Saturday’s Washington Post reported on newly-released employment data from the Department of Labor:
“In economic terms, a loss of 17,000 jobs is statistically insignificant. For example, the Labor Department’s initial report that only 18,000 jobs were created in December was later revised to a gain of 82,000 jobs, leading some economists to say that January actually may have created a few jobs.”
The only problem with this statement was not that it was false, but that it was buried deep in a story, which was headlined Decline in U.S. Jobs Could Prove Costly to GOP Nominee, and began:
“The economic storm clouds gathering over the 2008 presidential race burst open yesterday with news that the economy shed 17,000 jobs in January, the first job loss in 52 months and the clearest sign yet that the economy may be in a recession.
For Republicans already facing an economic headwind, the jobs numbers could prove punishing.”
Yes, but only if they’re spun that way by newspapers like the Washington Post. This is just the kind of statistical slight-of-data that powers the charge of liberal bias. Instead of creating a storyline about how favorable the numbers are to campaigning Democrats, the Post should be deconstructing the data.
January’s “loss,” for example, may yet turn out to be a gain, given that new unemployment claims declined – something not reported in the Post story – and that this tends to be interpreted by economists as a more definitive number.
And exports actually increased during January, due to the weak dollar, while manufacturing also increased.
This is not to say that job growth isn’t slowing, it’s to say that a change that represents 0.01% of the workforce – 130 million people – is too small to claim as “the clearest sign yet that the economy may be in a recession.” (Yes, clear as mud