CommBank Composite PMI

The CommBank Composite PMI (Purchasing Managers' Index) is a monthly gauge of Australian private sector business activity. It covers manufacturing and services, which together represent close to 75% of the economy. It is compiled by S&P Global from surveys of senior purchasing managers at 400+ manufacturing firms and 400+ services firms. Each index runs 0 to 100, with 50 (50.0) marking no change from the prior month.

Responses are usually collected in the 2nd half of each reference month. Flash estimates, the early advance readings, appear during the reference month on about 85% of responses. Final manufacturing results appear on the 1st working day of the next month, with services and composite on the 3rd working day.

Releases appear at 09:00 Sydney time with sector detail and commentary from publisher and bank economists. Holidays shift working day dates, and finals supersede flash estimates each month. Underlying survey data are never revised, but seasonal factors may be revised later.

The composite moves with its 2 sector gauges and the demand, work and price forces behind them.

• Manufacturing output: volume of units produced by factories.
• Services business activity: total services activity versus the prior month.
• New orders: inflow of new work, including export orders.
• Backlogs of work: outstanding business not yet completed.
• Employment: changes in staffing levels.
• Input costs and prices charged: purchase costs and selling prices.

The flash estimate published April 27, 2020 fell from 39.4 to 22.4, a record 17 point fall to a four-year survey low. New business, exports, backlogs and employment fell at record rates. Prices charged fell at the fastest rate since the survey began in 2016, with services hit hardest and manufacturing output also falling sharply. A 22.4 reading meant almost all firms reported deterioration, not a 22.4% fall in output, showing why both sectors matter.

1. Response weighting: Each reply is weighted by company size and sector importance, targeting about 80% response.
2. Diffusion index: For each question, the index equals % improving plus 0.5 times % unchanged. For example, 30% higher, 50% unchanged and 20% lower gives 30 + 25 + 0 = 55.0.
3. Manufacturing PMI: The manufacturing PMI averages 5 parts: new orders 0.30, output 0.25, employment 0.20, delivery times 0.15 and stocks 0.10. Delivery times are inverted so delays lift the index. For example, 54.0, 56.0, 52.0, 48.0 and 50.0 give 52.8.
4. Services activity: The services headline is the 1 question diffusion index for total services activity versus the prior month. It is not a weighted average like manufacturing.
5. Composite output: The composite blends manufacturing output and services activity using gross value added shares, which measure sector importance. For example, 52.0 and 54.0 with 0.25 and 0.75 shares give 13.0 + 40.5 = 53.5.
6. Seasonal adjustment: Adjustment normally uses the X12 statistical programme, which needs 5 years of history. Before that, an alternative method separates seasonal from underlying change using respondent reasons. Underlying data are never revised, but seasonal factors may be revised.
7. Flash estimate: Flash runs the same steps early on about 85% of responses, about 1 week before finals. Finals use the full response set and supersede flash estimates.