December unemployment situation, better, not best

Good morning everyone.  I hope that this post finds each of you doing well!

The Department of Labor announced just a short while ago the following data about the U.S. unemployment situation:

• Private sector jobs up 113,000 vs. expectations of 150,000 jobs to be created
• Unemployment rate down to 9.4% vs. an expected rate of 9.7%
• November private sector jobs revised up to 71,000 from 39,000

Analysis: I’m going to take these data in order.  That the private sector added 113,000 jobs in December is a good thing.  However, what is not a good thing is that the number missed expectations by a significant amount, 24.7%.  Economists also conduct their own surveys to derive their data.  This data is notoriously prone to revisions.

As a financial analysis tip, note that all of the data I am talking about are round numbers – 113k, 150k, 39k, 71k – I am not rounding the data, the data reporters are.  That is always a sign of estimated data.  Therefore, these data always have to be taken cautiously.  I tend to think of data like this as a “coming into focus” situation.  The more the data analyzers look at it, the more into focus the data will become.  But how long does that take?  Many months.

That second piece of data, the unemployment rate falling to 9.4%, the lowest level in 19 months, is evidencing a conundrum I have talked about on the blog before.  How can it be that the same economists who expected 150,000 jobs to be created also only expected a drop in the unemployment rate to 9.7%, whereas the actual data show fewer jobs added, but a more sizable drop in the unemployment rate?  At first glance these statistics would seem to contradict one another and be impossible.  Here’s how this happens…

First, the above data are gathered from two separate surveys.  So the surveys don’t necessarily have to coordinate in a mathematical, causality sort of way.

Second, the unemployment rate only counts those folks who are looking for work.  Thus, if a job seeker becomes very discouraged and stops looking she or he is excluded from the data.  That is clearly one of the things that has happened here that helps to reconcile the conundrum.  But those unemployed folks will eventually start looking again at some point in the future; after all, they have to eat!  When those folks start looking for work again they will add pressure to the unemployment rate figures all over again.  So this data, the unemployment rate, is something that will be difficult to bring down and to improve.

Let’s summarize what we have so far.  One statistic, the number of jobs added, missed expectations; a bad thing.  The second statistic, the unemployment rate, also is demonstrating a bad thing, people are dropping out of the job market entirely.  But what about that third statistic, the November revision?

November’s revision is the most important component of what was reported today.  The reason is that, when combined with the trajectory of other jobs data revisions lately, it clearly indicates an upward trend.  When all of the revisions are upwards it means that surveys are consistently under-counting.  It also is a strong indication that the job adds are happening at smaller businesses that are slower to report and likely missed by the big surveys.  That means the small “mom and pop” businesses are hiring.  And that typically means retail-type businesses.  And (still with me?) that means that consumers are spending more money.  This is the silver lining in a disappointing report.

Importance grade: 8; I am lowering the importance grade on these data because it is my intuitive sense that the labor market is not only mostly stabilized, but improving.  That is, the magnitude of job creation (103,000 jobs added) may be less than expected, but it is still strongly positive.  Second, all of the revisions lately are upward – a very strong sign.  So the jobs data is becoming less important – a very good thing.

Jason