Keeping Unemployment Numbers Real 17 comments
-
Font Size:
-
Print
- TweetThis
The weekly initial unemployment claims data this week showed a dramatic improvement.
Initial unemployment insurance claims is a straight forward indicator which is very useful in gauging where we are economically in our recovery cycle. If initial claims are falling, the economic conditions are improving (or at least getting less bad).
The only “tricks” absolutely necessary to review this data is the seasonal adjustment. Without seasonal adjustment the data wobbles so much it is worthless when you view it WoW.
But seasonal adjustment during labor freefalls also adjusts for events which may or may not be happening. My message is simple – stay with the 4 week moving average which has a better chance of smoothing over the potential errors building.
Overall, if this improvement cycle holds for the rest of October – we are again on the downward path in our job losses during this recession.
Keep in mind that this data only tracks people who apply for unemployment benefits – it does not include the mom and pop business which shuts its doors or the recent graduate who cannot get a job. Conversely, it does not show all the jobs being created.
Workforce Unemployment Rate is Completely Miscalculated
So many try to argue this is not the worst recession since World War II in terms of unemployment. Using the government’s U-3 unemployment rates (headline unemployment), they are correct.
On so many levels, the U-3 unemployment rate defies logic. But no matter how I approach unemployment, the root problem in the government’s data is its defective methodology for determining the size of the workforce.
The unemployment rate is simply the percent of the workforce which is unemployed.
If the size of the workforce is determined incorrectly, then the unemployment rate is wrong.
No one is smart enough to determine who should be included in the workforce. It is my position that unemployment needs to be based on an indisputable baseline – such as population.
The table below shows our last six labor contractions, and how large the contraction was against the population.
Instead of trying to size the workforce, the contraction percent of the workforce was determined by adding up the monthly growth of the gap between non-farm payrolls and population.
Workforce contractions continue longer than the recessions themselves as they do not stop until payrolls grow at the same rate as the population. Workforce contractions also start earlier than the recessions – contractions begin when the growth of payrolls drops below the population growth rate.
This method should produce slightly higher unemployment results than the government’s headline U-3 unemployment rate as it measures a larger area under the curve.
To test the logic, the graph below compares the methods converting population contraction percentages into unemployment rates.
Based on this population baseline unemployment methodology, the U-3 unemployment rates in the recessions in 1974, the early 1980’s, and the early 1990’s are validated.
You see that the U-3 unemployment methodology really breaks down big time beginning with the labor contraction in 2001. This is what many of the “fringe” naysayers have been telling. I too am in this fringe group telling you the unemployment numbers today are plain garbage.
According to John Williams at Shadowstats:
Up until the Clinton administration, a discouraged worker was one who was willing, able and ready to work but had given up looking because there were no jobs to be had. The Clinton administration dismissed to the non-reporting netherworld about five million discouraged workers who had been so categorized for more than a year.
I assume this was in response to the early 1990’s recession where it could be argued that unemployment was overstated by almost a full percent. This metamorphosis of definitions did not help Clinton, but the Bush administration profited from this change.
The 2001 recession was deadly with the convoluted methodology is significantly understating unemployment. In our current recession, this population baseline approach will continue to grow the unemployment rate long after the U-3 unemployment rate peaks – unemployment will grow until jobs growth rates equals population grow rates.
The population baseline approach forces the government to recognize population growth in determining the size of the workforce. It stops word definition games.
Addendum on Methodology
The “Contraction as % of Population” in the above table is the peak percentage of the population who were unemployed in a contraction cycle. The period begins when the growth rate of the workforce falls below the growth of the population, and ends when the growth rate equals or exceeds population growth rate. The U-3 unemployment was the peak unemployment rate of the workforce as determined by the BLS.
Roughly, workforce is half the size of population. If you double the percent contraction numbers shown above and add 5% (5% unemployment is full employment) – you can compare the contraction as a percent of population with the unemployment rate (e.g for Aug 1972 1.38% times 2 equals 2.76%, plus 5% equals 7.6% which compares to the 6.1 U-3 unemployment rate). In the above graph the exact payroll to population ratios were used.
The contraction percentage was determined simply by summing the MoM differences between population and non-farm payrolls during periods where payroll growth was below population growth. You may also conclude that in all other periods workforce grew equal to or faster than population.
This methodology not only calculates the jobs actually lost, but also adds the theoretical portion of the population which was trying to enter the workforce and could not find a job (like graduates).
As a side note, this methodology always forewarned of an impending recession – and can be considered a leading indicator for recessions.
A potential error is introduced by the addition of the full employment factor. Full employment can be argued is in a range between 3% and 7%. Therefore, I would not use this factor at all when expressing unemployment – but it is necessary when you compare population baseline unemployment to the existing unemployment numbers.
Hat tip to Steve at MEMETICS & MARKETING for editing support.
Disclosures: None
Related Articles
|
-
The Seeking Alpha 100
See allThe top 100 stock
market authors
selected for publication
Seeking Alpha 100 » -
Fastest Climbers
Authors with the largestSee all
increase in followers
in the last 7 days
Fastest Climbers » -
Top Commenters
Top commenters,See all
as ranked by their peers
Top Commenters » -
Top Instabloggers
The 100 most active
Instabloggers.-
1
David Fry
-
2
TraderMark
-
3
Don Dion
-
4
Cliff Wachtel
-
5
Mike Havrilla
Top Instabloggers » -
1
-
Top StockTalkers
TopSee all
StockTalkers.
Top StockTalkers »
This article has 17 comments:
To try to place this series in perspective, I did a literature search and based upon the last three recessions, initial claims have to get down to the 400K to 470K level to feel confident that we have passed the trough of the current recession.
Hey, isn't this the same thing that screwed us up already when we allowed FASB to change things in the middle, when we allowed the ratings agencies to change things in the middle, when we allowed the SEC to eliminate 70 year old rules in order to change things in the middle, etc. etc.
The only thing anyone should be questioning is how we want to screw things up further by CHANGING THINGS IN THE MIDDLE - AGAIN?
If it needs to be fixed, fine pull a Congress on it:
Have meetings.
Review the issues.
Discuss the issues.
Place the issues into committee.
Discuss the issues again.
Then do nothing!!!!!!!!!!!!!!!
But here's how I figure unemployment... I have two architect friends, a guy and his wife, working at 2 separate large architect firms in Boston. Were working, rather. They BOTH got canned, about a month apart this Spring. I've got a high school hockey buddy who's son-in-law worked at a big Boston law firm after working at Fidelity and getting his law degree nights. Top 10% biller at the law firm. Canned this summer... but gets to keep his desk and phone and office PC and pay for 6 months while he tries to find a job which after 4 months he hasn't. They figure if he keeps his space there it'll be easier to get a job because he can con the prospects into thinking he's still working. Problem is, there are no prospects.
I see a zillion empty storefronts in FL where I live half the year, and a 1/4 of a zillion in New England. I am yet to see one new storefront anywhere in my travels, so as far as I am concerned the birth/death numbers are the biggest crock of... since I don't know when.
Lawyers, architects, shopkeepers.. this ain't your father's inventory based industrial recession. This is ripping apart the middle class, top to bottom, and I really am tired of hearing about smoothing, and seasonality, and every freaking adjustment the economists can come up with to slap lipstick on a pig and conceal the true destruction to our jobs base, because many if not most of these lost jobs are never coming back anyway, except on some economic statistician's trumped up EXCEL spreadsheet.
Tell me about U-6. Tell me how many are at 26 weeks STILL unemployed and counting. Tell me how many have run out, or are on the brink of running out of benefits, tell me what a 17% U-6 number (and increasing) means for consumer spending in an economy where consumer spending is 70% of GDP and we don't MAKE anything anymore.
The freaking numbers are a whitewash.
All analysts seem to focus on indicators that favor signs of a recovery, which will lead to an even larger correction when everyone finally faces the fact that the world of the working class and middle class, which actually create value in the economy, rather than paper value (which gets the largest compensation), is becoming signficantly worse.
Ignorance is bliss.
Kept in the dark and fed bullsh*t.
Sadly, most citizens want it this way.
the break between 2004 and 2006 seems to smooth out the %unemployment to the upside at an increasing rate...Is this indicative of job recovery or worsening unemployment?
Steve,
Thanks for offering not just a critique of unemployment statistics, but also a useful solution.
Basing employment numbers on population is "So logical it will probably never happen" (to paraphrase political analyst Greg Valliere). Nevertheless, investors need a tool such as this to put the recession in perspective. I especially like the metric you use to measure the "Contraction as % of Population", since this accounts for college grads, immigration, and other sources of population growth.
One question: Does the workforce participation rate reflect this It has collapsed during the recession, and there is no masking its bearish message.
I believe that jobs are the key to this economic cycle, and to this presidential administration. Unfortunately, the massive monetary and fiscal stimulus have produced no jobs, nor are they likely to. Instead, we have a reflation of the global economy and continued outsourcing of American jobs. (The rest of my rant is in "The Deflation of the American Dream") seekingalpha.com/artic...
Excellent research--thanks for sharing.
Rob
Like most (all?) commenters I know people who have lost their jobs and have had to tighten their belts. But the consumer spending figures don't seem to have taken the sort of hit that you'd expect. Are consumer spending numbers another case of "slap lipstick on a pig" (fantastic description ain't no fortunate son).
Secondly, unemployment benefits are in the process of being extended yet again in those states with an unemployment rate above 10% (something like 14, if I remember correctly).
i have missed your comments for a long time.
no matter how the government cuts this, there are more unemployed this recession than any recession since WWII. the scary part is unrestrained globalization where literally entire economic sectors are outsourced overseas.
whenever the economy recovers, only a fraction of the jobs lost will return.
we are also on the doorstep of robotics. we will soon see the toll this takes.
we have terrible future labor fundamentals. and as you point out this reverberates through consumer spending.
What I do know is that all those short-term option traders/traitors out there who do not want to see more then 5 minutes into the future are leaches who are sucking the very life-blood out of all of us.
Revised Tax Rules:
1. Capital gains under <6 months - 55% tax on capital gains
2. Capital gains 6 > 12 months - 45% tax on capital gains
3. Capital gains 1 > 2 years - 35% tax on capital gains
4. Capital gains 2 > 5 years - 18% tax on capital gains
5. Capital gains 5+ years - 5% tax on capital gains
6. Most critical of all — Institute a capital gains tax of 55% on ALL short sales not directly tied to a long buy by a licensed hedge fund. I'm tired of paying for the pure shorts 3rd vacation home.
On Oct 09 11:25 AM TeresaE wrote:
> We are mushrooms.
>
> Kept in the dark and fed bullsh*t.
>
> Sadly, most citizens want it this way.
Your discussion is brilliant on so many levels. I am disappointed that some commenters indicate they didn't absorb the significance of what you have written here. On the other hand, a year or two ago my understanding of the details of the employment data was such that most of your discussion would have been lost on me too. So, anyone who takes offense from my statement of disappointment, no offense was intended unless you also accept that I was offending myself as well.
Some specific items:
1. You wrote: "No one is smart enough to determine who should be included in the workforce. It is my position that unemployment needs to be based on an indisputable baseline – such as population."
So fundamental that it seems obvious. Yet this has not been considered (to my knowledge).
2. You wrote: "But seasonal adjustment during labor free falls also adjusts for events which may or may not be happening."
Again, so fundamental that it shouldn't need to be stated. Unfortunately, it does need to be emphasized. The DOL recently announced that unemployment had been undercounted for 2008 through Q1 2009 by more than 800,000 people and final historical adjustments would be made this coming January. That is almost a million people!!!!
To paraphrase an old saying (I believe it was Everett Dirksen, talking about money): A million here and a million there and pretty soon you are talking about real people.
3. You make a connection between errors in measuring true unemployment and the change in the correlation of unemployment peaks with the bottom in economic activity. While I have not been able to make that correlation with data analysis, I still believe that your logical correlation is very insightful and my failures have been the result of not getting all the data.
4. You wrote : "You may also conclude that in all other periods workforce grew equal to or faster than population."
This is a fundamental factor that is very much unrecognized. This time is different and this is one of the most important differences. I have been writing about the incredible shrinking labor force. We don't know where this will take us.
5. You wrote: "As a side note, this methodology always forewarned of an impending recession – and can be considered a leading indicator for recessions."
This is perhaps one on the 2 or 3 most important statements in the article. "Employment is always a lagging indicator" is one of the most vacuous statements that is made. You can hear it several times a day most days on CNBC, which proves my point. I have written about employment measures that are clearly coincident indicators. I am sure that there are some that can be leading indicators. Most of the self promoting bozos who want you to give them your money to manage won't be bothered to evaluate detailed data (they are too busy writing advertising copy).
This was an article with some very fundamental concepts that can be a resource for reference on how to look at employment data. I hope some of the people who do this (altogether too few, in my opinion) will pursue some of these ideas and report back on their analysis.
I appreciate your comments.
I would like to add that I consider this methodology to usually being a leading (not coincident) indicator of our economy for forewarning a recession is coming.
in july1981 a recession began but the population baseline unemployment began contracting in july1979.
in july1990 a recession began but the population baseline unemployment began contracting in sep1990.
in mar2001 a recession began but the population baseline began contracting in aug2000.
in dec2007 the recession began but the population baseline began contracting in may2006.
even in the july 1990 recession where the population baseline unemployment followed the eventual recession start date - the population baseline was showing weakness before july 1990 - and the shift to negative growth would have been revealed in early october - long before the NBER revealed we were in a recession. Also consider that the 1990 recession was a financially induced recession.