Someone needs to tell the Associated Press's Jeannine Aversa and Christopher Rugaber that just because the number of unemployed people declines, it doesn't mean that they "found work."
That must be what the pair believes. Their error-riddled and suspect supposition-driven Friday afternoon report, whose title predictably focused on the unemployment-rate drop while ignoring the pathetic increase in seasonally adjusted jobs, actually made that claim (bolds and numbered tags are mine):
Unemployment falls to 9 percent, lowest since 2009
The unemployment rate is suddenly sinking at the fastest pace in a half-century, falling to 9 percent from 9.8 percent in just two months  - the most encouraging sign for the job market since the recession ended.
More than half a million people found work in January.  A government survey found weak hiring by big companies. But more people appear to be working for themselves or finding jobs at small businesses. 
... The Labor Department survey of company payrolls showed a net gain of 36,000 jobs in January. That's scarcely one-fourth the number needed to keep pace with population growth.
... the payroll survey (the source for the jobs number -- Ed.), about 140,000 businesses and government agencies send forms to the Labor Department showing how many people are on the payroll and how many hours they worked. The payroll survey can be slower than the household survey (the source for the reported unemployment rate -- Ed.) to recognize startup companies. 
... The government also said fewer jobs were created last year than first thought - a net 909,000, down from an estimated 1.1 million. 
-  -- The "sinking at the fastest pace in half-century" claim should have been framed as "the fastest two-month decline in a half-century." This phrasing was required, because in July 1983 during the Reagan presidence, the seasonally adjusted unemployment rate dropped from 10.1% to 9.4%. That month remains the fastest single-month decline since 1949.
 -- Aversa's and Rugaber's "more than a half million people found work" assertion is flat-out false. You obviously won't find it in the Establishment Survey, which was the source for January's reported seasonally adjusted gain of 36,000 jobs. It's not in the seasonally adjusted Household Survey either:
While the number of unemployed decreased by 622,000, the number of jobholders only increased by 117,000. The rest (504,000) left the labor force. The only way these folks would have "found work" is if they all began to devote more time to household chores. The not seasonally adjusted figures show a steep and expected decline in the number of people working, mostly because January is the month when retailers let go of employees they hired to get through the Christmas shopping season.
-  -- It seems that Aversa and Rugaber are clumsily attempting to lay the groundwork for some kind of future assertion that President Obama's "Startup America" initiative is why the economy and the job market came back. In writing that "The payroll survey can be slower than the household survey to recognize startup companies," they seem oblivious to the fact that the government's Bureau of Labor Statistics (BLS) already attempts to estimate business startups and terminations every month using its Birth/Death model, and that the government has been overly optimistic during the past two years in estimating job creation arising from small business startups. This is a primary reason why, as the AP pair wrote later, "fewer jobs were created last year than first thought." Per BLS, the total adjustment to prior figures spread over roughly the past two years "was 452,000 (483,000 on a seasonally adjusted basis)" -- something Aversa and Rugaber chose to underplay by solely identifying the 2010 effect.
If these guys tell you that the sun rose in the east this morning, I would suggest looking outside just to be sure.
Cross-posted at BizzyBlog.com.