Quant Memo
Foundational

Purchasing Managers Indices

PMIs are monthly survey-based gauges of whether business conditions are expanding or contracting, valued because they arrive weeks before official GDP data and split cleanly around the 50 line.

Official GDP figures arrive with a lag of a month or more and get revised repeatedly. A Purchasing Managers' Index (PMI) fills that gap: every month, survey firms ask purchasing managers at hundreds of companies whether output, new orders, employment, and a handful of other measures got better, stayed the same, or got worse versus last month, and compress the answers into one diffusion index.

The construction is what makes the number easy to read at a glance: each sub-question's index is built as %better+0.5×%unchanged\%\text{better} + 0.5 \times \%\text{unchanged}, so a reading of exactly 50 means conditions on average are unchanged from the prior month. Above 50 signals expansion, below 50 signals contraction, and the distance from 50 is roughly proportional to how widespread the change is.

A PMI is not a level of activity — it's a monthly diffusion index measuring the direction and breadth of change, so a reading of 51 for six straight months means output has been mildly but continuously expanding, not that output is at some fixed "51%" of capacity.

Worked example. In a manufacturing survey, 40% of firms report new orders increased, 45% report no change, and 15% report a decrease. The new-orders index is 40+0.5(45)=62.540 + 0.5(45) = 62.5, comfortably above 50 — a broad-based expansion in new orders that month. If the composite manufacturing PMI built from several such sub-indices reads 54.2 versus 51.8 the prior month, that's read as accelerating, not just continued, expansion, which is why markets often react to the direction of the PMI change as much as its level.

Related concepts

Further reading

  • ISM/S&P Global, Manufacturing PMI methodology notes
ShareTwitterLinkedIn