06 March 2008

February 2008 Non-Farm Payrolls Report

We are not going to try to forecast tomorrow's Non-Farm Payrolls Report. The work we have done so far with this report has made us believe that it is written in sand, and equally subject to change, and more adjusted in opaque ways than most numbers from official sources.

The consensus of economists is a net addition of 25,000 jobs, against a prior decline of 17,000 jobs in January. We fully expect January to be revised, and probably December as well. It will not surprise us to see this number come in revised to positive, with a potentially lowball number for February. Why? Because that is how the Bureau of Labor Statistics has been rolling their adjustments lately. They give us the 'bad news' but then show its unreliable, and probably not so bad.

We would like to show three graphs in anticipation of this report.

The first compares the seasonal adjustment with the actual numbers. As you can see February is a quite large downward adjustment, although not quite as large as the upward adjustment in January. There is obviously substantial room to play with adjustment in both months, especially when the adjustment method keeps changing and is 'proprietary.' The swings in the non-adjusted number are enormous this time of year, making the 25,000 net adds in the seasonally adjusted headline number almost a statistical rounding error. But Wall Street will react.

The second graph shows what we have been calling the Imaginary Jobs Component, more familiarly known as the BLS Birth-Death Model of small companies that are incognito. The net add in February should be about 100,000 jobs based just on prior history as we have shown here. We have heard that BLS will be issuing some major revisions in their methods, but cannot recall if its in this month or some coming month. This number is added PRE-adjustment, so its tossed in the wash in many months, and is not so big a deal as many have come to believe.

No matter how the number comes out tomorrow, highball or lowball, with a prior month twist, or a total restatement of everything back a couple of years (yes they do that too often we might add), we would like to emphasize that no matter what the stock market does in reaction to the headline numbers, there really is only one type of chart worth looking at for these types of volatile numbers: the moving average trend chart.

We have included the chart below to show how clear the the trend has been. The peak of Jobs Growth occurred in early 2006, and has been in a slow but steady decline since then. During the entire 'recovery' that the stock market was discounting we kept asking, "But where are the Jobs?" Well, they really were not there after the first few bubble years after 9/11 during what we like to call "The Great Reflation." Mask the unemployment as they might by dropping people from the counts as their unemployment runs out, the fact is that we entered a recession at the end of 2007 at the latest, and probably earlier than that if we could obtain a realistic Inflation number with which to deflate our bloated nominal GDP.

Don't be shocked if we do get a positive number tomorrow. Its well within probability given the revisions, and the history of Administrations messing with the numbers going back to LBJ at least. And don't be shocked if the Financial Talking Heads say "What Recession?!!".... or not. The number is capable of coming out any which way, and will be revised next month regardless. We're in the silliest of seasons here, as the proverbial piper comes to be paid, and the entire financial and political structure in the United States seems badly in need of adult supervision.