The Jobless

Via Paul Krugman, a chart and article by Floyd Norris contrasting the long-term trend in the unemployment rate (up and down) with the long-term trend in the proportion of prime-age men who don't have jobs (up and up). Naturally, this raises the question of what everyone's doing. One assumes that some portion of this is men taking on traditionally female roles as the personal primarily responsible for family care tasks. It also is my impression that there are more over-25 students than there used to be (certainly I know more than one person who was or is in law school at age 26 or higher). And the average age of retirement has tended to drop over time, so that must mean more men in their early fifties retiring.

On the other hand, for an older person the line between retirement and unemployment can be a fine one -- there are doubtless various retired people out there who would, in fact, be willing to work if there were more appealing job opportunities out there. But those kinds of thing aside, maybe there's been an increase in the number of people doing black market work at least part time? One trouble with official statistics is that trends are always ambiguous between whether or not something is actually not happening, or whether it's just not getting counted. Even during the very tight labor market of the 1990s, the jobless rate was way higher than it was in 1960 and it's a bit hard to believe that all those people were just doing nothing, and while the run-up since then very plausibly represents deteriorating labor market conditions, the job market was extremely strong back then.

Comments

A lot of these jobless men are in jail. Still others are doing things in the drug industry that do not count as jobs. These things provide a living but lead, eventually, to jail.

Posted by Peter Driscoll | April 13, 2008 10:06 AM

Naturally, this raises the question of what everyone's doing.

I believe most of them are blogging.

Posted by Stefan | April 13, 2008 10:21 AM

It's most likely a combination of different factors rather than one dominant one As you and a prior comment have noted, these factors include post-25 students, pre-55 retirees, men in jail and those in the underground economy.

I don't know if such statistics are available, but looking at men in the range of, for example, 30 to 49 would be more useful than 25 to 54. Setting the low end at 30 rather than 25 will eliminate most students and, probably, a good chunk of prison inmates, while an upper bound of 49 will eliminate most early retirees.

Posted by Peter | April 13, 2008 10:44 AM

Clinton changed the unemployment metric in the 1990s. That is part (but not all) of the problem right there. Numerically comparing unemployment rates now to those in the 1980s (or even worse, drawing a curve comparing them over time) is evidence of mathematical illiteracy.

There are similar problems with the changes in the CPI. You cannot compare inflation now to what it was in the 1980s without converting the metrics.

Posted by Walker | April 13, 2008 10:54 AM

Unemployment statistics don't include "discouraged" workers. I suspect that accounts for most of the discrepancy.

Posted by Ginger Yellow | April 13, 2008 10:54 AM

The whole phenomenon of men dropping out of the work force fascinates me, and has never been fully explained to my satisfaction.

One thing that's going on, though, is that good portion of those men are on disability. The number of men (and women) on disability has been rising for some time now. In fact, quite a few of the women who used to be on the welfare rolls have moved on to disability instead.

Posted by Kathy G. | April 13, 2008 11:25 AM

Re: One trouble with official statistics is that trends are always ambiguous between whether or not something is actually not happening, or whether it's just not getting counted.

The chart leaves out something major: half the population. I suspect that a chart showing these rates both men and women would answer the question: we would find that female jobless rates were very high 40 years ago, then fell to much lower levels, mirroring the ris in male jobless rates.

Re: Clinton changed the unemployment metric in the 1990s.

I know the CPI was changed but I had not heard that the unemployment calculation was. Do you have details? I believe that the household survey portion of it (Are you working? Are you looking for work) is still the same as it always was.

Re: Unemployment statistics don't include "discouraged" workers.

Discouraged workers must be doing something for an livelihood: underground economy, living off wives, girlfriends or parents, disability etc.

Posted by JonF | April 13, 2008 11:30 AM

Ginger Yellow's comments sent me to the source. My comment was wrong as the institutionalized (jailed) are not part of the population upon which the ratio is calculated. The underground economy still accounts for some of the discrepancy. And, of course, Bloggers.

Posted by Peter Driscoll | April 13, 2008 11:31 AM

One thing that's going on, though, is that good portion of those men are on disability. The number of men (and women) on disability has been rising for some time now. In fact, quite a few of the women who used to be on the welfare rolls have moved on to disability instead.

Excellent point.

It would be instructive to see whether the proportion of disability claims for conditions capable of being faked has risen over time. Psychiatric conditions and back pain, for instance, though there probably are others. Should there have been a significant rise in this proportion it would be strong evidence of increasing fakery.

Posted by Peter | April 13, 2008 11:36 AM

Trust fund babies? There seem to be quite a few of them in the 25-54 age bracket here in LA.

Posted by eb | April 13, 2008 12:16 PM

It's my impression(I'm a contractor) that there are huge numbers of people in the building trades whose work is legitimate but whose tax payments and official employment status are NOT legitimate.
Think house painters, carpenters, handymen, etc.
Other categories that spring to mind: House cleaners, various kinds of personal aides, ebay salespeople, food service workers.

Posted by Saul R | April 13, 2008 12:23 PM

I want to emphatically validate Matt's point that early retirement is often involutary, and often a disaster for the retiree. I worked for a company that closed in 2004, laying off the entire work force. A lot of my 50-ish colleagues looked hard for a job, and only "retired" when they couldn't find one. (Age discrimination in hiring is pervasive and brutal when you're over 50.) This often involved selling their houses to get cash to live on and moving somewhere where the cost of living was cheaper. It meant not putting their kids through college. It meant abandoning their friendship networks to live in (for example) a small town in rural Nevada.

And of course, if you retire early your monthly income is much, much less- this is true of both social security and pension (if one is lucky enough to have a pension).

Posted by Anne E | April 13, 2008 12:39 PM

A lot of these jobless men are in jail.

I doubt it, actually.

I agree that there are in fact a lot of men 25-54 years old in U.S. prisons and jails - close to 2 million of them, which is way above where it was a few decades ago. (Are we safer yet?)

But to the best of my ability to tell, all labor statistics published by the Department of Labor's Bureau of Labor Statistics (BLS) use what's called the "civilian noninstitutional population" as their sample universe.

That is, persons in the military, in prisons and jails, in juvenile detention, or in nursing homes, aren't counted in labor statistics.

I can say with certainty that this is true of labor statistics derived from the Census Bureau's Current Population Survey, which is the source of much (most?) of the BLS' statistics.

Posted by low-tech cyclist | April 13, 2008 1:21 PM

Re: Age discrimination in hiring is pervasive and brutal when you're over 50

How do they know your age? They aren't allowed to ask until you're hired and you have to fill out the forms. Many people in their 50s can pass for younger. And you don't need to include the distant past on your resume (like your first job out of college); seven to ten years of work history is usually adequete. Noting a college degree but not the year (again, until you fill out the forms after hiring) also would keep your age off the record at first. Also, is this above true in all careers? I know it's terrible in IT or engineering, but what about other fields? The people I know who have had to job hunt later in life were able to get jobs, but generally at lower wages and more meager benefits.

Posted by JonF | April 13, 2008 1:24 PM

Here's a paper (in PDF) that may shed some light on the increase in the jobless-unemployed gap from the late 1960s through the late 1990s: What Do Male Nonworkers Do? There's a lot going on in it, I haven't had time to read it thoroughly, and other than the statistic that 30% of 1997-99 male nonworkers had no income (which is certainly significant), I couldn't find any quick takeaways.

Posted by low-tech cyclist | April 13, 2008 1:30 PM

All this data on the employment status of individuals comes from the Current Population Survey which as has been pointed out only includes the non-institutional population. So jail is not the answer. Although the high jail population does result in many ex-convicts who often have trouble finding work and may well drop out of the labor force (and thus would be considered jobless, not unemployed by the BLS definition).

Other data from the BLS, like the monthly payroll employment, comes from a separate survey of businesses. There is also quarterly data on employment that comes from the unemployment insurance system which is generally considered the most accurate measure of employment (number of jobs).

Posted by bilowe | April 13, 2008 1:35 PM

Kathy G. and Peter are both correct about the dramatic increase in disability. I don't think this is due to fakery, however. The system assumes a lot of people are faking (even though there's very little evidence of this), and makes it very hard to qualify. I think that people who in previous eras were motivated to work, even though they had chronic medical conditions, are making some rational choices here. If you have a lot of medical problems, you may well be much better off on disability than working at a no-benefit job (even though the job would probably supply more income than disability would), because people on disability, regardless of age, automatically qualify for Medicare. In past eras, in which good jobs with benefits were much more plentiful, fewer people had to make these ugly choices. In other words, it's easier to ignore and/or cope with ones symptoms when one has a good job.

Posted by beckya57 | April 13, 2008 2:18 PM

What about househusbands? Those have been on the rise since the 70s as well.

Posted by Mo | April 13, 2008 2:35 PM

Kathy G. and Peter are both correct about the dramatic increase in disability. I don't think this is due to fakery, however. The system assumes a lot of people are faking (even though there's very little evidence of this), and makes it very hard to qualify. I think that people who in previous eras were motivated to work, even though they had chronic medical conditions, are making some rational choices here. If you have a lot of medical problems, you may well be much better off on disability than working at a no-benefit job (even though the job would probably supply more income than disability would), because people on disability, regardless of age, automatically qualify for Medicare.

Yes, those are good points, but on the other hand some disabled people should find it easier to work thanks to the accommodations which the Americans with Disabilities Act requires.

Posted by Peter | April 13, 2008 2:48 PM

And then there is the detail that, once your Unemployment runs out, there is no way to count you any more. Even if you are still looking (unsuccessfully) for work. Been there, had it happen. In short, the "unemployment" numbers at best show how many people have been losing jobs recently. Nothing more.

Posted by wj | April 13, 2008 3:26 PM

For a clear, non-technical description of changes in the calculation of employment numbers:

"Numbers racket: Why the economy is worse than we know", Kevin Phillips, Harper's, May 2008.
http://www.harpers.org

Phillips attributed evil motives to all these changes, which I believe overstates the case. High level economic data are abstractions. Counting "employed workers" for a large and rapidly changing economy is not like counting apples.

The US government’s data collection people do the best they can with the absurdly small funds we provide. If you want better numbers, I suggest asking your Congressmen to increase their budgets. In this, as in all things, we get what we pay for.

Posted by Fabius Maximus | April 13, 2008 3:44 PM

Following up on what Peter and Becky said, I doubt there's a lot of faking going on. I used to be a social worker for people with developmental disabilities, and one of the things I would help them do is apply for disability under SSI. This was almost always a difficult process; even though these folks were clearly disabled (and would not have qualified for my services if they weren't), almost all of them would routinely be denied the first time they'd apply. Then they'd have to appeal that decision, provide additional evidence, etc.

This was the mid-90s and maybe it's changed since then, but under Bush I doubt it's changed for the better.

Posted by Kathy G. | April 13, 2008 4:13 PM

The Employment-Population Ratio for men aged 25-54 years peaked in 1968 at 95% and has declined -- with large swings -- to present level of aprox 87%. As someone previously mentioned, there are many factors at work here.

You can create your own customized charts:
www.bls.gov

Click on "Get detailed statistics", click on the first icon "Create customized tables", then select your target group and criteria. You can show data as numbers of % change (under formating options), and as table and graph.

Posted by Fabius Maximus | April 13, 2008 4:33 PM

There are an increasing number of white collar fields -- like my fields of advertising and technology -- in which you're essentially done when you're forty-five or fifty. Or "freelanced," as we say. I know more than a few middle-class white men in their early fifties who will pretty much never work again, except maybe for minimum wage.

Posted by Alan in SF | April 13, 2008 4:38 PM

Re: I think that people who in previous eras were motivated to work, even though they had chronic medical conditions, are making some rational choices here.

During the 70s it was fairly easily to get on disability and more than a few people double-dipped-- got a disability check while working off the books. The Reagan administration put a stop to this by making it very hard to go on disability, so much so that the legitimately disabled sometimes couldn't get qualify. We're now at the opposite end of the pendulum swing and it's fairly easy to get disability at least for physical reasons. I am surprised the GOP, with its usual contempt for the poor and sick, did not go back to Reagan ways though. The comment about Medicare is spot-on though: that is a big draw for people with mild disabilities who may have trouble finding a job with good insurance.

Re: And then there is the detail that, once your Unemployment runs out, there is no way to count you any more.

This is often said, but it's not true. The household survey does not depend on who is collecting benefits. It asks two questions: Are you working? And if you answer No, then: Are you looking for a job? If you answer No to the first and Yes to the 2nd you are counted among the unemployed. I do wonder however how many people say that they are self-employed (hence, employed as far as the survey goes) when all they are doing is selling junk on E-Bay. Back in the tech bust, a lot of IT guys had sidelines businesses which no longer brought in an income, but which allowed them to claim to be still working, which is useful when job hunting since employers look askanace at anyone who has been out of work for too long.

Re: I know more than a few middle-class white men in their early fifties who will pretty much never work again, except maybe for minimum wage.

Does anyone even pay minimum wage? Maybe some greasy spoon out in Nowhereville. For sure in any populated are very few jobs (at least those on the books) are paying minimum: even fast food places usually start around $7 or $8 depending on the shift. That's not a tribute to employer generosity, but to the declining value of the minimum wage which has largely fallen below what pure markets inputs will produce.

Posted by JonF | April 13, 2008 7:02 PM

I work as a freelance IT support consultant. Who in the government knows or cares?

And, yes, in the IT field, if you're over 40 and not already ensconced in a position, you're history. Ageism is big time in the IT field.

"Does anyone even pay minimum wage?" Lots of places, trust me. When I was a student at San Francisco Community College doing work study, I got minimum wage (or at least minimum wage in California, which is higher than the mandated Federal minimum wage.)

Don't confuse the Federal minimum wage with the state levels. You're correct that $7-8 is more likely than $5-odd. But that's still minimum wage.

There's also been a hell of a lot of movement into a "non-traditional workforce". I remember reading a couple years ago that a significant percentage of the California workforce does not work in a "traditional job" - they either freelance, work several jobs, work off the books altogether, or are self-employed in some other sense. A lot of these people will fall through the cracks in the traditional accounting of "employment."

So Matt is right that not all of these people without "jobs" are starving homeless on the streets. But it does show that conventional employment is not working for a variety of economic and social reasons.

Posted by Richard Steven Hack | April 13, 2008 8:28 PM

wj--You don't have to be receiving unemployment insurance to be counted as unemployed. The unemployment and labor force statistics come from a monthly survey of households, the Current Population Survey.

So far as I know, there were no methodological changes in the CPS that would have led to any changes in the calculation of the unemployment rate, or of the labor force participation rate.

As many have noted, those in prison are excluded from the population counts. However, arithmetically, a rising encarceration rate will mean a lower labor force participation rate (the percentage change in the numerator will be larger than the percentage change in the denominator.

Finally, since the mid-1990s, the labor force participation rate of older (55+) males has risen substantially, after having declined since the Current Population Survey began in 1948.

Posted by Donald A.Coffin | April 13, 2008 9:34 PM

"One thing that's going on, though, is that good portion of those men are on disability. The number of men (and women) on disability has been rising for some time now. In fact, quite a few of the women who used to be on the welfare rolls have moved on to disability instead."

A few years ago, in Rio de Janeiro, I went out to dinner with Brazilian girl I met on Match.com. It was a double date with one of her Brazilian friends and another American guy. The other American guy had mentioned that he spent most of the year in Brazil. When the girls went to the bathroom together, I asked the guy what sort of work he was doing to support that lifestyle. He said he was on permanent disability from the NYPD. This was guy in his early thirties who looked perfectly healthy, but apparently he had some imperceptible shoulder injury that warranted disability. This injury didn't keep him from lifting weights, as he looked like he did some recreational weight training. Nice deal if you can get it, I guess.

Posted by Fred | April 13, 2008 9:57 PM

To the people making the point about disability being an attractive option because you qualify for Medicare, it should be kept in mind that there is a 2-year waiting period for this. So it probably wouldn't look that attractive to someone who really was able to work (although the point about it being difficult to find a job with benefits is true).

Posted by Lauren | April 14, 2008 7:02 AM

Re: To the people making the point about disability being an attractive option because you qualify for Medicare, it should be kept in mind that there is a 2-year waiting period for this.

Did this change recently? Didn't it used to be one year?
Anyway, bear in mind that many people on disability (at least middle class and the better paid working class folks) are not just on Social Security disability, they are also receiving income from private disability insurance via their former employers. That often includes health benefits at least for a limited period of time. A friend of mine went on disability in the 90s due to AIDS (when the disease was still a death sentence) and he was covered by his former employer's policy for ten years before having to convert to Medicare.

Posted by JonF | April 14, 2008 6:06 PM