It will be interesting to see how this divergence (below) plays out. Gallup’s polling corresponds loosely to the BLS household survey, the major differences being that Gallup’s results are not seasonally adjusted and cover adults of 18 or older, rather than BLS’ 16 year old cutoff. Otherwise the definition of “unemployed” is pretty much the same – available for work, looking for work, but not even an hour of work. Gallup’s poll is a weekly rolling average of about 30,000 interviews compared to BLS’ monthly 60,000. No need to observe that Gallup knows what they are doing and claims a margin of error of 0.7%. Clearly if this starts to show up in the BLS reports then taper talk will be silenced.
h/t zero hedge for the chart.