Factories reported stronger orders and easing price pressure. What did buyers actually say?

Two questions on the same questionnaire

The Institute for Supply Management’s October report put the Manufacturing PMI at 55.5, up from 54.5 in September. The new orders reading also finished at 55.5, but rose only 0.2 point. Their identical October levels conceal different changes. The survey’s input prices reading fell from 77.9 to 75.2. It remained well above the neutral reading of 50.

Source notes: [1] Institute for Supply Management · October manufacturing report

What a diffusion reading leaves out

Take an illustrative survey of 100 purchasing managers. Suppose 50 report higher prices, 40 unchanged prices and ten lower prices. Counting each higher response at full weight and each unchanged response at half weight gives an index of 70. Now move four respondents from higher to unchanged. The index drops to 68, although 46 respondents still report increases and only ten report decreases.

The size of the invoices never entered that calculation. Whether one respondent paid one percent more or twenty percent more, the direction was the same. That is why a decline in the October prices index cannot establish a decline in the average dollar price of purchased materials. It also cannot be plugged into a consumer inflation forecast as a monthly percentage change.

Source notes: [1] Institute for Supply Management · October manufacturing report

Which part of the report could move a market?

For a gold reader, the interesting question is which component changed the expected policy path. A trader watching industrial activity might focus on the strength in orders. Another might put more weight on the breadth of reported input cost increases. Both can read the same release accurately and choose different hedges because they entered the day with different forecasts.

ISM’s own dated commentary reported that written feedback was often downbeat even as the headline expanded. Managers can answer that output increased and still voice concern about margins. A count of positive or negative comments therefore answers a different question from the production index.

Source notes: [2] Institute for Supply Management · October manufacturing roundup

A record worth keeping

Before comparing gold with the next release, archive expectations for orders and prices separately. Write down which combination would change the policy interpretation. If yields and the dollar begin moving before publication, mark the pre-existing trend. A later gold quotation, however precise, cannot tell us which survey response changed a trader’s mind.