ISM Manufacturing Survey Improvement Continues

I do not put much stock in surveys.  They are entertaining, and have elements of truth if you are lucky enough to identify them.  As a former form-filler-outer of these surveys, my secretary was assigned to complete them.  I had real things to accomplish.

The ISM surveys tend to be particularly noisy.  The purchasing manager is normally not in the executive management loop – but admitted does have some feel of the organization after the two double martini lunch is finished.

The ISM manufacturing survey for January 2011 increased over 5%.  What does this mean?  The correlation to industrial production as demonstrated by the FRED graph below is my prima facie evidence:

Turning points are disassociated with hard data.  This month the data is relatively easy to believe.  The correlation between the ISM Composite Index and Industrial Production is sloppy on the matter of magnitude, but the correlation on direction is clear:  ISM leads IP direction by 3-9 months over the time interval shown in the graph.  On that basis, improved IP is indicated in the coming months.

Here are excerpts from the ISM press release:

“The manufacturing sector grew at a faster rate in January as the PMI registered 60.8 percent, which is its highest level since May 2004 when the index registered 61.4 percent. The continuing strong performance is highlighted as January is also the sixth consecutive month of month-over-month growth in the sector. New orders and production continue to be strong, and employment rose above 60 percent for the first time since May 2004. Global demand is driving commodity prices higher, particularly for energy, metals and chemicals.”


Of the 18 manufacturing industries, 14 are reporting growth in January, in the following order: Petroleum & Coal Products; Primary Metals; Apparel, Leather & Allied Products; Wood Products; Computer & Electronic Products; Transportation Equipment; Fabricated Metal Products; Machinery; Paper Products; Miscellaneous Manufacturing; Chemical Products; Furniture & Related Products; Food, Beverage & Tobacco Products; and Electrical Equipment, Appliances & Components. The four industries reporting contraction in January are: Textile Mills; Printing & Related Support Activities; Plastics & Rubber Products; and Nonmetallic Mineral Products.

  • “Continued weakness in the dollar is having a negative effect on the components we purchase overseas and increasing our material costs.” (Transportation Equipment)
  • “Lead times are increasing significantly, and commodity pricing is starting to increase.” (Chemical Products)
  • “January/February sales will be decent, and we see a strong March. We’re cautiously optimistic but reluctant to hire.” (Fabricated Metal Products)
  • “Business is still slow with no pick-up in sight.” (Furniture & Related Products)

The ISM reports have too many bells and whistles if your intent is to understand the economy.  The critical elements are new orders and backlog of orders.  With new orders, you can understand whether the Purchasing Manager believes the order booking ledger is improving over the previous month.  With backlog, you can understand if orders are coming in fast enough to support existing capacity.

New Orders

The new order index skyrocketed this month.   This is one of the more noisy elements of the ISM manufacturing index.  Some of the commentary from the headlines:

The 12 industries reporting growth in new orders in January — listed in order — are: Petroleum & Coal Products; Primary Metals; Computer & Electronic Products; Transportation Equipment; Wood Products; Machinery; Fabricated Metal Products; Miscellaneous Manufacturing; Chemical Products; Paper Products; Electrical Equipment, Appliances & Components; and Food, Beverage & Tobacco Products. The three industries reporting decreases in new orders in January are: Textile Mills; Plastics & Rubber Products; and Nonmetallic Mineral Products.

Backlog of Orders

This is normally a less noisy component of the ISM Manufacturing Index – but this month it had a violent improvement.   Commentary from the headlines:

The 10 industries reporting increased order backlogs in January — listed in order — are: Primary Metals; Apparel, Leather & Allied Products; Computer & Electronic Products; Machinery; Paper Products; Printing & Related Support Activities; Plastics & Rubber Products; Transportation Equipment; Fabricated Metal Products; and Chemical Products. The four industries reporting decreases in order backlogs during January are: Miscellaneous Manufacturing; Nonmetallic Mineral Products; Food, Beverage & Tobacco Products; and Electrical Equipment, Appliances & Components.

Related Article

Chicago Purchasing Managers Continue to see 21st Century High Business Conditions GEI News


My reply to a comment by OldProf needed a graph for explanation.   The Federal Reserve graph below plots the correlation between Industrial Production (IP) and the ISM since 1999.

5 replies on “ISM Manufacturing Survey Improvement Continues”

  1. The value of a survey can only be judged by looking at the explanatory and predictive value. My research shows the ISM data to be an important component of changes in payroll employment.

    The ISM has studied the long history of their data series as it relates to the overall economy. The press release includes the following statement, perhaps more important than the specific elements that you cite:

    A PMI in excess of 42.5 percent, over a period of time, generally indicates an expansion of the overall economy. Therefore, the PMI indicates growth for the 20th consecutive month in the overall economy, as well as expansion in the manufacturing sector for the 18th consecutive month. Ore stated, “The past relationship between the PMI and the overall economy indicates that the PMI for January (60.8 percent) corresponds to a 6.4 percent increase in real gross domestic product (GDP) on an annual basis.”

    This series has a long record and has been accurate in forecasting the recent GDP growth. Meanwhile, the CMI series is very new and experimental and has been completely wrong both on forecasting growth and explaining the reasons.

    My own inclination would be to favor the method with the long and accurate track record…..

  2. OldProf,

    I have done various studies of the ISM since 2000. I use 2000 because I believe there was an economic pivot point at 2000 where the economic dynamics changed. You will note that the ISM gave a false up-impulse during this recession. During the 2007 recession it hit the up-impulse point perfectly (1 for 2 for predicting recovery).

    Its predictive power for an economic decline (again using IP as a proxy) also 1 for 2. The ISM was a good predictor in 2000 – and then around 2004 lost complete correlation to IP.

    The whole purpose of surveys are to give a heads up before hard data. A 50% hit is not good enough. There are other dynamic indicators based on hard data which are timely (days after the reporting period closes) – and have better correlation, to not only IP, but the economy in general.

    I pick on the ISM because so many believe is has powers which it plainly does not have. Do I ignore it – no. It remains a tool in my bag, I use it to confirm other data I am seeing.

    I have added one more chart to the article which graphically explains the point I am trying to make.

    steven hansen

Comments are closed.