About a decade ago, Nicholas Eberstadt of the American Enterprise Institute published an alarming book called Men Without Work. In the mid-20th century, he showed, nearly all “prime-age” men (ages 25 to 54) were in the labor force—either working or at least actively looking for work. By the mid-2010s, more than one in ten such men were not even trying to find employment, with the problem especially pronounced for men with less education.
When the book came out, the trend showed no signs of slowing. But it seems things have recently leveled off (despite an unsurprising spike during the pandemic). A great new report from Steven A. Camarota and Karen Zeigler of the Center for Immigration Studies has the details and some solid analysis.
There was already a lot of debate about why men were leaving the labor force. Now we can also debate why they stopped!
Here is one of many ways the report breaks down the trend, separating immigrants from natives in the years the underlying data allow this:
Many different theories have been proposed to explain the long-term rise. Each has some combination of common sense and evidence backing it up. Many of them might fit the leveling-off in the past decade too.
Populists and the Left often focus on “demand-side” explanations: There just aren’t enough jobs, or attractive enough jobs, demanding the services of lower-educated men today. While “wage stagnation” is a debatable claim sensitive to measurement choices, long-term wage growth for less-educated men has been less than great by all accounts. The post-2015 period, meanwhile, has often seen wages grow faster for the less-educated and lower-earning than for the more fortunate.
Given its focus as a think tank, CIS unsurprisingly highlights a potential role for immigration in labor demand, noting the roughly 12 million people without a bachelor’s degree who claim to have immigrated since the turn of the Millennium. This, they hypothesize, in combination with low demand for less-educated labor, “is causing a growing number of less-educated American men to leave the labor force.” Migration to the U.S. has of course been in flux more recently, between the pandemic, the border crisis afterward, and the Trump administration’s recent moves.
Meanwhile, the nuclear family has declined since the mid-20th century, and non-working men are largely unmarried. Maybe guys are less likely to work because no one else is depending on them? Or maybe addiction and criminal records are a big part of the puzzle. Well, incarceration has coming down for a while now, marriage trends have shown signs of stabilizing since COVID, and drug overdoses might finally be turning around, too.
Yet another major theory centers on the safety net, especially the disability system (as many other programs are tailored to single mothers rather than to non-working men). As Eberstadt noted in his book, disability programs are a large source of support for otherwise out-of-the-labor-force men. Use of these programs rose even as jobs became less physically taxing. The Center on Budget and Policy Priorities, however, reports a turnaround in disability receipt starting in the mid-2010s.
Pondering these explanations, I recalled a 2017 Mercatus Center report on nonworking men, authored by Scott Winship, that supported a strong role for disability programs. One issue Winship highlighted is why men themselves said they weren’t working. Drawing on the same data through the present day, below I show the percentage of U.S.-born prime-age men who didn’t work for the whole calendar year, broken down by the explanations they chose from options given in the Current Population Survey.1 (The share of men who did work is not pictured, but such men are included in the denominators when calculating the percentages.)
Like Winship, I found that illness/disability is far and away the leading explanation. The rise in men not working for this reason peaked in the mid-2010s. Meanwhile, not being able to find a job is pretty rare even as a stated reason for long-term nonwork (though it spiked during the Great Recession period and is, obviously, far more common among men who are temporarily unemployed).
Ultimately, there are probably a lot of factors behind this shift, but I hope it continues.
From the Manhattan Institute
Jonathan P. Caulkins evaluates the important and wide-ranging topic of “prohibition,” from bans on drugs to prohibitions on surrogate pregnancy. If we prohibit something, will we necessarily see a violent illegal market emerge? And what are the options between draconian criminal laws and full-on legalization?
Other Work of Note
A study of Europe finds that “immigration increases the salience of birthplace identity along cultural lines, fostering a shift toward nativist identities among the native population.” The effect is “stronger in regions receiving culturally distant immigrants.”
Speaking of Europe, who actually pays for the generous welfare states there?
Medicaid spends too much money on long-term care for seniors who could afford to pay for it themselves, this Cato paper from Stephen A. Moses argues.
Which gun owners store their weapons most securely?
Whether a given mother will be given a C-section varies from county to county, apparently stemming from “provider behavior and local practice norms.”
When civilians claim that NYPD officers used force inappropriately, body-camera footage appears to increase the chance that the claims are substantiated, with an especially big effect for minority complainants.
Did watching The White Lotus cause viewers to try to get Lorazepam? (I won’t spoil the plot, but the drug is an anti-anxiety benzodiazepine.)
The Paris Agreement had an “aspirational target” of keeping global warming to 1.5 degrees Celsius. Humanity is likely to blow through that target sometime in the next few decades. What happens then?
If you’re curious about how the Congressional Budget Office does its modeling, check this out.
About a third of boys and young men, age 12 to 21, don’t have a male mentor they can turn to with questions about schoolwork and relationships.
Instead of just looking at kids’ overall standardized-test scores, should researchers and education administrators pay attention to exactly which questions each kid got wrong? Those more granular data seem to contain valuable information.
Enjoy your week!
The survey question pertains to work in the previous calendar year, so I subtract one from the survey year and look at men who were 26-55 when asked. CPS accessed via IPUMS: Sarah Flood, Miriam King, Renae Rodgers, Steven Ruggles, J. Robert Warren, Daniel Backman, Etienne Breton, Grace Cooper, Julia A. Rivera Drew, Stephanie Richards, David Van Riper, and Kari C.W. Williams. IPUMS CPS: Version 13.0 [dataset]. Minneapolis, MN: IPUMS, 2025. https://doi.org/10.18128/D030.V13.0