In February, the unemployment rate held steady at 9.7% as 36,000 jobs were lost. The consensus expectations were for the unemployment rate to climb to 9.8% and for 65,000 jobs to be lost.
Given the massive blizzards that blanketed major parts of the country during the survey week, the “whisper numbers” had been much more dismal. The revisions to jobs in December and January were a net positive of 35,000, with the new estimate for jobs lost in December falling to 109,000 from the previous estimate of 150,000 while January headed the other direction with the estimate increasing to 26,000 jobs lost from the original estimate of 20,000.
Of course, those numbers still indicate that the economy is losing jobs, just at a slower rate than we thought. Still, consider where we were a year ago, when the economy lost 726,000 jobs in a month — and that was a big improvement over the previous month!
As the first graph below (from http://www.calculatedriskblog.com/) shows, while the year-over-year rate of change in employment is still deeply negative — right on par with where it was at the worst points in both the Nixon/Ford recession and the Reagan recession — it is only about half as bad as its worst point this summer.
The unemployment rate appears to have peaked, but it is still far too early to know if that will prove to be the case. Typically the unemployment rate continues to rise even after the economy is adding jobs on a monthly basis as discouraged workers come back into the labor force.
Even though we have greatly slowed the speed of the hole we are digging, it is still an extremely deep hole, indeed the deepest whole since the Great Depression. Since this Great Recession started back in December of 2007, the economy has lost a total of 8.4 million jobs and brought the total number of unemployed up to 14.9 million.
In addition, when the number of underemployed people — such as those who want to work full time but have seen their hours cut back, or the only job they can find is a part time one — are considered, the rate (U-6) is 16.8%, which is up from the 16.5% rate in January, but still below the 17.3% peak in December.
This recession really has been different that anything that has gone before it (post WWII) it in terms of both depth and duration. Consider the second graph (also from http://www.calculatedriskblog.com/). It measures the cumulative jobs lost in each downturn as a percentage of starting employment (thus it controls for population growth).
Not only have the job losses been almost a full percentage point worse than the previous worst recession (1948), but by the 26th month after the recession had started not only had jobs stopped going away, in all but three of the cases (incidentally the last three cases) we had already recovered to the point where there were more total jobs in the country than when the recession started.
The last two recessions — the ones starting in 1991 and in 2001 — both had very long periods of a “jobless recovery” but were relatively mild affairs in terms of total job losses. Still, it took us almost four years from the start of the 2001 recession until we had more jobs than when the recession started.
Assume a very optimistic case, where starting next month we began to add jobs at a rate of 250,000 a month, which would roughly match the rate of job growth during the Clinton Administration (actually slightly higher), it would still take us almost 34 months until we made up the 8.4 million jobs that have already been lost. That means we would not be starting an actual expansion, rather than just a recovery, until December of 2013.
Beyond those headline figures, the internals of the report were very solid (it says a lot about our current situation when 9.7% of the workforce being unemployed can really be called solid, but at least it is generally moving in the right direction, even if the levels are still abysmal). The flatness of the unemployment rate was not due to people leaving the workforce.
The civilian participation rate, or the percentage of people both employed and unemployed in the workforce, actually rose for the second month in a row, to 64.8% from 64.7% in January and 64.6% in December. However, that is still well below where we were a year ago at 60.3%.
The percentage of people who are actually employed, or the employment rate (a very under-reported and significant figure) also ticked up for the second month in a row, reaching 58.5% from 58.4% in January and 58.2% in December (a little hard to see in a 60-year graph, however). Both measures were in a steady secular uptrend starting in the early 1960’s. That was driven by two very powerful demographic forces: the entrance of Baby Boomers into the workforce and women entering the world of paid employment in massive numbers.
Obviously the employment rate is far more volatile and cyclical than the participation rate. The ratio between the two corresponds to the unemployment rate. The participation rate topped out April of 2000 at 67.3%; after the 2001 recession it never really recovered the way it did following previous downturns.
While much more stable than the employment rate, there is a cyclical dimension to the participation rate in the form of discouraged workers. It was able to edge back up to as high as 66.4% in December of 2006, and when the recession started in December of 2007, it was still at 66.0%. Then it really started to fall off a cliff, and at its low point in December it was back at a level last seen in August of 1985.
The employment rate followed a similar but much more volatile path, rising from 55.0% in the early 1960’s and eventually hitting a high point of 64.7% in April 2000. It dropped sharply during the 2001 recession, (hitting a low of 62.0% in September 2003) but it always falls sharply in recessions.
The difference in the subsequent recovery was that it was very feeble, and did not come close to hitting a new high as it had following the recessions in 1960, 1969, 1973, 1981 (the recovery following the 1980 recession was too brief to do it before we double-dipped) and 1991. The highest it got was 63.4% in December 2006 before it started to fade again.
When the recession started in December of 2007, it had already fallen to 62.7%. Then it really started to fall off a cliff. At its low point in December, it was at the same level it was at in August 1983 and March 1974 (both also recessionary periods) and just above where it was (58.1%) in August 1969.
Think about that for a minute. We had the same percentage of people employed in December of 2009 as we did in August of 1969. In 1969 if you found a newspaper article where the words women and labor were used in the same paragraph, you were almost certainly reading about childbirth!
Will this uptick in both the participation rate and the employment rate last? Too soon to tell, but the fact that they are both up is a very positive sign. It is NOT the case that the unemployment rate held steady simply because people were leaving the workforce.
With the unemployment rate flat overall, one would expect mixed results when we look at the major demographic groups, and that is what happened. The unemployment rate for adult men was unchanged at 10.0%, but down from the 10.2% rate in December, though still well above the 8.4% rate of a year ago. The rate for adult women ticked up to 8.0% from 7.9% but was below the 8.2% rate in December, though well above the 6.8% rate a year ago.
This recession has been particularly brutal for men, in part because they are over-represented in some of the industries like construction that have been hit the hardest. There was a solid improvement in the unemployment rate for teens, but it is still abysmally high at 25.0%. Still, that is a nice improvement from the 26.4% rate in January and the 27.1% rate in December, though a year ago it was 21.8%.
But having a job is not as critical for most teens as it is for adults. For most, that money goes towards putting gas in the car and new clothes from Abercrombie and Fitch (ANF), rather than for paying the electric bill, the mortgage or groceries from Kroger’s (KR). Those jobs do teach basic work skills, however, and for many, those jobs also help put them through college.
Breaking things down by race, whites saw a tick up in the unemployment rate to 8.8% from 8.7% last month, but it remains below the 9.0% rate in December. It is also well below the national average. A year ago the unemployment rate for whites was 7.5%.
The rate for blacks, on the other hand, showed a big drop this month to 15.8% from 16.5% in January. Recessions always seem to hit blacks harder than whites. A year ago the rate was 13.5%. The rate for Hispanics ticked down to 12.4% from 12.6% in January and 12.9% in December. A year ago, the unemployment rate for Hispanics was 11.0%.
The good advice to stay in school is never more true than in a recession. The unemployment rate for high school dropouts (over 25 years old) was 15.6% in February, up from 15.2% in January and 13.0% a year ago. Just getting a high school diploma or GED drops your unemployment rate down to 10.5%, although that too rose from 10.1% last month and 8.4% a year ago.
Those who went on to college but didn’t finish or stopped with an associates degree saw the biggest improvement in their unemployment rate this month, falling to 8.0% from 8.5%. In December it was at 9.0%. A year ago it was at 7.1%. Those with a four-year degree (or more) saw the unemployment rate tick up to 5.0% from 4.9%, and is up from 4.2% a year ago. Still, even in the depths of the Great Recession the unemployment rate for college grads is about where the national rate is during periods of full employment.
Private Sector vs. Government Jobs
So where are the jobs coming from and where are they going away? The private sector only lost 18,000 jobs — half of the total — while the other half lost were government jobs. That is a bit of a surprise given the government is in the process of hiring lots of people for the Census. However, the 15,000 new census jobs were offset by lay-offs in the postal service and probably also by job losses at the state and local level.
States and localities are not allowed to run operating budget deficits, and the recession has caused tax revenues to implode. Raising taxes is neither popular or good economics at this time, so they have to reduce spending and the bulk of most state and local spending is for salaries and other direct employment costs. The Stimulus Act (the ARRA) only partially helped fill that hole, even though aid to the states made up almost a third of the total. Most states have already run through their “rainy day” funds. Look for employment to continue to fall at the state and local level.
Within the private sector, the goods producing sector continues to be hit hard, with a decline of 60,000 jobs. However, more than all of that was in construction. Construction is one of the industries that is most likely to be affected by the weather, but it is also one that has been absolutely devastated by this downturn.
In February, 64,000 more construction workers lost their jobs on top or the 77,000 who got pink slips in January. Since the recession started, 1.9 million construction workers, or 22.6% of the 8.4 million total, have lost their jobs. It’s not like the sector was that huge to begin with: when the recession started in December 2007, total employment in construction was just 7.491 million. Now there are just 5.555 million construction jobs, a decline of 28.8%.
In contrast, manufacturing actually gained 1,000 jobs, on top of 20,000 gained in January. This is sort of impressive since manufacturing employment has been in a long-term secular downtrend for decades now. As the fourth (Brown) graph shows, the absolute number of people working in manufacturing of both durable goods (green line) and non-durable goods (red line) is far below the levels of 1969, and showed no improvement at all during the last expansion (and for non-durable goods in the last two expansions). The graph also illustrates just how hard construction workers have been hit during this downturn.
The private sector actually added 42,000 jobs in the service sector. More than all of those (net) jobs, though, came from temporary help services like Manpower (MAN) and Kelly Services (KELYA).
Temp jobs increased by 47,500 in February. While those are not the best-paying or most desirable jobs in the country, they are a very important positive omen. The rise in temp jobs in February comes on top of gains of 50,200 in January and 49,700 in December.
Why is this a good omen? Because when businesses start to see things picking up, the first thing they are going to do is increase the hours of their existing workforce. This is particularly true if those hours had been cut back on earlier in the recession. However, they might not be confident enough to bring on new full-time employees, particularly if they pay benefits as well. So the first step is to call up Manpower and say send someone over.
Only when businesses feel sure that the increase in activity is going to last will they think, “It is crazy to be paying $20 an hour for an employee who is only earning $12 an hour (the difference going to the employment agency). Why don’t I bring someone on full-time and pay them $13 an hour and we will both be better off?”
The pink graph below does not break out temp jobs separately, but they are a major part of the professional and business services group (blue line). Note how that group traditionally weakens as we head into a recession and then recovers coming out. By contrast, most of the other major service sector industries, with the exception of retail (gold line), tend to be quite recession resistant — or at least had been up until the Great Recession.
Also note the scale of the pink graph and contrast it to the brown graph. We now have more people working in leisure and hospitality (i.e. hotels and restaurants) than we do in all manufacturing, both durable and non-durable. Also notice that the biggest service sector is education and health, and that sector has continued to add workers throughout the recession.
In 1969, we had more people working in non-durable manufacturing than we did in education and health AND financial activities combined. Now total manufacturing employees, both durable and non-durable, are less than 60% of the total for education and health alone!
The one fly in the ointment of the leading indicators within the employment report was the average work week, which ticked down to 33.8 hours from 33.9, the same level it had been in December. Normally I would consider this to be a troubling sign. However, this could easily be explained by the weather this month. But if it rises again next month, it would be a very bad thing.
I will post a separate blog later on the duration of unemployment numbers, which have been particularly troubling in this recession, but we did get a bit of good news on that front (at least in terms of direction, the levels are still awful) as the average duration of unemployment dropped to 29.7 weeks from 30.2 weeks in January, but up from 20.0 weeks a year ago. The median duration (which is always lower than the average) fell for the second month in a row, falling to 19.4 weeks from 19.9 in January and 20.5 weeks (an all-time off-the-charts record) in December, but remains far above the 11.4 week level of a year ago.
All things considered, this was a very good employment report. There is still a lot of work to be done, but the indications are that things are starting to get better. But it is likely to be a very long, slow and painful progress.
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.
