by Calculated Risk on 3/01/2011 03:29:00 PM

I've been sent some poor commentary today regarding the ISM manufacturing index. I won't embarrass the author, but I'd like to make a few points:

• The ISM reports several diffusion indexes. The indexes are calculated by adding the percent of positive responses plus one-half of those responding unchanged. Then the index is seasonally adjusted. As an example, for employment in February: 35% of companies surveyed said employment increased and 56% said employment was unchanged. So the ISM adds 35 plus 28 (half of 56) giving 63. Then the seasonal factor pushed this up to 64.5.

This is constructed so that any reading above 50 suggests expansion. This is based on companies, not number of employees, but the ISM employment index does track changes in BLS reported manufacturing employment pretty well over time.

• The ISM employment index can be strong - the 64.5 reported today was the highest since January 1973 - even though the overall employment situation is grim. This just means more companies are hiring today and says nothing about the current overall employment situation.

• Although this was the highest employment index reading since 1973, manufacturing employment is a much smaller percentage of overall U.S. employment now. In 1973, almost 30% of private payroll employment was manufacturing, today it is less than 11%. So the same reading today will have a much smaller impact on the overall U.S. employment.

• The ISM manufacturing index is from a private organization, the Institute for Supply Management. It is not a department of the U.S. government.

Prices were a clear concern in the ISM report - both with the prices index at 82, and from the comments. But for manufacturing production, orders and employment, this was a very strong report.