone mention the wage gap as an incentive to reduce workforce productivity. $\endgroup$ – Jason Nichols Dec 31 '14 at 22:06
  • $\begingroup$ You're right that desire to hire someone doesn't just depend on their productivity - what matters is productivity realtive to wages. But I don't see how this supports the conclusion that "we would not expect the economy to grow substantially" when many more women enter employment, unless you think that the marginal product of these women is zero or that they are somehow displacing men. $\endgroup$ – nominally rigid Jan 1 '15 at 0:29
  • $\begingroup$ @nominallyrigid My conclusion basically is that women don't necessarily need to do as good a job as the men because they are paid less. As shown in my answer, a woman could do the job less efficiently and still get hired. That would not increase productivity and, therefore, would not cause the economy to grow substantially. $\endgroup$ – Mathematician Jan 1 '15 at 6:18
  • $\begingroup$ @Mathematician, I don't understand. If women with on average 80% of the economy's existing labor productivity enter employment and increase the employment level by 1%, wouldn't we expect a 0.8% increase in output, not 0%? (Unless they displaced men, which doesn't seem likely either empirically or theoretically.) $\endgroup$ – nominally rigid Jan 5 '15 at 6:31
  • $\begingroup$ @nominallyrigid The 80% that I use in my example is the "wellness" that the individual can do the job. The employer hires a man that can do the job better than what they are paid (110% for \$100). The man wants to get paid more than what they can do for the job (90% for \$100). The equilibrium is at 100% for $100. Woman get paid less than men. So, the employer hires a woman that can do the job better than what they are paid (110% for \$80). The woman wants to get paid more than what they can do for the job (90% for \$80). The equilibrium is at 100% for \$80. $\endgroup$ – Mathematician Jan 6 '15 at 14:43
1
$\begingroup$

In the United States, women have been replacing men in the workforce, or rather were from sometime prior to 1976 until the mid-90's.

Women have not been replacing men. Women have just increased their labor force participation (LFP) rate.

Presumably, this occurred when a more qualified woman got a job over a male counterpart as women entered the workforce. Given that, one would expect the economy to grow substantially as overall productivity rose.

You are probably right at this point. At least you hold the similar view with Hsieh, Hurst, Jones, and Klenow (2019). However there have not been enough literature to make a consensus on the level of this effect.

Yet US GDP growth did not appear to outpace the world over that period, since not all the world liberalized their gender roles simultaneously.

I am not sure to what extent US lead the world in rising female LFP. However I am sure that a consensus in the literature is that there is a significant improvement in this dimension around the whole world. Therefore your reasoning can be wrong in a way that US haven't benefit from this trend too much comparing to the world.

Is there a flaw in the reasoning, or were the effects simply not visible at the scale I've looked because the data was too volatile or confounded by external factors.

Yes, there are always other factors that can matter. For example, if you believe that countries that are far away from technological frontier can enjoy a faster growth or that countries with lower wage can benefit from international trade, these gains can offset a better female LFP trend in the most developed countries like US.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.