Conflicting Reports On Manufacturing PMI's

One potential consequence of an economy that is growing at a slow pace is the higher likelihood that reported economic data has a tendency to conflict with some of the other economic reports. An example of this is the report on manufacturing activity today. Both the Institute for Supply Management (ISM) and Markit reported November Purchasing Manager Index data with the ISM PMI report coming in at 48.1%, down from October's reading of 48.3%. The Markit PMI was reported at 52.6% up from October's reading of 51.3%. As the below chart shows, the Markit PMI did not dip below 50 when the ISM PMI fell below 50.

There are several calculation differences between the Markit and ISM PMI calculations. Markit published a paper, Explaining Manufacturing Survey Divergences, that highlights some of the important differences between ISM's calculation and Markit's.

• "IHS Markit surveys just under 800 manufacturing companies (approximately double the size of the ISM panel size) from which an 80% response rate is typically received. However, unlike IHS Markit, ISM does not disclose actual numbers of questionnaires received."

• "ISM data are based only on ISM members, and as such are likely to only reflect business conditions in larger companies, with small and medium-sized firms under-represented. In contrast, IHS Markit’s survey includes an appropriate mix of companies of all sizes (based on official data showing the true composition of manufacturing output)." 

And an important difference,

• "Survey responses may relate to different markets: ISM also does not ask respondents to confine their reporting to US facilities/factories whereas IHS Markit specifies that all responses must relate only to metrics from US factories. ISM data could therefore be more heavily influenced by conditions of US-owned factories in China, for example, than the IHS Markit data."

Because of the above noted differences, and others in the paper linked above, the Markit data would seem to provide a more accurate picture of the manufacturing economy in the U.S. Also noted in the Markit paper is the fact the ISM data tends to be more volatile; thus, potentially providing a bit less reliable read on the manufacturing part of the economy. And finally, readings below 50 are contractionary but not recessionary. Recessionary levels on manufacturing occur when PMI readings are in the low 40% area.