The U.S. Bureau of Labor Statistics released new employment data, with the BLS reporting 225,000 new jobs created, according to the Establishment Survey. A survey of economists by Bloomberg had predicted an increase of 180,000 jobs, so this report beats expectations. The press, not surprisingly, is gushing over the good economic news.
To get a better sense of the jobs situation in context, though, we need to look beyond the headline data and delve more deeply into what the BLS is reporting.
Moreover, it's important to look beyond the government seasonal adjustments which can be employed to massage the numbers even more than the usually-employed methods.
So, a look at the non-seasonally-adjusted unemployment rate shows an unemployment rate of 5.1 percent for July 2016. That's compared to 5.6 percent for last year.
There's a little bit of an improvement there, but the unemployment rate is a function of both the number of people who say they're in the labor force, and the number of people who say they want jobs and have them.
So, we must look beyond the unemployment rate since changes in the size of the work force can push down the unemployment rate without any real gains in total employment.
For example, in spite of a falling unemployment rate, if we look at the actual number of unemployed people, we find that little has changed over the past year:
During July of this year, there were about 8.1 million unemployed people. 12 months earlier, during August 2015, the number was pretty much the same — and it's been hovering around 8 million all year.
June-July Growth Was Relatively Strong
Nevertheless, the jobs gains from June to July this year do fare rather well looking at the past twenty years. Since we're not using the government seasonal adjustments, we can account for seasonal issues by looking at the change from June to July of each year, compared. When we do that, we get this:
It turns out the June-July gain of this year was pretty good. As we can see, the economy usually sheds jobs from June to July for seasonal reasons, so we need to look at job growth within that context. And, we find that this year, the period was strong compared to the same period for each year over the past twenty years.
That's good news.
These short-term comparisons, however, hide just how weak the current expansion is compared to past expansions. If we look at total jobs gains from peak to peak in economic cycles, we find it looks like this:
The expansions just keep getting weaker and weaker, and total payroll employment today is only 3.5 percent above what it was back in 2008. That 3 percent compares to 20 percent growth from 1991 to 2000. And while it's not graphed here, the 1980s expansion from 1981 to 1990 also experienced 20-percent growth in payroll employment.
So, yes, there's an expansion going on right now, but it's a very weak one in historical context.
Moreover, the payroll survey measures only new jobs. It doesn't measure newly employed persons. In other words, you can add more jobs without adding any new people to the ranks of the employed. This is why the BLS uses two different employment surveys. The payroll/Establishment survey measures jobs while the Household survey measures workers. Using the same analysis of employed persons (as opposed to new jobs) we find a similar trend:
The 1980s expansion came in at 18.3 percent growth from peak to peak.
The Problem of Involuntary Part Time Workers
The picture becomes even more complicated when we consider that many of these workers might be employed part time involuntarily. They might be seeking full-time work, but can only find part time work.
Once we take that measure into account, the picture becomes even more dim for this most recent expansion compared to to others (in this case I've calculated the total number of employed persons and subtracted the number of involuntary part-timers:
In that case, the total number of employed persons is only 3 percent above what it was back in 2008. This compares unfavorably to the two expansions that came before it, and to the 1980s expansion.
We're looking at the worst expansion in more than thirty years, especially if we look specifically at the number of people who have managed to obtain full-time employment.
So, when we're looking at the current employment data, we should keep in mind the fact that the total number of unemployed persons has gone nowhere over the past year, and we should keep in mind that this expansion is especially weak compared to previous cycles.
The news outlets have noted that the job growth in July was particularly good compared to last May when the numbers were especially dismal. It's great there's been some small improvement in recent months, but it's also important to look beyond the last few months to get a sense of the larger picture.
Moreover, the current situation is especially alarming when we consider the fact that the Federal Reserve has been engaging in massive quantitative easing efforts and has kept the target rate at 0.5 percent or below for nearly eight years. The Fed likes to tout its Congressional mandate for maximizing employment as justification for its near-zero interest rate policy. And yet, this is all the Fed has to show for it.
Of course, jobs aren't the end-all-be-all of economic progress. What we really want is more wealth and more income, not more jobs. Unfortunately, the income data doesn't point toward much progress either.
Effects of Labor Force Size Over Time
Some astute readers may ask if the size of the labor force over time might be affecting these numbers. That is, if large numbers of retirees are leaving the work force, wouldn't we expect total job totals to get smaller over time?
There are a couple of ways to check on that. One way is to look at involuntary part timers and discouraged workers in relation to the work force overall. The Feds do track this in the number they call the "U-6" unemployment rate which looks at involuntary part times, but also looks at discouraged workers who haven't recently looked for work because they assume no work is available. If we keep track of those numbers, we find the unemployment rate remains quite high. For July 2016, the U-6 unemployment rate was 10.1. For context, we can note that the 2001-2001 recession maxed out at 10.8 percent. In other words, the U-6 unemployment rate during the current recover is about where it was at during the worst of the Dot-com recession 13 years ago:
In other words, when we take discouraged workers and involuntary part timers into account, we're still experiencing some of the highest unemployment numbers in 20 years. And this is taking changes in the overall labor force size into account. (The published data on U-6 goes back to only 1994.)
Note: All data analyzed here is not seasonally adjusted.
