842,000 Americans left the labor force in June

A truly massive number of people left the labor force in June. The unemployment rate fell 0.2 percentage points in June because even though the number employed fell by 301,000, the number unemployed fell by 350,000 since so many people had left the labor force and where no longer looking for work.

UPDATE: Here is a discussion from The Economist magazine.

THE best that can be said about the June jobs report is that it doesn't signal a return to recession. Total employment fell 125,000, but this was because of an end to some temporary jobs conducting the federal census. Private payrolls rose by 83,000. That's soft, and less than expected, though better than May's revised increase of 33,000.

Unfortunately, the guts of the report aren't much more uplifting. The private work week, after rising to 34.2 hours in May, fell back to 34.1 in June. In terms of the amount of labour being used throughout the economy, that's the same as a drop of 300,000 in payrolls. Average hourly earnings actually fell, by 0.1%. That may fan fears of deflation given the already low readings on inflation. I think that's premature. Hourly earnings have been generally rising, albeit sluggishly, and the June drop could be a one-time move.

Finally, while the unemployment rate did drop, to 9.5%, an 11-month low from 9.7%, it dropped for the wrong reason: a lot of people stopped looking for work. This might be because a Senate impasse led to the expiration of extended unemployment insurance benefits early in June. The number of recipients has been falling by about 200,000 per week since. Some of these people may have stopped looking for work (a requirement to qualify for benefits), and thus are no longer counted as unemployed. Census layoffs may have also played a part.

For what it's worth, the household survey (used to calculate the unemployment rate) showed a much bigger drop in employment, of 301,000, than the payroll survey. The rule of thumb is that when the two differ, go with the payroll survey. . . .