Comparing the unemployment rate (still bad) to the recovering GDP is sort of comparing apples to oranges, since total employment has to keep increasing in order to keep up with a growing population. Its easy to imagine a situation where our employment levels have tracked our GDP perfectly, and our high unemployment is due to a recovery that took too long rather than a shift in the economy. However, the actual data is consistent with his thesis, as shown by this graph he should have used: http://data.bls.gov/pdq/SurveyOutputServlet?request_action=w...