The purchasing managers index, or PMI, is a monthly survey that distils how companies say business is going right now into a single number. It is built as a diffusion index from 0 to 100. A reading above 50 means more firms report improving conditions than worsening ones. Below 50 points to contraction.
PMIs exist for manufacturing, services and a combined composite. Because they are quick to publish and based on what managers are experiencing today, markets watch them as early signals for growth, jobs and inflation pressures.
What exactly does the PMI measure?
Compilers ask a panel of purchasing managers whether key aspects of activity are higher, unchanged or lower than the previous month. The questions vary by sector, but common components include:
- Output or business activity
- New orders (often split domestic and export)
- Employment
- Supplier delivery times
- Backlogs of work
- Inventories or stocks of purchases
- Input prices and output prices
Manufacturing PMIs lean on production, orders, jobs, supply chains and inventories. Services PMIs replace factory concepts with business activity and new work, but the logic is the same. A composite PMI combines manufacturing and services, usually weighted by each sector’s share of the economy.
Different organisations publish PMIs for various countries and regions. The broad idea is consistent, though exact questions, panel sizes, weights and seasonal adjustments can differ.
How is the diffusion index built from the survey?
Each question becomes a sub-index using a simple rule: percentage reporting improvement plus half the percentage reporting no change. Mathematically:
Diffusion index = % higher + 0.5 × % same
Example: suppose for new orders, 40% of firms report higher, 45% the same and 15% lower. The sub-index is 40 + 0.5×45 = 62.5. That reads as broad expansion in new orders.
The headline PMI is then a weighted average of selected sub-indices. For manufacturing, output, new orders, employment, supplier delivery times and inventories are commonly used, with published weights that may differ by provider. Services PMIs typically centre on business activity, new business and employment. Supplier delivery times are treated inversely: slower deliveries are read as a sign of capacity strain and strong demand, so they push the index up. That convention can add noise when delays come from supply disruptions rather than busy order books.
Because the PMI captures breadth not magnitude, it tells you how widespread improvement is, not how big the change is in pounds or units. A move from 49 to 51 often signals a meaningful turning point even if output volumes have not yet jumped.
Manufacturing vs services vs composite: what is the difference?
- Manufacturing PMI: Focuses on factories, supply chains and goods demand. It is sensitive to the inventory cycle, global trade and commodity costs.
- Services PMI: Captures activity in services such as finance, transport, IT and consumer-facing sectors. Since services make up a large share of modern economies, this index can be closer to overall GDP trends than manufacturing alone.
- Composite PMI: Blends manufacturing and services into one figure. It is often used as a high-level proxy for monthly GDP momentum.
Many regions publish a flash estimate based on most responses received mid month, followed by a final release with the full sample. Flash numbers can move markets because they land earlier, but they can be revised as more replies arrive.
How traders and investors use PMI releases
PMIs influence expectations for growth, inflation and central bank policy, so they can move currencies, bond yields and equity indices within minutes of release. Common uses include:
- Rates and FX: A stronger than expected PMI can lift government bond yields and the domestic currency if traders see tighter policy as more likely. A weaker print can do the opposite.
- Equities: Cyclical sectors tied to demand for goods and services tend to react to surprises in the relevant PMI. Exporters watch new export orders closely.
- Inflation cues: Rising input and output price sub-indices hint at cost pressures that may show up later in consumer prices.
- Nowcasting: Analysts map composite PMIs to quarterly GDP to estimate growth in between official reports.
On an economic calendar, PMIs sit alongside other top tier releases such as non-farm payrolls. Together, they shape the narrative for the month.
Reading the numbers: levels, changes and sub-indices
There are three layers to sensible interpretation:
- Level around 50: The 50 line marks expansion versus contraction. A shift across 50 often matters more than a small surprise that leaves the index comfortably in expansion or contraction.
- Change from last month: A rising PMI within expansion still suggests momentum is building. A falling PMI within expansion hints that growth is slowing. The reverse holds in contraction territory.
- Sub-index detail: New orders lead output. Employment lags but signals firms’ confidence. Price indices show cost pressures. Supplier delivery times can be tricky: longer delays usually reflect busier conditions, but pure supply shocks can mimic that signal.
Cross country comparisons need care. Differences in survey panels, industry mix and methodology mean a 54 in one economy is not necessarily stronger than a 52 elsewhere. Focus on a country’s own history and turning points rather than the absolute gap with others.
Common pitfalls and limits of the PMI
While PMIs are timely and useful, they are not perfect:
- Diffusion, not volume: The index measures breadth. A high PMI might reflect many firms reporting small improvements, not a large jump in output.
- Seasonal and one off factors: Holidays, strikes or weather can skew monthly readings despite seasonal adjustment. Looking at three month averages can smooth noise.
- Supply chain quirks: The inverse treatment of delivery times can overstate strength when delays come from shortages rather than demand.
- Revisions and provider differences: Flash estimates can be revised. Method details, sector weights and question wording vary by compiler.
For all these reasons, traders rarely rely on a single headline print. They compare PMIs with other surveys and hard data, watch the direction over several months and check whether the price and employment sub-indices tell the same story as the headline.