What are the real unemployment numbers?

[Arraya]

http://www.chrismartenson.com/blog/fuzzy-numbers-jobs-report-bad-minus-533000-made-look-better/9760

But it would have been worse had it not been for the mysterious positive addition of 30,000 jobs by the so-called birth-death model. From here on out we’ll just refer to it as the birth-birth model because no matter how bad the underlying economic data this model somehow always adds jobs in every month except typically January and July when the model backs out a few of its more irrationally exuberant additions.

However, keeping regular track of the amount of fudging in the Birth-Death model is one of my favorite activities because I think it provides insight into just how unreliable government statistical models really are. For a nation that drives economic policy “by the numbers” it’s important that the numbers are good. The Birth-Death model provides a valuable caution flag for anyone with a tendency to “believe the numbers” that are being offered up for consumption.

[poorsaver]

I think Karl nails it on the head here:

http://market-ticker.denninger.net/archives/675-Quick-Ticker-Unemployment-12.2%25.html

[poorsaver]

I think Karl nails it on the head here:

http://market-ticker.denninger.net/archives/675-Quick-Ticker-Unemployment-12.2%25.html

[poorsaver]

I think Karl nails it on the head here:

http://market-ticker.denninger.net/archives/675-Quick-Ticker-Unemployment-12.2%25.html

[poorsaver]

I think Karl nails it on the head here:

http://market-ticker.denninger.net/archives/675-Quick-Ticker-Unemployment-12.2%25.html

[poorsaver]

I think Karl nails it on the head here:

http://market-ticker.denninger.net/archives/675-Quick-Ticker-Unemployment-12.2%25.html

[EconProf]

Socratt, good question. If 550,000 (net) jobs were lost last month, why did unemployment rate only rise by two-tenths of a percent? Answer: some of the unemployed stopped looking for work. The household survey compiles the # of unemployed. The # of jobs is more of a payroll total coming from employers.
By the way, your chance of being interviewed in any given month are about 1 in 3000.
Arraya, your reference to that blog is a good one, as it brings up one valid complaint to the way the BLS compiles their statistics. The Birth/Death allowance supposedly allows for the imputed jobs created or destroyed monthly by the birth of new businesses or the death of same. Their model is greatly flawed because it overstates job creation. It recently, for example, claimed that construction jobs are still growing in the U.S. A Barron’s columnist, Gene Epstein, has frequently mocked the BLS for these conclusions.
Still, the erroneous numbers are not large in the scheme of things, and are due more to bureaucratic sloth than dark conspiracy.

[EconProf]

Socratt, good question. If 550,000 (net) jobs were lost last month, why did unemployment rate only rise by two-tenths of a percent? Answer: some of the unemployed stopped looking for work. The household survey compiles the # of unemployed. The # of jobs is more of a payroll total coming from employers.
By the way, your chance of being interviewed in any given month are about 1 in 3000.
Arraya, your reference to that blog is a good one, as it brings up one valid complaint to the way the BLS compiles their statistics. The Birth/Death allowance supposedly allows for the imputed jobs created or destroyed monthly by the birth of new businesses or the death of same. Their model is greatly flawed because it overstates job creation. It recently, for example, claimed that construction jobs are still growing in the U.S. A Barron’s columnist, Gene Epstein, has frequently mocked the BLS for these conclusions.
Still, the erroneous numbers are not large in the scheme of things, and are due more to bureaucratic sloth than dark conspiracy.

[EconProf]

Socratt, good question. If 550,000 (net) jobs were lost last month, why did unemployment rate only rise by two-tenths of a percent? Answer: some of the unemployed stopped looking for work. The household survey compiles the # of unemployed. The # of jobs is more of a payroll total coming from employers.
By the way, your chance of being interviewed in any given month are about 1 in 3000.
Arraya, your reference to that blog is a good one, as it brings up one valid complaint to the way the BLS compiles their statistics. The Birth/Death allowance supposedly allows for the imputed jobs created or destroyed monthly by the birth of new businesses or the death of same. Their model is greatly flawed because it overstates job creation. It recently, for example, claimed that construction jobs are still growing in the U.S. A Barron’s columnist, Gene Epstein, has frequently mocked the BLS for these conclusions.
Still, the erroneous numbers are not large in the scheme of things, and are due more to bureaucratic sloth than dark conspiracy.

[EconProf]

Socratt, good question. If 550,000 (net) jobs were lost last month, why did unemployment rate only rise by two-tenths of a percent? Answer: some of the unemployed stopped looking for work. The household survey compiles the # of unemployed. The # of jobs is more of a payroll total coming from employers.
By the way, your chance of being interviewed in any given month are about 1 in 3000.
Arraya, your reference to that blog is a good one, as it brings up one valid complaint to the way the BLS compiles their statistics. The Birth/Death allowance supposedly allows for the imputed jobs created or destroyed monthly by the birth of new businesses or the death of same. Their model is greatly flawed because it overstates job creation. It recently, for example, claimed that construction jobs are still growing in the U.S. A Barron’s columnist, Gene Epstein, has frequently mocked the BLS for these conclusions.
Still, the erroneous numbers are not large in the scheme of things, and are due more to bureaucratic sloth than dark conspiracy.

[EconProf]

Socratt, good question. If 550,000 (net) jobs were lost last month, why did unemployment rate only rise by two-tenths of a percent? Answer: some of the unemployed stopped looking for work. The household survey compiles the # of unemployed. The # of jobs is more of a payroll total coming from employers.
By the way, your chance of being interviewed in any given month are about 1 in 3000.
Arraya, your reference to that blog is a good one, as it brings up one valid complaint to the way the BLS compiles their statistics. The Birth/Death allowance supposedly allows for the imputed jobs created or destroyed monthly by the birth of new businesses or the death of same. Their model is greatly flawed because it overstates job creation. It recently, for example, claimed that construction jobs are still growing in the U.S. A Barron’s columnist, Gene Epstein, has frequently mocked the BLS for these conclusions.
Still, the erroneous numbers are not large in the scheme of things, and are due more to bureaucratic sloth than dark conspiracy.

[socrattt]

EconProf I just can’t imagine that 320,000 +/- people (the percentage of people that would amount to the .3% of people in question) just stopped looking for work. It just doesn’t make sense. Even in a strong economy I can’t imagine that high of a percentage of the unemployed would just drop off the map. I think the government will see some major fundamental problems with the way they track unemployment if and when we get out of this mess.

[socrattt]

EconProf I just can’t imagine that 320,000 +/- people (the percentage of people that would amount to the .3% of people in question) just stopped looking for work. It just doesn’t make sense. Even in a strong economy I can’t imagine that high of a percentage of the unemployed would just drop off the map. I think the government will see some major fundamental problems with the way they track unemployment if and when we get out of this mess.

[socrattt]

EconProf I just can’t imagine that 320,000 +/- people (the percentage of people that would amount to the .3% of people in question) just stopped looking for work. It just doesn’t make sense. Even in a strong economy I can’t imagine that high of a percentage of the unemployed would just drop off the map. I think the government will see some major fundamental problems with the way they track unemployment if and when we get out of this mess.

[socrattt]

EconProf I just can’t imagine that 320,000 +/- people (the percentage of people that would amount to the .3% of people in question) just stopped looking for work. It just doesn’t make sense. Even in a strong economy I can’t imagine that high of a percentage of the unemployed would just drop off the map. I think the government will see some major fundamental problems with the way they track unemployment if and when we get out of this mess.

[Author]

Posts