MaxedOutMama
Friday, January 06, 2012
Odd, Very Odd
Okay, I fixed the blogging problem. New address maxedoutmama2 @ gmail.com.
I cannot yet make heads nor tails out of the US employment report, but it looks like we have trouble ahead. If you look at government jobs on Table B-1 vs. government jobs on Table A-8 you can see the difficulty. Table A-8 (household survey) has seasonally adjusted private sector jobs dropping and seasonally adjusted private sector jobs rising. Not good. Table B-1 shows the opposite.
Embrace the error bars! Confusing data means something.
In December, the seasonal adjustment factors are updated, which you can read at the end of this link to
the html employment report. No shift in these factors seems to account for the difference. Table A (household survey) shows that the labor force shrank in December. In a way, this makes sense - a lot of people retire in November and December, and these retirements are probably biased toward government jobs. The not-in-labor force number grew by 194,000 in December, which also makes sense given US demographics. The unemployment rate fell to 8.5% from November's upwardly revised 8.7%, which makes sense. The big drop in unemployment (only 176,000 new jobs are reported) thus probably comes mostly from retirements. Older people have a horrible time finding jobs in this labor market.
You absolutely cannot explain the disparity on the Birth-Death adjustment for the Establishment survey - if anything, it works the other way. The B-D number is negative for December, which I believe to be substantially wrong.
On the other hand, ISM services showed marginally dropping employment.
After looking at petroleum, especially gas, I think that the establishment report is showing the winner-take-all effect, and a much stronger seasonal effect on changing shopping patterns. Unless the petroleum reports are FUBAR, which I have no reason to believe, Americans are shopping online a whole lot more.
The temporary gains in courier and messenger (42.2 thousand) and about 18K of the retail gains are going to drop right out. They are not "real". The real gain in construction will reach its low in the first quarter. Auto manufacturing has probably come close to peak, and won't add much if anything in the first quarter. The big add in accommodation and food services, almost all of which was in restaurants, is probably going to retract a bit. Temporary help services fell considerably, which is not that favorable a reading, but is consistent with much else.
My first hunch is that the variance between government jobs in the establishment survey and in the household survey is probably closer to the reality in the household survey; I'm basing that on samples of advertising.
Over the course of the year, most of the differences between the two surveys are inconsiderable. We have gained between 1.5 to 1.6 million private sector jobs. This is much better than losing jobs, but barring the retirements of boomers, it does little to correct the employment gap, because the civilian non-institutional population grew by almost 1.7 million. But the civilian labor force grew by less than 300,000 over the course of the year, because the boomers are aging and retiring - the not-in-labor-force number grew by more than 1.4 million.
So we do have a slowly improving economy, combined with dropping or stagnant real incomes for probably about 50% of households. One completely depressing statistic is that over the course of the year, the unemployment rate for blacks did not change at all. Last December it was 15.8%, and this December it is 15.8%. As you would expect, part-time for economic reasons fell hard in December, but that is going to reverse.
If fuel costs rise sharply, the odds of the US going into a longer downturn this year rise sharply. This is rather tightly balanced income-wise, and I have already figured in the FICA tax cut and the COLAs for retirement/disability. I am counting on the warmer winter weather boosting GDP by about 0.3-0.4% in the first quarter, annualized.
The major wild card I see over the first quarter are commodity prices. The ECB is throwing money. Banks are putting the excess money back on deposit at the ECB. However the non-excess is going to pay off bank loans, and a good chunk of that money is going to find its way into commodities. Should fuel prices rise much more, it will shave 0.3% at least from GDP. That would be quite bad.
I cannot yet make heads nor tails out of the US employment report, but it looks like we have trouble ahead. If you look at government jobs on Table B-1 vs. government jobs on Table A-8 you can see the difficulty. Table A-8 (household survey) has seasonally adjusted private sector jobs dropping and seasonally adjusted private sector jobs rising. Not good. Table B-1 shows the opposite.
Embrace the error bars! Confusing data means something.
In December, the seasonal adjustment factors are updated, which you can read at the end of this link to
the html employment report. No shift in these factors seems to account for the difference. Table A (household survey) shows that the labor force shrank in December. In a way, this makes sense - a lot of people retire in November and December, and these retirements are probably biased toward government jobs. The not-in-labor force number grew by 194,000 in December, which also makes sense given US demographics. The unemployment rate fell to 8.5% from November's upwardly revised 8.7%, which makes sense. The big drop in unemployment (only 176,000 new jobs are reported) thus probably comes mostly from retirements. Older people have a horrible time finding jobs in this labor market.You absolutely cannot explain the disparity on the Birth-Death adjustment for the Establishment survey - if anything, it works the other way. The B-D number is negative for December, which I believe to be substantially wrong.
On the other hand, ISM services showed marginally dropping employment.
After looking at petroleum, especially gas, I think that the establishment report is showing the winner-take-all effect, and a much stronger seasonal effect on changing shopping patterns. Unless the petroleum reports are FUBAR, which I have no reason to believe, Americans are shopping online a whole lot more.
The temporary gains in courier and messenger (42.2 thousand) and about 18K of the retail gains are going to drop right out. They are not "real". The real gain in construction will reach its low in the first quarter. Auto manufacturing has probably come close to peak, and won't add much if anything in the first quarter. The big add in accommodation and food services, almost all of which was in restaurants, is probably going to retract a bit. Temporary help services fell considerably, which is not that favorable a reading, but is consistent with much else.
My first hunch is that the variance between government jobs in the establishment survey and in the household survey is probably closer to the reality in the household survey; I'm basing that on samples of advertising.
Over the course of the year, most of the differences between the two surveys are inconsiderable. We have gained between 1.5 to 1.6 million private sector jobs. This is much better than losing jobs, but barring the retirements of boomers, it does little to correct the employment gap, because the civilian non-institutional population grew by almost 1.7 million. But the civilian labor force grew by less than 300,000 over the course of the year, because the boomers are aging and retiring - the not-in-labor-force number grew by more than 1.4 million.
So we do have a slowly improving economy, combined with dropping or stagnant real incomes for probably about 50% of households. One completely depressing statistic is that over the course of the year, the unemployment rate for blacks did not change at all. Last December it was 15.8%, and this December it is 15.8%. As you would expect, part-time for economic reasons fell hard in December, but that is going to reverse.
If fuel costs rise sharply, the odds of the US going into a longer downturn this year rise sharply. This is rather tightly balanced income-wise, and I have already figured in the FICA tax cut and the COLAs for retirement/disability. I am counting on the warmer winter weather boosting GDP by about 0.3-0.4% in the first quarter, annualized.
The major wild card I see over the first quarter are commodity prices. The ECB is throwing money. Banks are putting the excess money back on deposit at the ECB. However the non-excess is going to pay off bank loans, and a good chunk of that money is going to find its way into commodities. Should fuel prices rise much more, it will shave 0.3% at least from GDP. That would be quite bad.
Comments:
<< Home
Statistics note from the BLS regarding this survey:
For example, the confidence interval for the monthly
change in total nonfarm employment from the establishment survey is on the order of plus or minus
100,000. Suppose the estimate of nonfarm employment increases by 50,000 from one month to the next. The 90- percent confidence interval on the monthly change would
range from -50,000 to +150,000 (50,000 +/- 100,000).
These figures do not mean that the sample results are off by
these magnitudes, but rather that there is about a 90-percent
chance that the “true” over-the-month change lies within
this interval. Since this range includes values of less than
zero, we could not say with confidence that nonfarm employment had, in fact, increased that month. If, however, the reported nonfarm employment rise was 250,000, then
all of the values within the 90-percent confidence interval
would be greater than zero. In this case, it is likely (at least
a 90-percent chance) that nonfarm employment had, in fact,
risen that month. At an unemployment rate of around 5.5
percent, the 90-percent confidence interval for the monthly
change in unemployment as measured by the household survey is about +/- 280,000, and for the monthly change in the unemployment rate it is about +/- 0.19 percentage point
http://www.bls.gov/news.release/pdf/empsit.pdf
For example, the confidence interval for the monthly
change in total nonfarm employment from the establishment survey is on the order of plus or minus
100,000. Suppose the estimate of nonfarm employment increases by 50,000 from one month to the next. The 90- percent confidence interval on the monthly change would
range from -50,000 to +150,000 (50,000 +/- 100,000).
These figures do not mean that the sample results are off by
these magnitudes, but rather that there is about a 90-percent
chance that the “true” over-the-month change lies within
this interval. Since this range includes values of less than
zero, we could not say with confidence that nonfarm employment had, in fact, increased that month. If, however, the reported nonfarm employment rise was 250,000, then
all of the values within the 90-percent confidence interval
would be greater than zero. In this case, it is likely (at least
a 90-percent chance) that nonfarm employment had, in fact,
risen that month. At an unemployment rate of around 5.5
percent, the 90-percent confidence interval for the monthly
change in unemployment as measured by the household survey is about +/- 280,000, and for the monthly change in the unemployment rate it is about +/- 0.19 percentage point
http://www.bls.gov/news.release/pdf/empsit.pdf
# posted by 
: 10:35 AM
: 10:35 AM
Ron - the over the year stuff is a lot more reliable, and over a couple of months it is pretty reliable.
been many years since my college stat class but looking at the BLS PDF today's number did not make the 90 confidence level if so then its noise but maybe I overstate the negative value.
# posted by : 2:10 PM
: 2:10 PM
I think the only statistics that matter anymore are the food stamp numbers. We may be adding jobs, but we keep adding even more to the assistance rolls. I assume the government statistics bureau will eventually start giving out bullshit "modeled" numbers for welfare et al, too.
Post a Comment
# posted by : 2:53 PM
: 2:53 PM << Home
