How commodities move: Who ships what to whom

Understanding Commodities / Module 7: How commodities move Part 1 of 5

Who ships what to whom

Only some commodities travel. Whether a material crosses a border depends on its value per tonne, on where deposits and climates sit, and on how hard governments protect home producers. This part measures how much of each commodity is traded, separates the view by weight from the view by value, and follows the few corridors that carry most of the tonnage.

1.1 The traded share of output follows geography, value per tonne and protection

Radetzki and Wårell compare world production with world exports for 18 commodities in 2018. The spread is wide. Natural rubber reached 88% of output, coffee 78%, cocoa 72%, iron ore (2017) and copper 66%, petroleum 51%. At the bottom sat timber at 7%, rice at 10% and maize at 14%, with wheat, steel and aluminium in the mid-twenties.

Three forces explain most of the ranking. The first is geography. Rubber, coffee and cocoa grow in the humid tropics but are consumed mostly in temperate rich countries, so nearly every bean must cross a border. Minerals that cluster in a few places behave the same way: Brazil mined 88% of the world's niobium in 2018. The second is value per tonne. In 2016-2018 coffee averaged about US$3,290 a tonne and tin US$19,380, against US$67 for iron ore and US$96 for phosphate rock. Freight is trivial on a coffee cargo and a large slice of an ore cargo; cheap, ubiquitous timber mostly stays home. The third force is protection and taste. Rice stores and ships well, yet OECD support for rice equalled 55% of farm receipts in 2016-2018, the highest of any product, and eaters are loyal to local varieties.

Weight and value give different rankings. By tonnes, petroleum (2,263 Mt in 2018), iron ore (1,640 Mt), hard coal and gas lead, and they set the fortunes of bulk shipping. By dollars, petroleum alone averaged US$1,114 billion a year in 2016-2018, more than world exports of cars.

The shares move over time. USDA's October 2026 estimates put 2025/26 wheat exports at 27% of output, soybeans at 43% and rice at 11%. Re-exports inflate traded shares, because a cargo counted into a hub can be counted again on the way out.

Radetzki & Wårell (2021), A Handbook of Primary Commodities in the Global Economy · pdf pp. 42, 46-49, 75, 199
USDA (2026), World Agricultural Supply and Demand Estimates, October 2026 web ↗ · World Wheat, Rice and Soybean Supply and Use tables

1.2 A handful of corridors carry most of the tonnage

The biggest flows share a shape: two or three exporters on one side and one dominant buyer on the other.

Iron ore is the clearest case. In 2024 Australia supplied 55.4% of world export value and Brazil 20.1%, while China took 73.6% of imports, with Japan and South Korea far behind. Smil shows the traded share of world ore rising from under 40% in the early 1970s to 69% in 2013. China's own ore averages about 30% iron against 62% in Australia and 52% in Brazil, and Humphreys adds a geographic reason for buying abroad: China's growth was on the east coast, where ore arriving by sea is easier to land than domestic ore hauled across mountains.

Crude oil runs from the Gulf and the Americas to Asia. Saudi Arabia shipped 18.8% of 2024 crude export value, the United States 12.4%, Canada 11.3% and Iraq 10.0%; China took 24.9% of imports and India 10.8%. Conway describes the loading at Ras Tanura, the world's largest oil export terminal, where tankers queue before sailing for China and India.

Soybeans run from the Americas to China. Brazil supplied 54.3% of 2024 export value and the United States 31.1%; China took 65.3% of imports. USDA puts Brazil's 2025/26 exports at 114 million tonnes of a world total of 186 million, and China's imports at 113 million.

LNG runs from Qatar, Australia and the United States to North-East Asia. Qatar shipped 30.5% of 2024 export value, Australia 23.9% and the United States 15.5%; China, Japan and South Korea took just over half of imports.

Concentration at both ends makes these corridors efficient and fragile. Huge ships and terminals pay off, but a policy shift in one buyer or one seller moves the whole market.

Smil (2016), Still the Iron Age · pdf p. 125
Humphreys (2015), The Remaking of the Mining Industry · pdf pp. 74-75
Conway (2023), Material World · pdf pp. 276-277
Morgan (1979), Merchants of Grain · pdf pp. 231-232, 311
USDA (2026), World Agricultural Supply and Demand Estimates, October 2026 web ↗ · World Soybean Supply and Use table

1.3 Industrialisation turned rich regions, and then China, into importers

Radetzki and Wårell trace the rich world's dependence on distant raw materials to two twentieth-century forces: falling bulk freight and heavy industrialisation.

The United States covered all its needs for metal ores in the first decade of the twentieth century. Self-sufficiency fell to 70% by the Second World War, and by mid-century the country was the largest importer of copper, lead and zinc. By 2000 it imported all its nickel, 52% of its refined copper and 40% of its primary aluminium. Japan industrialised later and faster. Its 1990 consumption of aluminium was 127 times the 1950 level, copper 131 times and nickel 160 times, and with little ore at home its import dependence became almost complete.

Population density shapes how far this goes. In 2018 Japan had 349 people per square kilometre, Western Europe 185, the United States 36 and China 151. Dense countries run out of domestic land and ore sooner. Food imports stayed smaller because Europe and Japan protected farmers.

China repeated Japan's metal-intensive phase at a much larger scale. Between 1990 and 2005 its demand for aluminium, nickel and copper rose six- to eightfold. By 2018 it accounted for 52% of world copper demand, 51% of coal and 49% of nickel. In 2015-2017 it ran a net commodity deficit of about US$371 billion a year, paid for with manufactures. Yet it is less import-dependent than Japan because it is itself a large miner: its ratio of energy production to consumption was still 0.80 in 2018, against Japan's 0.11.

A China share of 60-90% of imports is the normal shape of an ore or concentrate market today, not an anomaly.

Radetzki & Wårell (2021), A Handbook of Primary Commodities in the Global Economy · pdf pp. 49-53, 56-63

See it in the table

Oil Crude oil P Phosphate rock Cu Copper Al Aluminium Zn Zinc Ni Nickel Fe Iron ore St Steel Nb Niobium Mz Maize Wh Wheat Ri Rice Soy Soybeans Cf Coffee

Check your understanding

Rice is easy to store and ship, yet only about a tenth of world output is traded. What best explains this?

Radetzki and Wårell point to taste and protection: OECD support for rice equalled 55% of farm receipts in 2016-2018, the highest of any product. Value per tonne and storability are not the obstacle.

Which statement about the 2024 iron ore trade is correct?

Comtrade shows Australia at 55.4% of export value, Brazil at 20.1% and China at 73.6% of imports. Concentration at both ends is the defining feature of the big bulk corridors.

Why is China less import-dependent for raw materials than Japan, despite consuming far more?

Radetzki and Wårell note that China is a large miner in its own right; its energy production still covered 80% of consumption in 2018, against 11% for Japan.

← Module contents Mark read · Part 2: Reading trade data →