# Global Macro — Korea's chip-export boom meets China's order contraction and an energy import squeeze

**1 November 2026 — Brasília edition. Initial data cutoff: 05:00 BRT (08:00 UTC); updated after the official OPEC+ communiqué dated 1 November.** The Korean trade ministry's October merchandise-trade release, dated 1 November in Seoul, was available for the morning cutoff. US and European market prices refer to their last completed 30 October sessions, not to a fictitious Sunday settlement. The seven-country OPEC+ decision is dated 1 November and is separate from observed oil production or shipping volumes. [1][16]

South Korea reported **$115.28 billion in October merchandise exports, up 93.7% year on year**, and **$72.78 billion in imports, up 36.0%**. Its merchandise surplus was **$42.50 billion**. Semiconductor exports reached **$56.44 billion, up 259.4%**, close to half of all exports by value. Yet exports were lower than September's **$120.94 billion** (October 1 primary release), even as the year-on-year growth rate accelerated: a calendar comparison and an unusually strong base-sensitive technology cycle are not the same as a month-on-month acceleration in physical shipments. The ministry reports **76% growth in exports per working day**, below the unadjusted 93.7% annual increase. [1][15]

This new physical trade-value release changes the question raised by the three previous Global Macro editions. The 29 October edition examined a Federal Reserve pause and a Canadian rate increase; 30 October compared robust US GDP with weaker household purchasing power; 31 October documented China's manufacturing and export-order PMI falling below 50 and Europe's slowing but positive GDP growth. The 1 November evidence establishes a sharp divergence **within Asian trade**: a chip-intensive exporter records exceptionally high sales values while China's broader purchasing-manager survey reports contracting order diffusion. Neither result cancels the other, because they measure different products, units, survey populations and observation concepts. [1][2][3][4][5]

The key systemic issue is not whether all world trade is booming or collapsing. It is whether demand for compute infrastructure is sufficiently concentrated to support exceptional semiconductor revenue while the rest of industrial demand faces higher imported-energy costs, weak Chinese orders and tighter financing. A second test concerns the quality of the Korean surplus: record export receipts coexist with a larger energy-import bill and a **$7.35 billion** decline in the monthly surplus from September's **$49.85 billion** in the 1 October official release. That is an arithmetic change in a trade balance using the dated original releases, not evidence of portfolio outflows or a deterioration in the current account. [1][15]

## What the Korean trade record actually measures

Korea's October export value is the second-highest monthly figure in the ministry's historical ranking, after September's **$120.94 billion** in the original 1 October release. October therefore cannot be described as a new all-time monthly export record. The ten-month cumulative value reached **$930 billion**, but the ministry's expectation of surpassing $1 trillion in November remains a projection, not a completed transaction. The cumulative value cannot be annualised mechanically without accounting for the volatile chip cycle and holiday calendar. [1][15]

| Korean merchandise trade | October 2026 observation | Comparison / definition | Analytical caution |
|---|---:|---|---|
| Exports | $115.28bn; +93.7% y/y | Nominal customs value | Below September's $120.94bn |
| Exports per working day | +76% y/y | Calendar-adjusted value growth | Not a physical-volume index |
| Semiconductors | $56.44bn; +259.4% y/y | Product export receipts | Price and quantity both matter |
| Non-semiconductor exports | +34% y/y | Ministry aggregate | Not an independent world-demand index |
| Imports | $72.78bn; +36.0% y/y | Nominal merchandise value | Includes energy and capital equipment |
| Energy imports | $16.49bn; +63.1% y/y | Subset of imports | Strong price effect |
| Trade balance | +$42.50bn | Exports minus imports | Not the current-account balance |

The distinction between a nominal value and an observed quantity is decisive. Export receipts can rise because unit prices increase, more units are shipped, or the product mix shifts toward higher-value components. The ministry explicitly attributes semiconductor strength to both high prices and volumes, but does not publish a single aggregate volume index in its release that would justify treating the **259.4%** nominal semiconductor increase as equivalent growth in chip units or computing capacity. Nor does a customs value reveal the ultimate end user of the chips. [1]

```chart
type: bar
title: Korean October 2026 merchandise exports by selected product
unit: USD billion, nominal; selected categories only
Semiconductors | 56.44
Petroleum products | 6.92
Automobiles | 6.55
Computers | 5.85
Petrochemicals | 4.00
Electrical equipment | 1.99
```

The chart is deliberately not a decomposition of all exports. Selected categories do not exhaust the total, and computers, electrical equipment and semiconductors are distinct customs categories rather than a measured common 'AI' basket. The semiconductor value alone is approximately **49.0%** of total October exports (56.44 divided by 115.28); this is a calculated share of merchandise export *value*, not the fraction of Korea's employment, national income or productive capital tied to AI. The non-chip segment also expanded: autos rose **18.1%**, electrical equipment **74.0%** and cosmetics **57.5%** by value. Ship exports, however, fell **43.6%**, with the ministry identifying a large one-off 2025 offshore-platform delivery as a base effect. [1]

## AI hardware demand is real; its durability remains unproved

The October semiconductor number strengthens the evidence for substantial near-term demand for memory and related components. Computer exports of **$5.85 billion**, up **496.6%** year on year, are consistent with a powerful data-centre and server investment cycle. Electrical-equipment exports of **$1.99 billion**, up **74.0%**, are associated by the ministry with demand for transformers, wires and cables for AI data centres. These are recorded sales values, more concrete than a corporate capacity announcement. But they still do not measure whether installed data centres are fully utilised, whether customers earn returns above their cost of capital, or whether future orders are contractually guaranteed. [1]

```chart
type: bar
title: Korea merchandise exports in the last two completed months
unit: USD billion, nominal; ministry monthly totals
2026-09 | 120.94
2026-10 | 115.28
```

The year-on-year acceleration from September's officially reported **83.5%** to October's **93.7%** can coexist with this sequential decline. October had more working days than October 2025 because of the timing of the Chuseok holiday; the official **76% per-working-day** comparison removes part of that distortion but does not adjust for changes in chip prices or product composition. Claims that the global quantity of manufactured goods nearly doubled would therefore be unwarranted. [1][15]

There are two opposing economic forces. Higher chip prices improve exporter revenue and potentially finance new fabrication equipment and R&D; at the same time, high input prices and concentration in a handful of customers can increase exposure to an eventual correction in cloud-capital spending. The first force is directly consistent with customs receipts. The second is a conditional balance-sheet risk, not evidence that an AI investment crash has begun. The Korean customs release provides neither corporate debt maturities nor client-level project cash flows; those are necessary to quantify fragility.

A practical falsifier would be a sustained expansion of non-chip exports, better regional industrial orders and stable semiconductor receipts despite normalising unit prices. Conversely, falling chip export values together with reduced machinery imports, weaker Korean equipment orders and deteriorating credit spreads would support a cyclical slowdown. One month of record receipts cannot settle either trajectory.

## Energy imports: higher values can conceal lower physical volumes

Korea's October **energy imports rose 63.1% in value to $16.49 billion**. The ministry reports crude-oil import **unit prices up 62.0%** while crude-oil import **volumes fell 6.3%**. Consequently the crude import bill rose **51.9% to approximately $10.7 billion**. This is a direct example of a terms-of-trade burden: an importer can pay more for fewer barrels. It is not a measurement of a new 62% change in global Brent spot prices on Sunday. [1]

```chart
type: bar
title: Korea crude-oil imports, October 2026 versus October 2025
unit: percent year on year; distinct price and volume measures
Import unit price | 62.0
Physical import volume | -6.3
Import value | 51.9
```

Petroleum-product exports grew **80.0% by value** even as exported volumes fell **7.5%**; petrochemical export values grew **27.5%** while their volumes fell **6.8%**. Higher nominal trade turnover is therefore not evidence of rising physical throughput across the energy complex. Some producers capture higher selling prices while refineries, transport operators and final consumers confront more expensive feedstocks and freight. The distribution of the gain depends on pricing power, contracts, energy intensity and currency denomination. [1]

The International Energy Agency's **14 October** Oil Market Report estimated that global observed oil stocks fell **10 million barrels in September**, leaving stocks **509 million barrels below end-February**. It estimated September Strait of Hormuz flows of **10.9 million barrels a day**, a recovery relative to earlier disrupted conditions but not a guarantee of safe November deliveries. These are IEA estimates for *September* and should not be presented as measurements of the maritime situation on 1 November. The World Bank's **28 October** forecast put average Brent at **$91/bbl in 2026** and **$79/bbl in 2027**, conditional on no major intensification of supply conflicts. A forecast average is not a current quote. [9][10]

At their **1 November** virtual meeting, **Saudi Arabia, Russia, Iraq, Kuwait, Kazakhstan, Algeria and Oman** decided to maintain **September 2026 required production levels for December 2026**, according to the new official OPEC+ statement. The previous **4 October** communiqué had applied September requirements to November. The seven countries scheduled their next review for **29 November**. This is a verified production-requirement decision, **not a measurement of December physical output, November ship loadings or a spot oil price**; route security and compliance still determine deliveries. [12][16]

## China, Europe and the geography of uneven demand

China's official October manufacturing PMI fell to **49.3**, with new-export-orders at **48.7** and the non-manufacturing activity index at **49.7**. These diffusion readings describe the balance of survey responses and do not measure an 0.7% fall in national industrial output. Korean exports to China nevertheless reached **$26.98 billion**, up **133.6%** year on year. The apparent contradiction disappears once product mix and period definitions are separated: Chinese firms can buy expensive imported semiconductors for selected production chains while a larger share of domestic manufacturers reports weaker overall orders. [1][2]

```chart
type: bar
title: Korea exports by selected destination, October 2026
unit: USD billion, nominal; non-overlapping destinations
China | 26.98
United States | 24.90
ASEAN | 21.33
European Union | 8.10
Latin America | 3.00
Middle East | 1.67
```

The US destination rose **185.8%** and ASEAN **128.0%** by value; EU shipments grew **56.4%**. Yet exports to Latin America fell **37%** and to the CIS fell **32%**. These smaller destinations matter because a single global export headline hides asymmetric demand and market access. The Korean release does not provide enough physical-volume detail by destination to allocate these changes between prices, mix, calendar and quantities. Its Middle East export increase of **11%** also does not demonstrate that regional shipping constraints have disappeared. [1]

Europe's real GDP grew **0.4% quarter on quarter** in Q3, down from **0.6%** in Q2. The US Q3 advance estimate grew **3.4% at an annualised quarterly rate**; these conventions must not be compared as if they were identical. US September real disposable income fell **0.3% month on month**, while real consumption was essentially unchanged on a rounded basis. Euro-area unemployment was **6.5% in September**, up from **6.3% a year earlier**, despite the region's positive GDP growth. This is a labour-market rate rather than a direct count of lost jobs. The mixed demand evidence makes an indiscriminate 'global boom' interpretation of Korean exports implausible, but also contradicts a uniform global recession. [3][4][5][14]

## Monetary policy, credit, currencies and the distribution of gains

The Federal Reserve maintained its policy range at **3.75–4.00%** on **28 October**, while the ECB kept its deposit rate at **2.50%** on **29 October**. The US Treasury's indicative ten-year yield stood at **4.94% on 30 October**, below the preceding day's **4.98%**. Those are policy settings and sovereign benchmarks, not the rates actually paid by Korean chip firms, European industrial suppliers or emerging-market importers. Loan spreads, maturity schedules and currency hedges determine whether lower benchmark yields translate into cheaper project finance. [6][7][8]

A dollar-priced chip boom can lift Korean exporters' revenues without creating equal gains for Korean households. Imported energy costs rise for industrial users; some export-sector workers may benefit from overtime and hiring; consumers may face higher transport and electricity bills. Currency changes can amplify or dampen both sides, but this release alone does not establish a specific 1 November won or dollar exchange-rate move. Equally, a record merchandise surplus is not proof of a capital inflow: it excludes services, investment income and financial-account transactions.

The Investment Company Institute's US money-market fund asset series, last released on **29 October**, is a useful liquidity observation but not a bilateral flow from US investors into Korea. Weekly asset changes include subscriptions, redemptions and valuation effects. Korean export receipts are cross-border goods transactions, not a map of beneficial ownership of Korean equities or US technology firms. Fiscal policy may respond to expensive fuel through tax reductions, transfers or subsidies, but no new Korean or global fiscal package can be inferred from the customs data. [11]

## Food, logistics and the wider systemic exposure

The FAO Food Price Index averaged **136.0 in September**, up **1.5%** from August, with its next scheduled release on **6 November**. The World Bank expects food commodity prices to increase about **10% in 2026** and a further **7% in 2027** under its baseline. Those are distinct measures: the FAO's observed monthly international price index and the Bank's annual forecast. Neither translates mechanically into a household grocery bill in an importing country, where tariffs, exchange rates, subsidies, transport costs and domestic harvests intervene. [10][13]

Higher Korean electrical-equipment exports suggest demand for power infrastructure alongside chips, but they do not establish commissioned megawatts or grid connections. Rising oil-import costs can squeeze the same industrial supply chains that benefit from AI investment. For poorer energy-importing countries, fuel and fertilizer costs may transmit into food prices, real wages and fiscal trade-offs. The World Bank warns of this risk; no unobserved famine, new pandemic or imminent military escalation is asserted as a fact. [1][10]

```flow
Global AI infrastructure spending → Korean semiconductor and equipment export receipts → supplier investment and concentrated customer exposure
Energy-route constraints and crude prices → higher Korean import unit costs → margins and consumer prices → monetary and fiscal trade-offs
Chinese weak order diffusion → uncertain intermediate-goods demand → Asian supplier volumes → credit and employment risk
```

```map
title: Regional transmission and contrary evidence
South Korea | Chip receipts and costly imported crude | Customs values do not reveal capacity utilisation
China | Weak broad PMI, strong Korean chip purchases | Different survey and trade concepts
United States | AI demand and strong Q3 GDP | September real income softened
Europe | Positive but slower GDP and Korean imports | Financing and energy costs remain relevant
Latin America | Korean exports fell 37% by value | No volume decomposition or region-wide recession claim
Middle East | Shipping and energy supply exposure | September IEA flow estimate is not a November observation
```

## Tests for the next trading week

**Baseline, medium confidence:** world activity is uneven, with concentrated AI-related merchandise receipts and weak Chinese order diffusion coexisting with expensive energy imports. The strongest new observation is Korea's official October product and destination breakdown. The greatest uncertainty is whether nominal revenue gains reflect lasting physical demand and sustainable profitability or mainly high prices and holiday/base effects.

**Adverse conditional path:** Korean semiconductor export values and orders weaken over successive releases; Chinese customs volumes and November PMIs confirm broader demand weakness; oil shipping disruptions produce independently observed delivery shortfalls; credit spreads rise for energy importers. Such a combination could compress manufacturing margins and worsen food affordability. This is not a prediction of war or a declaration of recession.

**Resilience path:** chip demand stays strong after the holiday comparison normalises, non-chip volumes broaden, Chinese orders rebound, and verified energy deliveries lower landed costs. The ministry's **34%** increase in non-chip export value and positive European GDP are already evidence against a one-sector-only or worldwide-contraction narrative. The next tests are Korea's November trade release, China's customs and PMI data, US labour and manufacturing releases, the FAO update on **6 November**, and verified compliance with OPEC+'s new December production requirements ahead of its **29 November** review. [1][2][3][13][16]

The unresolved quantities are material: Korea's product-level constant-price export volumes, customer concentration and project returns in AI; independent current Gulf cargo and insurance data; country-level loan spreads; and ownership-resolved international capital flows. Until those series are available, the trade-value divergence should be interpreted as a measured distribution of receipts, not as proof of a durable redistribution of productive power.

## References

1. Ministry of Trade, Industry and Resources (Republic of Korea), October 2026 exports and imports, 1 November 2026: https://www.motir.go.kr/kor/article/ATCL3f49a5a8c/172360/view
2. National Bureau of Statistics of China, October 2026 purchasing managers' indices, 31 October 2026: https://www.stats.gov.cn/sj/zxfbhjd/202610/t20261031_1965718.html
3. Eurostat, preliminary Q3 2026 GDP, 30 October 2026: https://ec.europa.eu/eurostat/web/products-euro-indicators/w/2-30102026-ap
4. US Bureau of Economic Analysis, advance Q3 2026 GDP, 29 October 2026: https://www.bea.gov/news/2026/gdp-advance-estimate-3rd-quarter-2026
5. US Bureau of Economic Analysis, September 2026 Personal Income and Outlays, 29 October 2026: https://www.bea.gov/news/2026/personal-income-and-outlays-september-2026
6. Federal Reserve, FOMC statement, 28 October 2026: https://www.federalreserve.gov/newsevents/pressreleases/monetary20261028a.htm
7. European Central Bank, monetary policy decisions, 29 October 2026: https://www.ecb.europa.eu/press/pr/date/2026/html/ecb.mp261029~9a493adf17.en.html
8. US Treasury, daily Treasury par yield curve, observations through 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
9. International Energy Agency, Oil Market Report, 14 October 2026: https://www.iea.org/reports/oil-market-report-october-2026
10. World Bank, Commodity Markets Outlook, 28 October 2026: https://www.worldbank.org/en/news/press-release/2026/10/28/commodity-markets-outlook-october-2026
11. Investment Company Institute, Money Market Fund Assets, 29 October 2026: https://www.ici.org/research/stats/mmf
12. OPEC, seven-country production-requirement statement, 4 October 2026: https://www.opec.org/pr-detail/1891616-4-october-2026.html
13. FAO, Food Price Index for September 2026, published 2 October 2026: https://www.fao.org/worldfoodsituation/foodpricesindex/en/
14. Eurostat, September 2026 euro-area unemployment, 30 October 2026: https://ec.europa.eu/eurostat/en/web/products-euro-indicators/w/3-30102026-bp
15. Ministry of Trade, Industry and Resources (Republic of Korea), *September 2026 Exports and Imports*, 1 October 2026; exports $120.94bn (+83.5% y/y), surplus $49.85bn: https://www.motir.go.kr/kor/article/ATCL3f49a5a8c/172254/view
16. OPEC, *Seven OPEC+ countries reaffirm market stability and maintain September required production for December*, 1 November 2026: https://www.opec.org/pr-detail/620-1-november-2026.html