Monthly jobs report: A history of revisions
On the first Friday of each month around the Marketplace newsroom it's always the same scene: reporters, hosts and editors waiting anxiously and specutively for the monthly jobs report to be released by the Bureau of Labor Statistics. The minute the release hits the Internet, we join in the media frenzy with our analysis and insight about how many jobs were gained or lost that month, or how many long-term unemployed workers drop out of the workforce.
But as Marketplace's Heidi Moore explained a few months back, we should all remember that employment figures are revised at least once or twice after their release. So don't get too friendly with June's reported 80,000 jobs added. The original report is revised by the BLS, typically the following month. It's then revised a second time a month after that in what's called the "consensus."
To see just how diverse original estimates and revised numbers can be, we decided to look back at the past several months of reports and revisions to analyze the numbers (see the animated graphic above).
For example. In August 2011, the BLS reported a flat month for employment - no jobs added or lost. CNN Money led with this in its news story: "The economy added no jobs in August, the Labor Department said Friday. Zero, zilch, nada."
However, a month later that number was revised to 57,000 jobs added. Then finally, a second revision bumped it up to 104,000.
P.S. Here's what the BLS has to say about it:
The establishment survey revises published estimates to improve its data ... by incorporating additional information that was not available at the time of the initial publication of the estimates. The establishment survey revises its initial monthly estimates twice, in the immediately succeeding 2 months, to incorporate additional sample receipts from respondents in the survey.