What Conclusions Can You Draw on Increases in Unemployment by Age and Education?

Roosevelt Institute intern Charlie Eisenhood dug up this data on the unemployment rate by age and education from the Current Population Survey (CPS). Here it is in September 2010:

And here it was in December 2007 when the recession started:

Here is the difference between the two, along with the percent increase, so a (100%) is a doubling:

What jumps out for me? College educated 20-24 year olds have the highest percentage increase. This should hit against a structural unemployment story, as college educated people have the ‘freshest’ skills and incredibly high mobility. It’s worth pointing them out in particular because if their careers hit a rough spot hysteresis sets in and they’ll have serious wage losses years down the road (see this classic White House blog post on the subject by Peter Orszag on the research of graduating into a recession), and because the crisis is often seen as not a big deal for college educated workers.

The other thing that jumps out at me is that everyone 55-64 has more than doubled their unemployment rate. One thing we aren’t talking about enough is that someone who is 60 and has been unemployed for a year isn’t going to find a decent job again. Why don’t we temporarily lower the retirement age, conditional on a bunch of hoops? Other ways of looking at the labor search outcomes of 55-64 year olds are extra worrying. Why don’t we do that especially rather than raising the retirement age, as the December debate is likely to be over, when 55-64 year olds have had such a large jump in unemployment?

What jumps out at you when you look at this data?

This entry was posted in Uncategorized. Bookmark the permalink.

21 Responses to What Conclusions Can You Draw on Increases in Unemployment by Age and Education?

  1. Daniel Aronen says:

    One thing I did with this is invert it so that you’re looking at an employment rate and the change there, rather than the unemployment rate. The way I’m looking at it is say you’ve got two groups of people, one hundred each. One group starts out with 83 people working ( 20-24, < HS), one starts out with 97 people working (20-24, college grad). In the first group, seven more people lose their jobs and in the second, six lose theirs. Which is relatively worse off? The first group has 9% less people working and the second has 6% less people working.

    Anyway, my point here is that it's the top left portion of the table that is taking the worst of it. Not that the young college grads have it good or anything, but I think that when you need to be careful when you look at the relative change in unemployment rate and what it really means. If you look at those over 55, their relative change in employment rate isn't that bad compared to other age groups. Not that I disagree with the assessment that unemployed older workers are going to have a difficult time finding a decent job again.

    At any rate, this is one of those things that's pretty complex and it's difficult to make any conclusions without drilling deeper into the data. The one thing that I did notice is that those having some college seem to be faring just as bad or worse than HS grads without any college, depending on the age bucket. Not sure what that might mean, though.

  2. Mike says:

    I agree completely that a 25% unemployment rate for 20-24 year olds with less than a high school education is a travesty. That will lock those workers into a lifetime of decreased earnings, capabilities and opportunities. I didn’t mean to make this a comparison, though we do try and beat up structural arguments around here and 20-24 college educated shouldn’t have any problems associated with mobility, skills (or even locked in expectations of wages), hence me pointing them out.

    I need to start moving into the employment rate, since a lot of what I’m thinking about these days is how many workers are simply dropping out of the labor force. I’ll probably re-approach this in the near future. Thanks for commenting!

  3. Ken Houghton says:

    It certainly belies the argument that was being made a few months ago (and still is, by the people pushing Waiting for Superman) about how college-educated people aren’t being slammed by the Great Recession because their unemployment rate is much lower than anyone else.

    And it reinforces that older workers are being disproportionately affected.

  4. “Why don’t we temporarily lower the retirement age, conditional on a bunch of hoops?”

    It’s very inefficient to have a large number of people who are capable of being highly productive, and want to work, producing nothing.

    It would be much more efficient and growth promoting to just have strong economic stimulus, and much stronger age discrimination laws and enforcement. Plus, lowering the Medicare age to at least 50, so companies wouldn’t have high health care costs as a reason not to employ older workers.

  5. Also, a big thing is that older workers are a lot less likely to be unemployed, but when they are unemployed it’s a lot more serious in some ways. There’s a much bigger risk of not being able to find a job nearly as good as the last one (the higher your income, the farther it is possible to fall, plus skills can get very specific to just the current job, and they can become obsolete). There’s also a bigger risk, it seems, of very long term, or permanent, unemployment.

  6. I’ve seen an increase of people in the bottom left corner of your graphs just give up. They feel beaten down and think they are unable to get a job because of their age. Having had several off the record conversations with HR departments I know many companies are actually looking for older people to join the ranks in the down times. Seniors tend to be better workers.

  7. I had some recent thoughts on the problem of unemployment for older workers if you’re interested:

    “Why doesn’t the labor market “clear”? If demand falls in markets for other productive factors — say, wheat…— the price usually drops until the excess supply is mostly gone. What is unusual about … labor is that excess supply, which shows up as unemployment, can be prominent and persistent.

    Why? In short, the difference is morale. Factors of production like wheat or trucks or pumps don’t have morale issues. Human beings do. …”

    – Robert Shiller, at: http://www.nytimes.com/2010/10/03/business/economy/03view.html?_r=1

    I say this all the time. It’s very important.

    But there’s something new I’d like to add that I’ve been thinking about.

    Why is unemployment among older workers especially difficult and long lasting?

    I think an important reason is that due to human factors, their wages are especially sticky.

    Highly educated and specialized workers can peak in productivity in their 50s or 60s, or even later. One of the greatest mathematicians in history, Euler, did his best work in his 70s right up until he died at 76. The gains in knowledge and understanding over time can outweigh the loss in energy and endurance up to quite old ages.

    But most workers aren’t so educated and knowledgeable, and as they get into their 50s, except for people that take really good care of themselves (and have some luck), their health, energy, endurance can be much worse than that of a worker in his 20s or 30s. They can have far less time and energy to put into a job. Plus, a lot of their skills can become antiquated and a lot of older workers are reluctant to learn new technologies and skills, in part due to the fact that after all of the effort they won’t be milking those skills for nearly as long as younger workers will.

    And on top of this, the business’s health insurance cost for a 58 year old is way way higher than it is for a 28 year old (can we please have single payer type insurance like every other advanced country, or at least cut the Medicare age to 50).

    Thus, for these reasons, and others, many workers, if not most, are less valuable in their 50s and 60s than similar workers in their 20s – 40s. They create less net wealth, or a lot less.

    Now, using Shiller’s analogy, if this were wheat, then the price of 50-60 wheat would just drop a lot lower than 20-40 wheat, and the market would clear. There’d be no unemployment of wheat. It would all be utilized. The price would just keep dropping until it was.

    But try dropping the wage of a 58 year old by 30%. See what happens. See what morale is like. Ever hear of agency problems? There are tons of ways employees can underperform and screw you that are hard or impossible to detect, or pin on them.

    Try getting a 55 year old manager, highly specialized in a division of a company that was closed down, that was paying $120,000, to take a new job starting over in a new company, in a new division, with the 20-somethings, making $45,000. It would be very hard to get him to agree to take that job, and the new company would be unlikely to offer it to him knowing that after years of expensive training and seasoning, he would just retire not long after. Plus, he would constantly be looking for a job like the one he used to have, and his attitude and enthusiasm for a job much less prestigious and lower paying than what he had long grown accustomed to would be much worse than it would be if they just hired a young person who’d be very happy and excited to have the job.

    People think they should always be advancing and getting higher paid as they get older. It’s going to really hurt their morale and attitude to go way in the opposite direction, way backward, taking orders from people close to half their age. And employers know this, and so will have a reluctance to hire them.

    But due to age, and antiquation of skills, the actual amount of value that they can create in the market may really actually have dropped from $120K to $45K.

    Again, if they were wheat, their price would just drop from 120 to 45, and they would be utilized; the market would clear. But because they’re humans, that price will be very reluctant – sticky – to drop, and it may not drop enough at all, leaving the person unemployed at age 55 for the rest of his life, even though he could have still produced a great deal of value for many more years.

    So, bottom line, a big cause of protracted unemployment for older workers may be that in many cases their value creation drops a lot with age, but they think, and our society, culture, and human nature, encourages this, that they should keep making the same or more as they get older, and getting the same or more prestige and responsibility, not a lot less.

    Of course, so many freshwater economists just assume this away, as well as anything that says humans aren’t robots with super simple utility functions that only consider their own and current consumption.

  8. Daniel Aronen says:

    I’m not buying the structural argument either. I suppose there’s something to it, but that can’t be the primary driver here. We’ve just had a giant housing bubble wipe out how much wealth and all of the ramifications that go along with that and the problem is that there’s a misallocation of labor? Come on! Somebody’s trying too hard to find something that’s just not that difficult to see. People making structural unemployment arguments kind of remind me of people justifying bubble housing prices. Equilibrium hooey!

    Anyway, I was just doodling with the data for a bit and just trying to get a better grip on the percentage change. Something about that number just didn’t feel right to me and I wasn’t sure why at first. I was thinking if one group started with 0.1% unemployment and went to 1%, still extremely low, they’d have seen a 900% increase! So I flipped it over and then the percentage change in the employment rate made more sense to me.

  9. grooft says:

    I agree with Daniel Aronen on flipping to “employed” instead of unemployed to highlight issues. But I would also add some way to consider the population size of each cell.

  10. Pingback: Sustainable Prosperity and the American Dream | The Murninghan Post

  11. steve har says:

    To answer the question directly:
    What jumps out for me how hard it is to understand the tables. The visual view hides the data. It is for specialist table interpreters. There is no data that “JUMPS OUT” it is in some kind of data hole.

    Why perpetuate this kind of visual stupidity that hides the point?

    You have multiple scales in an apparent time series, the assertion that there is something to learn from studying tables. Maybe there is a trend line here somewhere? How many trend line data points. Is this a statistically testable like a range shift or a median shift or an average shift. Who built this data? What population. Where’s the data? Can it be localized to a Congressional District? Why not?

    You can see better reports of what is normal wether for, pick a place, Minneapolis. You can see shifts in what’s normal and not, temperature ranges, averages over time.

    Better Economic Data please. This is amateurs mumbling in their beer. Sorry

  12. Nathanael says:

    What jumps out at me? If people over 65 didn’t vote, this country would elect a very, very left-wing government in a heartbeat, based on those numbers.

  13. steve har says:

    Nathanael: I fit your age bias; I don’t fit your left right conclusion

    Some old dogs have new tricks;
    Some young pups miss the target when they pee

  14. Jim says:

    Thank you for your article about unemployment and older people. Check out this original song on You Tube about growing old during hard times.


  15. Pingback: Unemployment Has Doubled: Ratio Increases by State, Surprisingly Consistent. « Rortybomb

  16. Pingback: Jobs with the highest unemployment rates « Phil Ebersole's Blog

  17. Pingback: Comparing the Numbers on Subprime Mortgages and For-Profit College Loans | Rortybomb

  18. Pingback: Who’s Had the Worst Recession: Boomers, Millennials, or Gen-Xers? | chadjohannesen.com

  19. Pingback: ‘Occupy Wall Street’: What Should a Populist Movement Ask of Washington? | Freedom Developers

  20. Pingback: Signs of Dissent: Help Us Read Occupy Wall Street |

  21. Pingback: Signs of Dissent: Help Us Read Occupy Wall Street | My Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s