The economy lost 131,000 jobs in July, all of them in the government.  Private payrolls added north of 70,000 workers--anemic in the face of our current unemployment rate, but hey, at least it's something.  But government jobs fell by 200,000, mostly, as far as I can tell, due to the loss of the census jobs, though we also lost about 50,000 workers from state and local governments (I presume due to the end of the school year).

Unemployment is always a lagging indicator.  And it's clear now that the census jobs were pumping up the jobs figures.  But taken together with the new GDP figures, I think we can now dismiss any hopes of a summer recovery.

Does this mean we should have been doing more stimulus?  I'm still agnostic on this question.  Government jobs at the state and local level clearly got bloated during the long boom, and we can't just keep pumping money into these economies forever while we wait for a recovery.  Japan's experience with government-supported employment does not suggest great things--the government can ease some of the pain, but it can also end up with a terrifying debt-to-GDP ratio and no lasting recovery to help work off the debt.

On the other hand, the long-term unemployment number worries me greatly.  I had a conversation with a friend the other day who was pushing back on my support for 99+ weeks of unemployment benefits; at some point, he said, these people have to get jobs, and helping them grow the gap on their resume is not a good long-term strategy.  I'm not convinced that many of these people could find jobs--some people are undoubtedly coasting, but really, how far can you coast on $300 a week?

One thing's for sure, though; adding more people to that pool of job seekers is going to make it harder for those already in it.  How many of the people who haven't worked in two years will simply drop out of the labor force entirely?  Every extra week of unemployment past six months represents a destruction of human capital.