ISM Services PMI (Non-Manufacturing Index)
The "other half" that is 70% of the US economy: the factory line gets the headlines, services IS the economy — and both sides of 50 tell a story.
Monthly (10 AM ET, 3rd business day)
What it is
The ISM Services PMI (formerly the Non-Manufacturing Index), published by the Institute for Supply Management, is a monthly diffusion survey of purchasing managers across service industries (production, new orders, employment, prices, deliveries and more). Services represent roughly 70% of US output and employment — the manufacturing PMI describes the factory cycle, while ISM Services describes the body of the US economy. 50 is the boom-bust line; the prices-paid component is the core evidence on services-inflation stickiness, new orders the demand lead, and employment a cross-check on NFP.
Release schedule
| Item | Details |
|---|---|
| • | Frequency: monthly (covering the current month) |
| • | Release: 3rd business day of the month at 10:00 ET (23:00/00:00 Beijing time, DST-dependent) — two days after the manufacturing PMI |
| • | Contents: composite index plus new orders, employment, prices paid, supplier deliveries (diffusion indices, 50 = flat) |
| • | Frame: a services purchasing-manager survey — an independent survey from S&P Global's services PMI |
| • | Relation to ISM Manufacturing: same institute, different surveys — divergence reflects the manufacturing/services cycle mismatch |
Why it matters
Three reasons it deserves its own page: scale — services dominate GDP and employment, so "is the US expanding" is mostly a services question; stickiness — core services inflation (ex-housing) is the Fed's hardest problem, and prices-paid points straight at service costs; timing — the 3rd business day lands before most hard data, making it the month's first survey evidence. Historical monthly pulses in this print routinely steer the day's rate expectations and equity direction.
Impact across assets
Typical impacts (using a hot composite with rising prices-paid):
| Asset | Typical impact |
|---|---|
| US equities | Strong-but-not-overheating → earnings and soft-landing tailwind; strong enough to imply "overheating → hawkish" → growth pressured — calibration matters |
| US Dollar Index | Strong → firmer (growth differential plus rate expectations); weak → softer — a surprise contraction below 50 hits harder |
| Gold | Strong → pressured (rate expectations up); weak → benefits — same structure as the manufacturing PMI with larger elasticity given the scale |
| Crypto | Strong → tighter liquidity expectations → pressured; weak → rebounds — the day's vote on soft-landing vs recession |
| Treasuries | Strong + hot prices-paid → yields up (2s and 10s together); weak → down — a key front-end input |
How to read it
The standard read:
| Dimension | How to read it |
|---|---|
| The 50 line | Above 50 = expansion, below = contraction; deep-contraction territory in the mid-40s has historically coincided with recession zones |
| New orders | The demand lead — divergence between new orders and the composite often leads the composite's inflection |
| Prices paid | The services-stickiness evidence: persistently high = "inflation won't fall easily"; fast falls = supply-side relief |
| Employment | Cross-validates NFP's services hiring — when survey and hard data disagree, persistence decides who to trust |
The advanced frame: read ISM Services beside ISM Manufacturing — "manufacturing contracting + services expanding" is the structural norm; both contracting = a broad slowdown warning; services weakening alone = relief for services inflation (falling prices-paid). The components carry more information than the composite: prices-paid plus new orders nearly decide the day's narrative.
Limitations & common mistakes
- Treating it as "manufacturing, part two": different scale and logic — services decide the economy's body, manufacturing its cycle amplitude; mismatch is the norm.
- Reading only the composite: prices-paid plus new orders carry the narrative — a 52 composite with prices-paid at multi-year highs is an inflation story, not an expansion story.
- Confusing it with S&P Global Services: two independent surveys (different questionnaires, firm mixes, seasonal adjustment) — divergence is normal; cross-validate, don't "revise" one with the other.
- Misreading supplier deliveries: a low score (faster deliveries) means demand weakness in manufacturing, but was misread as demand collapse during supply-chain normalization — component semantics need context.
- Overreacting to one diffusion print: diffusion indices measure the share of firms improving, not magnitude — large single-month swings are often sample noise.
Related macro data
How it links to other macro data:
- With ISM Manufacturing: the same-institute "body and cycle" pair — only when both agree can you claim a broad slowdown or expansion. pmi
- With core PCE: prices-paid is the survey evidence on services stickiness — mutually corroborating with the PCE services trend. pce
- With the Fed decision: the services-growth-plus-services-inflation combination is a direct input to FOMC's "resilience vs stickiness" tradeoff. fed-rate
Symbols most sensitive to ISM Services PMI
Symbol pages that list this data as a factor to watch:
FAQ
Q When is ISM Services released?
The 3rd business day of each month at 10:00 ET (23:00/00:00 Beijing depending on DST) — two days after the manufacturing PMI. Holiday weeks shift.
Q Why does services matter more than manufacturing by weight?
Services are ~70% of US GDP and employment — "is the economy expanding" is mostly a services question. Manufacturing is roughly a tenth of output; its importance comes from cycle sensitivity, not size.
Q How do I use the 50 line?
Above 50 expanding, below contracting — with direction and distance from 50 mattering: chop in the 48–52 band is tepid; a surprise into the mid-40s has historically been recession-discussion territory.
Q Why does prices-paid matter so much?
It is the survey's cost-perception diffusion index — service costs (mostly labor) are sticky, so persistently high prices-paid is the core evidence that services inflation will not fall easily, feeding directly into the Fed's tradeoff.
Q When ISM and S&P Global disagree, who wins?
Two independent surveys with different questionnaires and coverage — divergence is normal. Markets price off ISM (longer history, fuller components), but the divergence itself informs; settle it with hard data (NFP, retail) rather than survey-vs-survey.
Turn MT5 / MT4 into a Tray Tool
Check quotes, manage positions, and hide in one click.
Download Camovia TrayThis page is educational content about macroeconomic data. It is not investment advice. Macro impacts involve multiple interacting factors — always combine them with your own risk management.