The Recovery Thread

[pinqy]

Payroll employment data from the establishment survey show employment growth in the private sector of less than 100,000 per month this year, leading to the conclusion that the economy is growing too slowly to reduce unemployment. In contrast the data from the less publicized household survey show employment in the private sector growing this year at almost 300,000 per month. Last month it soared 891,000 thousand.

Here is the data Employment Situation Summary Table A. Household data, seasonally adjusted

Where does it say that employment soared by 891K?

In Table A-8 under "Private industries."
To find information easier, always go to the full list of tables: (http://www.bls.gov/news.release/empsit.toc.htm) and that will include household (A tables) and establishment data (B tables). And you can tailor the historical data from that list as well. Once you select the historical data for private industries, seasonally adjusted, for table A-8 you can click on "more formatting options and select one month net change to see that for 2010 the monthly changes have been 603, -89, 505, 259, 200, -166, 129, and 891 for private industries. That averages to +291,000 but that's misleading considering how huge the July-August change was.

 

[Neubarth]

Payroll employment data from the establishment survey show employment growth in the private sector of less than 100,000 per month this year, leading to the conclusion that the economy is growing too slowly to reduce unemployment. In contrast the data from the less publicized household survey show employment in the private sector growing this year at almost 300,000 per month. Last month it soared 891,000 thousand.

Here is the data Employment Situation Summary Table A. Household data, seasonally adjusted

Where does it say that employment soared by 891K?

In Table A-8 under "Private industries."
To find information easier, always go to the full list of tables: (Employment Situation) and that will include household (A tables) and establishment data (B tables). And you can tailor the historical data from that list as well. Once you select the historical data for private industries, seasonally adjusted, for table A-8 you can click on "more formatting options and select one month net change to see that for 2010 the monthly changes have been 603, -89, 505, 259, 200, -166, 129, and 891 for private industries. That averages to +291,000 but that's misleading considering how huge the July-August change was.

As everybody can see and as I have pointed out for thirty years now, the Government Data = Total Fiction. They lie about the Population. They lie about the size of the Work force. They lie about jobs that are supposedly generated. Hell, they even lie about GDP.

 

[pinqy]

Here is the data Employment Situation Summary Table A. Household data, seasonally adjusted

Where does it say that employment soared by 891K?

In Table A-8 under "Private industries."
To find information easier, always go to the full list of tables: (Employment Situation) and that will include household (A tables) and establishment data (B tables). And you can tailor the historical data from that list as well. Once you select the historical data for private industries, seasonally adjusted, for table A-8 you can click on "more formatting options and select one month net change to see that for 2010 the monthly changes have been 603, -89, 505, 259, 200, -166, 129, and 891 for private industries. That averages to +291,000 but that's misleading considering how huge the July-August change was.

As everybody can see and as I have pointed out for thirty years now, the Government Data = Total Fiction. They lie about the Population. They lie about the size of the Work force. They lie about jobs that are supposedly generated. Hell, they even lie about GDP.

And yet you've never presented any evidence at all. Your assertions, and your view of what you see, does not equal evidence.

 

[antagon]

the OECD and the UN lie in similar fashion on our behalf?

 

[Neubarth]

In Table A-8 under "Private industries."
To find information easier, always go to the full list of tables: (Employment Situation) and that will include household (A tables) and establishment data (B tables). And you can tailor the historical data from that list as well. Once you select the historical data for private industries, seasonally adjusted, for table A-8 you can click on "more formatting options and select one month net change to see that for 2010 the monthly changes have been 603, -89, 505, 259, 200, -166, 129, and 891 for private industries. That averages to +291,000 but that's misleading considering how huge the July-August change was.

As everybody can see and as I have pointed out for thirty years now, the Government Data = Total Fiction. They lie about the Population. They lie about the size of the Work force. They lie about jobs that are supposedly generated. Hell, they even lie about GDP.

And yet you've never presented any evidence at all. Your assertions, and your view of what you see, does not equal evidence.

Suicide is painless. It causes subtle changes. As previously stated, you are a government Shill. Nothing you post is to be taken seriously. Amen.

 

[loosecannon]

Payroll employment data from the establishment survey show employment growth in the private sector of less than 100,000 per month this year, leading to the conclusion that the economy is growing too slowly to reduce unemployment. In contrast the data from the less publicized household survey show employment in the private sector growing this year at almost 300,000 per month. Last month it soared 891,000 thousand.

Here is the data Employment Situation Summary Table A. Household data, seasonally adjusted

Where does it say that employment soared by 891K?

In Table A-8 under "Private industries."
To find information easier, always go to the full list of tables: (Employment Situation) and that will include household (A tables) and establishment data (B tables). And you can tailor the historical data from that list as well. Once you select the historical data for private industries, seasonally adjusted, for table A-8 you can click on "more formatting options and select one month net change to see that for 2010 the monthly changes have been 603, -89, 505, 259, 200, -166, 129, and 891 for private industries. That averages to +291,000 but that's misleading considering how huge the July-August change was.

Thanks a bunch.

 

[loosecannon]

Arguing from or about BLS numbers is a waste of time.

There is usually some rhyme and reason to the BLS data. But the definitions are odd. in this case I have to wonder what distinction exists between "private" employment and "civilian non institutional" employment.

They have one category called "not in the work force" which defines a class of people who are neither in or out of the workforce at once.

 

[Foxfyre]

Arguing from or about BLS numbers is a waste of time.

There is usually some rhyme and reason to the BLS data. But the definitions are odd. in this case I have to wonder what distinction exists between "private" employment and "civilian non institutional" employment.

They have one category called "not in the work force" which defines a class of people who are neither in or out of the workforce at once.

While interesting, the numbers themselves can make your head spin. The bottom line is that millions of people who want work are out of work and millions more are underemployed. And our fearless leaders don't seem to give a damn about that and continue the destructive policies that got us into this.

 

[loosecannon]

While interesting, the numbers themselves can make your head spin. The bottom line is that millions of people who want work are out of work and millions more are underemployed. And our fearless leaders don't seem to give a damn about that and continue the destructive policies that got us into this.

True but that may be changing. I won't know unless I dig into the several definitions of the different categories of labor the BLS is collecting data on.

But a single month 891K employment gain may mean that we have turned some kind of corner, or not, depending on what the data really represents.

 

[antagon]

While interesting, the numbers themselves can make your head spin. The bottom line is that millions of people who want work are out of work and millions more are underemployed. And our fearless leaders don't seem to give a damn about that and continue the destructive policies that got us into this.

heya Foxfyre! we haven't crossed paths in a good while. hope all is well.

just had to break a lil' off on your contention that government leaders don't care about unemployment.

 

[pinqy]

There is usually some rhyme and reason to the BLS data. But the definitions are odd. in this case I have to wonder what distinction exists between "private" employment and "civilian non institutional" employment.

It's not too complicated. Looking at the population as a whole, there are a large number of people who cannot freely enter or leave employment due to legal or other external factors. Children, for example, prisoners, people in an institution for mental or medical issues, and the military. Including them in the numbers would throw things off, such as when the military was included in the Unemployment Rate (1984-1994) which was consistanly lower than the civilian UE rate and distorted the picture. So children under 16, prisoners, people in institutions, and the military are not part of the BLS population: the Civilian non-institutional population. For Employment, then there are Public workers (all levels of government employees) and Private Industry.

Does that make more sense?

They have one category called "not in the work force" which defines a class of people who are neither in or out of the workforce at once.

Not in the Labor Force is the term. These are people who are not working and not trying to work. Some don't want work: Retirees, full-time students, stay-at-home-spouses, trust-fund-babies etc. And some want to work but have problems with transportation, child-care, temporary medical problems, or feel that they would have so much trouble getting a job they've given up looking. None of these are Employed, because they're not working, but neither are they Unemployed, because they're not trying to work. Special attention is paid to those who say they want to work even though they're not trying to get a job, and they're the Marginally Attached: people who want work, are available to take work if offered, and have looked for work in the previous 12 months but not the last 4 weeks. A sub-set of the Marginally Attached are Discouraged Workers: those who say their reason for not looking is because they don't think they'll be successful.

 

[loosecannon]

There is usually some rhyme and reason to the BLS data. But the definitions are odd. in this case I have to wonder what distinction exists between "private" employment and "civilian non institutional" employment.

It's not too complicated. Looking at the population as a whole, there are a large number of people who cannot freely enter or leave employment due to legal or other external factors. Children, for example, prisoners, people in an institution for mental or medical issues, and the military. Including them in the numbers would throw things off, such as when the military was included in the Unemployment Rate (1984-1994) which was consistanly lower than the civilian UE rate and distorted the picture. So children under 16, prisoners, people in institutions, and the military are not part of the BLS population: the Civilian non-institutional population. For Employment, then there are Public workers (all levels of government employees) and Private Industry.

Does that make more sense?

They have one category called "not in the work force" which defines a class of people who are neither in or out of the workforce at once.

Not in the Labor Force is the term. These are people who are not working and not trying to work. Some don't want work: Retirees, full-time students, stay-at-home-spouses, trust-fund-babies etc. And some want to work but have problems with transportation, child-care, temporary medical problems, or feel that they would have so much trouble getting a job they've given up looking. None of these are Employed, because they're not working, but neither are they Unemployed, because they're not trying to work. Special attention is paid to those who say they want to work even though they're not trying to get a job, and they're the Marginally Attached: people who want work, are available to take work if offered, and have looked for work in the previous 12 months but not the last 4 weeks. A sub-set of the Marginally Attached are Discouraged Workers: those who say their reason for not looking is because they don't think they'll be successful.

great explanations, Thanks!

So the 891K number would actually be extremely significant. But not enough to offset layoffs in government and esp census positions.

 

[Foxfyre]

While interesting, the numbers themselves can make your head spin. The bottom line is that millions of people who want work are out of work and millions more are underemployed. And our fearless leaders don't seem to give a damn about that and continue the destructive policies that got us into this.

heya Foxfyre! we haven't crossed paths in a good while. hope all is well.

just had to break a lil' off on your contention that government leaders don't care about unemployment.

Well, if they really gave a damn, don't you think they would have addressed that instead of using hundreds of billions of dollars and thus mortgaging our children and grandchildren's future to pay off their union buddies, contributors, and others they owe for whatever? Wouldn't you think they would have actually focused on long term projects that would have provided long term employment for many people?

Or better yet, wouldn't they have at least given careful consideration to the very practical suggestions for tax and regulation relief that would have freed up tens of thousands of employers to get back to work and start hiring again?

It's just like saying somebody cares about his/her health while sucking on another cigarette, having another fifth drink, wolfing down mountains of french fries and bacon burgers.

 

[william the wie]

Arguing from or about BLS numbers is a waste of time.

There is usually some rhyme and reason to the BLS data. But the definitions are odd. in this case I have to wonder what distinction exists between "private" employment and "civilian non institutional" employment.

They have one category called "not in the work force" which defines a class of people who are neither in or out of the workforce at once.

Possibly you are right but I haven't trusted BLS (or NBER) numbers since I found out about that 1-4 check figure gimmick for 1st and 2nd significant figures and did a distribution table for UE numbers. By the way if you are investing or trading it is a great way to test for cooked books 1-4 is 60% probable for the first digit (50-70% range) and a tad over 50% (40-60% range) for the second digit if those numbers aren't about right start reading the footnotes carefully.

 

[pinqy]

I was saying he was taking data from some irrelevant column that had nothing to do with the question at hand which irrelevant column doesn't matter to me. Whether it is the household survey which undercounts unemployment

I'm curious as to how you think the CPS undercounts employment. Is it a disagreement over definitions, or are you claiming a methodological problem?

[qutoe] or the birth/death survey which overcounts job creation as far as I'm concerned the L in BLS is superfluous.[/quote]The birth/death model is a likely culprit for the overcounting of job creation in 2009, but that was corrected by BLS. But a birth/death model is a good idea to account for the changes in businesses that the survey just can't keep up with.

In unemployment numbers an honest count should result in the second significant number being 1-4 about half the time according to number theory and you can verify this with a logarithm scale where X.1-4 is about half the scale when you account for rounding. In UE numbers, particularly annual numbers, X.5 and X.9 account for about half of all annual unemployment numbers that the BLS endorsed for inclusion in "Historical Statistics of the United States" Arguing from or about BLS numbers is a waste of time.

I'm not sure that holds when we're talking about numbers that are rounded to the nearest thousand anyway.

 

[loosecannon]

Possibly you are right but I haven't trusted BLS (or NBER) numbers since I found out about that 1-4 check figure gimmick for 1st and 2nd significant figures and did a distribution table for UE numbers. By the way if you are investing or trading it is a great way to test for cooked books 1-4 is 60% probable for the first digit (50-70% range) and a tad over 50% (40-60% range) for the second digit if those numbers aren't about right start reading the footnotes carefully.

1-4 or 0-4?

 

[william the wie]

Possibly you are right but I haven't trusted BLS (or NBER) numbers since I found out about that 1-4 check figure gimmick for 1st and 2nd significant figures and did a distribution table for UE numbers. By the way if you are investing or trading it is a great way to test for cooked books 1-4 is 60% probable for the first digit (50-70% range) and a tad over 50% (40-60% range) for the second digit if those numbers aren't about right start reading the footnotes carefully.

1-4 or 0-4?

1-4. It is based on logarithmic distribution that much I know but the proof lost me on about the second sentence and strangely enough even in the second digit 1-4 is a more than 50% probability. I just keep an old slide rule around to double check the probabilities with a ruler because this rule is massively counter-intuitive. The question that was being answered was why numbers rackets always use the last 3-4 digits of published 7 digit numbers to pick winners. It turns out that numbers such as newspaper circulation or the betting totals at the local race track are non-random for their first three digits although the third digit is effectively worthless as a check figure (42% vs. expected 40% if memory serves.) Gangsters learned quickly to exclude the first three digits in order to stay in business and number theorists eventually found out why.

 

[loosecannon]

OK Time to revive this thread.

I've been skeptical of the recovery as of late, and have the bruises to prove it, having been beaten like a dog being short for the first part of September, which has turned out to be the best September in over 70 years. I'm still skeptical, given the awful data coming from the Fed regional surveys, but one must be open-minded and conclude that one might be wrong.

I try to look for contrarian ideas and scenarios, and maybe the most contrarian notion out there is that job growth is strong. Here is one take on strong job growth.

Payroll employment data from the establishment survey show employment growth in the private sector of less than 100,000 per month this year, leading to the conclusion that the economy is growing too slowly to reduce unemployment. In contrast the data from the less publicized household survey show employment in the private sector growing this year at almost 300,000 per month. Last month it soared 891,000 thousand.

As noted before in [Employment is Rising Fast] around economic turning point the household data are more reliable and usually lead the establishment data. For example, in seven out of eight turning points associated with the last four recessions — four downturns around the beginning of the recessions and four upturns around the ends of the recessions and the beginning of the recoveries — the peaks and troughs of the household employment data preceded the corresponding peaks and troughs of the establishment employment data.

Still, to lend credence to the private sector employment data from the household survey and particularly the outstanding figure for August, it might be useful to look for other, indirectly supporting data. In very general terms such data are also supportive of the notion of improved employment gains in August. Thus, the Challenger job cuts figures show a significant drop in August to a new post- recession low. Similarly, according to the Conference Board, the number of new online help wanted ads, marked a new post- recession high in August. Some weekly data also point in the same direction, including the un-interrupted downtrend of initial unemployment claims, from its peak in mid-August to a low in mid – September. All in all it appears that the data are what they are and not a fluke.

Mike Astrachan Follow up: Employment Is Growing Fast

So, Toro, do you think this private sector employment is the turning of a corner? Or some kinda fluke?

I tend not to take any gyroscopic reading of our recovery too seriously because every few months most indicators seem to reverse course, even if only momentarily.

 

[loosecannon]

Possibly you are right but I haven't trusted BLS (or NBER) numbers since I found out about that 1-4 check figure gimmick for 1st and 2nd significant figures and did a distribution table for UE numbers. By the way if you are investing or trading it is a great way to test for cooked books 1-4 is 60% probable for the first digit (50-70% range) and a tad over 50% (40-60% range) for the second digit if those numbers aren't about right start reading the footnotes carefully.

1-4 or 0-4?

1-4. It is based on logarithmic distribution that much I know but the proof lost me on about the second sentence and strangely enough even in the second digit 1-4 is a more than 50% probability. I just keep an old slide rule around to double check the probabilities with a ruler because this rule is massively counter-intuitive. The question that was being answered was why numbers rackets always use the last 3-4 digits of published 7 digit numbers to pick winners. It turns out that numbers such as newspaper circulation or the betting totals at the local race track are non-random for their first three digits although the third digit is effectively worthless as a check figure (42% vs. expected 40% if memory serves.) Gangsters learned quickly to exclude the first three digits in order to stay in business and number theorists eventually found out why.

Well statistics are often counterintuitive. But there are valid reasons for that.

There is a common sense explanation why numerals 1-4 would appear most frequently in any "randomly" generated number that represents real events or things*.

With 1 being the most likely followed by 2, 3, 4, 5, 6.....

But the prevalence of 1 vs 6 or 9 would be very slight. Tho measurable.

(* think about it)

 

[KissMy]

I warned Toro not to be short in the face of the Fed's POMO (Permanent Open Market Operations). They are flooding the market with cash & monetizing government debt.