There was a bit of confusion in the economics blogosphere on Friday following the employment situation release by the U.S. Bureau of Labor Statistics about the recent changes in the number of part-time and full-time workers. Let’s clear up what some of these numbers mean and how they differ.
From the release, the more commonly referred to employment status numbers are from Table A-9. This is a breakdown of workers who usually work either full-time, 35 or more hours per week, or part-time, less than 35 hours per week.
In August there were 116.208 million full-time workers and 27.999 million part-time workers. By comparing with the prior month, we can see that 118k full-time jobs were added and 234k part-time jobs were lost in August.
In 2013, the total number of full-time jobs gained have been 340k, while the total number of part-time jobs gained have been 497k. This means that of the 837k total jobs added in 2013, the portion of jobs added that are part-time is 59.4%.
Since July 2007, the number of full-time jobs have declined by 4.481 million, while part-time jobs have increased by 2.564 million.
Alternatively, Table A-8 from the release references the actual number of hours people were at work during the survey week. Only people who worked between 1 and 34 hours during the survey week are included in this table.
The number of workers who were employed part-time for economic reasons in August were 7.911 million, while the number of workers who were part-time for noneconomic reasons were 19.339 million, making total part-time workers of 27.250 million. By comparing to the month prior, total part-time workers declined by 123k in August. The total change in the number of part-time workers in 2013 has been an increase of 569k.
Since July 2007, the number of part-time jobs has increased by 2.864 million.
The actual worked hours figures show that about 750k less part-time jobs exist today than the usual work status numbers indicate. However, the actual hours worked figures also show that about 70k more part-time jobs have been added in 2013 and that 300k more part-time jobs have been added since July 2007.
But the actual hours at work figures listed here are incomplete, so it is difficult for us to find the change in actual at work hours at part-time jobs relative to the change in actual at work hours at full-time jobs. There is another factor included into the at work hours figures that the BLS does not publish in the employment situation release, and that is the “not at work” (but still employed) statistic.
The below table shows, not seasonally adjusted, the breakdown for August of how the at work hours employment count is calculated for both workers who are usually either full-time or part-time.
Because this table is not routinely published, or seasonally adjusted, or in the FRED database, the at work hours jobs numbers are not very practical for examining job growth. Additionally, a strong case can be made that the number of hours an employee usually works are a better economic gauge than the number of hours they happen to be working during the survey week. The change in part-time employment based on at work hours is better utilized as an indicator of why people are working part-time rather than trying to compare these figures to the total number of employed.
You can’t count part-time jobs added using Table A-8 (at work hours) and then compare that with the full-time jobs added from table A-9 (usual work hours). Additionally, you can’t use either of those numbers in comparison with the headline payrolls figure which comes from the establishment survey. Table A-8 and A-9 are both from the household survey.
The usual work status numbers from Table A-9 are the more credible data to be used when calculating the total changes in part-time and full-time employment.