In following trying to see whether we are really headed for a double dip I have been following the regional manufacturing surveys.

Why are these an important indicator?

Well there is a bit of a three-step but it goes like this:

1) Regional Surveys help us to forecast the national ISM

2) The National ISM helps us to forecast manufacturing payroll growth

3) Manufacturing payroll growth is “key” in recessions


I don’t have a graph to show you how closely the regionals match to the national ISM. Indeed, you would need a forecasting formula. But, suffice it to say lots of bad regional numbers do not add up to a good national number.

I can, however, show you national versus manufacturing growth over the last 30 years.

FRED Graph

Though you can see some gaps in recent years the match up is incredibly tight and the gaps represent a secular trend downward in manufacturing growth. If we zoom in the time horizon the strong correlation re-appears.

This is the last ten years

FRED Graph

Crazy tight huh? These aren’t even the same kind of units. One is a diffusion index and the other is a change in raw count. Its uncanny.

Now why is manufacturing so important. Because while manufacturing has declined without a recession, a recession has thus far never occurred without a decline in manufacturing.

Strikes in the early part of the century make the monthly data extremely bumpy and hard to read. When a strike is on manufacturing employment collapses and then when it is off manufacturing employment zooms.

But we can see the same thing by looking at year-over-year instead of month-over-month.

FRED Graph

You can see that since we have been recording manufacturing employment there has never been a recession without a decline in manufacturing jobs. Its to quintessential cyclical sector.

So we look at regionals to guess ISM. ISM basically tells us manufacturing payroll growth and a decline manufacturing payrolls has been a pre-cursor for every recession.