I think they might be more on the money than I because the are projecting just the working age population, but they don't publish their assumptions. My estimates are much more rough because within non-institutional population are retired, people going back to school/going to school, homemakers and so on and when break up the different age brackets, say 18-24 vs. 26-55 and so on, it's not a linearly relationship on who really is wanting needing a job, working versus doing other things. Suffice it to say there is around 4 million (consensus) missing workers, not already accounted for in the BLS report, who need a job and can't get one.
Another factor is seasonal adjustments and the assumptions, algorithms for. I forget when the yearly adjustment is, but it's been over 1 million recently and I keep meaning to dig around on this because that might also be a reason the ADP and the BLS diverge so much.
On the offshore outsourcing tally, yeah, that's from some studies done, where the numbers are estimated and I've just added them together. Problem is, frankly one of them is from a NBER paper, which has some seriously bogus, as in "I can't believe they allow this crap to be published" bogus paper trying to deny offshore outsourcing is affecting the U.S. workforce. Uh, duh, you don't need a 2nd grade education to see what has happened in town after town, particularly the midwest.
But I'll try to gather all of that and I have to do some raw data analysis because the NBER paper is so full of shit. If you're interested, you can pull some of this from the new Census data, but to really get it is near impossible. It's in various press releases and many corporations hide those numbers so close to the chest, you just can't get them. Also a job these days doesn't need to be directly offshore outsourced, more simply created in India or China instead of the U.S.
here is the sample size of the two surveys for the BLS:
The household survey provides information on the labor force, employment, and unemployment that appears in the "A" tables, marked HOUSEHOLD DATA. It is a sample survey of about 60,000 households conducted by the U.S. Census Bureau for the U.S. Bureau of Labor Statistics (BLS).
The establishment survey provides information on employment, hours, and earnings of employees on non-farm payrolls; the data appear in the "B" tables, marked ESTABLISHMENT DATA. BLS collects these data each month from the payroll records of a sample of nonagricultural business establishments. The sample includes about 140,000 businesses and government agencies representing approximately 410,000 worksites and is drawn from a sampling frame of roughly 8.9 million unemployment insurance tax accounts. The active sample includes approximately one-third of all nonfarm payroll employees.
But the numbers I really focused on are from the establishment survey, which "should" be the most accurate.
ADP has a sample size of 340,000 businesses, but I guess is also relying on them being "honest" maybe. These are all private.
There have been other times the ADP diverged from the BLS in slope and it seems to be when the monthly changes are greater. I think the correlation was at 0.94 or so.
I've got to write up more graphs and calculations on this. I fully expected to finally get some jobs and it's like we have something really awful going on here where America and Americans, in terms of the "two economies" has decoupled almost completely at this point.
Why the huge discrepancy between ADP and the BLS numbers? 297,000 to 103,000 when government only shed 10,000 jobs. Temp workers for the holidays? I wonder how many jobs U.S. companies are creating overseas? I'd like to see those numbers. At some point our congress and president need to reward companies for creating jobs in the U.S. or at least paying taxes in the U.S.
I am very disappointed in these numbers. I truly thought it was going to be a 200,000 plus blowout after the ADP report. Damn.
What is shows is a huge increase in people falling into the homeless, the permanent poor, the jobless who are not counted. It's also creating a flurry of articles on how arcane the BLS methods are and I have to agree. Of all of the overviews and analysis of various economic reports, this one takes me the longest because I must run a host of calculations to find the hidden jobless.
I'm referring to the random variable X, which in this case applies to the unemployed person sample set duration of TIME, or length of of being counted as unemployed before dropping out or finding a job. The median, defined at otherwise notated as .
In terms of "x w/ bar over it", uh, I can put as variables in an equation and it's valid. An equation is not the same as some nomenclature for a symbol implying an average.
uniform distribution is with the people, the unemployed, I'm saying they are all equal, no assumptions are being made in their distribution and only random variable X, that being TIME is being considered.
Just for you,
For a normal distribution, the probability density function (f(x) = dF(x) absolute continuous distribution function, oops F(X) is the cumulative distribution function where m is the median) becomes;
which is a Gaussian distribution.
the Median is usually denoted and for normal distributions, the mean equals the median and the correct mathematical symbols are:
Here we do not have that kind of distribution going on. It's a "fat tail", the mean or average does not equal the median.
Frankly, I do not expect most people to understand what I'm saying in this comment or even the previous one, because one is only going to see this (odds on) if they take an advanced class in probability ad statistics and remembered it and even used it. That said, I really don't expect to be questioned repeatedly when the above post is correct, from the layman terminology to the equations. It is showing that like wages, due to the super rich and now the long term unemployed, the median is diverging from the average and on top of things, we have more and more people who through no fault of their own can't find a frigging job and it's showing up in these two statistical measurements. Q.E.D.
Your 3rd sentence contradicted your 2nd sentence, and it's x bar(x with a bar over it), not x. I think you understand what you are trying to say, your just didn't do a very good job of explaining it. I'll stick with my comment.
Wall Street banksters bought the Fed, congress and regulatory agenices, so it only makes sense to now buy the White house. Corporatism is now in full bloom. Instead of Commander and Chief, we now have a Bankster and Chief. At least with GWB you knew you were getting an incompetent, bumbling Wall Street butt kisser, but I thought that Obama would be more than simply another Wall Street lackey. Didn't vote for him, but expected more.
50% is half. 50/100 = 1/2. Also, this is not a population sample set, it is those being counted as unemployed. The median means half of this sample set, is an assumed even distribution or uniform distribution of unemployed people. The variable is time, the distribution is the unemployed person, weighted evenly, i.e. uniform. The graph means that the duration where 50%, or when half of those unemployed found another job (or were not counted), has increased over time. The median, a singular data point defined by the variable w.r.t. the sample set, does not imply that all elements of the sample set's variable occurred at the median. That's part of the definition of term median.
Actually I did simplify the description for this particular distribution, but the description is correct. If you can't handle a summation symbol, I doubt the probability distribution functions are going to go over real well in a blog post.
The point of the post is not summation symbols or someone not understanding Statistics 101, never mind 501, the point is more and more people cannot get another @&*)$ job to the point of needing to document that fact for half a decade and this is a crisis and a crime.
The median is that point at which half the population, or half the sample of the population, lies below that point, and half lie above that point. So you description of the median graph is incorrect. At the median, half of the unemployed took less than or equal to that amount of time to find a job, and half of the population or sample took a greater or equal amount of time to find a job.
So you lost me before you got to the fancy Riemann Summation after your graphs.
I am aware of the outsourcing that is going on globally these days, however there are holes in your story.
First, outsourcing is done to countries with reliable network connections to the west, I doubt some of the countries you mentioned qualify, for example Armenia.
Second, law cannot be outsourced, BAR certification is needed, which can only be obtained in the United States. Perhaps legal assitance work can be outsourced.
I like how you bolded "officially." There are millions who are not working who are simply not obtaining unemployment benefits. This is a great article and I hope that it educates many.
"and yet our [per] capita income is 15% higher than Australia's ($46k vs. $39.9k)."
I was comparing 2009 wages with current exchange rates, so the 15% differential is inaccurate. Exchange rates are a moving target, but sufficed to say, Australia's per capita income is similar to our own, we just share it (or not) differently.
Good article Steve. Another benchmark would be to peg minimum wage at 10% of congressional salary of ($174,000), which divided by 2000 hour work year is $87. So $8.70 an hour, about where Washington State is as of the New Year ($8.67).
The "universal living wage" folks have an interesting idea, set minimum wage for each metro area based on housing costs x 3 (using HUD indexes). Some of them seem rather high, I guess you're still in Santa Barbara, which would be at $19.33/hr minimum versus Bakersfield, $11.96/hr.
http://universallivingwage.org/ulwformula.htm
Then again, the Australian minimum wage is $15/hr (US and Aus dollars at par currently). More than twice the US minimum wage and yet our capita income is 15% higher than Australia's ($46k vs. $39.9k).
As you can surely imagine, the Australian unemployment rate is catastrophic, from story earlier today:
The national unemployment rate in Australia now sits at 5.2, close to full employment, after more than 400,000 full-time jobs were created in the past year.
http://www.theaustralian.com.au/business/markets/stressed-labour-market-...
I blame the single payer healthcare system for their troubles. :o)
Anxiety over "might happen" is pointless. People, read the United States Constitution "No person shall be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation." That means the only way the government can steal your savings is through taxation or inflation. No matter what Alan Simpson wishes, the law is the law. Only a suicidal politician would support a law to confiscate your IRA. As for inflation, Uncle Ben keeps trying but his center of gravity seems to be stuck.
Workforce participation, U6 analysis based on recessions, and CPI adjusted average hourly wages. You'd think that you were a very high level policy analyst providing the latest Medici the absolute best information. In a way your are but they need to read it.
What's the point of an economy, or a political entity, unless it works for all of those involved. It's not like the Middle Ages where people were blind to better options.
The mindless focus on quarterly earnings as a function of someone's bonus is a crime and the graft and corruption on top of that is a crime against the state. Time to clean up and look at the obvious problems and solutions with these factors at the front: a) What works for the people? b) Is it being done with intellectual and professional honesty?
Where will the stimulus for inclusive, honest government comes from? Will it be a big ank failure in 2nd or 3rd Q? Will it take 1000 dead birds falling on Wall Street (specifically) instead of Beebe Arkansas? Does a camel need to fit through the eye of a needle;
Whatever it takes, we need a radical shift for the better and soon. The failures are getting to be too harmful and also, they've become tedious. It's the same old same old from the usual cast of incompetent characters.
I think they might be more on the money than I because the are projecting just the working age population, but they don't publish their assumptions. My estimates are much more rough because within non-institutional population are retired, people going back to school/going to school, homemakers and so on and when break up the different age brackets, say 18-24 vs. 26-55 and so on, it's not a linearly relationship on who really is wanting needing a job, working versus doing other things. Suffice it to say there is around 4 million (consensus) missing workers, not already accounted for in the BLS report, who need a job and can't get one.
Another factor is seasonal adjustments and the assumptions, algorithms for. I forget when the yearly adjustment is, but it's been over 1 million recently and I keep meaning to dig around on this because that might also be a reason the ADP and the BLS diverge so much.
On the offshore outsourcing tally, yeah, that's from some studies done, where the numbers are estimated and I've just added them together. Problem is, frankly one of them is from a NBER paper, which has some seriously bogus, as in "I can't believe they allow this crap to be published" bogus paper trying to deny offshore outsourcing is affecting the U.S. workforce. Uh, duh, you don't need a 2nd grade education to see what has happened in town after town, particularly the midwest.
But I'll try to gather all of that and I have to do some raw data analysis because the NBER paper is so full of shit. If you're interested, you can pull some of this from the new Census data, but to really get it is near impossible. It's in various press releases and many corporations hide those numbers so close to the chest, you just can't get them. Also a job these days doesn't need to be directly offshore outsourced, more simply created in India or China instead of the U.S.
that I was curious about from one of your comments:
"I've got a running tally on jobs offshore outsourced since 2008 and right now I've got 2.7 million."
here is the sample size of the two surveys for the BLS:
But the numbers I really focused on are from the establishment survey, which "should" be the most accurate.
ADP has a sample size of 340,000 businesses, but I guess is also relying on them being "honest" maybe. These are all private.
There have been other times the ADP diverged from the BLS in slope and it seems to be when the monthly changes are greater. I think the correlation was at 0.94 or so.
I've got to write up more graphs and calculations on this. I fully expected to finally get some jobs and it's like we have something really awful going on here where America and Americans, in terms of the "two economies" has decoupled almost completely at this point.
ADP on the other hand is
Why the huge discrepancy between ADP and the BLS numbers? 297,000 to 103,000 when government only shed 10,000 jobs. Temp workers for the holidays? I wonder how many jobs U.S. companies are creating overseas? I'd like to see those numbers. At some point our congress and president need to reward companies for creating jobs in the U.S. or at least paying taxes in the U.S.
I am very disappointed in these numbers. I truly thought it was going to be a 200,000 plus blowout after the ADP report. Damn.
What is shows is a huge increase in people falling into the homeless, the permanent poor, the jobless who are not counted. It's also creating a flurry of articles on how arcane the BLS methods are and I have to agree. Of all of the overviews and analysis of various economic reports, this one takes me the longest because I must run a host of calculations to find the hidden jobless.
I'm referring to the random variable X, which in this case applies to the unemployed person sample set duration of TIME, or length of
of being counted as unemployed before dropping out or finding a job. The median, defined at 
otherwise notated as 
.
In terms of "x w/ bar over it", uh, I can put
as variables in an equation and it's valid. An equation is not the same as some nomenclature for a symbol implying an average.
uniform distribution is with the people, the unemployed, I'm saying they are all equal, no assumptions are being made in their distribution and only random variable X, that being TIME is being considered.
Just for you,
For a normal distribution, the probability density function (f(x) = dF(x) absolute continuous distribution function, oops F(X) is the cumulative distribution function where m is the median) becomes;
which is a Gaussian distribution.
the Median is usually denoted
and for normal distributions, the mean equals the median and the correct mathematical symbols are:
Here we do not have that kind of distribution going on. It's a "fat tail", the mean or average does not equal the median.
Frankly, I do not expect most people to understand what I'm saying in this comment or even the previous one, because one is only going to see this (odds on) if they take an advanced class in probability ad statistics and remembered it and even used it. That said, I really don't expect to be questioned repeatedly when the above post is correct, from the layman terminology to the equations. It is showing that like wages, due to the super rich and now the long term unemployed, the median is diverging from the average and on top of things, we have more and more people who through no fault of their own can't find a frigging job and it's showing up in these two statistical measurements. Q.E.D.
Your 3rd sentence contradicted your 2nd sentence, and it's x bar(x with a bar over it), not x. I think you understand what you are trying to say, your just didn't do a very good job of explaining it. I'll stick with my comment.
Wall Street banksters bought the Fed, congress and regulatory agenices, so it only makes sense to now buy the White house. Corporatism is now in full bloom. Instead of Commander and Chief, we now have a Bankster and Chief. At least with GWB you knew you were getting an incompetent, bumbling Wall Street butt kisser, but I thought that Obama would be more than simply another Wall Street lackey. Didn't vote for him, but expected more.
50% is half. 50/100 = 1/2. Also, this is not a population sample set, it is those being counted as unemployed. The median means half of this sample set, is an assumed even distribution or uniform distribution of unemployed people. The variable is time, the distribution is the unemployed person, weighted evenly, i.e. uniform. The graph means that the duration where 50%, or when half of those unemployed found another job (or were not counted), has increased over time. The median, a singular data point defined by the variable w.r.t. the sample set, does not imply that all elements of the sample set's variable occurred at the median. That's part of the definition of term median.
Actually I did simplify the description for this particular distribution, but the description is correct. If you can't handle a summation symbol, I doubt the probability distribution functions are going to go over real well in a blog post.
The point of the post is not summation symbols or someone not understanding Statistics 101, never mind 501, the point is more and more people cannot get another @&*)$ job to the point of needing to document that fact for half a decade and this is a crisis and a crime.
Great, now that the banks got all of the money, here comes more bad trade?
I just watched CNBC documentary Facebook, where Facebook helped use social media to elect Obama. We were duped, what a waste.
The median is that point at which half the population, or half the sample of the population, lies below that point, and half lie above that point. So you description of the median graph is incorrect. At the median, half of the unemployed took less than or equal to that amount of time to find a job, and half of the population or sample took a greater or equal amount of time to find a job.
So you lost me before you got to the fancy Riemann Summation after your graphs.
I am aware of the outsourcing that is going on globally these days, however there are holes in your story.
First, outsourcing is done to countries with reliable network connections to the west, I doubt some of the countries you mentioned qualify, for example Armenia.
Second, law cannot be outsourced, BAR certification is needed, which can only be obtained in the United States. Perhaps legal assitance work can be outsourced.
I like how you bolded "officially." There are millions who are not working who are simply not obtaining unemployment benefits. This is a great article and I hope that it educates many.
"and yet our [per] capita income is 15% higher than Australia's ($46k vs. $39.9k)."
I was comparing 2009 wages with current exchange rates, so the 15% differential is inaccurate. Exchange rates are a moving target, but sufficed to say, Australia's per capita income is similar to our own, we just share it (or not) differently.
Good article Steve. Another benchmark would be to peg minimum wage at 10% of congressional salary of ($174,000), which divided by 2000 hour work year is $87. So $8.70 an hour, about where Washington State is as of the New Year ($8.67).
The "universal living wage" folks have an interesting idea, set minimum wage for each metro area based on housing costs x 3 (using HUD indexes). Some of them seem rather high, I guess you're still in Santa Barbara, which would be at $19.33/hr minimum versus Bakersfield, $11.96/hr.
http://universallivingwage.org/ulwformula.htm
Then again, the Australian minimum wage is $15/hr (US and Aus dollars at par currently). More than twice the US minimum wage and yet our capita income is 15% higher than Australia's ($46k vs. $39.9k).
As you can surely imagine, the Australian unemployment rate is catastrophic, from story earlier today:
The national unemployment rate in Australia now sits at 5.2, close to full employment, after more than 400,000 full-time jobs were created in the past year.
http://www.theaustralian.com.au/business/markets/stressed-labour-market-...
I blame the single payer healthcare system for their troubles. :o)
Anxiety over "might happen" is pointless. People, read the United States Constitution "No person shall be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation." That means the only way the government can steal your savings is through taxation or inflation. No matter what Alan Simpson wishes, the law is the law. Only a suicidal politician would support a law to confiscate your IRA. As for inflation, Uncle Ben keeps trying but his center of gravity seems to be stuck.
To help explain a math equation, an assumption and get people to be able to read a graph or two. If we can do it, then by golly, they can do it.
Workforce participation, U6 analysis based on recessions, and CPI adjusted average hourly wages. You'd think that you were a very high level policy analyst providing the latest Medici the absolute best information. In a way your are but they need to read it.
What's the point of an economy, or a political entity, unless it works for all of those involved. It's not like the Middle Ages where people were blind to better options.
The mindless focus on quarterly earnings as a function of someone's bonus is a crime and the graft and corruption on top of that is a crime against the state. Time to clean up and look at the obvious problems and solutions with these factors at the front: a) What works for the people? b) Is it being done with intellectual and professional honesty?
Where will the stimulus for inclusive, honest government comes from? Will it be a big ank failure in 2nd or 3rd Q? Will it take 1000 dead birds falling on Wall Street (specifically) instead of Beebe Arkansas? Does a camel need to fit through the eye of a needle;
Whatever it takes, we need a radical shift for the better and soon. The failures are getting to be too harmful and also, they've become tedious. It's the same old same old from the usual cast of incompetent characters.
Pages