The number of jobs added grew again in September reflecting the slow but steady growth and unemployment dropped below 8% for the first time since January 2009, the month President Obama took office. Republicans, including Jack Welsh the former CEO of General Electric, are going to argue that the Bureau of Labor Statistics somehow cooked the numbers to help the President. Ezra Klein explains why this couldn’t be the case.
We’ve hit that moment in the election when people begin to lose their minds. Case in point, within minutes of the jobs report, Twitter filled with Republicans claiming the books were somehow cooked, the numbers aren’t real, etc.
Let’s take a deep breath. Jobs reports are about the economy, not about the election. Confusing the two leads to very bad analysis.
This is a good jobs report in a still-weak economy. The 114,000 jobs we added in September aren’t very impressive. The revisions to the last two months, which added 86,000 jobs to the total, were much more impressive. Those revisions also suggest that September’s jobs could get revised up — or, of course, down. So be careful about reading too much into that number. Still, these are, at best, good, not great, numbers.
The chart shows the number of jobs added pretty consistent with previous numbers. So where is the controversy?
The controversy, if it’s worth using that word, is over the unemployment rate, which dropped from 8.1 percent to 7.8 percent. That’s three-tenths of one percent. That’s what all the fuss is about.
Let’s get one thing out of the way: The data was not, as Jack Welch suggested in a now-infamous tweet, manipulated. The Bureau of Labor Statistics is set up to ensure the White House has no ability to influence it. As labor economist Betsey Stevenson wrote, “anyone who thinks that political folks can manipulate the unemployment data are completely ignorant of how the BLS works and how the data are compiled.” Plus, if the White House somehow was manipulating the data, don’t you think they would have made the payroll number look a bit better than 114,000? No one would have batted an eye at 160,000.
The fact is that there’s not much that needs to be explained here. We’ve seen drops like this — and even drops bigger than this — before. Between July and August the unemployment rate dropped from 8.3 percent to 8.1 percent — two-tenths of one percent. November-December of 2011 also saw a .2 percent drop. November-December of 2010 saw a .4 percent drop. This isn’t some incredible aberration. The fact that the unemployment rate broke under the psychologically important 8 percent line is making this number feel bigger to people than it really is.
Wonkblog posted this chart soon after the numbers were released.
…the unemployment rate is a function of two things: the number of people
employed, and the number of people in the labor force. But the proportion of
people in the labor force actually went up, suggesting the fall in the
unemployment rate reflects a real improvement, rather than people stopping their
The Bureau of Labor Statistics looks at the unemployment rate in several ways as Klein explains.
Which leads to another argument: That U6, the broadest measure of labor-market pain, didn’t move, which should make us skeptical of the fact that U3, the normal unemployment rate, did move. That’s just misunderstanding what U6 is.
U6 is not an unemployment measure. It includes part-time workers who want full-time work. So it doesn’t count the increase in part-time work. But every measure of actual unemployment — U1, U2, U3, U4, and U5 — went down. You can see them all here. Again, there’s no mystery.
This is an encouraging report. What it tells us is that the labor market has been a bit better over the last few months than we thought, and that the recovery hasn’t slowed in the ways we feared. What the response to it tells us is that the election is driving people a little bit crazy.
You can see more charts by clicking the various links.