Real Unemloyment Rate: 22.5%?

[pinqy]

NBER definition
Hedonics are purported quality improvements in products that reduce the real price of the good or service. The metrics for determining quality improvement would have made the Edsel the cheapest car of all time since it was chock full of cutting edge technologies that customers neither needed nor wanted. That is the classic rebuttal of hedonics with fees galore free checking also being a top 10 pick of those who find hedonics dubious.

Consequences of hedonics.

Inflation is understated and therefore real economic growth is overstated two ways directly which may be marginal and indirectly by overstating investment. Hedonics overstates real investment not so much when quality improvements are attributed to new capital equipment but when physical depreciation to existing capital equipment is understated in terms of replacement costs. The GNP and later GDP have overstated Net National Investment for the past half century which also coincided with the hollowing out of the US manufacturing base and massive real estate leverage (housing is considered capital investment). Prior to hedonics the quarter Trillion dollar S&L disaster was one such consequence of overstatement. Although it is not currently known how large the current bill will end up being the most conservatively run major bank, Morgan Chase, has just under $2 T mortgage exposure as owner or agent. That is one of the reasons JPM is so admired it mostly stayed out of the real estate bubble and was given WAMU and Bear Sterns along with being offered Morgan Stanley for free but that deal was refused. (Treasury Secretary Paulson made that offer during the WAMU takeover.) So how badly has hedonics overstated investment? I would recommend contacting Morgan Stanley, which is still in existence because nobody who was in good enough to take it over wanted it.

I have no idea of which are your words and which are supposed to be NBER's definition. You forgot to include a link. In any case I'm very familiar with hedonics, though the math gives me headaches. I wasn't asking what they are or why you think they're wrong, but rather what you think NBER has to do with it. Again..I'm not familiar with NBER putting out anything where they would use hedonics in their calculations..t.hat's usually done by statistical agencies, which NBER is not.

Feel free to look it up online or in the dictionary.

Why would a dictionary tell me what you think NBER has to with hedonics?
I already said it, I'll go with bold this time "In any case I'm very familiar with hedonics, though the math gives me headaches. I wasn't asking what they are or why you think they're wrong, but rather what you think NBER has to do with it."

While the NBER may no longer be the chef in charge of this variety of book cooking just as they also backed out UE computation decades ago they are still the brain trust for coming up with new indices and ways of tweaking them.

No, they don't. And when were they ever involved in any official statistics? They're not a statistical agency, and never have been. They're not a government agency, and while they certainly do research and suggestions, they have NO official input into any government definitions or methodology. Where do you get these ideas?

 

[william the wie]

Do you know anything about the founding and operation of the NBER or read any of their statistical summaries of economic series that predate current government agencies? Just go to any college library and look in the economics reference section, you might learn something.

 

[loosecannon]

The NBER was founded in 1920. Its first staff economist, director of research, and one of its founders was American economist Wesley Mitchell. Russian American economist Simon Kuznets was working at the NBER when the U.S. government asked him to help organize a system of national accounts in 1930, which became the beginning of an official measurement of GDP and other related indices of economic activity. Due to its work on national accounts and business cycles, the NBER is well-known for providing start and end dates for recessions in the United States.

The NBER is the largest economics research organization in the United States.[2] Sixteen of the thirty-one American winners of the Nobel Prize in Economics have been NBER associates. Six of the past Chairmen of the Council of Economic Advisers have also been researchers at the NBER, including the former NBER president, Martin Feldstein. The NBER's current President and CEO is Professor James M. Poterba of MIT. NBER research is published by the University of Chicago Press.

wiki

 

[antagon]

the rate of 'real' unemployment depends on the definition of the term. if you track the state U3, then that's 'real' because it is consistently tracking the same definition. if you are a shadowstats fan, then you are just looking at a bigger #. if you do that consistently then maybe you could track our employment situation that way, although i'd think that source is less reliable than others with the resources to actually track unemployment.

why not just look at labor force utilization?

 

[loosecannon]

why not just look at labor force utilization?

is there such a stat? Where? BLS?

 

[william the wie]

why not just look at labor force utilization?

is there such a stat? Where? BLS?

I don't who publishes it I would check wiki however antagon is largely right about U3 which is generally a better number than U1 with only a few holes of unknown size:

Those who are not collecting UE insurance such as recent graduates or those whose UEI are exhausted are it is claimed missed with great regularity.

The homeless do get included in the check number of the job birth death estimate as employed but are much less likely to be counted as unemployed because they are much less likely to answer their cell phones when money is tight.

The job birth-death estimate has quality constraints that are budget based therefore reconciliation of job growth vs. unemployment is to some unknown extent guesswork.


While the first two holes combined are probably +/- 1% accurate in what they can readily measure the reconciliation process with the job birth death estimate that is supposed to uncover those missed by the employment survey is definitely the big problem area. Going from roughly 64.6% to 56% in labor force utilization means that 15% of the old labor is no longer employed or unemployed according to the BLS so tacking on an additional 15% to published UE is defensible, claiming 13% more non-PC unemployment than the BLS posts is quite defensible.

 

[antagon]

why not just look at labor force utilization?

is there such a stat? Where? BLS?

nah. they dont track institutional individuals.

if you are looking at how the employment situation compares to another point in time, the BLS will have you covered. i think the way they split the indexes based on the questions they ask are informative.

i like looking at working aged to working in a ratio. maybe all ages. that's economic efficiency. 'adhesion'... i dig the OECD statbank. cumbersome, but informative.

OECD Statistics (GDP, unemployment, income, population, labour, education, trade, finance, prices...)

 

[antagon]

why not just look at labor force utilization?

is there such a stat? Where? BLS?

I don't who publishes it I would check wiki however antagon is largely right about U3 which is generally a better number than U1 with only a few holes of unknown size:

Those who are not collecting UE insurance such as recent graduates or those whose UEI are exhausted are it is claimed missed with great regularity.

The homeless do get included in the check number of the job birth death estimate as employed but are much less likely to be counted as unemployed because they are much less likely to answer their cell phones when money is tight.

The job birth-death estimate has quality constraints that are budget based therefore reconciliation of job growth vs. unemployment is to some unknown extent guesswork.


While the first two holes combined are probably +/- 1% accurate in what they can readily measure the reconciliation process with the job birth death estimate that is supposed to uncover those missed by the employment survey is definitely the big problem area. Going from roughly 62% to 50% in labor force utilization starting at a base of 4% unemployment should result in 24% unemployment not 9.7+% unemployment as has been observed 2005-10. You did notice those discrepencies in Neubarth's 19 things thread well they are the result of the reconciliation mess.

the BLS has a specific fact that they are dishing. understanding it will not lead you to believe that it should reflect labor force utilization in that way. if someone is in jail or on welfare, they're not unemployed, so i can appreciate that. but that's why i look to utilization stats because they tell a broader story.

the BLS is also not hooked to UI directly.

 

[william the wie]

Quite true and I used your link and one of my own to do a quick check on the numbers I remembered and found the errors I made in my unedited post. Thanks for steering me to that source I always forget where the hell that data is.

 

[pinqy]

Explaining the BLS numbers:
First is Employment...a monthly survey of approx 140,0000 businesses (410 worksites). Reference period is the pay period that contains the 12th of the month. This is the non-farm payroll survey and counts jobs, not individuals (a person with 3 jobs is counted 3 times). Excluded are the self-employed, businesses that don't contribute to Unemployment Insurance, and domestic workers (those who work in other people's houses). Each worksite is asked how many people were on the payroll during the reference period. Because places go out of business and new businesses appear and neither will be captured in the survey until a full census of businesses is done, BLS uses a birth/death model to mathematically add and subract what the statistical model predicts are the changes in businesses and employment. Then in March a full count of all businesses is done and the next March adjustments are made based on the real data. The recession appears to have caused problems for the model and there was a large overcount in 2009.

The Employment data is in the B tables of The Employment Situation Report and also includes average hours and wages. The emplyment numbers are preliminary, and are adjusted the next 2 months for late reporting.

Labor Force and Unemployment data comes from the Current Population Survey, a joint Census/BLS monthly survey. This survey interviews approx 60,000 households (120,000ish individuals). The population is the Adult Civilian non-Institutional Population and includes all people in the US 16 years and older (15year olds are surveyed but not included) who are not in the military or in an institution (prison, mental institute, other long term health care). Its reference period is the week of the month that contains the 12th.

The CPS asks if the people in the household worked at all for pay during the reference period, owned their own business, or worked for at least 15 hours in a family-run business or farm. If yes, the person is Employed (note the differences in coverage between the CES and CPS..for Sep the CES has 130,201 jobs, and CPS has 139,391 people employed). Further questions determine part time vs full time, and if part time if the person wants full time and if yes, then if part time is for economic or non-economic reasons. Also if the person has multiple jobs, and hours for all jobs.

If a person didn't work, they're asked if they looked for work in the previous 4 weeks. If yes, they're Unemployed. Further questions determine why they're unemployed and for how long.

Employed + Unemployed = The Labor Force (people willing, able, and attempting to work). The Unemployment rate is the percentage of the the Labor Force that's Unemployed (NOT the % of the population).

Everyone else is Not in the Labor Force. Further questions are asked if the person wants a job. If they want a job they're asked if they are available to work and if they've looked in the previous 12 months. If so, they're Marginally Attached (potential labor...people likely to work if circumstances change...people who aren't available and/or haven't looked in over a year are not likely sources of labor regardless of saying they want to work). The Marginally Attached are then asked WHY they haven't looked. If they say they think they wouldn't be hired, then they're Discouraged Workers (not looking because of belief about the job situation).

To look at different aspects of labor utilization, BLS has 6 measures:
U-1 Persons unemployed 15 weeks or longer, as a percent of the civilian labor force
Gives us a picture of the long term unemployed, as opposed to frictional unemployment.

U-2 Job losers and persons who completed temporary jobs, as a percent of the civilian labor force
Gives us a picture of the layoff situation and how that changes.

U-3 Total unemployed, as a percent of the civilian labor force (official unemployment rate)
Tells us the real relative difficulty of getting a job (any job).

U-4 Total unemployed plus discouraged workers, as a percent of the civilian labor force plus discouraged workers
Tells us the real and perceived difficulty in getting a job.

U-5 Total unemployed, plus discouraged workers, plus all other persons marginally attached to the labor force, as a percent of the civilian labor force plus all persons marginally attached to the labor force
Shows us the percentage of the potential labor force that's not working.

U-6 Total unemployed, plus all persons marginally attached to the labor force, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all persons marginally attached to the labor force
Shows us the percentage of the potential labor force that's not used fully.

All of these are "real" unemployment looked at from different angles.

What shadowstats apparently does is add in all people who say they want a job, whether or not they could actually take one or whether they're serious about wanting a job. And then he adds more, I think. Using the formula of (Unemployed + want a job now + part time for economic reasons)/(Labor Force + want a job now) we only get 19%. He has to be using all part timers or some weird variation.

 

[william the wie]

The problem is that when people look around them and see beggars, empty shops and foreclosed homes reality proclaims that the published data is BS.

 

[loosecannon]

why not just look at labor force utilization?

is there such a stat? Where? BLS?

nah. they dont track institutional individuals.

if you are looking at how the employment situation compares to another point in time, the BLS will have you covered. i think the way they split the indexes based on the questions they ask are informative.

i like looking at working aged to working in a ratio. maybe all ages. that's economic efficiency. 'adhesion'... i dig the OECD statbank. cumbersome, but informative.

OECD Statistics (GDP, unemployment, income, population, labour, education, trade, finance, prices...)

A few years ago the BLS data indicated that only 60% of the working age group was employed. While only 80% of the working age group was in the labor force.

 

[antagon]

is there such a stat? Where? BLS?

nah. they dont track institutional individuals.

if you are looking at how the employment situation compares to another point in time, the BLS will have you covered. i think the way they split the indexes based on the questions they ask are informative.

i like looking at working aged to working in a ratio. maybe all ages. that's economic efficiency. 'adhesion'... i dig the OECD statbank. cumbersome, but informative.

OECD Statistics (GDP, unemployment, income, population, labour, education, trade, finance, prices...)

A few years ago the BLS data indicated that only 60% of the working age group was employed. While only 80% of the working age group was in the labor force.

trying to get data for my own mad science, i've been annoyed by the BLS tracking only civilian non-institutionals. could the fate of the 20% balance be outside this characterization?

i like that the idea of tracking that ratio if it were easy to lump non-civilians in too.

 

[loosecannon]

trying to get data for my own mad science, i've been annoyed by the BLS tracking only civilian non-institutionals. could the fate of the 20% balance be outside this characterization?

i like that the idea of tracking that ratio if it were easy to lump non-civilians in too.

Not that I fully understand what you are saying but when you have a 1% prison population they really are an important part of the employment picture. On the other hand I believe about half of those incarcerated in the US actually are employed. They should be counted either way.

 

[antagon]

trying to get data for my own mad science, i've been annoyed by the BLS tracking only civilian non-institutionals. could the fate of the 20% balance be outside this characterization?

i like that the idea of tracking that ratio if it were easy to lump non-civilians in too.

Not that I fully understand what you are saying but when you have a 1% prison population they really are an important part of the employment picture. On the other hand I believe about half of those incarcerated in the US actually are employed. They should be counted either way.

they should be counted if you are trying to look at the losers in a shit job market. it is my criticism of reagan's unemployment rebound after the forever recessions in the early 80s. folks just picked up and moved to prison cells and welfare rolls, and off the unemployment map altogether. i dont think the BLS figures should be designed to paint the worst picture of unemployment possible. i was saying earlier that this figure is only valuable as a relative term: it was like this, now it is like this. it should line up with other state's practices on unemployment, which are nearly all more opaque and exclusive than the US, that i am aware.

if some shadow stats guy pulls a figure from his hat and declares that it is 27%, that is no more real than the BLS figure. the reality is in the same methodology compared to the same at two different intervals. i wonder what people mean when they say 'real' unemployment if they are not looking at a raw picture of total population to hours worked/40.

 

[loosecannon]

My thinking is that employment data should not be geared to produce simple stats with complex underlying mechanisms.

It should be the inverse: employment data should provide a concise clear picture of what the overall state of our employment profile is, while employing almost no mechanizations involved in producing simple stats.

A simple breakdown of x% fully employed, y% employed part time, i% in prison and unemployed, etc.

About 20 such categories round out the picture pretty completely.

What could be worse than having 6 different definitions of unemployed?

And the prison population counts because they have been removed from the active workforce and are often employed in a captive work force. It makes a great deal of difference if that number is .4% or 4.2%.

 

[antagon]

i feel what you are saying, but that is just your opinion of how to look at unemployment. the BLS is convinced that it is a measurement of people trying to get work via a certain standard of effort who dont seem to be able to succeed. that's not so far fetched to me. some people are half-assin the pursuit of work for whatever reason, and rather than counting them in with everyone else or not at all, theyre in a deeper index. that seems ok to me too. this is different than accounting for labor force/pop util, or hard prison labor, or employment, which they also track separately.

 

[loosecannon]

But seriously, between the two: an index specifically contorted to produce a few slim data bites or a 1/4 page list that summarizes the entire nation's employment profile, which is more useful for more purposes?

To favor a slender definition of folks who are sufficiently involved in attracting employment vs a snapshot the nation's complete employment profile you are favoring a political term rather than an enlightening spectrum of results.

Think about how many possible purposes could be served by the profile approach. It would be an invaluable tool for a myriad of disciplines. Esp over the very long term.

 

[uscitizen]

But people are not supposed to understand. the few that care to find out anyway.

 

[loosecannon]

But people are not supposed to understand. the few that care to find out anyway.

But they don't really find out, because data is hidden behind the curtain. You can't reconstruct an honest survey by recollating the results of a flawed survey.