Almost everybody within earshot of a broadcasting device yesterday knows that the Bureau of Labor Statistics (BLS) reported a net loss of 4,000 jobs in the economy in August. Unemployment rate, at 4.6%, was unchanged.
Reporting, and misreporting, by the New York Times and Associated Press set Old Media's template for the story. Some reports, including this one by Vikas Bajaj at the Times, laid the entire onus of the loss on private companies:
Companies reduced their payrolls by 4,000 jobs in August, a sudden turnaround from the net increase of 68,000 jobs in July.
That is wrong, as the Associated Press noted:
The government actually sliced 28,000 jobs, while all private employers added 24,000, the fewest since February 2004.
The Times's Bajaj also managed to get in a mention of the BLS's downward prior-month jobs revisions to July and June. Old Media has almost always ignored similar or larger upward revisions to previous months in previous BLS reports.
While AP got the point about the private sector right, its coverage of the jobs report and the economy may have set a record for the highest ratio of scare words per net job lost. Here are a few examples from the "Fears Rise on Job Losses" report (words that especially annoyed me are in bold; items are not in the same order as they originally appeared):
- "..... raising new fears that a deep housing slump and a painful credit crunch could push the economy into a recession."
- "The ailing housing market and credit problems that have unhinged Wall Street are main culprits behind businesses' fresh sense of caution."
- "The new employment figures revealed the first major crack in the job market."
- "Under a worst-case scenario, the economy could slip into a recession this year. Earlier this year, former Fed chief Alan Greenspan had put the odds at one in three."
- "Mindful of political backlash heading into the 2008 elections, the Bush administration and Democrats on Capitol Hill have been scrambling to help millions of homeowners at risk of losing their homes and looking for other ways to limit the fallout.
- Factories led the way in job cuts; they slashed 46,000 positions last month, the most since July 2003. Construction companies eliminated 22,000 jobs, the most in six months. The carnage could turn out to be even worse because the report — based on information as of mid-August — doesn't capture the full brunt of the credit crisis which intensified during the month.
AP named no sectors with positive jobs growth, even though two of them had larger positive numbers than the negative ones AP identified. 60,000 jobs were added by service providers, along with 63,000 in education and health services.
As to the specter of recession, the best-kept secrets of the past week came from the Institute for Supply Management.
First, on Monday, the ISM's Manufacturing Index for August came in at 52.9%. Although the reading was a bit lower than the 53.8% recorded in July this means that the sector, representing about 12% of the economy, remained in expansion mode for the seventh month in a row, and the 49th out of the past 51 (any reading above 50% represents expansion). That 49-for-51 record is the best in over 40 years, going all the way back to December 1966, which was the the end of a 68-for-69 performance.
Then on Wednesday, the ISM's Non-Manufacturing Index, representing the other 88% of the economy, including the housing sector, came in at 55.8%, the same level as July. It was the 53rd month in a row of non-manufacturing expansion -- a record for that particular index.
So AP and the New York Times were clearly on the far side of hysteria in their reports -- the same reports that provided grist for the woe-is-us broadcasts.
Baloney. In addition to the fact that the economy as a whole is still expanding decently, as shown in the ISM reports just cited, some context to the reported job losses is in order. A NewsBusters e-mailer has come through with just that, in the form of this BLS chart showing job losses and gains since the beginning of 1997:
If August's net job loss holds up or ends up worse (I expect that it will), the 47-month streak of job gains (September 2003 through July 2007) will still be longer than any other in the past 10 years.
I reviewed my e-mail archives, and found that BLS initially reported job losses in May of 2000 (-116,000, after taking out 357,000 census workers), July (-108,000), and August (-105,000).
The figures in the chart are different because BLS makes adjustments in the next two reports, and applies two comprehensive annual revisions when it has more complete data.
If you don't remember any gnashing of teeth over reports of job losses during 2000, you're not alone.
So why is an initial report of 4,000 job losses in the backdrop of a still-expanding economy cause for such alarm?
Cross-posted at BizzyBlog.com.
