EconoMonitor

Ed Dolan's Econ Blog

September Jobs Report is the Strongest in Months; Unemployment Falls to 7.8 Percent

After a spring and summer when the monthly jobs reports have brought nothing but gloom, the September data are strong across the board. The headline numbers—114,000 new payroll jobs and an unemployment rate of 7.8 percent—are themselves encouraging enough. In many respects, the details behind them look even better.

Let’s begin with the payroll jobs numbers. As shown in the following chart, the September gain of 114,000 nonfarm payroll jobs is respectable, although unspectacular, especially compared with the 202,000 new jobs created in the same month a year ago. But the best news lies not in the figure for September, but in the revisions for July and August. Recall that July payroll jobs were originally reported at 163,000, then revised down to 141,000. That number is now revised up to 181,000. The August job gain, originally reported at just 96,000, is revised up to 142,000. If we add the September preliminary number to the upward revisions, it would be accurate to say that the economy has a full 200,000 more jobs than we thought it had a month ago.

Next consider the unemployment rate, which fell to 7.8 percent. That is a magic number for Democrats because it brings the rate down to where it was when President Obama took office. Unemployment is still well over most economists’ estimate of the natural rate, but at least the White House no longer has to answer to the charge that joblessness has gotten worse on its watch.

Furthermore, the September unemployment figure was a “good” decrease. Remember, the unemployment rate is the ratio of unemployed workers to the labor force. Sometimes the ratio falls only because labor force participation decreases. That was the case in August, when the unemployment rate fell but so did the number of employed workers. In September, all indicators pointed up. The labor force increased by 418,000. The number of employed workers grew by a robust 817,000, and the number of unemployed fell by 456,000. The labor force participation rate and the employment-population ratio both increased in September, after falling in August.

Note that the 817,000 increase in the number of employed workers reported in the household survey greatly exceeds the 114,000 new jobs from the establishment survey. The two numbers can differ partly because the two surveys use completely different samples and methodologies, and partly because the household survey includes farm workers and self-employed persons.

Even in good months for employment, it is possible to find some less favorable news, just as it is possible to find some good news in the bad months. Among September’s disappointments:

  • The BLS’s broadest measure of employment stress, known as U-6, remained unchanged at 14.7 percent. U-6 differs from the headline unemployment rate by including discouraged workers and people who are involuntarily working part time while looking for full-time work. The number of involuntary part-time workers actually increased compared with August.
  • The percentage of the unemployed who were out of work for 27 weeks or more increased slightly in September, and both the mean and median duration of unemployment increased. However, all of these indicators are healthier now than they were a year ago.

There will be one more report on the employment situation just before election day, on November 2. It is anyone’s guess as to whether it will bring more good news or a reversion to a gloomier mean. If the jobs data for October are bad, expect a barrage of last-minute GOP ads to ensure that everyone gets the news—along with a lot of Democrats giving silent thanks for the fact that at least a third of the electorate is expected to vote early this year.

 

12 Responses to “September Jobs Report is the Strongest in Months; Unemployment Falls to 7.8 Percent”

RollsWRanglerOctober 5th, 2012 at 6:08 pm

Maybe there should be a reciprocal subtraction, the opposite of "discouraged workers" namely the number of individuals remaining in the workforce who would otherwise retire if they trusted the government.

wtt1011October 6th, 2012 at 10:42 am

THE HOUSEHOLD # IS A MIRAGE, ITS A BAD SAMPLE, AND WILL CORRECT THE OTHER WAY NEXT MONTH. BESIDES, THE ONLY # THAT MEANS ANYTHING IS U-6 AND THAT # IS STILL 14.7%. THIS REPORT DOESNT CHANGE THE FACT THAT 23 MILLION PEOPLE ARE STILL UNEMPLOYED OR UNDEREMPLOYED. HOUSEHOLD INCOME FOR MIDDLE AMERICA ARE STILL DOWN $5000.00 FROM 4 YRS AGO. SO TELL ME AGAIN WHATS SO GOOD ABOUT THIS REPORT-YOU HAVENT SOLD ME YET !!

Ed Dolan EdDolanOctober 6th, 2012 at 11:57 am

I agree, the household survey is subject to sampling error, so it is also subject to mean reversion, and it could well go down again next month.

What interests me is the emotional intensity of your comment. Do you react to it in the same way when it has a bad month, for example, in August when the household survey showed a decrease in jobs even though the establishment survey showed an increase? Why didn't you comment last month to say that because of sampling error, the economy was probably stronger than the survey number made it look?

Same thing about U-6. Last month it dropped by three tenths of a point. Did you attach equal importance to it then? Or did you dismiss the drop in the August U-6 as a statistical mirage, and see the decrease in reported jobs as the "only thing that mattered" that month?

Yes, you have a point that we can't attach too much importance to any one month's figure. That's why I like to include charts in the discussion because they help to put each month's figure in the context of the trends. Here are a bunch more charts that I didn't include in the post: http://tiny.cc/24tplw Check them out.

wtt1011October 6th, 2012 at 11:53 am

YOU ARE GOING TO HAVE TO EXPLAIN TO US HOW AN ECONOMY GROWING AT 1.3% IN GDP CAN LOWER UNEMPLOYMENT .3. IT DESNT MAKE ANY SENSE !!! I THINK FIGURES LIE AND LIERS FIGURE.

Ed Dolan EdDolanOctober 6th, 2012 at 12:08 pm

I am tempted to say TONE IT DOWN PLEASE, ENOUGH WITH THE ALL CAPS!!!! (just joking)

Seriously, since you pointed out before that wages are falling, why is it logically inconsistent for the number of jobs to be growing faster than aggregate income?

sharpratioOctober 9th, 2012 at 7:58 am

Economists should support the BLS and the efforts it makes to ESTIMATE the labor force, employment, etc on a monthly basis. This is a really difficult task. The BLS produces estimates that get things more or less correct in the long run. Let's not confuse that with the media emphasis on one number (the unemployment rate) rather than others such as U6 or the employment rate as well as the emphasis on individual months rather than longer term trends. That is NOT the BLS's problem.

MikeOctober 9th, 2012 at 8:51 am

So in other words, after nearly 4 years in office the Obama Administration can boast that at least things are no worse. Please-that is pathetic!

JohnOctober 10th, 2012 at 3:59 am

The Fix is in. This is so convenient for Obama. Just like how Q2GDP was initially 1.5%, then revised up to 1.7% which the Obama administration touted until it got revised down to 1.25%. I had never seen a GDP revision first up, then down like that. Looks like someone is playing lose and fast with the data for political gain. Obama is more corrupt than Richard Nixon ever was.

Ed Dolan EdDolanOctober 10th, 2012 at 5:41 am

Series such as unemployment, jobs, and GDP are subject to sampling error and thus reversion toward the mean (or perhaps better to say toward the trend). Revisions first down, then up, or vice versa would be just what we expect. They are not evidence that "someone is playing lose and fast with the data for political gain." In countries like China, Iran, or Argentina where it is much more plausible to believe that someone is fiddling the data, the relevant numbers tend not to jump around so much from month to month.

It will be interesting to see what happens if sampling error and mean reversion cause the unemployment rate for October to rise. Will the conspiracy theorists who populate the Commentosphere simply fall silent, or will they attribute the shift to a new set of conspirators, this time a team that is working to advance the GOP rather than the Democratic cause?