More Breakdown of BLS Numbers - Analyst Blog

EDITOR'S NOTE: We conclude our in-depth analysis of today's employment report from the Bureau of Labor Statistics (BLS) with Part 2.

Participating More Fully

So why is the unemployment rate steady at 9.7% if we are actually adding jobs? In part because the numbers are derived from two different surveys. The unemployment rate numbers come from a survey of individual households, while for the employment numbers, they ask the employer.

The job growth shown by the household survey is actually much stronger than that shown by the establishment survey, with a total of 264,000 jobs created in March on top of 308,000 created in February. Based on the household survey numbers, employment is down 1.38% year over year, or 1.95 million jobs, because a higher proportion of the population is actually in the labor force.

The civilian participation rate  rose to 64.9% from 64.8% in February and 64.7% in January. A year ago it was 65.6%. This is the percentage of people actually in the labor force, either employed or unemployed. Unless we seriously relax out child labor laws, and get rid of the idea of retirement completely, that number is never going to get close to 100%. As the second graph shows, the highest it ever reached was in April 2000 when it hit 67.3%. Prior to that time it had been on a broad secular increase.

The participation rate first went over 60% in December of 1955, but did not stay there, and only moved above on a permanent basis in 1969. While during that secular increase there was a tendency for the participation rate to flatten out or even decline slightly during recessions, it always made a much higher high in the next recovery. That didn’t happen in the last recovery, and the best we could hit was 66.4% in January 2007.

There were two powerful demographic forces behind the secular rise in the participation rate, the Baby Boomers and women entering the workforce. Note that the participation rate started its increase around 1964. The Baby Boom started in 1946, so that meant that the first of them were turning 18 and entering the workforce at that time (unless, of course, they were patrolling rice paddies in the Mekong delta).

The other factor was that back in the 1960’s if you saw an article that had the words women, and the word labor in the same paragraph, the odds were that it was about childbirth. That is clearly no longer the case today, as there are almost as many women (20 years or older) working 63.495 million as there are men working, 70.913 million.

The effect of the rise in the participation rate can also be seen in the first graph. Note how much higher the job growth had to be to have an effect of driving down the unemployment rate. In the late 1970’s, sustained year over year employment growth of over 3% barely budged the unemployment rate, but in the early part of this century year over year employment growth of under 1% managed to shrink the unemployment rate.

Employment vs. Unemployment Rate

The red line on the graph shows the percentage of the population that is actually employed, or the employment rate. The unemployment rate can be seen as the ratio between the participation rate and the employment rate. Like the participation rate it had been on a secular increase since the mid 1960’s, but in a much more volatile way.

After all, the unemployment rate has varied quite a bit over the years. Still, each peak was higher than the previous one.

During the 1950’s and 1960’s you needed an economic boom to get the employment rate up over 57.5%, and we did not get above 58% on a sustained basis until the late 1970’s. The employment rate peaked out in April 2000 at 64.7% and following the 2001 recession we did not come close to setting a new high, only reaching 63.4% in December of 2006.

Then the floor started to fallout on it, and we plunged to a low of 58.2% last December. However, we are working our way back now, with the employment rate increasing now for three straight months to 58.6%. Without a doubt, that is still a very ugly level, about where we were in November 1983.

Still, the key point is that the steadiness in the unemployment rate in recent months is not due to people getting discouraged and dropping out of the work force; quite to the contrary, the overall percentage of people in the workforce has started to rebound, as has the percentage that are actually working. The participation rate and employment rate (also known as the empop ratio) are two very under reported numbers, and should really figure much more prominently in news reports about the employment situation.

Demographic Breakdown

The unemployment rate was not just stable overall, but it was also pretty much unchanged for most of the major demographic groups. The unemployment rate for adult men was unchanged at 10% for the month but up from 8.9% a year ago. For adult women, the rate was also unchanged, but at 8.0%, up for 7.1% a year ago.

The rate for whites was also unchanged at 8.8%, up from 8.0% a year ago. However, conditions did worsen for Teens, Blacks and Latinos. The rate for Teens rose to 26.1% from 25.0% in February, but still below the 26.4% level hit in January, but well above the 22.0% rate a year ago. For Blacks, last month’s improvement was erased as the unemployment rate rose to 16.5% from 15.8%, equal to the rate in January and well above the 13.5% rate of a year ago.

The rate for Latinos ticked up to 12.6% from 12.4% in February, and also matched the January level.  A year ago the Latino unemployment rate was 11.6%. Recessions usually hurt the disadvantaged more than the privileged. Conversely, in a recovery, they tend to make up some ground in terms of employment faster (their unemployment rates are more volatile).
 
The uptick in the unemployment rates for teens, Blacks and Latinos is not a good sign in that regard.  However, there was a major improvement in the unemployment rate of one of the most disadvantaged of groups, those that have less than a high school education, where the rate dropped to 14.5% from 15.6% last month, but up from 13.8% a year ago.

The unemployment rate goes down as the level of education goes up. In March, the rate for people with a high school diploma was 10.8%, but that was up from 10.5% in February and 9.1% a year ago. For those with some college or an associates degree, the rate was 8.2%, up from 8.0% last month and 7.3% a year ago. The unemployment rate for those with a Bachelor’s degree or higher had an unemployment rate of 4.9%, down from 5.0% last month but up from 4.4% a year ago.

There is a much bigger gender divide in the unemployment rate among blacks than among whites. The rate for black men rose to 19.0% from 17.8% last month and 15.6% a year ago. The rate among black women was much lower at 12.4%, which was up from 12.1% last month and 10.1% a year ago.

Among whites, men were still more likely to be unemployed that women, but the rate was only 8.9%, down from 9.0% last month and up from 8.1% a year ago.  For white women, the unemployment rate held steady at 7.3% and is up from 6.5% a year ago.

The racial divide in unemployment among teens is huge, but shank a bit last month. The black teen unemployment rate is still at levels one associates with third-world countries, at 41.1%, but that was down for the second month in a row falling from 42.0% in February and 43.8% in January, although up from 33.1% a year ago.  Among white teens the rate rose to 23.7%, up from 22.5% in February, but only marginally higher than the 23.5% rate in January and up from 20.3% a year ago.

Digging Out from a Very Deep Hole

While this report is generally positive, we should not forget just how much damage this recession has done, and just how hard it has hit. As the graph below shows, this has been by far the worst downturn in terms of employment since WWII. Yes the unemployment rate rose higher during the Reagan recession of 1981 through 1983, but that was in the face of the demographic headwinds discussed above.

The third graph shows the jobs lost in each downturn as a percentage of the employment level at the start of the downturn. At the worst point of that recession, total employment was just a little bit over 3% less than when the recession started. In this recession we have lost over 6% of the total number of jobs we had in December of 2007 before the downturn started.

Furthermore, it has now been 27 months since the downturn started. By that point in the Reagan recession, not only had jobs turned up convincingly, but had actually exceeded the previous high. The economy was not just in a recovery, but an expansion. The same is true for every single recession since WWII, with the exception of the 1990 and the 2001 recessions, and we only have a few more months before we pass the full recovery point of the 1990 downturn.

The 2001 downturn took an extremely long time to get back to the full recovery point -- almost four years after the downturn started. That, however, was a relatively shallow downturn, with total employment down just 2.0% at its worst point. Even with very optimistic assumptions it is going to take us even longer to get back to the full recovery point this time around.

There are now 8.2 million fewer jobs than there were when the downturn started. Suppose we were to continue to add jobs at the rate we did in March, 162,000 (remember that includes temporary census jobs created). In that case, it would take us 50.3 months from now to get back to late 2007 employment levels. That would be May of 2014.

Maybe that’s too pessimistic. The economy was able to add 23 million jobs under Clinton, or an average of 240,000 per month. At that rate, we would be back at peak rates in just 34 months, or January 2013. In the meantime, though, the population is growing, so unless the civilian participation rate starts to fall again, we will still be dealing with a relatively high unemployment rate, even after we get back to the previous high in terms of total jobs in the economy.

Unemployment Duration

If there is one thing that surpasses the overall duration of this downturn in terms of employment, it is unemployment duration. This time around, if you get a pink slip, you are likely to be out of work for a very, very long time.

The next chart shows the average and median duration of unemployment. Since you can’t be unemployed for less than 0 weeks, the average will always be higher than the median. However, both measures are simply off the charts.

On average, if someone is out of work, they have been out of work for 31.2 weeks. Half of all the unemployed have been out of work for 20.0 weeks or more.

The government breaks down the unemployed into four groups, based on how long they have been out of work. Short-term unemployment (less than 5 weeks) is actually pretty stable over time, regardless of the state of the economy.

What really separates good times from bad is long-term unemployment, particularly joblessness of over 26 weeks. That is the point at which regular state paid unemployment benefits run out. In March, 6.547 million people had been out of work for more than 26 weeks, up from 6.133 million in February.  In percentage terms, they now represent 44.1% of all the unemployed, up from 40.9% in February and just 24.6% a year ago.

The ratio of short-term to long-term unemployed rose to 2.47, an all-time record. Prior to this downturn, the highest that ratio had ever reached was 0.72, towards the end of the Reagan recession. If it were not for the extended unemployment benefits in the ARRA, those people would have been left absolutely destitute.

In Summation

Overall, this was a very positive report, even if the top-line numbers were less than had been whispered going into the report. The key leading indicators of the average workweek and temporary employment were positive. The shortfall from expectations was entirely from fewer census workers being hired than expected.

There is still a constitutional mandate to do a full counting of the population, and the Obama administration clearly has a political interest in having as full a count as possible, so that just means that there are still many more census jobs to come. The job increases were widespread by industry, even including the hard-hit construction industry, although that was probably weather affected. Even more importantly, we saw an actual increase in the participation rate and the employment rate.

All that being said, it was not a perfect report. The unemployment duration numbers not only remain deeply troubling, but are getting worse again. The damage that was done is very, very severe, and is going to take a very long time to repair.

There are many groups that are not yet seeing any real recovery. Almost one in five black men are out of work, and black teen unemployment rates are at levels that one associates with Africa, not African Americans. Yes, the recovery is starting to take hold, but this is no time to be complacent about it.

The Federal Reserve still needs to keep interest rates near zero, and we still need more from the Federal government in terms of stimulus to get people working again. The steps taken so far have helped, but the problem still remains, and is a big one.

Dirk van Dijk, CFA is the Chief Equity Strategist for Zacks.com. With more than 25 years investment experience he has become a popular commentator appearing in the Wall Street Journal and on CNBC. Dirk is also the Editor in charge of the market beating Zacks Strategic Investor service.

More about Zacks Strategic Investor >>

Zacks Investment Research