Unemployment Has Doubled: Ratio Increases by State, Surprisingly Consistent.

Ryan Avent, reporting live from the AEA meeting, mentions a panel where Robert Shimer gives an excellent overview of the labor market (my bold):

Rob Shimer, who’s well known for his work on labour markets, offered additional thoughts on the employment situation, beginning by laying out a few key labour market facts. Take any given group within the labour force, and the crisis has essentially generated a doubling of the unemployment rate. Turnover among the unemployed has also been quite low since the initial decline. It hasn’t been a slump in which many different groups rotate through joblessness; instead a lump of labour fell into unemployment and has struggled to return to the work force.

That’s a great way to phrase it.  Charlie Eisenhood has looked at this with breakdowns of age and education here and there’s been roughly a doubling in each case. And it isn’t just unemployment. Arjun Jayadev and I looked at this with underemployment, or people working involuntary part-time jobs, and we saw that it doubled in every occupation and in every career. To me this is strong evidence that our current employment problems are primarily cyclical and demand focused.

This plays out at the state level. Greg Mankiw quotes Raghu Rajan citing the work of Erik Hurst saying that “structural unemployment may account for up to three percentage points of total unemployment. In other words, were it not for construction, the US unemployment rate would be 6.5% – a far healthier situation than today.”

I very much doubt that, and the video linked to by Rajan doesn’t mention 6.5%, and I don’t see a working paper anywhere that quantifies that. The video focuses on the state of Nevada, which has the highest unemployment in the country. It wouldn’t surprise me that the state of Nevada is in particularly bad shape given that it also gets a large share of its trade through tourism, an industry that suffers in a recession, alongside a large housing bubble. (We’ll try and get into the specific points he brings up in another post.)

But let’s just say that Nevada is uniquely messed up. What about all the other states? Again, following the Shimer rule above, they’ve basically all doubled their unemployment rates:


That’s the ratio of state U3 unemployment rates, taking the 11/2010 rate and dividing it by the 11/2007 rate.  (It’s consistent with other dates.)  The average is 1.93, and the median is 1.87. So roughly a doubling. So to me Nevada having a higher unemployment isn’t nearly as interesting as why Nebraska has a 4.6% unemployment rate when it used to have a 3% unemployment rate – a 50% increase. Why has North Carolina’s unemployment rate doubled from 4.8% to 9.7%? Yes Nevada is an important story, but it’s clearly an outliner outlier in what is a national trend.  They didn’t all have housing bubbles.

This entry was posted in Uncategorized. Bookmark the permalink.

16 Responses to Unemployment Has Doubled: Ratio Increases by State, Surprisingly Consistent.

  1. Mike,

    I’m late to your call for suggestions, but here’s one I’m really interested in; posts/research on unemployment and reemployment for older workers (40’s and up, but even 30’s as compared to 20’s).

    It appears that losing your job is much more serious for an older worker than a younger one. They’re a lot more likely to be unemployed for a long time and to have to accept a job of much lower prestige and pay. One reason is that their skills can become highly specialized and then their specialized position ends, or those skills otherwise become obsolete. Another is that a young worker is less willing to think a job is beneath his dignity; he’ll work low level, or even at Starbucks when he was an entry manager before, rather than a senior one like an older worker may have been. When people get older they expect to have gotten to higher positions and to not be taking orders from people much younger than them, and society expects this too.

    In addition, the young have more energy and time, and more of a future with the company, which makes it more worth it for the company to invest in them.

    Here is a good article on this from the WSJ:


    And note in this article the talk about older law school graduates:


    I did some initial looking and I found this:

    – Displaced men age 50 to 61 are 39 percent less likely to become reemployed each month than those age 25 to 34, and men age 62 or older are 51 percent less likely.

    – For displaced men who become reemployed, the median hourly wage on the new job falls 20 percent below the median wage on the old job at age 50 to 61. For those reemployed at age 62 or older, the new median wage falls 36 percent below the old median wage.
    By contrast, men’s median wages fall only 4 percent at age 35 to 49 and 2 percent at age 25 to 34.

    From Richard W. Johnson and Corina Mommaerts at The Urban Institute, Prepared for the 12th Annual Joint Conference of the Retirement Research Consortium, August 5-6, 2010, at:

    Click to access II-b%20Johnson-Mommaerts.pdf

  2. geaugailluminati says:

    stop paying attention to the unemployment rate…the headline on the jobs report was that unemployment had fallen to 9.4%, but that was largely because of the shrinking of the denominator, as 260,000 americans no longer count in the official statistics because they’ve dropped out of the workforce, which is now at the lowest it’s been in over 27 years, back to the time before women were forced to enter the workforce by anti-labor policies that made it impossible to support a family with just one wage earner…only 103,000 jobs were added, less than needed to make up for population growth…..since the recession ended in july of 2009, we’re now well into our second year of recovery; over the past year the economy has only added back 1.1 million of the 8.4 million jobs that were lost during the recession; however, just to create jobs to make up for the increase in the population, we should be adding 1.5 million jobs a year; thus at the rate jobs are being added in this recovery, the labor force participation rate will continue to shrink, and we’ll never get out of this hole…


  3. geaugailluminati says:

    mr serlin: here’s a good breakout of employment for men by age group:


  4. ScottB says:

    Thanks, great article. Very similar chart if you use U-6 instead of U-3.

  5. Jason says:

    I guess I’m going to be that guy. Your comment about Nebraska is probably not statistically significant.

    The 12-month change in November in the total non-farm payrolls was 0.5%, but the standard error for Nebraska for 12-month changes total non-farm payrolls is 14 797 / (944.4 * 1 000) = 1.6%.

    Dividing rate numbers three years apart is probably even more unreliable.

    I plotted out what annual job gain/loss change would be with this uncertainty for Nebraska and the data points (with error bars) look like they could be consistent with no change, a loss or even a gain from Nov 2007 to Nov 2010.


    I do wish the BLS would do a better job of providing information about errors instead of this change in the number of payrolls number.

  6. Mike says:

    Ha don’t worry Jason, this blog is a place for those guys.

    I’ll check it out your way later, but quickly in passing, Nebraska seasonally adjusted unemployment rate average, standard deviation and +2 * and – 2* are:

    Average 2000-2007, Average 2010:
    3.438947368 4.79
    0.47940021 0.152388393
    4.397747789 4.485223215

    If I use number unemployed, same:

    33186.94737 47110
    4679.049039 1748.745709
    42545.04545 43612.50858


    Will think through your approach and check out data.

    • Jason says:

      Ok, cool. Thanks for the pointer to the data. I did a LOESS regression through the employment data (labor force and unemployed) to subtract out the secular trend, and the subsequent standard deviation in the labor force data was about 3200 for Nebraska, comparable to the BLS error number for 1-month changes. Anyway, long story short:

      Nebraska increase = 1.5 +/- 0.2 at 90% confidence, so it checks out.

      Nevada increase = 2.8 +/- 0.3
      North Dakota increase = 1.2 +/- 0.1

      The typical (two sided) error bar seems to be on the order of your vertical grid spacing.

      ps I wasn’t doubting the overall trend.

  7. Pingback: some rudimentary employment stats this m… « D-O-A

  8. Pingback: FT Alphaville » Double or quits? The structural unemployment question

  9. Pingback: More on Erik Hurst and FT Alphaville on Structural Unemployment « Rortybomb

  10. Pingback: Double Trouble

  11. Pingback: ZIRP and ZMP - NYTimes.com

  12. Pingback: Blog Carnival: Weekly Link Round-up for January 7 and 14, 2010 | Blueleaf

  13. Pingback: Raghuram Rajan verdreht die Zahlen « Kantoos Economics

  14. Pingback: NV unemployment rate nearly triples since 2007 : Blog – Work News Now

  15. Pingback: Economists mull how “Nevada is uniquely messed up” | Workforce Connections

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 )

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