It is amazing that most jobs stories featured upfront the revision to November’s Non-Farm Payrolls. November was revised to +4,000 from the original -11,000. A net gain. However, October was revised downwards – a complete wash.
The fact is that we lost a surprising 85,000 Non-Farm Payroll jobs. A widely respected economic consulting group expected +50,000. The market expectation was about flat.
I am convinced the situation is much worse than we are led to believe.
Non-Farm Payrolls data comes from the Establishment Survey. It showed a decrease in payrolls of 85,000 in December.
The widely reported Unemployment Rate comes from a different source – the Household Survey. The rate is defined as the number of unemployed, 15.267 million, divided by the civilian labor force, 153.059 million, which equals 9.997% or 10.0% as reported. To be clear, the Non-Farm Payrolls data do not enter the calculation of the Unemployment Rate.
But the Household Survey shows a far worse situation. In December, employment dropped by a staggering 589,000. But, confusingly, unemployment also dropped by 73,000. Overall, the size of the civilian labor force shrunk by 661,000.
What is going on?
First, the decrease in the civilian labor force “explains” why the unemployment rate did not change. We are dividing the number of unemployed by a smaller number.
Second, and more importantly, is to understand the reason for the shrinking size of the civilian labor force. It is shrinking because people are becoming discouraged and have dropped out of the labor force. These people are not counted in the numerator (unemployed) or the denominator (labor force).
Let’s take a closer look.
The Marginally Attached and Discouraged
There are 2.486 million workers in December 2009 that are classified as “Marginally Attached”.
The BLS definition of “Marginally Attached” is: “These individuals were not in the labor force, wanted and were available for work, and had looked for a job sometime in the prior 12 months.” These individuals are not counted as “Unemployed” because they failed to search for work in the past four weeks.
In December 2007 (the beginning of the recession), there were only 1.395 million “Marginally Attached”. In November 2009, there were 2.323 million “Marginally Attached”. Over the last month, 153,000 additional people became “Marginally Attached”.
There are 929,000 individuals that are classified as “Discouraged” in December 2009. The “Discouraged” are a subset of the “Marginally Attached”.
The BLS definition of “Discouraged” is: “Persons not currently looking for work because they believe no jobs are available for them.”
Over the past month 68,000 additional people were classified as “Discouraged”.
Again to be clear – neither “Discouraged” nor the broader “Marginally Attached” are included in the numbers. Does that make sense to you?
Not to me.
If we add the Discouraged into the calculation the unemployment rate is 10.5%. If we include the broader category of “Marginally Attached”, the rate goes to 11.4%.
Finally, there are a large group of persons that are working part time — but don’t want to work part time. Suppose you work 10 hours a week but you really want to work 40 hours a week. It seems to me that you are 3/4 unemployed. However, the widely reported numbers do not include these underutilized people.
The BLS does report the U-6 rate which includes all “Marginally Attached” and all Part-Time that want full time work. They don’t do the fractional calculation that I suggest for the part-time persons. As a result, this rate is somewhat overstated.
In December 2009, the U-6 or “all in” rate, was 17.3% up from 17.2% in November.
So, the situation is worse than one would think.
More Sobering Data
The number of people unemployed for more than 26 weeks rose in December 2009 to 6.130 million persons from the previous month’s 5.901 million. At the beginning of the recession, this statistic was only 2.612 million.
The average number of weeks unemployed increased to 29.1 weeks in December 2009 compared to 28.6 weeks in the previous month. At the beginning of the recession, this statistic was only 16.5 weeks.
Why Analysts are Fooled by the Trend in Initial Claims for Unemployment Insurance
The trend in Initial Claims for Unemployment Insurance has been favorable. The most recent week saw “only” 434,000 new claims. The four-week average was 450,250. This is a big improvement from January to June 2009 where the rate was averaging well above 600,000 per week (actually more than 700,000 in January).
While this is good news, it does not immediately translate into lower unemployment. This is despite the strong historical correlation between initial claims and unemployment.
Why? There is an “omitted variable”. Claims can decrease but unemployment can increase. Reduction in unemployment depends on people getting jobs. The statistics I quoted above suggest people are not getting jobs.
Yes, fewer people are being added to unemployment rolls — but this just increases the level of unemployment at a slower rate.
Here is another way of looking at it. While it is good news that fewer are applying for unemployment insurance, it is bad news that those already getting insurance can’t find work – or are taking longer to find work. The good news is nixed by the bad news.
The way to see this is to look at people rolling off the state insurance programs into the Federal Emergency program. While there are reporting delays, in the week of December 19, the EUC had 5.143 million beneficiaries. This was up 235,626 from the previous week and sharply higher than last years’ 1.922 million.
Finally, there are people that have exhausted EUC and now have no benefits. These people are not tracked in the data (or at least I don’t know if these data exists).
Don’t Forget the Growth in Population
As I mentioned in my last posting, the working age population is growing by at least 100,000 persons per month. This means to simply hold the unemployment rate, we need to add 100,000 to Non-Farm Payrolls. In addition, we have 7.2 million jobs to create, to get back were we started in December 2007.
No matter how you cut the data, this is bad news. It will likely take five years and no hiccups to get back were we started.
Is there any Good News?
- Initial claims are decreasing.
- A number of surveys are showing improving conditions
- Temporary employment has been increasing which is a prelude to permanent employment growth
This good news is consistent with an economy bottoming out. However, 2010 will be a transition year and not much of a recovery year. Unemployment will remain subbornly high. We will see a surge in mortgage and other debt defaults. While economic growth might be above average, it will not be enough to quickly get us back to pre-recession levels.