BLS Jobs Report Covering December 2011: The Improvement That Wasn't
The Bureau of Labor Statistics' report covering December 2011 is out. The number of jobs created from the Establishment survey is 200,000, an excellent number, but one which reflects strong holiday, hence likely temporary, hiring. In the Household survey, the unemployment rate dropped two tenths of a percent to 8.5%. However, last month's unemployment was revised upwards from 8.6% to 8.7%.
The BLS is once again involved in both year end (thus the change in last month's unemployment rate) and longer term revisions that affect both this month's and next month's data. This month, it has revised its Household survey (unemployment) data for the year. Next month, it will revise its Establishment survey (jobs) data. On top of this, next month's Household data will begin to incorporate data from the 2010 Census. What this means is that BLS data which has been getting increasingly erratic will become even more so this month and next, and Household data from January 2012 on will not be directly comparable to earlier data at all. Staggering its revisions as it does essentially isolates this month's Household (unemployment) data. The previous 2011 (and earlier) data have been revised and future months will be based on a different census. And as I said, while this month's Household data have been revised, this month's Establishment data have not. In broad terms, what this means is that all the numbers are going to be wrong although the rates, barring major errors, are likely to remain reasonably close.
Diving into the numbers with those caveats in mind, the civilian non-institutional population over 16 (the potential labor force or NIP) increased 143,000 in December from 240.441 million to 240.584 million. The employment-population ratio was unchanged at 58.5%. The product of these .585(143,00) is 83,655, the number of jobs needed in December to account for population growth. So the economy did considerably better than population growth in December.
However, the actual labor force shrank by 50,000 in December going from 153.937 million to 153.887 million. The number of employed increased by 176,000 from 140.614 million to 140.790 million and the number of unemployed decreased by 226,000 from 13.323 million to 13.097 million. Essentially, change in employed plus change in unemployed equals change in labor force (176,000 – 226,000 = -50,000). So while employment increased and unemployment decreased, both good, the overall labor force decreased somewhat, not good.
These are seasonally adjusted numbers. Interestingly, the seasonally unadjusted numbers show a very different picture. These show that employment declined by 389,000 in December, unemployment increased by 79,000, and the BLS dialed 310,000 out of the labor force.
By and large, if we were looking at recovery, we would like the seasonally adjusted and unadjusted numbers, while not the same, at least headed in the same direction.
The broader U-6 measure of un- and under employment, made up of U-3 unemployed (13.323 > 13.097 million), the marginally attached (2.591 > 2.540 million), and involuntary part timers (8.469 > 8.098 million), decreased from 15.6% to 15.2% (24.383 > 23.735 million, a decrease of 648,000). As I noted above, the U-3 seasonally adjusted and unadjusted unemployment numbers were severely at odds. This is also true for involuntary part timers where the unadjusted number showed an increase of 157,000 as opposed to the seasonally adjusted one which showed a decline of 471,000.
What I take to be the BLS' own measure of its undercount: Not in Labor Force, Want a Job Now (seasonally adjusted) decreased 210,000 from 6.595 million to 6.385 million.
My own calculation of the BLS undercount, those unemployed it chooses to stop counting, is the difference between the labor force in a good expansion .67(NIP) minus its current measurement by the BLS. That is [.67(240.584 million) - 153.887 million] or [161.191 million - 153.887 million] or 7.304 million, an increase of 170,000 from a recalculated 7.134 million for last month.
Taking this number, we can add it to the U-3 unemployment figure of 13.097 million to produce a true number of unemployed of 20.401 million or a 12.7% unemployment rate. We can in turn add the number of involuntary part timers (8.098 million) to arrive at a revised U-6 measure of un- and underemployment, that is disemployment, of 28.499 million or a disemployment rate of 17.7%. My calculation of this rate, albeit with somewhat different numbers, was 18% in November, so a 0.3% decline. Basically, while the BLS undercount increased, its counts of the unemployed and involuntarily part time decreased more, and the denominator, the labor force size increased.
In other Household data, white unemployment edged down from 7.6% to 7.5% White male unemployment dropped faster than white female unemployment (0.2% vs. 0.1%) but still remained higher (7.1% vs. 6.8%). African-American unemployment is on the way back up going from 15% in October to 15.8% in December. White teenage unemployment is 20.3%. African-American teenage unemployment is 42.1%, up 4.6% since October. So not only is the African-American unemployment rate twice as high as that for whites. It is currently trending both opposite to white unemployment and doing so at a faster clip.
In the Establishment data, average weekly private non-farm hours did increase slightly in December from 34.3 to 34.4 hours/week, a good sign. Average wages for this group increased 4 cents an hour, also a good sign. They have increased 2% over the last 12 months from $22.77/hr to $23.24. However, the consumer price index increased 2.9% over the last 11 months so workers continue to fall behind.
The 200,000 job increase represented a 212,000 growth in the private sector and a 12,000 loss in government. Seasonal hiring was particularly noticeable in transportation and warehousing which increased by 50,000 in December.
Looking at the last 12 months, the jobs picture looks like this:
Total nonfarm (public and private): 1.640 million
Total private: 1.920 million
Retail trade: 239,700
Professional and business services: 452,000
Food services and drinking places: 230,400
Total public: -280,000
Although it is from a different survey, the non-institutional population increased by 1.695 million in 2011. The employment-population ratio averaged an abysmal 58.4 for the year meaning the economy needed around 990,000 jobs to keep up with population growth. It did about 650,000 better than that. But this needs to be measured against a real unemployment number of 20.401 million. At this rate, it would take 31 years to reduce unemployment to zero. 20 years to reduce it to 5%. Also if you look at the breakdown of these jobs and that wage gains aren't even keeping up with inflation, what jobs are being created are not quality jobs, jobs that individuals and families can live reasonably well on.
The short form of the year in review is that the economy did not cliff dive. Some net jobs were created but their quality and wages indicate that American labor continued its slide, or, in other words, good jobs and enough of them are not being created. The shortfall of jobs both in quality and number remains enormous and is simply not being addressed and nothing is being seriously done.
One final note, the February revisions could contain several surprises both good and bad. We will just have to wait and see.