BLS Jobs Report Covering March 2012: Screeching to a Halt
It looks like the seasonal early year surge in jobs is, as we predicted, wearing off. The big news this month is that the BLS is reporting 120,000 jobs created in March. Traditional media sources were forecasting more in the 200,000 area, and this was itself a weaker forecast than in previous months. Nor was there any good news to be found in the near term revisions for the past two months. It was a wash. The January job creation was revised down 9,000 from 284,000 to 275,000 and the February number was revised up 13,000 from 227,000 to 240,000.
The official U-3 unemployment rate edged down from 8.3% to 8.2%.
With that let's dive into the numbers. Looking at the Household survey, the potential labor force as measured by the Non-Institutional Population over 16 (the NIP) increased 169,000 from 242.435 million to 242.604 million. The Employment-Population ratio declined 0.1% to 58.5% seasonally adjusted. So the number of jobs needed to keep up with population growth was .585(169,000) or around 99,000. The March jobs number of 120,000 indicates that the economy created jobs only marginally better than what was required for population growth.
Seasonally adjusted the employed declined 31,000 from 142.065 million to 142.034 million. Unadjusted, employment grew 728,000 from 140.684 million to 141.412 million. In other words, the pendulum of seasonal adjustment which pumped up numbers in the first quarter is now swinging the other way. Note too that the unadjusted numbers for the employed are about 1.5 million smaller than their seasonally adjusted counterparts.
Seasonally adjusted, the unemployed decreased 133,000 from 12.806 million to 12.673 million. Since the employed decreased by 31,000, this means that some 164,000 workers left/got defined out of the labor force in March. And this is, in fact, what we see when we look at the seasonally adjusted labor force (the employed plus unemployed) which decreased by this amount from 154.871 million to 154.707 million. This decrease (because they are based upon them) is reflected in the decline I noted in the employment-population ratio (employed vs. NIP) to 58.5% and in the labor force participation rate (the actual labor force vs. the potential labor force, i.e. the NIP) to 63.8% from last month's 63.9%.
Unadjusted, unemployment declined 526,000 from 13.430 million to 12.904 million. Taking the 728,000 unadjusted increase in employment this nets out to an increase of 202,000 in the unadjusted labor force: 154.114 million to 154.316 million. Again we should expect these movements between adjusted and unadjusted numbers as the unadjusted numbers are now playing catchup to their seasonally adjusted counterparts.
The BLS uses a highly restrictive definition of what it means to be unemployed. You need to be without a job and have looked for one in the 4 weeks before their survey was conducted to qualify. If not, then the BLS defines you out of the labor force. I call the difference between the current labor force as defined by the BLS and where the labor force should be, the BLS undercount. The BLS itself recognizes this problem to some extent in its category Not in Labor Force, Want a Job Now. I take this to be its measure of its undercount. This is one category where the BLS emphasizes the seasonally unadjusted number. For March, it decreased 335,000 from 6.376 million to 6.041 million.
A subset of these is the marginally attached, those who have looked for work in the last year. In March, the marginally attached declined 256,000 from 2.608 million to 2.352 million. The marginally attached are along with the unemployed and those working part time but who would work full time if they could one of the three components which make up the broader U-6 measure of un- and under employment.
In March, involuntary part timers decreased 447,000 from 8.119 million to 7.672 million.
Adding the unemployed (12.673 million), the marginally attached (2.352 million), and involuntary part timers (7.672 million) gives 22.697 million for the U-6 or a rate with rounding of 14.5%.
As I have said in the past, I have problems with the BLS undercount because it does not seem to track with economic events so I have developed a different way to measure it. I compare the current labor force to where we would expect it to be in a solid economic expansion: labor participation rate of 67%. The difference between these two is my measure of the undercount. For March, the expected labor force would be .67(242.604 million) or 162.545 million. The difference between this and the current labor force (154.707 million) is 7.838 million, an increase of 278,000 from February.
You can see the problem I have here. Per the BLS, both the number of employed and unemployed, that is the labor force, decreased in March so we would expect the BLS undercount to increase but instead it decreases too. If you calculate it the way I do, the labor force contracts and you see an expected increase in the undercount. The BLS approach is premised on job seeking. If you are not job seeking for as little as a month, you are out of the labor force. In the U-6, only the marginally attached, those who have looked for work in the last year are counted. I think this is mistaken. My view is what we should be looking at is how many Americans would be working if jobs were available to them. I calculate analogues to the U-3, real unemployment and the U-6, disemployment using the whole of my measure of the BLS undercount, because these are Americans who would be working if there were jobs out there for them.
Real unemployment for March was 12.673 million + 7.838 million or 20.511 million, giving a real unemployment rate of 12.6% up from 12.5% for February.
Disemployment for March, that is adding in involuntary part timers, was 28.183 million. yielding a disemployment rate of 17.3%, down from 17.5% last month. The driver in this monthly change was the decrease in involuntary part timers.
By race, white unemployment remained unchanged at 7.3%. African American unemployment overall improved slightly from 14.1% to 14.0%. Unemployment among African American males over 20 decreased from 14.3% to 13.8%. Unemployment among African American teens jumped from 34.7% to 40.5%.
In March, average hourly wages in the private sector rose 5 cents (0.2%) to $23.39. The average workweek, however, declined 0.1 hours to 34.5 hours. So average weekly wages actually declined 60 cents from $807.56 to $806.96
Briefly, from the Establishment data, manufacturing gains 37,000 jobs, burger flipping 37,000, and healthcare 26,000. Retail lost 34,000. Manufacturing is a bright note although manufacturing jobs do not have the quality they once had. Most of the other jobs added in March are not high quality employment.
To conlcude, in March the economy ground to a halt. In the Household data, seasonal adjustments that painted a rosier picture in the first quarter are gone. Real unemployment, my measure, worsened. In the Establishment data, first quarter inventory rebuilding looks to be over, and job creation barely exceeded the demands of population growth. In other words, another green shoot bites the dust.