Don't be misled by the drop in the Unemployment Rate

November Labor report

The November labor report indicated that employers added 120,000 jobs in the month of November.  The unemployment rate, measured by the household survey, dropped to 8.6% from 9%.    Superficially, this seems like evidence of an economic expansion, or at the least the beginning of one.  However, the leading indicators within the labor report indicate continued stability in the US labor force with no indications of either a meaningful improvement or a recession.   

The “break even” number for maintaining the unemployment rate is generally accepted to be 150,000 new jobs.  How could a mere 120,000 new jobs cause a four-tenths of one percent decline in the unemployment rate?  The easy answer is “technical factors” which are discussed below.  The key leading indicators within the report do not show signs of an improving labor market.

For reasons discussed below the best that can be said about this report is that it indicates stability in the labor force.  It is NOT indicative of a “double dip” recession, but it is also NOT indicative of an incipient expansion.  Within the report the “leading indicators” were all flat, not up.  All of these indicators show a labor market that has improved since the end of the recession but is now largely stable with perhaps a tiny bias toward improvement.  A focus on the improvement in the unemployment rate, to 8.6%, would be misleading.

Some key elements of the report.

1.     Hours worked for production workers declined slightly to 33.6 from 33.7.  This is a leading indicator of an improving economy and labor market.   In an expanding economy, employers respond to increased demand by lengthening the hours worked in the initial phase since uncertainty is high and the cost in overtime is justified by flexibility.  Normally, production work is the most variable part of the work force and is very sensitive to changes in demand.  I wouldn’t attribute any significance to the small decline but merely point out that this is NOT indicating further expansion.

2.     The average hours worked for all employees remained flat at 34.3 hours, the same as in September and October.  Generally this is a leading indicator of a labor market improvement.  Similarly Overtime hours remained flat compared to October at 4.1.  Overtime is also an immediate source of flexibility for employers in responding to increases in demand. 

3.     The number of Temporary workers (seasonally adjusted) increased 1% in November compared to October and 7.8% compared to prior year.  In October, the increase over prior year was 8.1%.  Year over year gains have moderated throughout 2011. 

4.     The penetration rate of Temporary workers increased slightly to 1.77% of the workforce from 1.76% in October.  The penetration rate is highly influenced by production workers.  The rate has been improving since the bottom of the recession.  This small improvement is consistent with the trend of a stable labor market. 

5.      Unemployment among college graduates remained flat at 4.4% compared to October, although a significant improvement from the 5.1% of prior year.  This is a key statistic.  It is partly a leading indicator since these are the most employable people in the labor force and partly because purchasing power is highest in this group. 

These are important signals in the November Labor report that there is little risk of a double dip recession, which was greatly feared as recently as October.  In previous recessions, these indicators turned down for many months before a recession was evident.  On the other hand there is little evidence, in this data, of the economy entering into an expansionary phase strong enough to reduce the record unemployment rate. 

The most encouraging data, the decline in the unemployment rate, is misleading as an indication of future activity.  In fact, the unemployment rate is likely to increase in coming months.  The decline in the unemployment rate was a function of two factors in the household survey that is used to measure it.  First, the reduction in the total labor force in the household survey accounted for about half of the change in the unemployment rate or 0.2%.  This reduction is due to discouraged workers dropping out of the workforce.  Hardly a good sign; and perhaps influenced by seasonal factors.   Second, the household survey contained a larger increase in the number of people employed compared to the establishment survey; this accounted for about half of the improvement or 0.2%.  For many reasons the Household survey and the Establishment survey frequently differ; in this environment, the Household survey is probably a better view of new job formation, so let assume the 0.2% is real.  Nevertheless, the nominal 0.4% improvement in the unemployment rate is unlikely to be sustainable.  As the labor market recovers, more individuals will be drawn back into the work force, which will, paradoxically, cause the unemployment rate to increase.