Calculated Risk and U Michigan

One of the interesting correlations the Calculated Risk does on a regular basis is the correlation between unemployment and housing starts. As seen here. The comment is unemployment lags housing by 12 to 18 months. So does this mean that we should see unemployment be reduced significantly in the next 24 to 36 months?  The economists at the University of Michigan don't seem to think this correlation will hold as they predicted the following:

           Housing starts, expected to total just 605 thousand units this year, recover to 807 thousand in 2011 and 1.13 million units in 2012.
 
I understand that this includes multifamily which I will assume to 20% so that would be in the neighborhood of 600-650 single family or on the twelve month lag in the neighborhood of 8.2 - 8.5% unemployment. This isnt too far from their 2012 estimate of 9.3%.  My bigger concern is that in 2013 will we be at 7.5-7.7% unemployment? I don't think so.  Therefore the lag may be a little more pronounced closer to 18-24 months, will not correlate as we did in 2001 or the housing starts is a little optimistic.  I see some merit in all three but feel that the housing starts are significantly optimistic due to the low rates of household formation and still large supply of existing homes.

Comments