by Tyler Cowen
on January 14, 2014 at 2:03 am
in Current Affairs, Data Source, Economics, Uncategorized
There is more here from Matt Yglesias.
I didn’t look but I presume he is calling for more discrimination against men?
No. Try reading.
No. Nobody in his or her right mind has ever called for that. The “end of men,” to the extent it’s happening at all, is a symptom of inescapable economic realities, not some conspiracy cooked up in a Barnard ovular.
Go back to ROK. Grownups are talking here.
No conspiracies needed. It’s in codified law. Grownups indeed.
The graph tells you something only if you assume that the size of the working-age population has not changed since 2003, right?
Indeed. Without being corrected for population growth, this graph doesn’t mean a great deal.
That was a strange choice of y-axis scale to plot it in absolute numbers than percent employed.
The general point made in the title of the post will stand unless you believe that the gender ratio of the American population has skewed massively away from its value in 2003.
I don’t think that’s true if the population is changing, even if the gender ratio is constant.
Does the absolute change in working-age population matter if the male/female composition of that population hasn’t changed?
I think if the population was increasing and the male and female employment level expressed as a percent were constant, then men would appear to be pulling away from women on this graph because the gap between their absolute employment levels would be rising.
So the graph is poorly designed and misleading, but probably in a way that understates the problem for men.
Also, why would you design the labels such that you need to look at the bottom right and then look at the title to find out what that means? It’s like this was created by a child…
W.r.t. the claim being made, the relevant data points are “male employment pre-recession vs. male employment post-recession” and “female employment pre-recession vs. female employment post-recession”. As long as the male/female ratio of the working-age population stays constant I don’t see how an increase in the working-age population should matter w.r.t. those two data points.
Strictly speaking, that’s correct. It’s just that that isn’t a complete way to talk about the question. The problem should also be discussed per person if you want to understand the relative effects of the recession and recovery on men and women. That’s why changes in population mess up the graph.
Oversimplifying, if there are 20 people in the US, 6 of whom are working men and 4 of whom are working women, and we double the population keeping the employment rates constant (i.e., move to 40, 12, 8), on this graph it would look like the employment gap between men and women went from 2 to 4, when really, nothing happened differentially to the genders.
Agreed. I thought we were just talking about which gender has “recovered better” post-recession, not whether the gap between the two has increased or decreased from pre- to post-recession.
Hmm…Basically employment looks similar for the age groups over 25. The difference since 2009 appears to be almost totally from the younger age group. Interesting.
If you look at past recessions this seems fairly typical: (1) Male employment rate falls faster & deeper during a recession. (2) Female employment rate returns to it’s pre-recession value faster (3) 2-4 years are needed before male employment rate matches & rises above the female employment rate.
Am I making a mistake?
I don’t know that you’re making a mistake, but part of the reason the male unemployment “recovers” is that males leave the workforce. The male labor force participation rate has been dropping since 1950. See chart. OTOH female labor force participation rate has been rising, but now might be stalling.
Are these all employment levels or just private sector? If it includes government jobs, I would like to see the difference. I think there might be one reason.
Also do NGO jobs count as private or government jobs?
Also what about the salary? The same graph might be even more interesting to have with a statistic on salary levels and their trend.
NGOs? Of course private, it’s called the aid industry 😉
Pretty sure those are total.
Under the hood, private employment has grown by 8 million jobs in the past four years, while (all levels of) government employment is down 600,000 jobs, so…
That’s a popular way to disguise the truth. Anyone living near the imperial capital knows better. There are millions of Americans working at firms that exist solely due to government invention. Pretending that these people are “private sector” is simply not honest. If we cut Federal spending back to 2000 levels, Maryland would experience a depression making the 1930’s look like good times.
My point is not “popular”, it’s scarcely heard. Your Beltway point gainsays a tiny part of my point. It’s a big country out here.
healthcare and service sector
Here is a another related article from last fall: http://www.iwpr.org/press-room/press-clips/u.s.-women-last-month-regained-all-jobs-lost-to-the-great-recession-men-still-2.1-million-short-september-12-2013. Excepts from that link:
“Lower-wage industries, like retail, education, restaurants and hotels, have been hiring the fastest. Women are predominant in those areas. Men, by contrast, dominate sectors like construction and manufacturing, which have yet to recover millions of jobs lost in the recession.”
“Since the recession officially ended in June 2009, education and health services have helped drive job growth: That sector added nearly 1.6 million jobs, the second-most of any category. And women gained nearly 1.1 million of them.”
“Women also make up more than half of the workforce in hotels and restaurants, which has produced the third-largest job gain of any industry.”
“Heidi Shierholz, an economist with the Economic Policy Institute, says the lackluster economy has limited the growth of good jobs — the kind traditionally held by men. Low-paying jobs, more typically held by women, have been growing instead.”
“Even though women’s employment has recovered faster than men’s, there are still more men with jobs than women. And more men than women have found work since the recession ended. Yet men still haven’t recovered all their losses because the cuts were so deep in sectors such as manufacturing and construction.”
“Female workers have regained their job losses in the recession even though they were hit hard by government budget cuts, especially by states and localities. Since the end of 2007, state and local governments have cut 689,000 jobs. In a report last year, the liberal Economic Policy Institute calculated that women accounted for 70 percent of those lost jobs.”
Also interesting: http://www.nber.org/digest/jul12/w17951.html mainly on how men are in more cyclical (affected by boom-bust) industries
In short, men and women have never held the same mix of jobs, looking at trends in employment by gender is mostly highlighting the way shocks hit the economy and how it is (not) recovering. The mix shift here is important and probably behind the 20-25 age chart in Yglesias piece, but I do find that one disconcerting. It will be awhile before they ‘age out’ of the labor market. But hey, Average is Over and The Great Stagnation FTW?
+1 Workforce composition by sex makes a big difference, eg, construction, manufacturing,
All kind of potentially interesting inferences from this. So let’s assume there were many educated high quality female workers who were “sitting” out due to choice pre-recession. Now economic pressures are pushing more of these women into the workforce and crowding out less qualified men (who do not as typically make that choice). Men do not have a similar reserve of educated potential employees sitting out who would be more likely to work during a recession so their share of the workforce falls.
On small factor might be that men are more likely to work for cash and continue to collect UI than are women.
What makes you think that? Just adding up all the maids, nannies, lawnmowers, and snowplowers in my head doesn’t give me much intuition as to which gender comes out ahead in cash earnings. Not to be argumentative, I was just curious if you had a source or if this was intuition.
Henry Siu’s work is relevant here, I think. His thesis is that jobless recoveries are the result of employment in “routine” jobs having declined over time. Start here:
If it turns out men make up the bulk of workers in routine jobs, and those have disappeared never to return (shutting out young men who would have otherwise taken them), that would solve the mystery. I confess I haven’t looked at the split of routine jobs between men and women to be more specific than this.
Far fewer women are employed. 20 million women lost and regained employment. 50 million men lost employment that 40 million men have regained.
The most obvious explanation is that women corresponding to the men who haven’t regained employment were never employed in the first place.
The phrase “women corresponding to the men” does a lot of the work in that sentence, and I don’t think I really understand what you mean by it. Can you unpack how those women would “correspond to” those men and what the implications of that correspondence are?
Is it possible that the decline we are seeing in work force participation, which might be tied to retirement, can potentially be tied back to this, or vice versa? I would think that the disparity between the sexes for those who have been in the workforce for 40-odd years would differ significantly from the disparity (if any) of those in the workforce less than 20 years….
Comments on this entry are closed.
Previous post: Measure for Measure (affective computing as a means of social control)
Next post: Inequality and the Masters of Money
Email Tyler Cowen
Follow Tyler on Twitter
Email Alex Tabarrok
Follow Alex on Twitter
Subscribe in a reader
Follow Us on Twitter
Marginal Revolution on Twitter Counter.com
Get smart with the Thesis WordPress Theme from DIYthemes.