Forex Economic Analysis: Making sense of the Initial Claims Data. Don't necessarily believe what you see.

The low Initial Claims data is confounding given the reports and expectations for the economy.  However, the recent data strength may be more about faulty seasonals than overlall strength in employment.

Economic figures like the Initial Claims have seasonal factors in an attempt to smooth out the swings due to things such as holidays.  The Christmas/New Year period for instance shows large movements in claims historically.  The chart below shows the spikes in unadjusted Initial Claims over the last 3 years.

Over the last 3 years the Spread between the Seasonally Adjusted Initial Claims and the Adjusted Initial Claims have shown quite a different picture.  Over the 2006 year end, in the thrid week of January, the difference between the Unadjusted Inital Claims and the Seasonally Adjusted release was 32,926.  In 2007 the difference waw 39,583.  This week the difference between Seasonally and Unseaonally Adjusted was a robust 107,333.  The 60,000+ difference may be faulty seasonal which may have understated the "real" number by that same amount (see table below).

In addditon, if you look at last years Unadjusted Claims, they came in at 367,583 and the Adjusted figure was 328,000.  This year the Unadjusted was a higher 408,333 while the Adjusted was only 301,00 (see table). 

The conclusion is the Inital Claims seem to be suffering from faulty seasonals which may underestimate jobless claims by about 60,000 when compared to the last few years.  As a result, the Non Farm Payroll may not necessarily be as strong as what the Initial Claims suggest.