The most important chart is the 12 month moving average of the changes in US non-farm payrolls directly below.
It should have been apparent to any economist, as it was to us, that the US was falling into recession at the end of 2007. The actual start of the recession is a formality, but no dating for the start past June 2008 seems justifiable, especially when all the other coincident non-jobs indicators are consulted.
The primary argument for a later dating to September 2008 is based on 'real GDP' number which in our view is distorted by a significantly understated rate of inflation. The traditional coincident indicators do not agree with that dating as well.
The headline or seasonally adjusted payroll number turning negative does not necessarily imply a recession in and of itself. It could be a response to a transitory exogenous shock. However, when one looks at the longer term trend as we show in the chart above, and the many other coincident indication as we have been pointing out this year, the implications of an endogenous recession is obvious.
Much is made of the Birth-Death Model, which the BLS uses to account for the jobs created by small business that are not in its survey. As you can see, every year the pattern repeats with some regularity and revision. The numbers are added to the payroll number from the surveys, to the actual number, which is then seasonally adjusted to create the 'headline number.'
We hope it is obvious that the seasonal adjustment factor is large, and often far more significant than the birth death model. This is why they choose to 'adjust' the birth death model lower during periods of extreme seasonal adjustment. It does seem to be statistically useless at best, and at best a tool for very short term data manipulation at the worst. It can have an effect in months where seasonality is slight.
It should be noted that this are the numbers that the BLS shows in its database today. They have been revised, and sometimes significantly so, from their original introduction to the public in the Wall Street headlines.
The economic luddite will ask, "What good does this do to me now? Of course I know that the US is in recession!"
The answer of course is that this is the same conclusion we presented as early as February, when it was more easily ignored.
The better question now is, how deep will the recession go, and when will the recession end? Questions with answers not so obvious as of yet without some informed insight. Common historic averages are just that: common, average and old.
