Unemployment Rate Falls to 10% - Duration Still Rising

In another positive sign for the U.S. economy, fewer-than-expected job losses in November and an unexpected drop in the unemployment rate helped send markets opening higher this morning.

The Labor Department has determined that 11,000 jobs were lost due to unemployment in November, considerably fewer than the estimated 125,000 unemployment losses from a survey of economists at Bloomberg News. The unemployment is also far below ADP's figure of 169,000 job losses earlier this week, which had been cited in a blog Wednesday by Zacks senior strategist Dirk van Dijk, CFA.

In addition, overall unemployment reduced from 10.2% to 10% in the month -- a further signal that the U.S. economy is finding its footing.

That's the first word that comes to mind -- wow! The Unemployment Report from the Bureau of Labor Statistics (BLS) was far better than expected, with a loss of only 11,000 jobs in November. Consensus expectations were for the unemployment report to show a loss of 125,000 jobs for the month.

The good news does not stop there. The unemployment rate fell to 10.0% from 10.2% last month, and is half-way back to the 9.8% level in September, according to the employment report.

In addition to the lower-than-expected unemployment report numbers for November, the job losses for both October and September were revised sharply lower. As of last month we thought we had shed 190,000 jobs in October -- now we find out that the economy "just" lost 111,000. We thought we had lost 219,000 in September, but now we find out that we "only" lost 139,000 jobs.

Add the revisions to the surprise for this month to the unemployment report, and we have 273,000 more jobs in the economy than we thought we had as of last night. That is very good news. In addition, as the first graph below (from http://www.calculatedriskblog.com/) shows, the year-over-year percent change in job losses has finally turned the corner. As you can see from the graph, historically that has been a very significant inflection point.

However, even though the unemployment report indicates we have turned the corner, the year-over-year change in unemployment is still further in negative territory than at the nadir of any previous recession. On just about every metric, this is a massive improvement from where we have been. However, we should not forget that this is a cumulative damage-type situation. While we only lost 11,000 jobs over the last month, over the last year we have lost 4.759 million jobs.

The unemployment report showed that improvement in the unemployment rate was widespread, with a decline in every major demographic group. For adult men, there are now 10.5% unemployed, down from 10.7% last month; the rate for adult women also dropped by 0.2 points, to 7.9% from 8.1%. For teens, the unemployment rate fell to 26.7% from 27.6% last month. The unemployment rate for whites fell to 9.3% from 9.5%, the rate for Blacks fell to 15.6% from 15.7% and for Hispanics it dropped to 12.7% from 13.1%. While the unemployment report direction is encouraging, the rates are still far too high, and for every major group except Hispanics (unchanged) the unemployment rate is still higher than it was in September.

The biggest improvement the unemployment report illustrated is happening at the low end of the skill scale. The rate for people who never finished high school fell to 15.0% from 15.5%, the rate for people with just a high school diploma fell to 10.4% from 11.2%. On the other hand, people who had an associates degree, or who went to college and never finished was unchanged at 9.0% and the unemployment rate for people with a Bachelors degree or more actually rose to 4.9% from 4.7%, according to the employment report. Still, the more than 3:1 ratio between college grads and high school drop-outs still points to the value of education, particularly in economic downturns.

Some of the more deep-in-the-weeds internal numbers in the unemployment report were also much better than expected. The most important of these was that the average work week lengthened to 33.2 hours from 33.0 hours. While 0.2 hours a week (aka 12 minutes) might not sound all that significant, there were still 131 million non-farm jobs in the country. That extra 12 minutes a week translates to the equivalent of an extra 794,000 more jobs in the economy (33.2/33.0 * 131 million).

In manufacturing, the average work week rose even more, to 40.4 hours from 40.1 hours. The average amount of over time rose to 3.4 hours from 3.3 hours. That rise came on top of a increase from 3.0 hours in September.

This is a very encouraging sign looking forward. When business starts to pick up, the first thing a company is going to do is restore the hours of its existing workers that have been cut back. It is good for morale, and it is also less costly to the employer than hiring new people since most of the time-benefit costs are the same if a worker is working 35 hours a week instead of 33 hours a week. The cost savings are much smaller, and sometimes non-existent if the employer has to pay time and a half overtime rates. So if employers see that they are having to pay a lot of overtime, it is time to start posting positions on Monster (MWW - Analyst Report).

In another positive development for unemployment, the number of people who were working part-time for economic reasons fell by 34,000 to 9.194 million. The sharp rise in these people who want to work full time, but have had their hours cut back, or the only job they could find was a part-time one, has been particularly sharp in this Great Recession relative to past downturns, as is shown in the second graph below (also from http://www.calculatedriskblog.com/). When the recession started, there were about 4.5 million such people, and a year ago there were 7.2 million. While the improvement on that front might be a false dawn, it is also an encouraging sign.

The other thing an employer will do when business picks up is that they will start hiring temps. In November, temp jobs expanded by 52,000. This is the third straight month in a row that temp jobs have increased, rising by 44,000 in October and by 17,000 in September. Once employers become convinced that the upturn is real, they start wanting to have their own employees on the job, rather than handing over a big chunk of their payroll to Kelly Services (KELYA - Snapshot Report). While temp jobs might not be people’s dream job -- and the improvement on that front might explain why the drop in the unemployment rate was so much more dramatic at the low end of the education spectrum in November -- temp jobs are a (good) canary in the coal mine for the economy.

The much slower rate of unemployment from the employer survey is more significant than the drop in the unemployment rate that comes from the household survey, but there is not much conflict between them. Part of the reason for the drop in the unemployment rate is that the civilian participation rate dropped to 65.0% from 65.1% in October and 65.2% in September. In other words, fewer people were unemployed, not because they found new jobs, but because they dropped out of the work force.

The participation rate will of course never hit 100% -- toddlers are not expected to be looking for work, and people do retire as well. Over time, the current demographics suggest that the participation rate should be declining as the Baby Boomers start to retire. However, one way or another it is people with jobs that have to support those who are not working. Sometimes it is wanted support, as in parents supporting children, sometimes it is less welcome support, as in taxpayers paying out unemployment benefits to the jobless.

Perhaps a more important number than the unemployment rate is the employment rate or the percentage of the population that is actually bringing home a paycheck. Technically, this is referred to as the employment population ratio, and is shown in the third of the graphs from http://www.calculatedriskblog.com. In November, the employment rate was unchanged at 58.5%. The stabilization of this is a very welcome development since the employment rate has been in an absolute free-fall during this recession. Note that there was a secular rise in it, underlying very distinct cyclical effects, from the early 1960s through the end of the century. The last economic expansion, though, did not come close to bringing it back to its prior peak.

The secular rise in the unemployment ratio was particularly sharp during the late 1970’s and most of the 1980’s. This was due to two powerful demographic forces. First, the Baby Boomers started to enter the work force (people tend to get their first jobs between ages 18 and 24, depending on how much education they get; the "baby boom" started in 1946, so the first effects would have been seen in the mid-1960’s).

The second demographic force was that back in the 1960’s, if you saw a magazine article about women and labor, the odds were that is was about childbirth. Clearly that is no longer the case today, although the employment ratio for adult women is still well below that of adult men. In November it was 55.6% for women vs, 66.7% for adult men. A year ago the rate was 70.5% from men and 57.6% for women.

The decline in the unemployment ratio over the last year shows that this downturn has been much tougher on men than on women. Probably because, on average, women earn less than men, so employers would rather get rid of the high-cost help first. However, a bigger part is probably due to the mix of industries that men and women tend to work in. Men still dominate goods producing industries like manufacturing and construction much more than they dominate the service sector, and particularly the one area that has seen consistent growth in employment during this downturn, health care.

The participation rate has also fallen for both genders over the last year, from 75.6% for men to 74.6% now, and from 61.0% from women a year ago to 60.4% now. The difference between the participation rate and the employment rate roughly corresponds to the unemployment rate. It is noteworthy that not only has the unemployment rate for men risen much more than for women in this downturn, a greater proportion of men have also decided to leave the workforce altogether.

While the unemployment data in the report were generally much better than expected and offer substantial cause for hope, we should not lose sight of just how nasty this recession has been and just how much work is left to be done to repair the damage. The employment losses as a percentage of peak pre-recession levels for every recession since WWII. The only other recession that has come close to the Great Recession in terms of depth of downturn is the 1948 recession as the economy struggled to demobilized from the WWII effort. But that was simply an unpleasant side effect of a generally good situation.

There is nothing comparable going on this time around. Indeed, by this point after the peak, back in 1948, not only had the economy stopped losing jobs, but it had regained everyone that it had lost an then some. In fact, every postwar recession but four (and next month it will be three) had already fully regained all the jobs lost by 23 months after peak employment. The three that will remain after next month are a cause for concern, since they are the last three downturns -- and each has taken successively longer to get back to pre-recession peak levels.

In the case of the 2001 recession it took 46 months, or almost four whole years to get back to a new high in terms of total jobs. That was up from 30 months in the 1990 recession of unemployment. Both of those downturns were extremely mild in terms of the depth of the downturn. This suggests that we will be extremely lucky if the total number of jobs in the economy surpasses the November 2007 peak by October 2011.