How the Unemployment Rate is Calculated

People who have exhausted all of their unemployment benefits; that is, people who no longer receive unemployment benefits (people previously referred to as the "99ers"),  ARE counted as unemployed if they are STILL LOOKING FOR WORK.

Was there a change in the way the unemployment rate was calculated under Obama?  Has the way unemployment is calculated changed?    Read below.... 

 

As of January 2014.  More details HERE.

Repeat after me: People who have exhausted all of their unemployment benefits are counted as unemployed if they are still actively looking for work (within the past four weeks).  This is the biggest myth about the unemployment rate that is out there. 

Repeat after me:  People who have just entered or re-entered the labor force are counted among the unemployed as long as they have actively looked for work within the last four weeks.

Repeat after me:  People who have NEVER been eligible or collected unemployment are considered unemployed if they are actively looking for work (within the past four weeks).  

Repeat after me:  People who are self-employed or people who are/were 1099 contractors ARE considered as unemployed if they are actively looking for work (within the past four weeks).


There have been NO changes in the way that the unemployment rate has been defined or calculated for decades.  Despite what you might have read in the right-wing media, President Obama has made NO changes in the way the unemployment rate has been calculated.  The Bureau of Labor Statistics, which calculates the unemployment rate and puts out the Jobs Report, is a non-partisan agency that was actually headed by a Republican appointee until 2013.  

Repeat after me:  The last time that any changes were made to the way the regular official unemployment rate was defined or calculated was 1994, and those changes resulted in an increase in the unemployment rate compared to the prior method.  (Read towards the bottom of the page for more info about this.)

For those who are not familiar with the term,  "the 99ers" referred to people who were laid off in the Great Recession, who had maxed out their unemployment benefits, and who still had not been able to find work.  The 99 weeks was the maximum number of weeks of unemployment available in most states through 2011.  It included 26 weeks of regular state benefits, four tiers of federal Emergency Unemployment Compensation totaling up to 53 weeks, and up to 20 weeks of Extended Benefits   In some states and in some situations, people maxed out their unemployment benefits before 99 weeks. 

Update 12/28/2011:  A two-month extension of federal unemployment extended benefits was signed just before Christmas.  However, this extension will begin to reduce the total benefit weeks available to unemployed Americans to 79, depending on the state in which they reside, and depending on exactly when they became unemployed.  The Republicans in the House are attempting to reduce the maximum number of benefit weeks even further, down to 59.

Update 2/20/2012:  A compromise to extend unemployment insurance throughout 2012 was passed by both houses of Congress on February 17, 2012.  The deal is rather complicated in terms of weeks of unemployment extensions that will be available.  More details and a chart of available weeks of unemployment insurance can be found HERE.

Update April 2013:  The unemployment extensions which were passed in early 2012 and which are discussed above were extended through December 2013.  This does not mean that there are ADDITIONAL weeks of unemployment insurance available, but it means that people newly unemployed in 2103 will be able to get the same number of weeks of unemployment insurance as people who became unemployed in late 2011 or 2012.  The maximum number of weeks of unemployment insurance now available in the various states varies from 40 weeks in 14 states or territories up to 73 weeks in 7 states or territories.  86 weeks is available only in Alaska.  More details and a chart of available weeks of unemployment insurance can be found HERE.   

Update February 2014:  The unemployment extensions which were passed in early 2012 and were extended through December 2013 have expired.  As of February 10, 2014, no states has any extended benefits, either federal or state.  Another attempt to extend these benefits failed in the Senate this past week.  It is unclear if the Congress will again try to extend these benefits. More details and a chart of available weeks of unemployment insurance can be found HERE. 

Update January 2015:  No unemployment extension was worked out with the Republicans.  As a result, there are only 26 weeks of unemployment insurance available.  

Now, let's see if we can answer a few other common questions about the way the government counts the employed/unemployed and jobs....

Q.  How is the unemployment rate calculated?


Let's look at a couple of definitions:

The Current Population Survey:  This is a monthly survey of 60,000 households, about 110,000 people, that is used to calculated unemployment rates, labor and employment participation rates.

Basically, the "base" of the labor market is the civilian non-institutional population age 16+, that is, everybody 16 or over who is NOT in the military and NOT in an institution, such as prison or a nursing home.

The civilian labor force is anyone in the "civilian non-institutional population age 16+" (as defined above) who is either working or who has been actively looking for work in the past four weeks.  This includes students age 16 plus who are working part-time by choice, people who are working part-time but want full-time jobs, people who are retired but want part-time work, and people entering the labor market for the first time.

The key is that to be considered part of the civilian labor force, a person must either be working or have actively looked for work (an application, a resume, a an interview, a contact) in the past four weeks.

People who are not working and haven't looked for work in the past four weeks because they are ill, they have children to care for, they are discouraged, they are retired, they are in school or training, or for any other reason are NOT counted among the civilian labor force.

The unemployment rate is the percentage of people in the civilian labor force who are actively looking for work and who do NOT currently have either a part-time or full-time job. 

Q.  Are part-time workers counted among the unemployed?

Anybody who is unemployed and actively looking for work, whether they were laid off or quit, whether they have previously worked or are just entering the labor market, whether they were previously working full-time or part-time, is considered "unemployed".  If someone is working part-time now but wants full-time work, he/she is NOT counted as unemployed.. Read on.

Q.  If I'm working part-time, but I need a full-time job and I'm looking for a full-time job.. am I counted as "unemployed"?

No.  If you are now working, you are not unemployed, even if you don't make much money or work many hours.  However, there are alternate unemployment measures, including a "U-6" measure of underemployment put out by the Bureau of Labor Statistics that do include part-time workers who want full-time work.

Q. Since I no longer get unemployment compensation, I'm not counted among the unemployed.

A. The monthly unemployment rate is completely separate from the weekly jobless claims reports. They use different inputs. That weekly questionnaire that people need to submit if they are getting unemployment benefits is NOT input into the monthly unemployment numbers. There's NO connection.

The monthly unemployment report is based on a survey sample of 60,000 households in the United States. Statistical samples are used for almost every number out there, as it is not possible to count every person in the United States. We only attempt to count every person in the United States every ten years, in the decennial Census. It's a mammoth undertaking, which is why most numbers and statistics are based on a sample.

Statisticians have been designing and calculating samples for generations now; it's a science, not an art. Though every statistical study has a margin of error, not only is this the only way of getting any reasonable statistic, but many statisticians feel that carefully chosen and conducted samples may be more accurate than counting every last individual.

Q.  The government will not approve more unemployment benefits because they don't want people to know how bad things are.  If they approve more unemployment benefits, the unemployment rate will go up.

A.  Sorry, but this simply isn't true.  If the government approved 14 more weeks tomorrow, which would be welcomed by so many, the "regular" UI rate would not go up because it is NOT related to the number of people getting unemployment benefits.

The government, including the Democrats, aren't stalling on extending UI benefits for this reason; they either don't think they can get any more benefits passed this Congress or they don't think the 99ers are a big enough voting bloc.  Or both.

Q. Nobody has ever asked me whether or not I am unemployed as part of this survey. How do they know?

A. There are approximately 125,000,000 households in the United States and the sample is based on 60,000 of them. Therefore, less than one in two thousand households will be interviewed for this survey. One in two thousand means that your chances of being interviewed for this survey are very small indeed. By comparison, most political polls use samples of only a couple of thousand people. But they are still pretty good at predicting the outcome of most political races.

Q. This is all just a lie to make people feel that they are being counted. Nobody has counted me.  Nobody has counted my friends.

A. The Bureau of Labor Statistics has been calculating the unemployment rate based on such a survey sample for decades now. If anything, their methods have improved as new and faster computers are used to crunch data. If you don't believe that a survey sample can effectively represent a population, there's not much I can say.

You may wish to read about statistical sampling. There's a good basic intro article here at Wikipedia. The first paragraph presents a very basic overview.

Q. This seasonal adjustment stuff is hogwash.

A. There are times of the year when employment always goes down and unemployment always goes up.  Vice verse, there are times of the year when employment always goes up and unemployment always goes down. 

Christmas is the easiest of those times to understand.  Employment is always higher in December due to Christmas retail hiring, in good economic times and in bad.  Unemployment is always down in December.  In January, when all of those retail establishments cut back, employment is always down and unemployment is always up.  

There are also ups and downs due to seasonal variations such as outdoor construction work, staffing for summer and vacation venues, and layoffs associated with the end of the school year.  In March, usually 500,000 up to 900,000 new employees are added to non-farm payrolls, but the numbers are adjusted downwards.  In March 2011, about 900,000 "real" people were added to non-farm payrolls, but the number of jobs increased in seasonally adjusted numbers by only about 200,000.  

Seasonal adjustments (made by statisticians using these usual variations) are attempts to even out these normal ups and downs so that we can all have a better understanding of what is happening with the employment situation aside from the usual seasonal ups and downs.

"Raw" unadjusted numbers reflect what is actually happening in the labor market in any given month or week.  Those numbers can tell us who actually is getting jobs, and what kind of jobs people are actually getting.  That's important to know if you are looking for work.  But "seasonal adjustments" round things out so that comparisons can be made between one month and the next.  Leave a comment if this still isn't clear. 

Q. My sister has been laid off for a long time but is now selling stuff on eBay. She keeps looking for work, but she does make a few hundred dollars a month. Is she counted as unemployed any more?

A. If she is surveyed and answers the questions honestly, she would no longer be counted as unemployed if she made money during the survey month.

That is a real drawback to the unemployment rates. People who are working small businesses, even if they make only a few hundreds bucks a month, are counted as self-employed and not included among the unemployed... even if they are looking for work.  If they are self-employed, not currently earning any money, and they are looking for work, they are considered as unemployed.

Q. My uncle was laid off last year and he has not been able to find a job. As he is over 62, he decided to take early retirement. He has stopped looking for work now, but he might see if he can find something later this year if things look better. Is he counted as unemployed?

A. If he hasn't looked for work in the past month, he is no longer counted as unemployed. If he looked for work sometime in the past year but not during the past month, he may be counted as a discouraged worker. But he would no longer be in the "labor force".

This is another shortcoming of the current survey. As far as I know, they don't ask questions to determine what the true status is of someone who has retired; that is, if the person is retired because they can't find work or if the person has retired because they genuinely were ready to retire. They also don't ask if a retired person intends on returning to the work force in the future.

Q. Someone told me that the Feds have changed the way they calculate the unemployment rate. This person said that the Feds drop people from the "seeking work" category every month so that the unemployment rate keeps dropping.

A. Well, "someone" is wrong. To repeat, the Bureau of Labor Statistics bases its counts on the answers to survey questions from the "Household Survey". The BLS doesn't drop anybody. If you answer the questions one way, you are counted as "not in the labor force". If you answer the questions another way, you are counted as unemployed. Read a few paragraphs down, and you can check out the BLS links yourself.

And the BLS has not changed the way they calculate the unemployment rate in decades.  In the early 90's, some changes were made to questions and procedures and to some alternate rates.  When they ran parallel tests, the changes actually increased the unemployment rate. 

Q. This is all tomfoolery and balderdash. The government is cooking the books so that we don't know how bad things are.

A. Hmm... If the government is cooking the books, they are doing a bad job of it. If you were Obama and wanted to cook the books in terms of unemployment numbers, wouldn't you have cooked them down below 8% before the November 2010 elections?  (* This was originally written in early 2011.) 

There are alternate rates of unemployment, such as the U-6 number that adds back discouraged workers and part-time workers who want a full-time job. This rate (as of January 2011) is about 16% compared to 9% for the "regular UI rate". 

Q. How do you know all of this stuff? Are you just theorizing or making stuff up?

A. No. The Bureau of Labor Statistics does a pretty darned good job at explaining what they do. First of all, the complete monthly situation report is available online HERE. And here's another good link that explains in detail how the BLS collects data and calculates the unemployment rate. Also, the monthly report is available online back to the early 90's at the ARCHIVES.

Updated 1/11/2015

Updated 2/2014

Updated 4/22/2013 

Updated 12/28/2012

Updated 2/29/2012