A Closer Look at Unemployment

All eyes are on Friday’s NFP Report. With the markets desperately rooting for a Fed cut, especially after yesterday’s rather punk ADP Employment Report, tomorrow’s BLS release is eagerly awaited.

In the past, I have frequently mentioned that I was not a big believer in the headline Unemployment number. Its methodology — the Household self-reported survey — is suspect, and it simply ignores too many people who are unemployed.

It is not that the unemployment rate is so very bad — its just not nearly as good as the 4.6% we have heard the BLS report. If there were that much of a labor shortage, than real wages have likely risen much more than they did over the past few years.

This line of argument is often greeted by skepticism (what are you, a conspiracy theorist?).

However, what many people seem not to realize is that the source for alternative measures of unemployment comes from none other than the Bureau of Labor Statistics themselves. They actually have quite a few official measures of Unemployment — not that you ever hear about the others.

There are actually six BLS reported measures of Unemployment. (They can be found here: Table A-12: Alternative measures of labor underutilization)

The headline number that you will hear tomorrow — what BLS calls "the official unemployment rate" — is called U3. It is defined as the "Total unemployed, as a percent of the civilian labor force."

For the mathematically inclined, that looks something like this:

Unemployment rate = (unemployed) / (employed + unemployed)

Take those termed unemployed, divide that into the civilian labor force, and you get a percentage.>

Let’s look at the other measures of Unemployment BLS reports that are more inclusive than U3.


Alternative Measures of
Labor Underutilization
Measure: Definition:
U1 Persons unemployed 15 weeks or longer, as a
percent of the civilian labor force
U2 Job losers and persons who completed
temporary jobs, as a percent of the civilian labor force
U3 Total unemployed, as a percent of the
civilian labor force (the official unemployment rate)
U4 Total unemployed plus discouraged workers,
as a percent of the civilian labor force plus discouraged workers
U5 Total unemployed, plus discouraged workers,
plus all other marginally attached workers as a percent of the civilian labor force plus
all marginally attached workers
U6 Total unemployed, plus all marginally
attached workers, plus total employed part time for economic reasons, as a percent of
the civilian labor force plus all marginally attached workers (the "real world" unemployment rate)
Note: Marginally attached workers
are those who are neither working nor looking for work but indicate that they
want and are available for a job and have looked for work sometime in the recent past.
Discouraged workers, a subset of marginally
attached workers, have given a job-market-related reason for not currently looking for a
Persons employed part time for economic
are those who want and are available for full-time work but have had to settle for
a part-time schedule.


How do these various unemployment measures shake out?

U4 = 4.9%

U5 = 5.5%

U6 = 8.3%


Here’s a graphic comparison:


(thanks, Oregon!)


Now you know . . .



Table A-12: Alternative measures of labor underutilization

Measuring Available and Underutilized Labor Resources
Federal Reserve Bank San Francisco
Economic Letter, 2000-06; March 3, 2000

How the Government Measures Unemployment
U.S. Department of Labor, Bureau of Labor Statistics. 1994.
Report 864 (February)

Alternative Measures of the Unemployment Rate
Eric Moore
Oregon Employment Department, Mar-29-2006   

What's been said:

Discussions found on the web:
  1. Grodge commented on Sep 6


    Comparos for other business cycles for the other measures of unemployment?

    Is U6 at 8.3% high or low relative to other times?

    Two data points are difficult to interpret as a trend, but since 2005, the U6 has decreased by 6.7% (from 8.9% to 8.3%).

  2. bjk commented on Sep 6

    All of this focus on unemployment, has it ever actually resulted in a profitable investment or trade?

  3. Michael Donnelly commented on Sep 6

    BR I don’t get this post at all.

    Sure if we used something other than U-3 the unemployment rate would be higher, but so what?

    All the past recessions would have much higher unemployment rates as well, so we’d be in the same boat in terms of comparison with the past.

    The only useful item is comparing the US to Europe. Europe’s unemployment is always higher than the US. Is that because they use something more like U6?

    It’s the same with gasoline prices. Everyone’s focus is on the unleaded regular rate, if you said let’s start looking at the price of 93 Octane and viola suddenly gas prices are higher. Should we be worried?

  4. Mikael commented on Sep 6

    Heres a link giving a graph of what Michael above just said.:
    It basically shows that both U3 and U6 run parallel to one another, so it really makes no difference which one you use, when U3 is up, the fact of the matter is that unemployment is up, no matter which was you choose to look at it.

  5. jj commented on Sep 6

    excellent post

  6. Stuart commented on Sep 6

    it’s added enlightenment that in absolute terms the real unemployment rate is higher than stated and in all likelihood, always has been. Ergo, the public should take stated unemployment figures with a grain of salt recognizing they are subject to the same bias of always putting the most optimistic figure “foot” out. I do not know if Europe uses the U6 equivalent, but I know they do not simply use U3. The big benefit from being aware of these real statistics is in extracting consequences on future economic growth. When one hears 4.9% unemployment, they assume 95.1% are employed and as such the economic growth potential of 95.1% employment. That would be false as 95.1% are not employed. That’s where this type of analysis is extremely helpful.

  7. maximo zeledon commented on Sep 6

    Very useful Barry! I actually will need these sources for a class project.

  8. Steve commented on Sep 6

    The only useful item is comparing the US to Europe. Europe’s unemployment is always higher than the US. Is that because they use something more like U6?

    No, they use something similar to U-3.


    Unemployed persons are persons aged 15-74 who were without work during the reference week, were currently available for work and were either actively seeking work in the past four weeks or had already found a job to start within the next three months.

  9. Barry Ritholtz commented on Sep 6

    That reads very differently to me — In Europe, unemployed means:

    a) without work during the reference week, i.e. neither had a job nor were at work (for one hour or more) in paid employment or self-employment;

    b) currently available for work, i.e. were available for paid employment or self-employment before the end of the two weeks following the reference week;

    c) actively seeking work, i.e. had taken specific steps in the four week period ending with the reference week to seek paid employment or self-employment or who found a job to start later, i.e. within a period of at most three months.


    That is a far more inclusive measure than the US version, as BLS U-3 excludes:

    a) all marginally attached workers,
    b) total employed part time for economic reasons
    c) all marginally attached workers

    The definition in the US appears to be narrower.

  10. matt m. commented on Sep 6

    “All eyes are on ……” fill in the blank.

    That expression is the most overused financial phrase in the media today. When I hear it I just laugh…so I get to laugh every day. Don’t mean to sound like a jerk….it’s just such a BS phrase.

  11. Mike Nomad commented on Sep 6

    Nice story, Barry.

    The unemployment numbers have a way too powerful effect on the _perceptions_ of the average US consumer (that engine thingy that’s driving what’s left of the economy). That is why people using U-3 as the official word need to be called on it.

    A couple of posts have mentioned how European unemployment is measured. It would be interesting to see if:

    1) They have something similar to U-6

    2) If so, does their “spread” (difference between U-3 and U-6) match ours?

  12. Steve commented on Sep 6

    a) all marginally attached workers,
    b) total employed part time for economic reasons

    Marginally attached workers are not counted as part of the labor force in Europe either, since they are not “actively seeking work.”

    If you want to compare how UE rates compare across countries, you don’t have to speculate. The OECD tracks something called the Standardized Unemployment Rate (SUR). You can see it here.

  13. michael schumacher commented on Sep 6

    much in the same way that the NAR report “caused” the “massive selloff” yesterday.

    These figures belong in the garbage as that is abut as useful to normal people as they can get.

    I can only imagine how many construction jobs were “created” last month.

    Big Ben has had three money drops already today….think he was anticipating anything??


  14. michael schumacher commented on Sep 6

    like the record foreclosure information released today???

    At least try to make it look like you’re not holding up the market with “boat money”…

    ECB pushed almost $60b at it’s brokers…but it had bids for almost $150b……basically saying that’s not enough….and Big Ben to the rescue..


  15. techy2468 commented on Sep 6

    Consumer is still holding up (if we can trust the retails sales data).

    Employment is up, if we think ADP data is enough to interpret this trend.

    i think this is a bit of positive news, even though things may get ugly in the next few months.

    but guess what…the bulls dont like this good news….they just want some cheap money instead (rate cut)….

  16. techy2468 commented on Sep 6

    MS..i am sorry but you sound like a person who has a lot of short interest.

    so you get too upset when the FED/Central bankers dump liquidity into the market.

    if you had to choose between recession and inflation, which one will you chooose?

    i would go for inflation any day.

    right now thats what FED/CBs are trying to do all over the world, make cheap money available so that we can keep the negative sentiments from hitting us too hard.

  17. curious commented on Sep 6

    does that total labor force include our prison population? not trying to be a crank, but i’ve always wondered how much of an impact that has on our low #s vs. other first-world economies with lower incarceration rates. about a decade ago, i read a paper that suggested it could be between 1 and 3%, depending on how you looked at people in prison, how they would work, and how being in prison excluded them from the regular labor market.

  18. roscoe ii commented on Sep 6

    Barry, how about a 5 or 10 year chart of U6 versus U3.

    I would guess a sharp rise in U6 over past 24 months compared to flattish U3.

  19. Karl Smith commented on Sep 6

    Barry, how about a 5 or 10 year chart of U6 versus U3.

    I have a chart of that here

    As I still maintain there is nothing special about U6 and there is nothing misleading about our current unemployment numbers.

  20. Mark Hessel commented on Sep 6

    Great Post!

    So I have to ask.

    What was the “official” unemployment rate in 1980? 1990? 2000?

    Also from previous discussions by Barry —
    What was the “official” inflation rate in 1980? 1990? 2000?


    What was the “official” GDP in 1980? 1990? 2000?

    If everything has been jiggered to look better for whatever reason
    what is the point in measuring these things?

  21. M.Z. Forrest commented on Sep 6

    When people talk about a bad job market they are saying that they can’t easily (and in some case with much difficulty) find work and a salary that interests them. Unfortunately data tunnel vision causes people to keep looking at a measure and variations of the same measure to force it to answer something it can’t. When need only look at the wage stangation for anyone making under $100K a year to understand the job market is incredibly weak. We don’t have a people starving in the streets weak labor market. We have a I’m-not-sure-how-much-longer-I-can-continue-paying-my-bills-if-prices-keep-rising-and-my-wage-doesn’t-continue-to-rise job market.

  22. Nick commented on Sep 6

    Unemployment rate = (unemployed) / (employed + unemployed)

    Take those termed unemployed, divide that BY the civilian labor force,and multiply by 100, and you get a percentage.>

  23. Moe commented on Sep 6

    Actually, this job market is one where it’s incredibly hard to find anything other than minimum wage jobs. This job market is the worst I’ve ever seen. You can send out an endless number of resumes and barely get any responses. Starting a business is hard too because no one is spending so they can buy anything.

    You’re talking a 50%-75% (or more) cuts from previous income.

    This is chronic long term unemployment / underemployment.

    Canada on the other hand has been growing 10%+ year-over-year and have begun limiting US people from entering Canada.

  24. Moe commented on Sep 6

    Here’s something from another site about Mass Layoffs Soar for August.

    “Announced lay-offs surged 85 percent to 79,459 in August from 42,897 in July, according to Challenger, Gray & Christmas Inc, an employment consulting firm. August’s job cuts were the highest since February, when they totaled 84,014.”

    ” Private payrolls grew by 38,000 in August, following a 41,000 gain in July, ADP said. “A deceleration of employment may be under way,” the company said in a release. The report suggests nonfarm payrolls may have grown much less in August than the 120,000 anticipated by economists.

    “The slowdown in hiring was not related to last month’s credit market turmoil””

  25. Winston Munn commented on Sep 6

    In 2000, this was the age breakdown.

    174.1 million, or 62 percent, were age 18 to 64.

    Doesn’t matter how you slice or dice it, this is a rough guage of the total number available to work.

    The only true guage of the aggregate strength of an economic system is how many of the total work-eligible population is working?

  26. Juan commented on Sep 7

    May be worth noting that BLS methodologies and definitions have not been static but have periodically undergone more and less substantial revisions, the last major overhaul was, i believe, completed in 1993. Shortly thereafter the range of alternative measures of unemployment was reduced and the headline measure shifted from U-5 to U-3 while what had been the most inclusive, a U-7 measure was ended in favor of a broadened U-6.

    This particular complete revision also created problems of comparability between pre and post-1994 time series, which may have been corrected but if so not until the last two or three years. If not corrected, erroneous cycle peak and trough comparisons can result.

    At the same time, unemployment rates fail to take account of the quality of employment, e.g., longer run and at least partially disaggregated changes in real wages.

    It’s late, so, let me add a clip from a recent paper by David R. Howell, Milano Graduate School
    The New School Beyond Unemployment: Measuring Labor Market Performance Across Countries:

    “Recent wage trends in the U.S. underscore the current need to take changes in the quality of jobs into account in an assessment of labor market performance. Comparing business cycle peaks, the unemployment rate showed an impressive decline over the last two decades, from same period, the real earnings for less-educated workers collapsed and earnings inequality exploded, requiring more household members to hold more jobs and work more hours to maintain a constant standard of living. Real annual hourly wages for production and nonsupervisory workers fell at an average rate of .2% per year from 1979 through 1999, and weekly wages fell even faster, at .4% per year (Ibid., Table 2.4). Workers at the 10th percentile took home 9.3% less in 1999 than in 1979, and those with just a high school degree show an average wage decline of about 15% over this twenty year period, from $15.65 to $13.34 (Ibid.,table 2.6; table 2.19). As a result, the share of all full-time, full-year workers with poverty-level earnings rose from 14.4% in 1979 to 17.5% in 1998 (ibid., Figure 5G). For those without jobs or threatened with job loss, declining unemployment rates are the critical dimension of labor market performance.”

  27. The Big Picture commented on Mar 6

    ADP, NFP Numbers

    Construction and Financial Employment Changes Source: ADP, Macroeconomic advisors Tomorrow we learn what BLS data shows for the Non-Farm Payroll data for February. Consensus is for a meek 25,000 new jobs created. However, given the present concerns abo…

  28. The Big Picture commented on Mar 6

    More on Unemployment Rates

    Over the past few days, we’ve been discussing job creation and the various ways to think about unemployment. This has been a long standing theme around here (Augmented unemployment rate, as well as the NILF issue — Not In Labor Force). See the list at…

  29. The Big Picture commented on Apr 12

    Jobless vs. Unemployed

    In today’s NYT, Floyd Norris hits on a subject that has been a favorite of ours over the years: Finding the true measure of the economy’s labor situation.The unemployment rate is low. The jobless rate is high.Those two seemingly contradictory statement…

Posted Under