One of the alleged "green shoots" perking up the economy was that unemployment was getting worse more slowly. Economic and financial journalists have been calling this the "second derivative" argument: measuring the state of the economy by the rate of change in the rate of change. There's something to this, but not as much as was made out of it--if unemployment continues to grow more slowly for another two years, that's still bad news.
Then, of course, how are you supposed to feel when the initial jobless claims figures start going back up, as they did this week? I'm more inclined to credit jobless students flooding the market than some disastrous turn in our economy's fortunes. But then, I was never that cheered by the second derivative in the first place. I'll feel cheerful when the unemployment figures start going down.
Copyright © 2010 by The Atlantic Monthly Group. All rights reserved.
The up movement is most likely noise. The four week moving average was up just 500 jobs. The week to week stuff is fairly noisy. For example we hit the peak of 674K in the last week of March only to dive to 605K by the first week in May and then back up to 643K by the second week in may.
Moreover, you can't claim unemployment if you weren't just laid off. So students don't count.
There's also the fact that first and second derivatives are unrelated in the sense that it is quite easy to construct a decreasing function with a second derivative that is positive everywhere in the domain of the function. Speaking of 'second derivatives' when one means that the rate of job losses are decreasing strikes me as obfuscation, not clarification. It gives the sense that there's a lot more going on than what's really the case.
Oh, be still my heart! Someone who really learned their calculus.
The author said: "I'm more inclined to credit jobless students flooding the market"
Jobless grads do not file unemployment claims. You have to actually have been "employed" before you file a claim.
Ed beat me to it.
Of course, maybe there are states that allow students to apply- New York, California, Maryland, Vermont, or Massachusetts come to mind as possibilities.
The unemployment rate as measured by new jobless benefit claims is a dubious statistic. What I think we are trying to measure is the number of people out of work and wanting to work. What we are really measuring is new people applying for benefits. Some of those people just lost their job and need a hand but I bet there are a significant percentage that are gaming the system. Here is my own example:
In my youth I worked at a stone quarry in northern Ohio as a summer job (dirty but awesome job paying $6.50 an hour in 1984). The quarry was unionized and had about forty full time employees. Every winter the quarry would shut down and "lay off" everyone. This was a scheduled shutdown and happened every year to coincide with the reduction in stone demand during the winter. The schedule was fully coordinated with all the workers to maximize the state unemployment and union benefits. Basically they got four months off every winter at about 75% pay. Of course most of them worked side jobs during the shutdown for cash. Come spring the quarry would restart just as benefits expired. They had the system down and I bet they still do.
How many people claim benefits and don't want to or need to work?
The second derivative is almost entirely shameless spin. It was touted by entities that, under GWB, treated every "disappointing" drop in joblessness as sign of gloomy economy.
I love that you throw some obviously safe predictions in now and then to offset the major ones you get completely wrong. Any forecast on whether or not the sun rises tomorrow?
In general, or in your world specifically?
How about the jobs that the trillion dollar stimulus has saved? Imagine without that, the unemployment rate will be 100%
How many people claim benefits and don't want to or need to work?
And how many people can't claim benefits because they were self employed or independent contractors? If you want to look at the number of people who want work and can't then take a look at U-5 or U-6 unemployment rates instead of the published U-3. U-6 is currently 15.9%
http://www.bls.gov/news.release/empsit.t12.htm
U-6 isn't an unemployment rate. It counts people who ARE employed. U-5 is in a sense a sort of unemployment rate, but since it includes people who aren't actively seeking jobs it isn't what is normally thought of when we use the term "unemployment rate", and isn't comparable with past unemployment rates or unemployment in most other countries. (unless your using U-5 or something very similar for those times and places)
U-3 isn't based on jobless claims. So the fact that some people can't claim benefits doesn't distort it.
------
"...Some people think that to get these figures on unemployment, the Government uses the number of persons filing claims for unemployment insurance (UI) benefits under State or Federal Government programs. But some people are still jobless when their benefits run out, and many more are not eligible at all or delay or never apply for benefits. So, quite clearly, UI information cannot be used as a source for complete information on the number of unemployed.
Other people think that the Government counts every unemployed person each month. To do this, every home in the country would have to be contacted—just as in the population census every 10 years. This procedure would cost way too much and take far too long. Besides, people would soon grow tired of having a census taker come to their homes every month, year after year, to ask about job-related activities.
Because unemployment insurance records relate only to persons who have applied for such benefits, and since it is impractical to actually count every unemployed person each month, the Government conducts a monthly sample survey called the Current Population Survey (CPS) to measure the extent of unemployment in the country. The CPS has been conducted in the United States every month since 1940, when it began as a Work Projects Administration project..."
http://www.bls.gov/cps/cps_htgm.htm#where
"Other people think that the Government counts every unemployed person each month."
That would create at least a million jobs for counting jobs, no?
The typical figure I see for number of self employed workers in the US is around 30-33% of the workforce. Without getting a handle on how many of these people are no longer employed, it is hard to figure the "true" unemployment numbers. I realize the claims number is easily available data, but it is a 1950's way of looking at the economy. Were contract employees stripped off early, before full-time? Probably. Are some firms hiring part-time contract personnel to replace former full-time positions? Possibly.
I don't know what numbers you could look to for better data, I just don't see a ton of usefulness in looking at the first time claims delta.
I also think first time claims data is only modestly useful. I was just pointing out that first time, or all, claims data, isn't part of the unemployment rate. (either the standard U-3 or any of the other U's 1-6).
Byrk seemed to think that because of the problems with and limitations of first time claims data that we should look to U-6 for the "real unemployment rate". But those problems have no connection to any of the unemployment rate, and also U6 counts many employed people as "unemployed".
Basically they got four months off every winter at about 75% pay.
I know of at least one school district that does precisely that for its bus drivers. So while the current bump probably isn't students, it might be seasonal school employees.
How many people claim benefits and don't want to or need to work?
What does that have to do with anything - it's insurance.
I know of at least one school district that does precisely that for its bus drivers. So while the current bump probably isn't students, it might be seasonal school employees.
The numbers are seasonally adjusted so they shouldn't be impacted by seasonal layoffs that happen every year.
The problem with using derivatives is that as you go up in order, noise is enhanced.
The number varies over the years, but under 50% of the unemployed are eligible for unemployment compensation.
You know you are amongst econo-nerds when the title of this posting makes sense.
Next week, on the physics forum, people will be referring to odious posters as "third derivatives..."
I told my girlfriend she was average. She said "that's mean!"
I am self employed. My business fell apart over a year ago due to housing bust. I have earned about $75.00 in 2009. I cannot collect unemployment, although I did legally collect disaster unemployment after the last hurricane around here. I shall now call myself a first derivative or a non derivative. I have not decided which.
Sorry, you are the proverbial soul who passed thru the crack. But feel free to pay a higher tax rate when you eventually found employment.
For anyone trained in the hard sciences, employment statistics look like gibberish measurements.
Back in college, my professor brought in a binder that contained the hundreds of arbitrary definitions of what constituted unemployment. For example, a family decides that given the cost of child care, transportation, effects on the children etc that one of the spouses should quit their job and stay home. Are they truly unemployed? In a period of inflation, such as back in the seventies, cost outstripped wages so some people dropped out of the work force. What if they would and could reenter the workforce if compensation rose by 10%? Are students unemployed when they quite their summer job? How do you measure freelancers?
Worse, the definitions change over time such that an person counted as unemployed in 1970 would be counted as employed in 1980 and vice versa.
At best unemployment statistics can only give us hindsight information in any given time span of less than 10 years (because of shifting definition.) In shorter terms, only the grossest, integer point movements have any predictive value.
People who pay attention to the minutia of this imprecise metric might as well be reading tea leaves.
Unemployment is fairly well defined: not currently working, actively seeking work and available to start immediately. Certainly, your test cases pose absolutely no problem for the standard definition.
http://www.bls.gov/cps/cps_htgm.htm
Lets look at your test cases:
No. They are not actively seeking work, and are therefore out of the labor force.
I presume these people are not actively seeking work? Then they are not unemployed.
Did they do work for pay or profit in the survey week? If so, they are employed.
If they did not do work for pay/profit in the survey week, did they search for work in the past 2 weeks? If so, they are unemployed.
If they did not work and did not seek work, they are not in the labor force.
I was never tremendously charged up by the word 'derivative' in the first place. Recent events have done nothing to reduce that standing.
Is it possible for people to be TOO smart? The evidence seems increasingly conclusive.
Jimminy,
I think you mean "too educated". The smart people are off having lives and not arguing about possible misinterpretations of mathematical terms on blogs...
Oh hang on, that rules me out of the smart category. D'Oh!