- Joined
- Jan 2, 2006
- Messages
- 28,185
- Reaction score
- 14,274
- Location
- Boca
- Gender
- Male
- Political Leaning
- Independent
Total nonfarm payroll employment increased by 321,000 in November, and the unemployment
rate was unchanged at 5.8 percent, the U.S. Bureau of Labor Statistics reported today.
Job gains were widespread, led by growth in professional and business services, retail
trade, health care, and manufacturing.
This is probably one of the strongest reports in the past two years, given the upward revisions from previous months and the rise in earnings. YoY, average hourly earnings have increased 2.1% year over year, which is actually higher than CPI inflation of 1.6% in the same time frame.
The rest of the report can be found here.
As usual, you (and other Keynesians/big government lovers) refuse to look past the headlines.
Check the household survey.
Number of Additional Americans employed...JUST 4,000!!!
Employment Situation Summary Table A. Household data, seasonally adjusted
Then, check the number of full time Americans employed.
150,000 fewer Americans were full time employed.
Table A-9. Selected employment indicators
This is not a great report...it is, IMO, a lousy one.
labor force went up 119,000, employed went up 4,000, unemployed went up 115,000. Where did they get the 321k number?
You're using OLD math... you need to use common core math you see. :joke:
labor force went up 119,000, employed went up 4,000, unemployed went up 115,000. Must be nice to pick which report you pull your data from for you headline.
It's always the same. The official jobs numbers ("new jobs" or "jobs created") comes from the Current Employment Statistics (CES) a monthly survey of "about 144,000 businesses and government agencies, representing approximately 554,000 individual worksites..."
This is a payroll survey and excludes agriculture, the self employed, those working in other's houses, unpaid family workers, and those working under the table. The reference period is the Pay Period that includes the 12th of the month.
The unemployment and other labor force data (including total employed) comes from the Current Population Survey (CPS) a monthly survey of 60,000 households (around 110,000-120,000 individuals) that includes everyone age 16 and older not in the military, prison, or an institution. The reference period is the Week that contains the 12th.
So the CES employment data, while more limited, is much more accurate and is benchmarked each year to the Quarterly Census of Employment and Wages. The CES has always been the official jobs data.
You know full well that your statement is impossible to prove.
Do you know how to respond to a counter-position without invoking an argument from ignorance fallacy? The CES survey is more accurate because it draws from a much larger sample size. This is a fundamental principle of statistics; as the sample size increases, the ability to accurately depict the central tendency increases.
While not 100%, because there are always flaws in every measuring techniqe, the Quarterly Census of Employment and Wages is a COUNT (not a survey) of businesses covered by Unemployment Insurance and of government agencies, so it's as close to 100% as is possible.Oh, come on now.
You know full well that your statement is impossible to prove. You would have to have the actual, 100% accurate number to compare it to, but that is presently impossible to calculate.
Much larger sample size, and benchmarked to the actual numbers from a near total census. How could it not be more accurate?The BLS says it is more accurate based on it's calculations...which means virtually nothing since they came up with the calculation process and it is NOT an exact science.
Each month, approximately 554,000 worksites are surveyed. In order to aggregate that to the total number of employed, you have to use the total number of worksites in each industry. But new businesses pop up every month and other businesses fail, so how do you know how many businesses there are? You can just ignore births and deaths and use the number from a set date, but that's obviously inaccurate. Or you can use a birth/death model based on previous observed patterns and trends. This is better, though obviously not perfect (which is why benchmarking to the QCEW is still necessary).Plus, the CES uses all kinds of models and guesstimates (like the Net Birth/Death model...which even the BLS admits is often wrong).
That's a differnt survey. The CES asks "How many people were on your payroll for the pay period that contains the 12th of the month". And that's aggregated out based on a model of current number of businesses.Asking a bunch of businesses how many people they hired or fired and then putting your own estimates and models to those numbers is hardly an accurate way to measure employment.
While not 100%, because there are always flaws in every measuring techniqe, the Quarterly Census of Employment and Wages is a COUNT (not a survey) of businesses covered by Unemployment Insurance and of government agencies, so it's as close to 100% as is possible.
Much larger sample size, and benchmarked to the actual numbers from a near total census. How could it not be more accurate?
Each month, approximately 554,000 worksites are surveyed. In order to aggregate that to the total number of employed, you have to use the total number of worksites in each industry. But new businesses pop up every month and other businesses fail, so how do you know how many businesses there are? You can just ignore births and deaths and use the number from a set date, but that's obviously inaccurate. Or you can use a birth/death model based on previous observed patterns and trends. This is better, though obviously not perfect (which is why benchmarking to the QCEW is still necessary).
During the recession, the birh/death model, which was updated yearly, was clearly off, so BLS changed to a quarterly basis of updating the model.
The Household survey does a similar thing, using estimates of population growth. Those are adjusted every January, but the previous numbers are not revised....it's just a jump between December and January data.
That's a differnt survey. The CES asks "How many people were on your payroll for the pay period that contains the 12th of the month". And that's aggregated out based on a model of current number of businesses.
What do you suggest would be more accurate?
You said it 'is' more accurate...and that is impossible to know.
U.S. factory orders fall for third straight month
'(Reuters) - New orders for U.S. factory goods fell for a third straight month in October, pointing to a slowdown in manufacturing activity.
The Commerce Department said on Friday new orders for manufactured goods declined 0.7 percent after a revised 0.5 percent drop in September.'
U.S. factory orders fall for third straight month | Reuters
So, factory orders have fallen for three months in a row?
Oh yeah...America is just humming along.
Btw, according to the NOT seasonally adjusted household survey, there were actually 270,000 FEWER Americans employed in November compared to October.
Plus a whopping (again NOT seasonally adjusted) 735,000(!) less Americans employed in November compared to October.
Table A-9. Selected employment indicators
This report just gets uglier and uglier.
You did no say it is probably more accurate (which I still do not agree with).
You said it 'is' more accurate...and that is impossible to know.
I am not going to explain statistics and mathematics and physics to you...your matter-of-fact statement is IMPOSSIBLE to prove. Period. This should be obvious.
It's not a question. It IS more accurate. But go ahead, and show your math that a sample of .05% of the universe is not for sure less accurate than a sample of 32% of the universe. Or that the standard errors are wrong.Come on man, you went too far. You exaggerated. We all do it. No big deal. If you had said 'I believe it is more accurate'...I might not have said a thing.
In November, the average workweek for all employees on private nonfarm payrolls rose
by 0.1 hour to 34.6 hours. The manufacturing workweek rose by 0.2 hour to 41.1 hours,
and factory overtime edged up by 0.1 hour to 3.5 hours.
I'll just this, regardless of the report this last election, the voters didn't look at all the rosy reports or buy into the economy is improving. What they did was look around, at themselves, their families, their neighbors and friends and assessed their own situation and voted accordingly.
45% of those who voted said the economy was their number one issue. 78% of those who voted said they were worried about their and this nation's financial future. Now those are how the people look at these things, how they feel. Not how a statistical report said or was spun if it was. This last election was decided upon each individuals own pocket book and how they saw the future, their own and friends and family financial future.
No it's not impossible to prove. The Household survey samples approximately 0.05% of the population. the Establishment survey samples approximately 32% of employed people. There is no doubt that that much larger a sample is going to be more accurate. That's basic math. Standard error for Household Survey employment is about +/- 300,000 and for the Establishment survey it's +/- 74,000 While statistics is not an exact science, it is a science, and that large a difference in sample size is inarguabley more accurate.
And that's not even including non-sample error, which will be more prevelant in the Household Survey.
It's not a question. It IS more accurate. But go ahead, and show your math that a sample of .05% of the universe is not for sure less accurate than a sample of 32% of the universe. Or that the standard errors are wrong.
That is untrue. A larger sample size has a smaller margin of error than a smaller sample size (up to a certain point). This is true regardless of anything else.I will say this one final time because you seem a decent chap and you are usually polite.
It is impossible to know how accurate a measurement of a number is compared to another measurement of the same number is unless you know what that number you are measuring actually is.
Since it is presently impossible to determine the actual number of unemployed persons in America with a 100% certainty, then it is impossible to determine how accurate a measurement is compared to another measurement.[/qutoe] First, you're changing the topic from Employed to Unemployed. Second you are again ignoring the issue of different sample sizes. Third, you're ignoring that, for the Establishment survey we do have a near exact count from the QCEW.
No, it is nothing like that. It's like there is a giant jar with about 1 million marbles in it of red, white, and blue. You pick a sample of 500 marbles, count the number of each color, and use the percent of your sample to estimate the total number of each color (for example, out of the 500 you pick, there are 100 white, which is 20%....20% of of 1 million is 200,000 so that's your estimate). I pick a sample of 320,000 marbles and get a count of 48,000 white marbles (15%) giving me an estimate of 150,000 white marbles.It's like if I think of a number and you come up with two methods to estimate what that number is. Then you say that method 'a' is more accurate then method 'b'...even though you have absolutely no idea what the number is.
It is impossible to know which method is more accurate since you do not even know the number you are trying to determine.
Whose estimate is more likely to be closer to the true number?
Do you really want to claim sample size doesn't matter and we could go with a sample of 100 households and get the same accuracy as the current 60,000 households?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?