# Global Macro — China's orders retreat as Europe's output slows and energy risks outlast the bond rally

**31 October 2026 — early Brasília edition. Information cutoff: 05:00 BRT (08:00 UTC).** The Chinese National Bureau of Statistics released its October purchasing managers' surveys at 09:30 Beijing time on 31 October, equivalent to 22:30 BRT on 30 October. European GDP and unemployment releases and US Treasury yields refer to 30 October or earlier completed observations. No November policy decision, physical shipping recovery or future economic outcome is assumed.

China's official manufacturing purchasing managers' index fell from **50.1 in September to 49.3 in October**. The new-export-orders component declined from **50.0 to 48.7**, while production fell from **51.7 to 49.8**. The composite output index slipped from **50.7 to 49.7**. These are diffusion indices, not percentage changes in factory production, and they establish a broadening loss of momentum rather than a measured 0.7% decline in output. The national survey is unusually revealing because the largest manufacturers remained just above the expansion threshold while smaller firms contracted more sharply. [1][2]

Europe supplied a second new observation: Eurostat's preliminary third-quarter estimate put euro-area real GDP growth at **0.4% quarter on quarter**, versus **0.6% in the second quarter**, and EU growth at **0.5%**, versus **0.7%**. The euro area still expanded, but September unemployment stood at **6.5%**, above **6.3%** a year earlier. In the United States, the latest quarterly GDP advance estimate remained strong at **3.4% annualised**, yet the official ten-year Treasury par yield declined again, from **4.98% on 29 October to 4.94% on 30 October**. Growth, credit benchmarks and orders are not moving in one global direction. [3][4][5][7]

The incremental question is whether weakness in Chinese industrial orders will propagate through Asian suppliers and European manufacturers while oil and financing costs remain elevated. Previous editions addressed the Korean memory supplier's contracted AI demand (28 October), the Fed–Canada policy split (29 October), and US GDP versus household income (30 October). The October Chinese survey and the European GDP flash estimate add direct evidence about the *breadth* of industrial demand and the uneven geography of growth. They do not establish a world recession, a new trade-war escalation, or a collapse in Chinese export volumes. [1][3][7]

## Chinese manufacturing: weaker orders, uneven firms

The official manufacturing PMI is a monthly survey of purchasing managers. A reading below 50 indicates that more respondents report deterioration than improvement after the survey's aggregation and weighting, not that output itself contracted by the numerical distance below 50. China's manufacturing index was **49.8 in August**, **50.1 in September** and **49.3 in October**. Its production subindex moved from **50.4** to **51.7** to **49.8** over those same months. October therefore reverses a brief September improvement and shows that the survey's output and demand components both moved below the neutral threshold. [1]

```chart
type: line
title: China official manufacturing PMI, completed monthly surveys
unit: diffusion-index points (50 = neutral)
2026-08 | 49.8
2026-09 | 50.1
2026-10 | 49.3
```

The order breakdown is more informative than the headline. New domestic-and-export orders together fell from **50.5 in September to 49.4 in October**. New export orders fell from **50.0 to 48.7**; imports from **49.2 to 47.0**; and purchasing volumes from **51.0 to 48.9**. The latter two measures suggest weaker procurement demand, but a survey cannot establish whether firms reduced purchases because of final sales, inventory policies, financing constraints or precautionary responses to uncertainty. Finished-goods inventories were **47.5**, while raw-material inventories were **48.4**; these are diffusion measures, not observed stock levels in tonnes or yuan. [1]

```chart
type: line
title: China manufacturing new-export-orders PMI
unit: diffusion-index points (50 = neutral)
2026-08 | 50.1
2026-09 | 50.0
2026-10 | 48.7
```

Large firms reported **50.2**, medium-sized firms **48.4** and small firms **47.5**. The **2.7-point** large–small gap is a difference between survey readings, not a measured productivity or credit-access gap. A plausible explanation is that larger manufacturers have more diversified buyers, stronger balance sheets and greater exposure to high-technology orders, but those characteristics are not directly established by this monthly survey. The statistical agency separately reported that production and new orders in rail, shipbuilding, aerospace, computing and communications equipment remained at or above **53.0**. Textiles, non-metallic minerals and non-ferrous processing were below 50. This sector contrast makes an economy-wide collapse an implausible reading of the evidence. [1][2]

```chart
type: bar
title: China manufacturing PMI by firm size, October 2026
unit: diffusion-index points (50 = neutral)
Large firms | 50.2
Medium firms | 48.4
Small firms | 47.5
```

The non-manufacturing business-activity index also declined, from **50.2 to 49.7**. Construction fell from **50.3 to 48.0**, while services eased to **50.0** from **50.2**. Non-manufacturing new orders were **44.4**, including construction new orders of **40.0**. These figures reinforce a demand-side concern but should not be converted into a quarterly GDP forecast: survey composition, holiday timing and actual output weights differ. The NBS explicitly cited the National Day holiday and advance release of some demand as possible influences, an alternative to interpreting all of the change as persistent structural deterioration. [1][2]

## Europe: positive aggregate GDP conceals a fragile distribution

Eurostat's preliminary flash estimate is based on incomplete national data. Euro-area GDP rose **0.4% quarter on quarter** in Q3, down from **0.6%** in Q2; EU GDP rose **0.5%**, down from **0.7%**. The euro area's year-on-year growth rate was **1.3%**, and the EU's **1.5%**. The quarterly figures are *not annualised*, unlike the US **3.4%** GDP release; direct numerical comparison of the rates would exaggerate European underperformance. A consistent cross-region comparison requires harmonised growth conventions and sector composition. [3][7]

```chart
type: line
title: Euro-area quarterly real GDP growth, preliminary flash
unit: % quarter on quarter, not annualised
2026-Q2 | 0.6
2026-Q3 | 0.4
```

The country distribution matters for transmission. Ireland registered **+5.5% quarter on quarter** and should not be treated as representative of household demand throughout Europe, given the sensitivity of Irish GDP to multinational activity. Spain and the Netherlands each recorded **+0.6%**; Germany **+0.2%**; France **−0.1%**; and Lithuania **−1.0%**. These are preliminary observations with different national industrial structures and revision risks. A European aggregate can grow even as a subset of countries contracts. [3]

The labour-market observation also qualifies the growth narrative. Euro-area unemployment was **6.5% in September**, unchanged from August but above **6.3% in September 2025**. Eurostat estimated **11.555 million** unemployed persons in the euro area and **13.773 million** across the EU. The monthly number of unemployed increased by **122,000** in the euro area; seasonal adjustment and survey methods mean that the change should not be read as the exact number of jobs lost. Youth unemployment reached **15.4%** in the euro area, against **15.1%** in August. Positive GDP and a weakening employment margin can coexist when productivity, hours, sector composition and participation change. [4]

The ECB held its deposit rate at **2.50%** on 29 October. September euro-area HICP inflation was **3.8% year on year**, up from **3.2% in August**, with energy contributing **1.71 percentage points** to the annual rate. The October flash inflation estimate was *not yet released* at the information cutoff. Thus, stronger-than-expected Q3 growth cannot be used to assert that inflation had normalised; nor can September's price level prove the full effect of the latest energy shock on future wages or credit. The relevant tension is positive but uneven real growth alongside still-elevated prices and financing costs. [6][8][9]

## The US yield decline is a price signal, not proof of cheap credit

US Treasury's official indicative par yield curve gives a sequence of **5.05% (28 October), 4.98% (29 October) and 4.94% (30 October)** for the ten-year maturity. The two-year fell from **4.67% to 4.61% to 4.56%**; the thirty-year from **5.37% to 5.32% to 5.31%**. The two-year-to-ten-year spread moved from **38 basis points on 28 October**, to **37** and then **38**. Falling benchmark yields did not produce a decisive flattening or inversion of this portion of the curve. [5]

```chart
type: line
title: US Treasury ten-year indicative par yield
unit: % at completed daily observations
2026-10-28 | 5.05
2026-10-29 | 4.98
2026-10-30 | 4.94
```

The decline could reflect weaker expected activity, changed policy-rate expectations, duration demand, reduced inflation compensation or shifting risk premia. The observed curve cannot separate those explanations. In particular, an exporter or small manufacturer seeking working capital does not borrow at the sovereign par yield: bank margins, collateral, currency exposure and lending standards intervene. The Federal Reserve held its policy target at **3.75–4.00%** on 28 October. Neither the unchanged overnight target nor a four-basis-point daily ten-year decline establishes an easing in credit availability for firms facing weaker orders. [5][10]

US activity remains a contrary observation to a global-demand-collapse thesis. The BEA advance estimate reported **3.4% annualised** third-quarter real GDP growth, compared with **2.2%** in Q2. Its September income accounts nevertheless showed **−0.3%** monthly real disposable personal income and essentially flat real consumption. Quarterly output and monthly household purchasing power are different objects. The latest China and European readings make the distribution of demand across countries and firms more important than any single aggregate growth figure. [7][15]

## Oil, trade routes and the risk of a two-sided shock

The World Bank's 28 October Commodity Markets Outlook projects a **32% increase in energy commodity prices in 2026** and **10% in food commodity prices in 2026**, followed by another **7%** for food in 2027. It forecasts Brent averaging **$91 a barrel in 2026** and **$79 in 2027** under its baseline. These are annual projections, not Friday settlement prices. Its adverse scenario of **$105 Brent in 2027** is conditional on a combination of conflict and supply disruptions, not an observed price or probability-weighted central forecast. [11]

Reports on 30 October described renewed threats to Gulf shipping and energy infrastructure and divergent oil settlement quotations across providers and contract references. A reported incident, an actor's claim of responsibility, verified cargo throughput and a futures settlement are four different evidentiary objects. Without independently comparable vessel volumes, insurance quotations and loading data, the analysis cannot claim a measured fresh loss of global oil supply from Friday's headlines. The more defensible conclusion is that geopolitical uncertainty continues to create an upside risk to import costs even as some goods-demand indicators weaken. [11][12]

This is the mechanism behind a possible two-sided shock. If weaker Chinese orders translate into fewer intermediate-goods purchases, suppliers in Southeast Asia and commodity exporters may face lower volumes or weaker bargaining power. If Gulf supply or shipping costs remain constrained at the same time, those economies may pay more per unit of imported energy despite selling fewer manufactured goods. Margins narrow, currency pressures may raise local-currency fuel costs, and governments face difficult choices between subsidies, fiscal restraint and monetary tightening. The October releases do not establish that these downstream effects have already occurred.

```flow
China new orders weaken → Asian intermediate-goods demand may slow → supplier revenue and working-capital needs change → credit spreads and investment decisions respond
Gulf shipping risk persists → energy freight and insurance may rise → importer costs and consumer prices face pressure → fiscal and monetary choices become harder
Europe grows unevenly → national employment and purchasing power diverge → demand for imports varies by market → exporters face uneven order books
```

```map
title: Geographic channels under observation
China | October manufacturing and export-order diffusion below 50 | Survey, not measured export-volume decline
Southeast Asia | Potential intermediate-goods demand transmission | No October cross-border volume estimate established
Euro area | Q3 GDP positive but slower; employment margin weaker | Preliminary GDP and September unemployment
North America | US output strong while benchmark yields ease | Credit spreads and borrower access not inferred
Gulf and import-dependent economies | Energy-route and import-cost exposure | Physical throughput change not independently quantified
```

## Cross-asset and fiscal reading: different clocks, different claims

| Observation | Latest verified vintage | Measure | Analytical constraint |
|---|---|---|---|
| China manufacturing PMI | Oct: 49.3; Sep: 50.1 | Survey diffusion points | Not an industrial-production growth rate |
| China new export orders | Oct: 48.7; Sep: 50.0 | Survey diffusion points | Not actual customs export value |
| China large / medium / small firms | Oct: 50.2 / 48.4 / 47.5 | Size-specific diffusion points | Not direct financing or productivity measures |
| Euro-area GDP | Q3: +0.4%; Q2: +0.6% | Real q/q, preliminary | Not annualised; may be revised |
| Euro-area unemployment | Sep: 6.5%; year earlier: 6.3% | Seasonally adjusted rate | Not a direct payroll count |
| US GDP | Q3: +3.4%; Q2: +2.2% | Real annualised quarterly rate | Different convention from Eurostat |
| US Treasury 10Y | 30 Oct: 4.94%; 29 Oct: 4.98% | Indicative daily par yield | Not corporate loan rate or term-premium estimate |
| ECB deposit / Fed target | 2.50% / 3.75–4.00% | Policy settings, 29/28 Oct | Different mandates and currency areas |
| World Bank Brent baseline | 2026: $91; 2027: $79 | Annual forecast, USD/barrel | Not spot or futures settlement |

Credit conditions and currency channels deserve scrutiny precisely because the indicators are not interchangeable. The Treasury curve prices dollar duration; the ECB and Fed set policy instruments; the Chinese PMI describes reported operating conditions; the World Bank provides conditional annual commodity forecasts. Their coexistence does not identify a single global shock. A weakening order book can lower demand for credit even while a higher energy bill raises the *need* for working capital. Actual loan volumes, rates charged to firms, delinquencies and trade-finance terms would be required to determine which force dominates.

Portfolio evidence is similarly incomplete. The Investment Company Institute's **$7.901 trillion** in US money-market fund assets for the week ending 28 October, a **$43.29 billion** weekly decline, is a previously released US fund-asset observation, not a real-time measure of international money leaving China or Europe. It cannot be combined with Friday equity prices to infer beneficial ownership, forced sales or global capital flight. Equity-sector declines described in Friday's market coverage may signal a reassessment of technology earnings, but share-price losses are valuation changes, not automatically cash transfers to another region. [13]

Fiscal policy is part of the transmission but not a measured new spending decision. Governments facing higher fuel costs may reduce taxes, subsidise energy, expand targeted transfers or tolerate household real-income losses. Each option has different distributional and financing consequences. In highly indebted importers, a stronger dollar or higher local risk premium can offset a small decline in the US Treasury benchmark. No new cross-country fiscal package is attributed to the Chinese PMI release; such a claim would require published budgets or decisions.

## What would change the assessment

The baseline interpretation is **uneven global expansion with weakening Chinese order diffusion, positive but slower European output, robust measured US activity and persistent energy-cost risks**. It has medium confidence: the Chinese and European releases are primary and timely, but one month's survey and an early quarterly estimate are insufficient to determine persistence or causality.

A more adverse path requires several confirming observations: Chinese manufacturing and export orders remain below 50 in November; customs export volumes and industrial production weaken; European employment and orders deteriorate beyond the preliminary GDP vintage; and energy-route constraints show up in verified shipments or sustained delivered prices. Under those conditions, manufacturers could face lower volumes and high input costs simultaneously. This is a conditional scenario, not a forecast of war or recession.

A less adverse path would be supported by a rebound in Chinese orders after holiday effects dissipate, stronger customs shipments, resilient European final demand, and verified easing in energy transport and insurance costs. The high-technology strength reported by China's statistical agency and positive Q3 European GDP already provide contrary evidence to a uniformly contracting world economy. [1][2][3]

The next decisive observations are China's customs and industrial-production data, the euro-area October inflation flash on **4 November**, US trade data on **4 November**, the Eurostat GDP revision scheduled for **13 November**, and evidence on actual credit and shipping costs. Forecasts and market narratives should be revised against those observations rather than repeated with a new date. [3][9][14]

## References

1. National Bureau of Statistics of China, *October 2026 China Purchasing Managers' Index*, 31 October 2026: https://www.stats.gov.cn/sj/zxfbhjd/202610/t20261031_1965718.html
2. National Bureau of Statistics of China, *Interpretation of October 2026 PMI*, 31 October 2026: https://www.stats.gov.cn/sj/sjjd/202610/t20261031_1965719.html
3. Eurostat, *GDP up by 0.4% in the euro area and 0.5% in the EU*, 30 October 2026: https://ec.europa.eu/eurostat/web/products-euro-indicators/w/2-30102026-ap
4. Eurostat, *Euro area unemployment at 6.5%*, 30 October 2026: https://ec.europa.eu/eurostat/en/web/products-euro-indicators/w/3-30102026-bp
5. US Treasury, *Daily Treasury Par Yield Curve Rates*, observations 28–30 October 2026: https://home.treasury.gov/resource-center/data-chart-center/interest-rates/TextView?field_tdr_date_value=2026&type=daily_treasury_yield_curve
6. European Central Bank, *Monetary policy decisions*, 29 October 2026: https://www.ecb.europa.eu/press/pr/date/2026/html/ecb.mp261029~9a493adf17.en.html
7. US Bureau of Economic Analysis, *GDP Advance Estimate, Third Quarter 2026*, 29 October 2026: https://www.bea.gov/news/2026/gdp-advance-estimate-3rd-quarter-2026
8. Eurostat, *Annual inflation up to 3.8% in the euro area*, 16 October 2026: https://ec.europa.eu/eurostat/en/web/products-euro-indicators/w/2-16102026-ap
9. Eurostat, *Release calendar for euro indicators*, accessed 31 October 2026: https://ec.europa.eu/eurostat/en/news/euro-indicators/release-calendar
10. Federal Reserve, *FOMC statement*, 28 October 2026: https://www.federalreserve.gov/newsevents/pressreleases/monetary20261028a.htm
11. World Bank, *Commodity Markets Outlook*, 28 October 2026: https://www.worldbank.org/en/news/press-release/2026/10/28/commodity-markets-outlook-october-2026
12. Reuters, *Oil jumps on fears of escalation in Iran*, 30 October 2026 (reported events and settlements; not independently verified throughput): https://www.reuters.com/business/energy/oil-rises-supply-concerns-houthis-attack-saudi-arabia-2026-10-30/
13. Investment Company Institute, *Money Market Fund Assets*, release 29 October 2026: https://www.ici.org/research/stats/mmf
14. US Bureau of Economic Analysis, *Release Schedule*, accessed 31 October 2026: https://www.bea.gov/news/schedule
15. US Bureau of Economic Analysis, *Personal Income and Outlays, September 2026*, 29 October 2026 (real DPI −0.3%, real PCE 0.0% m/m): https://www.bea.gov/news/2026/personal-income-and-outlays-september-2026