Pethokoukis, Economics, U.S. Economy

The September jobs report: Here’s what it would have been


Thanks to the partial government shutdown, the US Labor Department’s Bureau of Labor Statistics will not be releasing the September jobs report today. This is a huge bummer for economic data junkies everywhere — and for the Federal Reserve which has been closely watching employment data as it decides when to begin tapering its bond-buying program.

But what might have the September jobs report told us about the health of the labor market? Well, there is some other data that can help us make a rough guess:

– In a pre-shutdown survey, according to The Wall Street Journal, economists expected September nonfarm payrolls to increase by 181,000 and the jobless rate to hold at 7.3%.

–  Initial jobless claims for the week ending September 28 rose by 1,000 to 308,000 which lowered the four-week moving average by 4,000 to 305,000. That’s the lowest level since May 2007.

– Over the past quarter, employee tax withholding revenues have increased at a 4.0% annualized rate, down from a recent peak of 7.0% growth this past March. At its current pace, according to Deutsche Bank, tax receipts are broadly consistent with 2.2 million per year gains in nonfarm payrolls or about 180,000 jobs per month.

– US businesses added 166,000 jobs in September, according to payroll processor ADP, just below its moving average of 175,000.

When Deutsche Bank looks at the data we do have, particularly the ADP and tax data, it concludes “whenever the September employment data are released—we are starting to think it will be toward the end of the month, we look for a similarly sized increase as ADP. Going forward, if claims (recent distortions aside) remain near 300k, then the rate of job growth should pickup to at least a 200k per month pace.”

1. Now let’s say Deutsche Bank is correct. Even at an accelerated job creation pace of 200,000 jobs a month, it would take more than five more years to return to pre-recession employment levels.

2. Remember the current jobless rate probably understates the true health of the labor market. Let’s say the consensus was correct, and the September jobs report showed a 7.3% unemployment rate. That number is based on a shrunken labor force participation rate where people who stop looking for work are disappeared from the data.

3. If the labor force participation rate were back at January 2009 levels, the unemployment rate would be just shy of 11%. As Minneapolis Fed president Narayana Kocherlakota recently put it: “Most of the declines in the unemployment rate since October 2009 have occurred because the fraction of people who are choosing to look for work has fallen.”

4. But what accounts for that drop in labor force participation? There have been many studies on this subject. Some part of the decline reflects the aging of the population. But when the Goldman Sachs econ team looks at the balance of the evidence, it suggests to them that “a bit more than half of the observed decline since end-2007 (around 1½pt) is related to the cyclical weakness of labor demand. The implication of this estimate is that the labor market is much further from full employment than the unemployment rate alone would suggest.”


5. And if you plug the Goldman Sachs numbers into the BLS data, you get a “real” unemployment rate of 9.5%. That is, by the way, far above the 2009 White House forecast of 5% unemployment by this point in 2013. Long-term unemployment — more than 4 million Americans have been out of work for more than 27 weeks –is particularly dreadful.

So what would the September jobs report have really told us? The US labor market remains deeply depressed.

2 thoughts on “The September jobs report: Here’s what it would have been

  1. The published reports are false anyway, to assume that the nation gets honest number is absurd. The first thing the cross dressing Emmanuel did when he became “chief of staff” was to remove the numbers from Commerce, for that reason Senator Gregg (N.H) refused to join the cabinet, knowing this outlaw crowd was going to falsify all data, which they did.

  2. I’d be interested to see the composition-adjusted labor force applied to U6, which has declined more steeply from its high (17.4% to 13.7%, a difference of 3.7%) than U3 has (10.1% to 7.3%, a difference of 2.8%).

    Disproportionate increases in part-time employment probably account for most of the discrepancy, but I wonder how much U6 would decline if you changed the “marginally attached” classification from people having looked for work in the past 12 months to people under 65 who’ve looked for work in, say, the last 60 months.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>