9/06/2013

Numbers continue to be Stunning. 96 percent of net jobs added this year have been part-time jobs

Updating the job numbers with the new numbers that are out today continues to show that virtually all the jobs created this year have been part time jobs.


So far this year there have been 848,000 new jobs.  Of those, 813,000 are part time jobs (for both economic reasons and noneconomic reasons).   As can be seen in the above screen shots from the BLS.gov, the total part-time jobs changed from 26.437 million in January to 27.250 million in August.  To put it differently, an incredible 96% of the jobs added this year were part-time jobs.  (It isn't clear to me why we see the differences in part time for economic and noneconomic reasons.)  However, these numbers raise a question about whether the employer mandate that has already been forcing some companies to move to part-time workers so as to avoid government imposed penalties is having a significant impact.  Might the relationship between the recent part time jobs and overall job growth be stronger as we were getting closer to what people thought would be the implementation of 
Obamacare's employer mandate?  Might we now see a divergence with the very recent one year delay in the employer mandate?  Or was the 2012 election the defining event for firms to begin to study the impact more?  Data from the BLS.gov.

UPDATE 4: Using the sum of the part-time economic and noneconomic reasons.  The relationship that appears to exist this year seems interesting.



Using the numbers from Table A6



UPDATE: The BLS.gov also provides a sum of these two numbers (available here) where they get a total of 532,000 (thanks to the "Department of Numbers" for pointing this out and to Nick for emailing me about their post).  That implies that 63% of jobs added this year were part-time jobs, which is still an amazingly high number.  
Department Numbers uses this number Total Part Time Workers (LNS12600000) rather than the sum of the Part-Time for Economic Reasons (LNS12032194) and Usually Work Part Time Noneconomic Reasons (LNS12005977).  At least part of what is going on has to do with the seasonal adjustment of the components produces a different result than seasonally adjusting the total.  I can't see any reason that looking at the total makes more sense than looking at the sum of the two components.  Indeed, a strong case could be made that it is better to seasonally adjust the components and not just the total.  In any case, whether it is 63% or 96% or someplace in between, it is still a large portion of the increase in jobs that is due to part time employment.



Steve Bronars, a friend and a labor economist, argued that the most useful comparison might be to look at only Usually Work Part Time Noneconomic Reasons (LNS12005977).  In that case, there was an increase in part-time workers of 875,000, but only an increase in jobs of 848,000.  Thus, the increase in part-time jobs equals 103% of the increase in jobs from January to August.

UPDATE 2: Ben at the Department of Numbers now says that the percentage using Table A-6 is 59%, not 63%, but that is using a different time period.  It is easy to verify the 63% using the "total part-time" number for the change from the beginning of the year (January through August) (63% = 532,000/848,000).  532,000 = 27,999-27,467.  848,000= 144,170-143,322.  It seems pretty clear to me that the number to compare the 96% to is 63%, not 59%.

UPDATE 3:  Tyler Cowan writes: "As Ben Engebreth points out, based on these BLS charts, the correct number seems to be 59% not 96%, though the higher estimate does still seem to hold in Lott’s (more cumbersome and less transparent) sources."  But he provides no explanation for why this should be correct nor does it appear that he has checked the issues discussed in this post.

Labels: ,

12 Comments:

Blogger marmico said...

You are conflating changes in part-time/full time composition of jobs with changes in the level of jobs.

Nowhere does the BLS state: 96% of the jobs added this year were part-time jobs.

On an unadjusted year over year basis, the composition of part-time jobs for economic reasons are down about 2%.

9/06/2013 11:59 AM  
Blogger doc said...

Unles I’m reading the tables incorrectly, and comparing August 2012 with August 2013 (in thousands):

PT Emp 8/12: 26,899
PT Emp 8/13: 27,250
% Change in PT: +1.34%

FT Emp 8/12: 115,275
FT emp 8/13: 116,920
% Change in FT: +1.43%

Tot Emp 8/12: 142,164
Tot Emp 8/13: 144,170 %
Change in FT: +1.41%

PT Emp as a % of Tot Emp, 8/12: 18.9%
PT Emp as a % of Tot Emp, 8/13: 18.9%
PT employment growth as a & of Total employment growth: 18.0%

This looks like a nothingburger to me.

9/06/2013 2:21 PM  
Blogger Josh Sher said...

This seems like a rediculous computation to me:

Suppose
10 Jobs Created Full Time
8 Distroyed Full Time
10 Created Part Time
11 Distroyed Part Time

(For instance 10 part time workers moved to new part time jobs, 8 full time workers moved to new full time jobs, 1 part time worker moved to full time, and 1 unemployed moved to full time)

Your computation:
2 Net Jobs Created Full Time
-1 Net Jobs Created Part Time
1 Net Job Created Total

So, 200% of "net jobs created" were full time
And
-100% of net jobs created" were part time

I am not sure why these are meaningful numbers. In my example, 50% of all jobs created were full time, and 42% of all jobs distroyed were full time....How this nets out to +200% is either a work of art, or a calculation that just doesn't represent a meaningful concept....

9/06/2013 5:54 PM  
Blogger John Lott said...

Dear Doc:
The importance of part-time jobs as a percentage of total job growth has been increasing over time.

9/06/2013 7:17 PM  
Blogger Kevin O'Neill said...

Ummm.... who's in charge of the math here?

If you start with 10 million full-time workers, create 1 million new full-time jobs, and reduce 1 million existing jobs from 40 hr/wk to 36 hr/wk you end up with 100% of new jobs created were part-time jobs. Which is completely misleading.

As an indicator of job growth, job creation, or *anything* meaningful this metric appears a complete waste of time.

9/06/2013 10:42 PM  
Blogger marmico said...

The importance of part-time jobs as a percentage of total job growth has been increasing over time.

This FRED Graph shows the percentage of 3 types of part-time employment to civilian employment:

1. The green line is "Usually Work Part Time Noneconomic Reasons"

2. The blue line is "Slack Work or Business Conditions"

3. The red line is "Could Only Find Part-Time Work".

Since 1995 the green line has trended downwards slightly, the blue line spiked upwards during the Great Recession and has trended down since, the red line spiked upwards during the Great Recession and slightly trended upwards since.

Conclusion: Involuntary part-time employment has not been increasing over time. Voluntary part-time employment increased during the Great Recession and in combination (red and green lines) has been slowly declining since.

This much ado about nothing, indeed.

9/07/2013 10:37 AM  
Blogger Per Kurowski said...

Why should we be surprised? Bank regulations which allow banks to earn much higher risk-adjusted returns on equity when lending to “The Infallible”, than when lending to “The Risky”, can only produce fatty tissue, and not the muscles required to engineer a new generation of sturdy jobs… that requires risk-taking and does not allow for excessive risk-aversion.

If Mark Twain resurrected, he would opine as follow about bank regulators

“A bank regulator is one who likes the bank to lend out the umbrella when the sun shines, even more than what a banker likes to do that, truly amazing, and one who wants the banker to take that umbrella back when there is the slightest indication it could rain, even faster than what the banker would like to do, equally truly amazing.”

These regulations should not be allowed under Regulation B, but no one wants to enforce that.

http://subprimeregulations.blogspot.com/2012/07/complaint-i-presented-to-consumer.html

9/08/2013 10:10 AM  
Blogger "O" Anonimo said...

John Lott your post is completely dishonest. Shame on you.

9/09/2013 12:10 AM  
Blogger John Lott said...

Dear "O" Anonimo:

Would you mind explaining what exactly is dishonest about the data that I present?

9/09/2013 1:57 AM  
Blogger spencer said...

Weird data.

If you look at the seasonally adjusted data you get your results.

But if you look at the not seasonally adjusted data from January to August part time employment actually fell
-1,833,000.

I suspect there is some problem with the seasonally adjusted -- especially since the seasonally adjusted January observation is lower than every single observation in 2012.

9/14/2013 1:52 PM  
Blogger onlylolz said...

The importance of part-time jobs as a percentage of total job growth has been increasing over time.

10/10/2013 1:38 AM  
Blogger spencer said...

How can you talk about job growth in 2013 without once using any 2012 data.


I always thought growth rates were computed from one year to the next, not within a given year.

Second, the January observation appears to be a flukish data point.
Part time employment in January 2013 was lower than in any month of 2012.

If you compare part time employment in 2013 to the same month in 2012 it shows that part time employment in 2013 was lower than in most months
of 2012. This is particularly true if you look at January. January 2013 part time employment was 483,000 lower than in January 2012. January 2013 part time employment was less than in every month of 2012.

It sure looks like January 2013 is an outlayer observation . While your math may be correct, Your results are strictly a function of the January data point being a weird number. It looks like the seasonal adjustment for January 2013 created the weird observation.

By the way, unpublished BLS data show that federal employees accounted for over 100% of the 2013 increase in part time employment. From January thru August of 2013 private part time employment was less than in the same months of 2012. So the 2013 jump in part time employment was due to the sequester, not Obama-care.

Ill be happy to send you the unbublished data since the BLS is shut down.

10/12/2013 11:46 AM  

Post a Comment

Links to this post:

Create a Link

<< Home