ISM Manufacturing Index Expands, But Less Than Expected

by: Doug Short

Today the Institute for Supply Management published its February Manufacturing Report. The latest headline PMI at 51.3 percent is the fourth month of expansion following a month of contraction. However today's number was below the consensus of 54.0 percent.

Here is the key analysis from the report:

Manufacturing expanded in March as the PMI™ registered 51.3 percent, a decrease of 2.9 percentage points when compared to February's reading of 54.2 percent. This month's reading reflects the fourth consecutive month of growth in the manufacturing sector. A reading above 50 percent indicates that the manufacturing economy is generally expanding; below 50 percent indicates that it is generally contracting.

A PMI™ in excess of 42.2 percent, over a period of time, generally indicates an expansion of the overall economy. Therefore, the March PMI™ indicates growth for the 46th consecutive month in the overall economy, and indicates expansion in the manufacturing sector for the fourth consecutive month. Holcomb stated, "The past relationship between the PMI™ and the overall economy indicates that the average PMI™ for January through March (52.9 percent) corresponds to a 3.3 percent increase in real gross domestic product (NYSEMKT:GDP) on an annualized basis. In addition, if the PMI™ for March (51.3 percent) is annualized, it corresponds to a 2.8 percent increase in real GDP annually."

As I've repeatedly pointed out, I'm reluctant to put very much focus on this index for various reasons, but they are essentially captured in's Big Picture comment on this economic indicator.

This [the ISM Manufacturing Index] is a highly overrated index. It is merely a survey of purchasing managers. It is a diffusion index, which means that it reflects the number of people saying conditions are better compared to the number saying conditions are worse. It does not weight for size of the firm, or for the degree of better/worse. It can therefore underestimate conditions if there is a great deal of strength in a few firms. The data have thus not been either a good forecasting tool or a good read on current conditions during this business cycle. It must be recognized that the index is not hard data of any kind, but simply a survey that provides broad indications of trends.

The chart below shows the Manufacturing series, which stretches back to 1948. I've highlighted the eleven recessions during this time frame and highlighted the index value the month before the recession starts.

For a diffusion index, the latest reading of 49.5 indicates contraction. What sort of correlation does that have with the months before the start of recessions? Here are the eleven data points for the months before recessions arranged in numeric order with the latest data pointed inserted in the sequence (highlighted in bold).

42.1, 44.8, 45.7, 47.2, 47.8, 48.5, 49.2, 50.5, 50.7, 51.3, 53.2, 66.2

nine were lower than the latest data, and two were higher (the recessions that began in 1969 and 1973).

But how revealing is today's 2.9 point change from last month? There are 783 monthly data points in this series. The average month-to-month point change is 2.01 points. So today's headline PMI number is 44% greater the mean change.

To reiterate the assessment: "The data have thus not been either a good forecasting tool or a good read on current conditions during this business cycle." The ISM reports nevertheless offer an interesting sidebar to the ongoing economic debate.

Note: I use the FRED USRECP series (Peak through the Period preceding the Trough) to highlight the recessions in the charts above. For example, the NBER dates the last cycle peak as December 2007, the trough as June 2009 and the duration as 18 months. The USRECP series thus flags December 2007 as the start of the recession and May 2009 as the last month of the recession, giving us the 18-month duration. The dot for the last recession in the charts above are thus for November 2007. the "Peak through the Period preceding the Trough" series is the one FRED uses in its monthly charts, as illustrated here.