Measuring Employment: The Civilian Employment-Population Ratio

Posted By PK Last updated February 21st, 2013 28 Comments

8.3%.

The significance of that number? In this instance, I’m talking about the most recent BLS unemployment report for January 2012. The top-line number, U3 (total unemployed as a percentage of the labor force), is the rate most often quoted in news stories and reports. Let’s continue the trend and say that the top-line number improvement is an encouraging sign. As recently as September of 2011 the unemployment rate was 9.0%. Of course, drawing a trend line from a few months of data isn’t the most honest graph you can make, but you can’t call this drop anything except what it is: an encouraging sign.

Why Look at The Civilian Employment-Population Ratio

Remember, U3 is defined as the total unemployed (over age 16) divided by the total civilian labor force. That means you’ve got two variables which, theoretically, can be messed with to make numbers appear a certain way.

First, you’ve got the total unemployed. To determine ‘unemployment status’, people have to demonstrate they are looking by contacting companies, filling out applications, sending resumes, placing or answering job ads, contacting union or professional registers, and/or some other means in the last 4 weeks. Even though this seems pretty cut and dry, it’s not always easy to place a person in a category. 15 hours of part time work for a family owned enterprise? Employed. Any part time paid work? Employed. Self-employed, even if no money was made in a week? Employed. (I point these out just as corner cases, but you could make an argument either way…).

Secondly, you’ve got the nebulous ‘labor force’. It is politically advantageous to over-count labor force, as increasing the denominator will lower the top-line number. However, it isn’t exactly easy to estimate how large the labor force is (you can’t survey everyone… well, except during the census). That’s why you see lots of complaints when the report includes large swings in the estimated labor force (this report had an incredibly high 1.25 million person adjustment).

From Carter to Obama… 35 Years of Employment Summarized

On the other hand, you’ve got the civilian employment-population ratio. The denominator in this one is hard to count, although the numerator might be open to some tricks… basically, take the number of employed, non institutionalized civilians over 16, and divide by the total population fitting those criteria. Let’s see how it looks (St. Louis Fed data):

The Civilian Employment-Population Ratio 01/1977 - 01/2012

The Civilian Employment-Population Ratio since 1977

We’d expect the ratio to drop a bit since the Baby Boomers are retiring faster than people are entering the workforce (currently – this won’t be true forever)… but that graph is still worrisome. The number currently sits at 58.5%, where it hasn’t been since September of 1983 (it went to 58.6% the next month and hasn’t looked back until recently). Consider this: in April of 2000, the number was as high as 64.7%, while U3 sat at 3.8%. If (and that’s a huge if… note that picking different dates would yield different results) participation rates were the same? Unemployment wouldn’t be 8.3%. It would be 10%.

And there you have it – another thing to monitor in the jobs report!

What do you think about this ratio? Is it more or less useful than the top-line unemployment numbers? Do you think a lot of Baby Boomers left the workforce during the recent recession?

If you enjoyed this post, let others know!

Filed Under: Economics Tagged With: bls, civilian employment, employment ratio, government statistics, jobs, unemployment

http://www.thadthoughts.com/ Thad Puckett

It is that second variable that gets left out of so much reporting. The unemployment rate is dropping, but mostly (only?) because of the number of people dropping out of the workforce totally. Not a good picture.
- http://www.dqydj.net/ PK
  
  Some of the drop is certainly legitimate – Boomers leaving the workforce for retirement. However, the fact that the rate was approaching 65% (and the steep drop) means that some of it may have been ‘forced retirement’. Also, with unemployment insurance compensating for 99 weeks, some of the drop may have been people who remain on unemployment insurance but haven’t fulfilled the job search criteria (even though you are supposed to be searching on UI).
  
  So yes, I agree. With the Boomers, I expected a slow decline in the participation rate, not a precipitous drop. However, I think the rate itself is certainly an improvement – and if any of the forced retirees started to search for jobs again it would also be reflected in the rate. I think in a few more months we’ll have a clearer picture of the environment.
http://www.thefreefinancialadvisor.com/average-joes-money-blog/ AverageJoe

My mind is actually on where this graph is headed….long term. When the baby boomer spike ends in several years (decades?) we should have (in theory) a nice rubber band. I remember Harry Dent projecting some of this early in the century.
- http://www.dqydj.net/ PK
  
  Well, we know we can’t just draw lines extending off the edge of it, am I right?
  
  Yeah, I can recall as recently as 2005 a heated argument about what would happen when Baby Boomers started exiting the workforce. Generation X isn’t big enough to carry water for the Boomers, but Y is pretty large. However, having a large generation hit retirement all at once, with current life expectancies? It will be interesting to see where this number ends up – it is entirely possible that ~60% is the ‘new new normal’.
  
  Thanks for the comment!
http://www.investitwisely.com/ Invest It Wisely

I don’t really have much faith in the unemployment numbers, and I like charts like these more. The unemployment numbers also don’t really say much about underemployment, discouraged workers, jobs not created due to excessive regulation, etc…
- http://www.dqydj.net/ PK
  
  “[J]obs not created due to excessive regulation” – I would love to see this chart, actually; something going back to Carter, perhaps? I’ll try to find something that isn’t too biased… but that’s a good one.
  
  Minimum Wage: excessive regulation? News at 5!
http://myuniversitymoney.com/ My University Money

Has the effect of women entering the workforce in greater numbers during the years in the graph come into play at all? One might say that there are greater competition for jobs thanks to that social change right?
- http://www.dqydj.net/ PK
  
  Yeah, it has raised the ratio quite a bit. The sexes are roughly in balance, but there are a bit more females in the Unites States. Any reading over 50% means there has to be more females in the workforce – and the drive for more female workers started en masse around World War 2 (you know, Rosie the Riveter, haha).
  
  Under Reagan you had the yuppie movement (young urban professional) and DINKs (dual income, no kids) which drove up the number even more (all coinciding with rising productivity and quality of life). However, since that second wave was Boomers? We might be seeing a generational shift. Thanks for the comment!
- freeby50
  
  From the Census site, they give civilian labor force participation rates per gender of :
  
  Women :
  1970 = 43.3%
  1980 = 51.5%
  1990 = 57.5%
  2000 = 59.9%
  2010 = 58.6%
  
  Men :
  1970 = 79.7%
  
  1980 = 77.4%
  1990 = 76.4%
  
  2000 = 74.8%
  
  2010 = 71.2%
  - http://www.dqydj.net/ PK
    
    Can you find any data older than 1950? This would be an interesting article if so – changing perceptions in the Post WW2 era. This 1950 data says female participation was 33.9%.
    - freeby50
      
      Yes there is census data going back to 1900. I talked about it in this article :
      http://www.freeby50.com/2010/10/historical-look-at-womens-participation.html
      
      Its interesting to note the large difference in labor force participation between married and single women.
      - http://www.dqydj.net/ PK
        
        Nice article - from your graph it looks to me like mothers/married women juiced the returns in the WW2 & beyond era. Single women stayed steady until the 1960s (Maybe coinciding with the sexual revolution? How much effect do you think The Pill had on the female participation rate?)
        
        I feel like we’re just peering into the rabbit hole. Want to collaborate on something with this topic?
http://www.moneyspruce.com/ Jeffrey Trull

Interesting! I’ve never though of this before, and I think it’s useful to consider this instead of simply looking at the typical unemployment figure. Kevin has some good points about unemployment, too.
- http://www.dqydj.net/ PK
  
  Yeah, it’s a good number – we just have to make sure we consider generational shifts instead of painting the number as solely good or bad. Bad? Maybe if the drop was Boomers leaving the workforce before they wanted to. Good? Maybe in terms of clearing out top-heavy employment rolls.
  
  I’ve got an article coming up about Social Security coming up, which would paint the picture as ‘bad’ when it comes to solvency for the younger generations. However, the fact that we are (as a country) able to weather the Boomer storm without missed payments until the 2040s? That’s actually impressive – you know, for government.
  
  Thanks for the comment!
http://thecollegeinvestor.com/ The College Investor

I’m with Invest It Wisely. I think the government has too much play with these figures.
- http://www.dqydj.net/ PK
  
  Which one do you like better?
  
  I didn’t like the 1.2 million revision in the latest report – I’m not saying it was dishonest, but it means that some of the other months were pretty sloppy. However, U-3 should be a reasonably honest report of what it purports to be (the narrow definition given), even if it misses a lot of cases that only get picked up by U-4, U-5, and U-6. The truth is people don’t generally view the economy through the numbers anyway – if they have a bunch of neighbors without jobs? We’re still in a recession.
  
  I think what is happening on the ground matters much more than the numbers we are reported.
freeby50

In 2000 the participation rate was 64.7% and unemployment was 3.8%. That was the peak. Today the unemployment is 8.3% and the participation rate is 58.5%. If you sum the numbers you get 68.5% in 2000 versus 66.8% today. The difference is 1.7%. About 0.7% of that is due to aging of the population and a higher % of retired people. College enrollment went up about 5 million from 2000 to 2010, so I imagine a decent % of the population is in college instead of working.
- http://www.dqydj.net/ PK
  
  Do you have a link on the .7%? I wonder if we can take a guess as to how much of the rate was ‘forced’ retiring… I’m thinking Boomers who wanted to work a while longer but, well, no longer can due to the job market.
  
  Something weird is going on with the adjustments too, because as recently as January of last year the Employment/Pop ratio was 58.4% and unemployment was 9.1%.
  - freeby50
    
    The 0.7% was just a quick calculation based on the total adult population in 2000 vs 2010 versus the # of people in different age groups. Unfortunately a day later I can’t even remember how I did the math. I do know the Census website was involed. One interesting point is that the % of people in the labor force among people older than 65 is actually higher now than 10 years ago. So Boomers are hanging on to their jobs more now than before, I assume thats cause some of them are broke and have no money to retire with.
    - freeby50
      
      Here’s the Census page with participation rates broken down by age and sex :
      http://www.census.gov/compendia/statab/2012/tables/12s0597.pdf
      - http://www.dqydj.net/ PK
        
        I also found these stats from The World Bank… interesting stuff worldwide. Those ones go back to 1980. Maybe that could make an interesting presentation? I’m thinking something on IBM Many Eyes.
freeby50

Additional… I wonder what % of the population is in prison or serving in the military in 2000 versus 2010. I think those groups are excluded from the civilian population and labor fore #’s. May not matter at all, but either of those could have a 0.1% impact too I wouldn’t be suprrised.
- freeby50
  
  The % of the adult population in prison only went up 0.01%
  0.69% in 2000 versus 0.70% in 2010.
  Military personel in 2010 was around 1M total. It probably didn’t vary enough to chagne the total employment % noticably either.
  - http://www.dqydj.net/ PK
    
    Thanks for tracking that down! I was actually under the impression that the prison population has been skyrocketing from the articles I’ve read, heh. That puts it in perspective.
    - freeby50
      
      THe prison population number actually did go up a lot. It went from something like 1.3 million to 1.5 million. Thats a pretty big jump for 10 year period. But in relation to the total population it doesn’t seem as big cause we’re talking about 200+ million adults.
      - http://www.dqydj.net/ PK
        
        That reminds me of this quote from James Bovard:
        
        “America needs fewer laws, not more prisons.”
Changeiscoming1

This is very interesting information as it appears to display how easy it would be to manipulate the unemployment numbers by impacting the labor participation rate. It appears to show a significant number that were close to retirement chose retirement in lieu of actively seeking employment. It would suggest that the “drain” on the system will only increase as these individuals will become “users” of government programs and not “payers” for government programs.
- http://www.dqydj.net/ PK
  
  There is definitely some incentive to manipulate the unemployment rate – it’s a hard number with lots of observers, and one of the common measuring sticks to use ranking presidents. At some point, your numbers would become so unbelievable you’d lose observers, sort of like that Abe Lincoln quote (“You can fool some of the people all the time, and all of the people some of the time, but you cannot fool all of the people all the time.”).
  
  You make a good point on the users vs. payers – we just did an article on Social Security discussing that very phenomenon – and the fact that only 2.9 people pay into Social Security for every person who draws from it.
  
  Thanks for your comment!