The declining labor force participation rate for middle-aged males

Here is for ages 40-44, since 1976:


You sometimes hear there is no evidence of automation putting people out of work, but arguably the automation of manufacturing, plus IT-enabled foreign competition, are significant factors behind this trend.  This picture also casts doubt on the common view that there are hidden real wage increases, not picked up by standard data and wage deflators and the like.  You would expect higher real wages, if indeed they were in place, to be reflected in a more positive labor supply response, but we don’t see that.

That picture is from Bill McBride.


I'm reading the book by journalist Martin Ford "Rise of the Robots", and it's true what TC says. The white collar crowd has been largely immune from automation and outsourcing, but when it comes, expect all kinds of political barriers to be put up, since they vote more than the blue collar crowd does.

Not only do they vote more than the blue collar crowd, they make much larger campaign contributions and possess more intellectual tools to dress up their private interest as the public interest.

So you are saying that Hilary's supporters are less honest than Trump's supporters? Trump's working class voter base is pilloried because they lack the intellectual sophistication Hilary's Upper Middle Class supporters do in hiding their self-interest?

If the Trump crowd could focus on their self interest (ignoring the strong likelihood that retrenching into protectionism would probably be very destructive for America in a fairly broad economic and geopolitical sense) rather than parading around their hate/disrespect for various minorities (this does not characterize all of his supporters), I might feel a little more that way.

I think people support Hillary for different reasons. I don't trust the motives of her Wall Street donor base, but a lot of her potential electoral base has very different reasons to support her (e.g., stance on health care, belief that having a female president would be good for the status of women).

I think Sanders' anti-establishment positions can be taken more credibly than Trump's, and anyways his more radical (to American ears) plans would mostly be blocked by Congress.

Shorter Nathan W: I'll tell you what your interests are plebes. Now this would be bearable if Nathan W were from some ancient line of aristocrats or even a half decent business owner, but instead he probably teaches at a Community College.

Shorter Sam: saying things people don't say and strawmanning them.

And anyways, would you be unable to recognize an argument based on its own merit for the fact of whatever way that person happened to find to pay the rent?

Shortest Nathan W comment ever. And they say you can't teach a communinty college teacher new tricks!

What I was saying is that your ridicolous comments would be more acceptable if you weren't personally such a loser. Kind of like how pretty girls can be boring. Community college sociology professors should be meek and too the point.

Wow. It is now an insult to "demean" someone for belonging to a profession which trains the vocational abilities of the future, including the right wing white working class.

You must have a truly sad life to be getting something out of going around name calling like that. I suggest getting a hobby. Perhaps rugby? You might get a little extra attention on the field for your generally poor attitude, but maybe you'll learn eventually.

The blue collar workers that I have know want more jobs but less welfare. They feel like they work everyone should have to work. I think that is why they like Trump he promises more jobs and less welfare.

On your first point, I don't think it's really debatable that automation reduces the number of workers required in many manufacturing processes. This is unlike the 1960s where this implied higher possible wages for workers, in a fortunate context where expansion in many new areas of economic activity, namely the services sector, enabled ongoing job gains. It is not clear that such job gains are occurring with the exception of some highly paid tech people. Also, it's not really debatable that IT-enabled competition is relevant, given that people can price check on eBay, etc., and order from anywhere in the world to save money. How large the effects are is debatable, but the existence of the effects is not.

On the second point, I would point out that too much attention can be paid to the median or average real wage. For some tens of millions of Americans at least, the evolution of real wages across the entire income distribution is entirely relevant. The minimum wage, for example, is much lower in real terms in many places than it was 10, 20 or 30 years ago.

Countless barriers already exist. The white collar crowd (managers of the corporative state) is all about gaming entitlements in the rent seeking game. Most white collar workers earning salaries in corporations (other than engineers and the like) play this game for a living. They aren't the last to be automated for any technical reason. On the contrary, according to legend, Alan Turing once sat in a meeting with officers of some corporation (AT&T maybe). The CEO of the corporation asked Turing, "What sort of jobs can we expect your automated 'computers' to replace?" Turing responded, "Jobs like yours, I imagine." I don't remember where I read this story, and I can't find it online, but Andrew Hodges' biography of Turing seems the most likely source.

The problem with Turing's prediction is that he was talking to people entitled to decide which jobs are replaced with technology in a particular corporate context, and there really is no competitive process whereby officers of large corporations, made large by statutory barriers to market entry, replacing themselves with technology gain an advantage over other corporate officers who don't. We might as well imagine apparatchiks of some socialist central planning bureaucracy automating their own jobs out of existence. Of course, they don't.

We might expect corporate officers to automate secretaries and graphic artists out of jobs, by using tools like Word and Power Point themselves to prepare documents and spiffy presentations, but in reality, as managers adopt this technology, they only produce more documents and spiffier presentations, and the new norm for this sort of self-promotion demands more management labor, so the profits of increasing productivity are "reinvested" this way.

When the rabble was holding on by their fingernails and suffering in silence, Tyler and his ilk were picking over the signs with a fine tooth comb, reserving judgement against negative indicators and dredging up doubts and alternative explanations. Now that the rabble's got a champion and it looks like they're going to catapult him to power, all these sceptical little bean counters are shifting their scepticism ever so slightly. Suddenly, they seem almost sympathetic.

Berniebots are numerous.

And you continue to drool nonsense...he's long pointed out the divergence between U6 and the reported unemployment.

What is this, a picture for ants?

+1. Made me smile. There is a much more readable version at the link...this one is indeed a bit on the small side.

+2 LOL, nice one Derek Z.

High resolution here:

If workers are being replaced by automation, should that not be reflected somewhat in productivity growth ? Is it ?

It could take a while to show up in data until the costs of developing automation are fully paid for, and such technologies become more widespread and cheaper, even if there is a genuine productivity growth happening.

Recent productivity growth has resulted from reducing inputs more than increasing outputs

A significant portion of automation is waste reduction. Even when a given company doesn't have the sales to increase production it can still increase profit by reducing waste. Automation is a reliable & straightforward way to do so.


Why not call it thoughput improvements, which reduces inventory and time to completion.

95% to 90% over 40 years? Does not seem to be a large issue. It's so minor that their are other potential reasons: more overall wealth, easier disability insurance, it is a tiny percentage.

5% is not tiny in the labor market context. Think of the difference between 3% unemployment and 8% unemployment. Here, we are looking at people who are dropping out of the labor market instead of people who are looking for work and cannot find it but the principle is the same. There are very few men in their early 40s who can afford early retirement and it is much more difficult to qualify for disability if you are under 50.

Labor force participation of men who are between 30 and 50 is a very good measure of the overall health of the American economy for these reasons. No reason to focus specifically on 40-44 but the general principle is that the number of people in this group who can support themselves in the long-term through sources outside of employment or rely solely on a spouse's employment income is so small that any protracted shift in employment patterns in this group should be concerning.

I have been skeptical of other participation charts, but this one centered on a prime working age, over a longer span is more convincing.

Not just automation though. When GM was the largest company in America cars were made in the US. Now and Apple is the largest company phones are made in China.

According to the article the absolute number of "missing" workers is 177,000. In roughly the same period the number of stay at home dads has increased by well over a million. I would guess men age 40-44 are less likely to be stay at home fathers because their children are probably older, but then again couples are waiting longer before having children. If a guy is a first time father at 35, and he has two kids, the second one is still quite young when he''s 44.

I doubt it. If prime age male LFPR is down 4%, you're talking about a couple million guys between 30-50 on the sidelines.

Wikipedia said there were 140,000 stay at home dads in the US in 2008. That number may have jumped due to the recession, but most of the LFPR drop was already in place by then.

Pew says there are 2 million stay at home dads.

Hazel, Pew defines "stay-at-home dad" as any man between the ages of 18 and 69 who has his own minor children living with him and is not working. Their survey of this group of people in 2012 showed only 21% told a surveyor the reason they were not working is in order to take care of the family. So the real number is closer to 400,000. Pew also notes that many of the men they surveyed are toward the top of their age range and are cases of early retirement or disability. That again is the value of looking at narrower age ranges between 30 and 50: these are people who are mostly too old to still be studying but mostly too young to be independently wealthy, sick, disabled, or to face age discrimination from potential employers.

Assuming there is no great discrepancy in age between spouses this is also the age (~40-something) when wives are likely done with child-bearing and are back in the workforce, if they have careers at all. So I can see the bulk of this being stay-at-home dads.

Clearly the answer to this problem is more immigration.

So is there a matching inclining lfpr for middle aged women since 1976. A graph overlaying male and female lfpr, age 40-45, might be instructive.

It may be automation. It may not. This has also been the period of family break down and declining marriage rates. There is no point for men to go out and work hard all day if there is no home to come to. Or if marriage does not offer the sort of social status and respect that makes up for those long hours.

Of course as marriage declines, men will cut back on their hours. Look at the Black community. Why would men bother?

You are doing the same analysis as Tyler. Stating a prior and pointing at a downward sloping graph. But consider that hispanic male marriage rates show consistent decline, like everyone else, but hispanic male participation rates only started dropping from the Great Recession.

That would explain an observation of lower hours, or less interest in the super hard work required for career advancement, not people dropping out of the labour market altogether.

Natah: For those who started with very low motivation there is no where lower to go than out of the labor market.

Re: There is no point for men to go out and work hard all day if there is no home to come to.

Unless you are into homelessness, there's every reason to work at that age if you are single: you need to put a roof over your own head. You're not likely to have parents to move in with (they are either six feet under, in a nursing home or living it up in their seniors condo down in St Pete), so you really do need to have your own income.
Again, I suspect that the bulk of these guys are living off the income of spouses or girlfriends (or boyfriends for a small fraction of them)

FFS... the series started around 94% end up at 91%... does that tell you something? Probably not. I think Tyler didnt even look at the figure...

You know, you could subtract the numbers from 100% ...

Yes, it increased to double, but who knows, it is due to automation or something else? A lot of things changed since 1976.... For instance, from 1976 to now, the prison population among african-americans of this age group probably increased a lot, would that explain smth?

Of the cuff, perhaps 0.5 percentage points, not at all irrelevant when considering a total change of 3 percentage points, or the other millions with a records who do not enjoy full access to the labour market.

Why did it go up during the recession? Shouldn't it go down during a recession?

If it's a story of men not working because their wives are the bread-winners, an increase in the participation rate during the recession could suggest their wives saw wage or job losses, so they entered the labor force to supplement the family income.

Like other commenters, I'm confused about the link to automation job losses. Male participation has fallen since the 1950s. But you wouldn't paste any labor chart with a downward trajectory and claim it shows the effects of automation. Does this post require previous posts for context?

And yet they are all still eating. Somehow they found a more attractive means of support than work.

My last landlord was disabled.

He also re-roofed and re-wired the entire rental property himself, then moved to Washington to his retirement horse ranch at age 55.

I think people are getting smarter about farming benefits.

Cops and firefighters have been brilliant at staging back injuries magically 6 months before they retire.

"back injury" "disabled" cop I knew was retiring could still hike Cindercone mountain in Shasta....

Tyler's automatons go by another name: women.

Women have been entering the work force in greater numbers over the past 45 years. That creates two phenomena that could easily explain a meager six percentage point decline without turning to invisible robots.

The first is Mr. Mom. A small percentage of men, often married to women with well-paying jobs, have dropped out of the work force to become primary caregiver. I've known a couple guys like that.

The second is a little harder to explain but probably responsible for the bulk of this micro-phenomenon. Because the country has more two-earner households, there is less pressure to return to the workforce after a layoff. The family still has cash flow - maybe not enough to make ends meet, but enough to keep things going. That means the laid-off man can spend more time looking for the right job (and that job is harder to find because the wife's income makes it harder to relocate). Spending, say, two months between jobs instead of one would pretty easily explain the tiny, tiny decline we are fretting over here.

Come to think of it, the normalization of the layoff itself could explain the decline. Companies were much more reluctant (and in the case of union shops, sometimes unable) to fire workers without cause. Greater layoffs mean more workers in transition, which would, again explain this chart.

"maybe not enough to make ends meet, but enough to keep things going."

Please explain the difference.

Debt is increasing but you can still make the monthly payments to keep the lights on, etc.

These days the time between jobs may be more like six months to a year instead of two or three months.

And you might not be willing to move for a job because of your wifes job.

How much of this is driven by the fact that the average 40-44 year old is getting older? I.e., the critique of Case and Deaton?

Now, is the youth labor participation increasing?

If not, then jobs are disappearing.

The greatest falling off by far in LFP has been among those under 25 years.
However there has been an increase among the oldest workers (fewer of whom can afford to retire)

How much of this is incarceration?

These comments (which are good) reveal the glaring weakness of most economic research: dots on a graph. Dots don't tell a story, and it's why Desmond's book is so riveting, because it tells a story about real people not dots on a graph. The same goes for George Packer's book, The Unwinding. It may not take a village, but it does take a narrative to get behind the statistics and the dots on a graph. Of course, academics avoid the narrative, preferring dots on a graph so as not to influence with stories of sympathetic, or unsympathetic, real people. Before going on the bench, Louis Brandeis was famous for his pioneering briefs which relied on a combination of a narrative and a compilation of scientific information and social science rather than simply citations of authority. The brief, which came to be known as a "Brandeis brief", provided a vivid picture for the court, the real people involved and the social science that helped explained their actions. As with academics, few judges today will accept a Brandeis brief, preferring instead an antiseptic recitation of authority.

This chart does not show what it appears to show.

Notice there is a rolling five year cohort through the entire period.

There is little reason to conclude that participation by a 40 year old declines over the five years they are in this chart when just as easily participation by an incoming 39 year old may be lower then the 40 year old. This may instead be a chart showing opportunity varies directly with age.


Sorry, I mean, "varies directly with birth year".


If automation is responsible, then why is it not happening in other advanced economies? The second chart on this page is the Employment-to-Population Ratio, for men 15-64 for five OECD countries

Since 1990, the male employment ratio has been relatively flat in Canada, Japan and the U.K., it has improved in Germany, and dropped significantly in the U.S. This does not look like technological change, but policy variation.

Counterargument: Canada was able to reallocate some labour to resource industries (now in bust years), Japan's broadly declining population means you can have both automation and stable labour force participation rates, while Germany (sorry guys) simply makes higher quality high-end manufactured goods and so can increase manufacturing employment and introduce automation at the same time.

Even if those explanations are correct, they are policy explanations, not technology. The U.S. has other policy priorities than maximizing employment. The next chart at my link shows that the U.S. led all five countries in 1990 for female employment-ratio, and by 2014, the U.S. was last.

The graph is classic "How to lie with statistics", because the y axis goes from 86% to 97%. The huge fall in participation is just a few percentage points. The axis should go from 0 to 100, or at least 50 to 100 to be reasonable.

Maybe there is a story here, but the graph is a lie.

Yes, from now on, all graphs of the unemployment rate should also be on an axis that goes from 0 to 50 "to be reasonable." Also, graphs of the daily temperature should be presented on the Kelvin scale with absolute 0 at the bottom lest we "lie with statistics" about temperature changes.

Seriously people, we can't expect the mere common man to actually read the numbers! Commoners can't possibly understand such things!

Obviously, I'm being sarcastic.

Most people receive a woefully inadequate education surround statistics, or even fractions. So if the audience of this chart were in fact, everyone, many might misunderstand the chart and JackPQ, as much as I hate to admit it, has a point. But this chart is posted on an economics blog - a place where I suspect the level of mathematical sophistication is above average. Judging from the comments, most readers noted that this was a small absolute, yet large relative change.

That said, I believe JackPQ is a troll, so forgive me for feeding the trolls...

Nope, not a troll, just grumpy. Okay, so Ricardo doesn't like 0 to 50. Fine. The point is not to cherry-pick your axes to inflate your point.

Another factor that could be leading to the decline is people working off the books.

The link above claims that undocumented labor has increase substantially since 1970, although the increase isn't smooth.

I think the growth in female participation + Incarceration + undocumented income could explain the LPR decline without resorting to a "robots are taking our jobs" story.

Bill McBride is a leftist - also big fan of Krugman. Like El Krug, he wants to make Obama look good. Here he cherry picks a very small slice of data (a 5 five year age range of one gender which should probably be verified before taking at face value) and puts it on a graph with a very narrow Y axis range. This is an attempt to make the drop in labor participation look like some macro mega trend that's been going on for 40 years.

However, when you zoom out it's quite clear this labor participation issue is a recent phenomena. It's not entirely Obama's fault, but it's clearly accelerated under his administration. McBride doesn't want to discuss that fact. He wants to brush it under the rug. I quit reading his site ~5 years ago for this reason. Appears he and his site haven't changed much, although he does have a nice sexy profile pic on the front of his blog now. That's new.

'Bill McBride is a leftist'

Who just happens to have an enviable track record in looking at data and making informed judgments about it. Yep, must mean a retired technology executive and son of a Navy fighter pilot is a leftist. Well, OK, he did think the Bush administration lied when it came to invading Iraq, but then, considering the data, McBride was right there, too.

My comment is an answer to the obvious question of "Why take a very small slice of data, put it on a very small axis, and use it to draw a conclusion different than one would draw looking at the entire set of data on a bigger axis?"

There may well be 40 or even 100 year macro trends that affect the labor participation rate, but the dominant trend is a cyclical one.

I would like to interrupt my pity-laden laughter to point out that prior_pest just made the argument that you can't be a leftist if your dad was a fighter pilot.

No harm in pointing out such an obvious fallacy, but the pity is for those with so little sense of self worth that they have to run around passing off insults to feel like something. And there's nothing funny about it.

Too cute. Pithy and one point beats obtuse and verbose ever time Professor- or is it teacher. I don't know community college protocol.

I'm a gas station attendant.

Does that change the validity of anything I say?

Only if you're too dumb to evaluate for yourself, and instead require simplistic categorizations and shortcuts because you can't be bothered to think for yourself.

Seriously, go find something that makes you feel worth something that doesn't rely on childish name calling. I can only imagine what a loser your girlfriend would think you are if she knew you were running around saying such things ...

Disability rolls increased.

ACA also strongly encourages households to game their income to get subsidies.

So, a man might quit his low-paying job so the household qualifies for subsidies for the ACA.

Lots of scenarios to get a few percentage points less participation.

The increase in disability is mainly about over-50 people who can credibly (and often legitimately) claim disability due to the problems of increasing age. Claimants younger than that, absent catastrophic injuries or terminal illness, are very likely to have their claims denied.

LFP for men HAS been falling for some time. This is not a recent thing at all.

Why make everything about partisan politics?

"You would expect higher real wages, if indeed they were in place, to be reflected in a more positive labor supply response"

Why? Higher real wages could led to both more labour supply (via substitution effect) or less labour supply (via income effect).

"You sometimes hear there is no evidence of automation putting people out of work, but arguably the automation of manufacturing, plus IT-enabled foreign competition, are significant factors behind this trend."

Good one Tyler. If you look at that chart more closely the participation rate is *flat* until you get to the mid 80s. That's when it really begins to drop.

As with a lot of labor market questions, it helps to think about labor supply at the household level. You should look at what is happening to the participation rate of women 40-44 over the same period.

In particular, the participation rate of women went from about 60% to about 76% over the same period.

Thanks I asked that question in the comment above. i don't see how such a non occupation specific chart like lfpr can have a question about automation imposed on it. It would need an overlay of increase of automation. Then when you also overlayed the increase of lfpr for women you could only conclude automation was good for women but not for men. However that would also be a false conclusion. It is more about wages for women are lower than men, I think.

I am surprised to see a comment from Professor Cowen blaming automation. I thought that his entire meme is that innovation creates more jobs. I thought that maybe he was pointing to short term data, but these data go back to 1976.

Or, is he saying that men age 40-44 have some particular inability or unwillingness to get work or seek employment in the jobs newly created by automation? Is this what he is getting at with this statement: "IT-enabled foreign competition."

Otherwise, it seems like Professor Cowen is agreeing with President Obama: ATM machines destroyed jobs.

Technology has something to do with this trend, but I doubt that the trend indicates a Luddite nightmare with legions of unemployed workers standing in lines at soup kitchens. Presumably, periods of unemployment, between jobs, are increasing, both because wealthier workers are more selective when changing jobs and because they effectively spread retirement over their working lives rather than saving it all for their waning years. Workers are wealthier both because they earn more while employed and because welfare state programs cushion their periods of unemployment.

This choice makes eminently good sense to me. Why work forty hours a week, fifty weeks a year, for forty years continuously only then to enjoy some leisure in your sixties and seventies, as opposed to taking a few months off every few years and continuing to work into your sixties and seventies? Is that a rational choice or a rut that people commonly fall into?

My last tax preparer said his employees all like the job because its enough work to qualify for unemployment for the rest of the year. This kind of farming of the benefit state is increasing. Leisure time is valuable, too.

At a guess those employees must have a working spouse or some other form of income. Unemployment benefits are very miserly* and not something you can live off of without some sort of supplemental income.

* On the average they replace about 35% of previous income. Also, only about 1/3 of laid off workers qualify at all.

I find the discussion concerning the graph fascinating, and though I intend no defence of just how sloppily Prof. Cowen cites an economic colleague, one should be aware that for the last decade, calculated risk notes the following in such cases, just as now - 'This graph shows the 40 to 44 year old men participation rate since 1976 (note the scale doesn't start at zero to better show the change).'

This graph is intentionally so presented, with a clear explanation on calculated risk's part why. That Prof. Cowen merely displayed the graph provides an interesting insight into how many people bother to actually visit the link.

Good graphical analysis is SUPPOSED to zoom in highlight the change, not zoom out to the max to minimize it. We labels axes so that intelligent/educated people are not misled. It annoys me any time people get upset at seeing graphs which present information they don't like which makes it as easy as possible to observe what is changing, because they would simply prefer a graphical presentation which minimizes things they wish didn't exist.

No one has mentioned age discrimination, which as. 58 year old can tell you is real. Second, the older one is the less likely one will uproot themselves and move elsewhere.
Automation creates job losses but it also creates new jobs; it's just that you are able to just shuffle people around. Too many companies put the burden on retraining on the workers to do on their own time without assuring them that they will get a return on their investment.

Is age discrimination bad for people in their 40s? Maybe in selected fields but I would not overall it would be.

What else has happened the last 40 years? Mexican immigration:

Who do Mexican immigrants compete against? Blue collar males. Both at the low end (why teen unemployment keeps going up) and likely at this 40-44yo range as well, right around when a man's body starts really breaking down and family responsibilities increase, making a 30yo with no family that's incapable of talking back due to limited English a very appealing hire.

Also, many went back south during the great recession as housing related jobs dried up. Would explain the uptick in this chart during the recent recession.

Yes. This is the graph. Note that declining LFPR for prime age males has been declining for four decades.

There are 1-2 million prime working age men not in the labor force who would have been in the labor force 40 years ago.

Obviously, the devaluation of brawn is a big part of the story. For some men, this is the big thing they bring to the table, and it used to be rewarded, but not so much anymore.

OTOH, so many more women are now in the labor force, so the employment to population ratio is at or near an all-time high.

It's not that jobs are going away so much as a certain kind of job has been going away.

Those with excess brawn always have the option of donning brownshirts and lobbying for laws to put golden stars on shirts of all members of their most hated minority. Can't imagine how that could ever possibly go wrong ... no need to look to history at all. Move along folks ...

Exactly. These people must be demonized and kept down at all costs. Or indulged and told that it's some guy beyond the curtain fucking with them.

Can't imagine how that could ever possibly go wrong.

People never backlash when you demonize them, of course. That's why we need to keep on heaping scorn on minorities and welfare bums. It's tough love. They will learn.

Seriously though, it would be nice if some GOPer could legitimize their concerns without appealing to the very worst in our human nature. I've been watching videos of Trump rallies, and they do NOT look like a nice place to be, and Trump's definitely not the one who's going to shush them as they manhandle minorities or become angered at the mere presence of a Muslim wearing a shirt that say "we come in peace". If I didn't understand a word of English and attended both a Trump and Sanders rally, I'm pretty sure I would easily pick Sanders as being the leader of the kind of country most people would want to live in. Anger and hate rarely lead anywhere good.

Short Nathan W. Brawny men of little education are Nazis and should be feared and hated.

Except for minorities. They're the best. I wonder why they're not at these awesome Sanders rallies.

That you can see such vast, and vastly incorrect, generalizations does not suggest clear thinking.

It is indeed a concern that those with excess brawn in a period where their brawn is of little value to the economy could be troublesome for public security. it is a tried tested and true political strategy to scapegoat foreign powers and/or minorities and to promote anti-something sentiment for electoral gain.

Being concerned that some of them "have the option of donning brownshirts" and observing the reality that minorities get manhandled at Trump rallies without Trump so much as wagging his finger at them is indeed a cause for concern.

But to suggest that I think all people of brawn will resort to such thinking in no way reflects anything I said, whatsoever. I merely expressed the concern that some can go that way (see to be going in that general direction RIGHT NOW).

I expect smarter from you. You're not dumb or given to logical fallacy, as far as I can tell from anything else you've written. When the KKK explicitly endorses Trump and 20% of Trump supporters think we'd be better if we'd never ended slavery, we should not be shushing those who express concerns.

I don't know how we ended up at Trump. Look, reread my original comment. It's an observation, not a judgment. It applies to humans generally.

You're the one who had to introduce normative speculation into the conversation. Your mind immediately jumped to brownshirts (bad). Nice derailing.

After I replied, it occurred to you that some of these brawny men might be minorities (good), so you haul up your slacks and start bashing the GOP.

You should make a better effort at responding to comments as written rather than indulging wherever the comment leads in your imagination and then writing about that.

Wouldn't it just be crazy and unbelievable if it were a combination of a whole lot of things - some thought to be good, some thought to be bad - that together resulted in a small but noticeable drop in employment for middle-aged men?

Mentioned here: automation, stay-at-home dads, kept men (sorry for the denigrating term - there must be a better one), increase incarceration, the effect of criminal records on employment records, easier attainment of disability (for both legit and fraud cases)' off the books earnings for tax avoidance and ACA subsidy gaming.

I'd add another possibility concerning incentives: it may be that men who make substantially less than their wives find the work incentive doesn't offset the subsidy gains for the safety net. I'm sure many people here have had the experience/sticker shock of being a young parent earning a relatively low age discovering that child care costs nearly consume an entire parent's earnings.

I love the search for causes (because presumably we could address the bad things, or at least stop addressing the non-causes that we've mistakenly identified), but as Cowen has pointed out, we usually know way less than we think we know about complex systems/problems, no?

I doubt the ACA has much to do with this-- that's still too new, and this is a long standing trend. And if you live in a state that did not expand Medicaid you won't get anything unless your income is high enough to qualify for subsidies-- you just end up uninsured.

Labeling foreign competition "IT-enabled" is a very weak prop for an argument that automation is to blame. Automating away a job is one thing, offshoring it to cheaper labor is another. The former dramatically increases productivity, the latter somewhat increases productivity and spreads it from front-runner countries to catch-ups.

Besides, if you're going to make the argument that labor participation fell due to loss of blue collar jobs, you should roll out some data to back it up. I suspect it's no more than a third of the explanation. Competition from immigrants, Mr Moms, affluent early retirers, rising numbers on disability are some other factors.

Wait -- isn't this figure influenced by the recently-publicized increase in the death rate for males in roughly that age range (much of which is alcohol and war and suicide related)?

I'll supply a single data point as I'm part of the data on the graph. I'm probably atypical, but I "retired" at 42 a few years back, and the main reason, besides severe burnout, was because I could. All my debts were paid, and savings and retirement sufficient for an otherwise frugal lifestyle. My wife is a self-employed writer who built her career while I worked a corporate job, so as long as her modest income holds up while investments grow we should be fine. In previous generations I wonder if such an option would've even been possible?

Again, I don't know how typical this is, but it's certainly possible today for one to work hard at a good job and save for about 20 years with intent of retiring early to pursue your interests. It is not lost on me how fortunate I am to live both when and where I do. (This is not to dismiss the real struggles of many - but I do think for at least some, the simple decision to retire early is a real option).

The graph shows that, over a period of 40 years, the participation rate has declined from about 95% to about 90%.

For 40-44 year old males (note the site this is taken from looked at other cohorts, and it's likely the selected graph is the most extreme one).

Among the MANY things that have happened over this generation-and-a-half period is the increased participation of women in the labor force, and the consequent beginning of acceptance of househusbands / males who stay home with the kids while they are young. Not many, but we're not talking about a large effect here.

Looks like another argument for a Basic Income Guarantee.

I'm sure someone has already said this, but didn't you just post an article about associative mating? If my wife made 150K I'd be home with the kids. Of course maybe she could automate my primary duty...

If more automation means fewer workers it would be nice if meant the same number of employees working fewer hours each. My wife works at a hospital, she would like to work fewer hours but her department is has too few employees. I wonder if benefits and overhead are the reason.

It's wonderful that you are getting thoughts from this
article as well as from our argument made here.

Comments for this post are closed