Today's initial claims number printed at 357K, on expectations of 355K, a number which next week will be revised higher once again, likely to 362K. The game here is simple - just show a decline in claims, as what happened to last week's number, also revised higher, this time from 359K to 363K, just so it can show a 6K decline and allow the idiot media to blow such headline as "Weekly US unemployment benefit applications fall to 357,000, lowest in 4 years" from AP and "Jobless Claims in U.S. Decrease to Lowest Level in Four Years" from Bloomberg [5]. In reality, this is the third consecutive miss of consensus in a row. Give us a break - funny then when one considers that last week's consensus was 350K, which has since been revised to 363K. Or what about that 348K print the week prior which ended up being a more realistic 364K. In other words, the headlines were 348K, 359K, 357K, and somehow this is indicative of anything more than outright and endless data manipulation. Needless to say, when next week the number is revised to a far greater miss, nobody will care as the embargoed headlines will once again say "Jobless Claims in U.S. Decrease to Lowest Level in Four Years" and the sheep will keep on buying it over and over and over. What is also notable, is that just like yesterday's ADP number, today's claims data gives no hint what to expect from tomorrow's market holiday NFP.
Below is a true apples to apples chart showing how real claims accounting for continuing, and extended, are not really at 2008 levels.
[6]