Payroll data came in this morning showing zero net new jobs. Actually, it came in at +17,000 private payroll jobs, -17,000 public payroll jobs. The government continues to drop jobs, shedding around 600,000 since the beginning of the recession. This has become a major disaster that has increased the ranks of the unemployed and put downward pressure on demand at the worst possible time.
The terrible private payroll jobs numbers need to be taken into account along with two other worrisome signs. Firstly, in the previous two months job numbers were revised down, making the anemic recovery even worse. June dropped from 46,000 to 20,000 jobs created, and July went from 117,000 down to 85,000. The labor market is even weaker than we thought.
The second sign is that major indicators surrounding those who do have a job are weak. Let’s look at the total amount of hours worked in the economy, production and non-supervisory hours. The Bureau of Labor Statistics adds up all the hours worked in the economy and plots them on a graph, divided by a previous year to set a baseline. This is total hours worked in the entire economy, not hours per employee or anything like that. And that number declined in August:
We haven’t seen such a small number of hours worked since November 2004 or March 1999. And as you can see (if you squint), it dropped during August, as it did for total aggregate weekly payrolls.
Average weekly earnings and average weekly hours both dropped slightly. We need these numbers to be taking off, not holding steady or declining. So the economy isn’t working even for those with a job. Also, declining weekly earnings mean there are going to be even less inflationary pressures in the economy than we thought.
Meanwhile, the average duration of unemployment has dropped while the median has increased. It’s too early to tell, but that’s a troubling sign with weak job growth — it means that we are likely seeing more and more unemployed people simply dropping out of the labor force instead of finding a job. This will continue to make the unemployment rate a less important indicator than the employment-to-population ratio.
One thing terrible jobs numbers could do is galvanize public and elite opinion in a way that weak jobs numbers wouldn’t. In the next few weeks, there could be movement in the three major initiatives needed for recovery. President Obama is going to lay out an agenda for fiscal stimulus, the Federal Reserve is having a two-day meeting to determine monetary stimulus, and numerous things appear to be underway in the housing market, including FHFA suing the major banks and several attorneys general breaking from the cover-up-like proposed settlement to investigate foreclosure fraud. We’ll see if it is all too little, too late for the economy as it is.
Rortybomb 
Pingback: Friday links: zero daze | Abnormal Returns
Great stats & insight. Quick question for you or other readers: is there any evidence that “hours worked” really means “hours paid,” and hours worked and hours paid can diverge?
If you think that there are lots of employers making employees work off the clock, it might be the case that more work is being done in the economy, but fewer people are being paid.
Also, if anyone knows of a good reference paper on the difference between the productivity of workers, and the dollar value of what they produce, I’d love to see it.
Pingback: Still waiting for the employment recovery « Phil Ebersole's Blog
You can never solve a debt problem with more debt anymore than you can cure a drunk’s addiction with more booze. Goldman Suck’s 54 page secret report on the pending collapse of our economy which was written for institutional investors lays it all out quite clearly, and the public was never meant to see a damn word of it. Trillions already spent which ended up swirling the bowl like it was an event horizon falling into a black hole. I can’t wait for the encore.
Pingback: Sunday Reading « zunguzungu
Thanks for great stats Mike!
I second Frank’s question on hourly worked vs hourly paid. I’m wondering if those workers with jobs are being pushed either explicitly or implicitly (unsaid fear factor) to work longer hours without pay.
Second – might be cool to take a look at the aggregate hours worked divided by the total population. It would be similar to the BLS “Employment-Population ratio” except it would be “Aggregate Hours- population ratio”
Pingback: The Jobs Problem « thecurrentmoment
Pingback: Unemployment: the tip of the iceberg
there are a number of statistics the bureau of labor fudges on. the most obvious is they don’t count short term or long term unemployed that have had their unemployment benefits run out. when they are counted, the actual unemployment is 23%. they also count managers, owners & executives as “labor”, meaning that hourly wages are artificially inflated. they count hedgefund managers making $2 billion a year as “labor”. next they count everyone employed by a company that has gone out of business as still employed. they also don’t take into consideration people working shorter hours as anything but “employed”. they are a lot more deceptive practices, but the point is unemployment is worse than the BOL says it is by quite a bit.