In May, the economy added a total of 431,000 net new jobs and the unemployment rate fell to 9.7% from 9.9%. Good news, right? WRONG.
Only 41,000 of those jobs were from the private sector, a sharp slowdown from the 218,000 added in April and well below the 158,000 private sector jobs added in March. Almost all of the net job growth came from the addition of 411,000 temporary workers for the Census. While those are real jobs, and provide real pay checks, and are needed to fulfill the requirements of the U.S. Constitution, they are only going to last a few months, and those people will be back looking for new jobs before the leaves start to fall.
As for the decline in the unemployment rate, that is mostly due to a decline in the civilian participation rate, or the percentage of people who are either working or looking for work. While that rate will never get close to 100%, unless we want to do away with the idea of retirement entirely and totally get rid of all child labor laws, the rate fell to 65.0% in May from 65.2% in April and is down from 65.8% a year ago.
In the last cycle, the participation rate peaked at 66.4% and the all-time high was in April of 2000 at 67.3%. The percentage of the population that is actually employed, the employment-to-population ratio, or the “employment rate,” also fell — to 58.7% from 58.8%, and is well above the year-ago level of 59.6%.
Like the participation rate, it never fully recovered from the 2001 recession. The unemployment rate is the difference (well, technically the ratio) between the participation rate and the employment rate. The participation rate is generally much more stable than the employment rate, and that shows up as changes in the unemployment rate.
The first graph below shows the history of these two data series back to 1948, along with the unemployment rate (green, right scale). The decline in both of these numbers reverses a very promising upward trend (after a sickening decline) that started late last year.
The unemployment rate gets all the headlines, but it is important to understand the dynamics behind it. A rising participation rate makes it harder to keep the unemployment rate down, while a falling participation rate makes it relatively easy to show a low unemployment rate.
From a longer-term perspective, the secular rise in the participation rate that started in the mid-1960’s and lasted though the end of the 20th century was due to two demographic factors. First was the entrance of the Baby Boomers into the workforce (the leading edge of whom turned 20 in 1966), and secondly by the integration of women into the workforce.
In the mid-1960’s if you saw an article that mentioned women and labor in the same paragraph, it was a good bet the article was about childbirth. Now the number of women working almost equals that of men (63.51 million jobs held by women, 65.78 million jobs held by men), although the participation rate is still significantly higher for men than it is for women at 74.5% versus 60.6% for women. The employment rate is also higher for men at 67.2% for men and 55.6% for women. Women over 20 years old outnumber men in the civilian workforce by 114.2 million to 106.4 million.
Breakdown by Demographics
This downturn has hit men harder than it has hit women. The unemployment rate for men fell to 9.8% from 10.1% in April and is back to where it was a year ago. The rate for women was 8.1%, down from 8.2% in April, but up from 7.5% a year ago.
Regardless of gender it is easier to find work as an adult than as a teen. The teenage unemployment rate rose to 26.4% in May from 25.4% in April and 23.2% a year ago. Summer jobs are going to be exceptionally hard to find this year. While generally teen jobs mostly generate discretionary income for teens to spend at the malls, rather than income that is needed to pay the mortgage and the electric bill, they are often important to funding college educations. They also teach important job skills that will be needed in the future.
A summer jobs program should be passed as soon as possible. It would be a very good idea to effectively restart the Civilian Conservation Corps, which proved to be a very effective program in the 1930’s, and put the teens to work cleaning up the mess on the Gulf Coast this summer. It would not add to the deficit, since the government could send the invoice to BP (BP).
The unemployment rate did decline across ethnic groups, with the biggest improvement among blacks, where the rate fell to 15.5% from 16.5% in April, but still up from 15.0% a year ago. The rate for blacks still far exceeds that for whites, where the rate fell to 8.8% from 9.0%, but up from 8.6% a year ago.
Latinos saw the smallest decline with their unemployment rate falling to 12.4% from 12.5%, and actually down from 12.7% last year. The rate for black men has been particularly high in this downturn, but there was significant improvement in May, with the rate falling to 17.1% from 18.0% last month but up from 16.7% a year ago.
The rate from black women also declined sharply, falling to 12.4% from 13.7% last month, but up from 11.3% a year ago. The unemployment rate for black teens is at Depression levels, rising to 38.0% from 37.3% in April but down from 39.9% a year ago. Considering that the participation rate for black teens is just 26.0%, up from 25.3% in April but down from 27.4% a year ago, this unemployment rate should give anyone pause.
The unemployment rate for white men fell to 8.8% from 9.2% last month and 9.0% a year ago. The rate for white women was unchanged at 7.4% on the month but up from 6.9% last year. The white teen unemployment rate was 24.4% in May, up from 23.5% in April and 20.7% a year ago.
Unemployment Duration Still Climbing
While the unemployment rate might be showing some improvement (mostly due to the decline in the overall participation rate) once people lose their job, it has been extremely difficult for them to find a new one. The duration of unemployment continues to increase.
The average unemployed person has now been looking for work for 34.4 weeks now, up from 33.0 weeks in April and 22.9 weeks a year ago. Now, that number includes some people who have been out of work for years now, and that can skew the numbers (since you can’t be out of work for less than zero weeks). However, the median time people have been out of work set yet another new record, rising to 23.2 weeks from 21.6 weeks in April and 14.9 weeks a year ago.
It is not like the year-ago figures were particularly low, either. On both the mean and the median, they were at the time all-time record highs. Prior to the Great Recession, the highest the mean duration of unemployment had ever reached was 21.2 weeks in July of 1983. The previous record on the median was reached in May 1983 at 12.3 weeks. The second graph shows the history of the unemployment duration metrics back to 1967 (the start of the median data).
There are now a total of 6.763 million people in the country that have been out of work for more than 26 weeks, constituting 46.0% of all the unemployed in this country. Keep in mind that 26 weeks is when regular state-based unemployment insurance expires. After that point, people have to move over to extended claims, which are paid for by the federal government and is part of the Stimulus Bill (the ARRA).
Just a year ago, there were “only” 4.030 million long-term unemployed, representing 27.7% of the unemployed. It is the change in long-term unemployment that really makes the difference between the economy doing well and the economy in a deep slump. Over time, the number of short-term unemployed tends to be fairly stable.
The third chart below shows the four duration groups that are tracked by the BLS. These numbers are not adjusted for population growth, so there should be an upward tilt to the numbers over time. Normally, the short-term unemployed (gold line) are the largest group of those looking for work, while long-term unemployed are fairly rare. During recessions and their immediate aftermath, the number of long-term unemployed (blue line) tends to increase dramatically (as does, to a lesser extent, the number of people unemployed between 15 and 26 weeks).
One measure of the intensity of a recession then is the ratio of the number of short-term (under 5 weeks) to long term (over 26 weeks) unemployed. Before this downturn, the highest that ratio had ever hit was 0.78 in March of 1983, right at the tail end of the Reagan Recession. The ratio in May stood at 2.46 — over 3x the previous record high. A year ago the ratio stood at 1.25, or roughly half the current level.
The Effect of Long-Term Unemployment
Being out of work for more than six months is a very different experience than being out of work for less than a month. The latter is sort of like an unplanned vacation. It might not be convenient, but as long as you can get back to work, it is not going to be a disaster. You will in all likelihood find work are the same salary or better, and your savings, credit cards and unemployment insurance can tide you over without causing long-term damage to your financial position. Your skills are still intact, and your Rolodex is still fresh.
Being out of work for more than six months means that those savings have probably already been depleted and the credit cards maxed out, especially if you were expecting to be back at work after a few weeks. If it were not for extended benefits, you would have no income at all.
While the rates vary by state, generally unemployment insurance will pay about 60% of your pre-unemployment income, usually with a cap of about $400 per week. What that blue line represents is the crashing of the long-term financial futures of millions of Americans. While they sit at home and send out resume after unrequited resume, their skills start to deteriorate and their contacts get stale. In short, they become less desirable employment candidates. When they do get a new job it is likely that it will be for a substantially lower income than what they were earning before they got their pink slip.
There is a danger in extending unemployment benefits for too long. If extended indefinitely they would, in effect, become a new welfare system, with all the dependency issues that entails. However, with long-term unemployment at record highs, now is not the time to end it. Those people would be left with no income at all, and no real job prospects.
With no other financial resources left to fall back on, they would fall into abject third-world style poverty. Not only would that be a humanitarian tragedy, it would be very bad news for the rest of the economy. With no income those people would not be able to spend anything, not even on the most basic items at deep discount stores like Wal-Mart (WMT) or Big Lots (BIG).
That would mean less business for those stores, leading to layoffs and yet more unemployed people — who in turn would eventually run out of money. Lather, rinse, repeat. It is for this reason that the non-partisan Congressional Budget Office scores extended unemployment benefits as among the most effective stimulus programs on a jobs-saved-per-dollar-spent basis.
A Few Bright Spots
While the overall report was disappointing, there were a few bright spots that should be noted. First, the average work week increased to 34.2 hours from 34.1 in April and 33.9 a year ago. It was the third straight month of an increasing average work week.
While an additional 6 minutes a week on average does not sound like that big a deal, it is when you consider that there are still 107.6 million private sector workers in the country, and that many people working just a little bit longer can add up to a lot of economic output.
It is also a good leading indicator of future employment trends. When business starts to pick up after a recession, particularly a deep recession like this one, employers tend to be cautious, not sure if it is just a blip or the start of a real turnaround. So to meet the increased demand, the first thing they will do is ask their existing workers to stay on a little later to get the job done. If those employees have already seen their hours cut back during the recession, they will be more than happy to do so.
However, eventually they get tired, remember they have families they want to see, etc. So the next thing the employer will tend to do is call up a temp agency like Kelly Services (KELYA) or Manpower (MAN). This gives them the flexibility to reduce their workforce again quickly if the increase in demand does turn out to be just a temporary blip.
In May there were 31,000 new temporary workers hired, up from 26,600 in April, and a decline of 12,900 temps in May last year. These should not be confused with the temporary census workers. Temp workers are important, not because they are the greatest or highest-paying jobs in the world, but because they are an important leading indicator of permanent employment.
Service Sector Job Creation
Most of the net job creation in the private sector came from the service sector, which increased employment by 37,000 in May, which was a big slowdown from the addition of 156,000 in April but one heck of a lot better than the loss of 109,000 jobs in May of last year.
In addition to the increase in temp jobs (which are all counted as service sector jobs, even if the temp agency sends someone to work on an assembly line) the other big source of jobs in the month was health care — which has been impervious to job losses in this downturn and which added 13,100 jobs, although that was a slowdown from the addition of 23,100 jobs in April.
Those gains were offset by losses in the financial industry (-12,000) and retail (-6,600). The goods producing sector increased jobs by only 4,000 in May, a huge slowdown from the addition of 62,000 jobs in April. The big swing was in construction jobs, which plunged by 35,000, effectively wiping out all the gains in the previous two months.
Construction & Manufacturing
The construction industry has been particularly hard-hit in this downturn. Since December 2007, the peak employment month for the economy as a whole, the construction industry has lost a total of 1.944 million jobs, or 25.8% of all construction jobs as of November 2007. The economy as a whole has lost 7.381 million jobs, or 5.4% of the December 2007 total.
Put another way, the construction industry, represented 5.46% of all jobs when the recession began, but has represented 26.3% of all jobs lost.
Manufacturing jobs have fared better, particularly recently. The economy added 29,000 factor jobs in May on top of 40,000 in April. Factory jobs got hit hard early in the recession, but have been on the way back. That is particularly true of durable goods manufacturing, which added 34,000 jobs in May, while non-durable goods manufacturing lost 5,000 jobs.
Since the start of the recession, durable goods jobs have fallen by 1.532 million of 17.6%, but are up 124,000 from December. Non-durable goods manufacturing was not hit as hard, falling by “only” 534,000 or 10.6% since the recession started, but have not been seeing the same sort of rebound either, with a gain of just 6,000 jobs since the December low.
Of course, manufacturing jobs have been in a steep downtrend for decades, regardless of the overall state of the economy, as is shown in the third graph (which also shows construction employment). Durable goods employment is at the same level it was in 1950, and non-durable goods jobs are even lower today than they were in 1939, when the Great Depression was still going on and the population of the country was less than half of what it is today.
A Bit of Perspective
While this report was not as good as people were expecting going into the report, it is also important to keep in mind just what a huge improvement it is from a year ago, and especially from where we were 18 months ago. Then the economy was in absolute freefall, leading to the deepest recession the country has seen since the Great Depression.
Now employment is starting to recover with five straight months of job gains. The next graph (from www.calculatedriskblog.com) shows just how bad things were relative to previous postwar recessions. Not only was the percentage decline relative to the starting employment levels deeper than any other previous recession (the only one that even came close was in 1948 as the WWII demobilization was underway) but it has lasted much longer.
It has now been 29 months since the recession began. By 29 months after the start of the recession in the past, not only had employment started to recover, but total employment had gone on to make a new record high. The two exceptions to this were the 1990 recession (which made it to full recovery after 30 months) and the 2001 recovery. Both were pretty mild affairs in terms of job losses.
While we have clearly made the turn, both in terms of total jobs (solid red line) or excluding the Census effect (red dotted lines), we still have a long way to go to get back to the total number of jobs we had in the economy at the end of 2007. Also note that it is not particularly unusual for the economy to be in a recession in Census years, and each of those years would have also benefited from government hiring to count people.
So far this year, the economy has added a total of 982,000 jobs (including Census jobs), or a rate of 196,000 per month. At that rate, it would take almost 38 months from today to get back to the total level of jobs at the start of the recession. In other words, the middle of 2013.
The Census jobs are not going to last, let alone be repeated for the next 38 months. The duration of unemployment for the country is just as bad as the record duration of unemployment for the unemployed.
In such an environment, especially with no inflation to speak of, it is absolute folly to be talking about tightening up monetary policy by raising the Fed Funds rate. It is equally misguided to talk about tightening up fiscal policy by cutting spending to create jobs and raising taxes. Longer term, the budget deficit will need to be addressed, but it can wait. Getting the country back to work needs to be Job 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.
