10. Bulk Commodities
Teaching rewrite of the bulk commodity markets: coal, iron ore, and freight — the raw materials and shipping behind the steel supply chain. Original wording; worked numbers recomputed. Prices are teaching values. Not investment advice.
Introduction
“Bulk commodities” is a loose label for markets linked by steel: iron ore and coking coal go into blast furnaces, and dry bulk ships carry both — about 60% of Capesize vessels haul iron ore. Some scale (Singapore Exchange, 2019):
- In tonnage, steel is roughly 20 times all other metals combined.
- About 98% of iron ore goes into steel.
- About 15% of coal is used in steelmaking (as coking coal).
- China takes about two-thirds of seaborne iron ore.
Banks organise these desks differently: a consumer-focused bank may put iron ore under base metals, while a producer-focused one may group coal, iron ore, and freight together.
This lesson covers:
- Coal: types, supply and demand, the supply chain, price drivers, and derivatives (swaps, swaptions, covered calls)
- Iron ore: grades, CFR pricing, and exchange-listed contracts
- Freight: vessel classes, voyage vs time charters, Baltic indices, Worldscale, and forward freight agreements (FFAs) and exotic structures
10.1 Coal basics
Coal is a combustible sedimentary rock — mainly carbon, hydrogen, and oxygen — formed from plant matter compressed and heated over millions of years. Longer formation and greater heat and pressure give higher rank coal: more carbon and energy, less moisture.
| Rank | Type | Share of reserves | Main uses |
|---|---|---|---|
| High (hard coal ≈ 53%) | Anthracite | ~1% | Domestic and industrial heating |
| Bituminous — thermal (steam) | ~52% (with coking) | Power generation, cement | |
| Bituminous — metallurgical (coking) | Coke ovens → blast furnaces → pig iron | ||
| Low (≈ 47%) | Sub-bituminous | ~30% | Power, cement |
| Lignite (brown coal) | ~17% | Local power generation, briquettes |
- Coking coal is harder and scarcer than thermal coal, so it trades at a premium.
- Hard coal has enough energy per tonne to justify international shipping; lignite is usually burned near the mine.
10.2 Coal demand and supply
Coal supplied about 27% of primary energy in 2019, second only to oil. World Coal Association figures:
- ~41% of global electricity comes from coal
- ~70% of steel uses coal
- ~200 kg of coal per tonne of cement
- ~50% of the energy used to make aluminium comes from coal
Reserves (BP 2020): Asia Pacific 42.7%, North America 24.1%, CIS 17.8%, Europe 12.6%, Middle East and Africa 1.5%, South and Central America 1.3%.
Reserves-to-production (R/P) ratio = reserves at year end ÷ that year’s production → how many years reserves last at current output. Coal ≈ 132 years, versus ≈ 50 for oil and for gas.
| Largest thermal exporters (2019) | Mt | Largest coking exporters | Mt | Main importers | Mt |
|---|---|---|---|---|---|
| Indonesia | 435 | Australia | 179 | China | 295 |
| Australia | 203 | USA | 56 | India | 240 |
| Russia | 173 | Canada | 29 | Japan | 185 |
| Colombia | 81 | Russia | 26 | South Korea | 142 |
Coal grew because energy demand rose, it is cheaper per kWh in many regions, it is safe and cheap to transport, and reserves are large.
Since 2019 coal’s role has shifted: Europe and the US have closed many coal plants, while China, India, and Indonesia have kept consumption near record highs. 2022 saw extreme price spikes (Newcastle above USD 400/t) as Europe replaced Russian gas.
10.3 The physical supply chain
Production: underground mining (most output) or surface/open-cast mining (recovers more of a deposit, usually cheaper). Coal is crushed, cleaned, and moved by truck, rail, barge, and ship.
| Participant | Concern | Typical solution |
|---|---|---|
| Mining companies | Falling prices; financing linked to coal and FX | Swaps, collars, price-linked loans |
| Utilities | Volatile fuel cost; carbon cost | Financially settled swaps (separate price from delivery) |
| Banks / trading houses | Liquidity, client hedges, structured notes | Make markets, offset other banks’ exposures |
| Industrial users (cement, steel) | Rising costs | Buy swaps; steelmakers using thermal coal as a proxy take basis risk |
The source says financial markets trade only thermal coal. Today coking coal futures exist too (e.g. SGX and CME contracts on premium hard coking coal, FOB Australia), so steelmakers can hedge more directly.
Price drivers
| Driver | Effect |
|---|---|
| Benchmark location | Indices at Central Appalachia, Powder River Basin, ARA (Amsterdam–Rotterdam–Antwerp), Richards Bay, Newcastle |
| Electricity demand | Electrification (e.g. India) lifts demand; China largely uses its own coal |
| Generation margins | Gas-to-coal switching via spark vs clean dark spreads (lesson 8) |
| Technology | Cleaner preparation, scrubbers, more efficient boilers |
| Environment | Methane from mines; CO₂, ash, SOx, NOx from burning — regulation hurts prices |
| Freight | Delivered cost includes shipping |
| Economic growth | Chinese steel boom → coking coal demand |
| Quality | Calorific value and sulfur; low-sulfur, open-cast Powder River coal is cheaper and cleaner than Appalachian |
| Mining | Capex, disruption, costs, depth; long rail hauls (Russia), derailments, port logistics |
| Competing fuels | Liquefaction (coal → liquids) gains when oil is dear; gasification → syngas for power; cheap gas or LNG displaces coal |
| Investors | Flows can lift prices and deepen liquidity |
Calorific value: GAR vs NAR
Contracts specify energy content, checked by lab assay (like a crude assay).
| Basis | Meaning |
|---|---|
| GAR — gross as received | Gross (higher) heating value: counts the heat recoverable by condensing the water vapour produced |
| NAR — net as received | Net (lower) heating value: excludes that latent heat — closer to what a boiler actually gets |
GAR > NAR for the same coal; the gap is the heat of the water vapour.
The source describes GAR as measured “after all the moisture has been removed”. “As received” means including the coal’s moisture as delivered; a dried basis is “air-dried” or “dry basis”. The GAR–NAR difference is the latent heat of the water vapour, not the moisture removal.
Checkpoint 1: coal basics
Checkpoint — sections 10.1 to 10.3
1 / 5Which coal is used in coke ovens to make iron?
Pick an answer to see if you are right.
10.4 Coal derivatives
Market size: 8.1 billion t × USD 60 ≈ USD 486 billion of physical coal (2019), versus about USD 100 billion of derivatives notional — small, as derivatives usually exceed the physical market.
Indices
“API” here means All Publications Index — an average of several published assessments (Argus/McCloskey). The number only distinguishes locations.
| Index | Basis | Location | Role |
|---|---|---|---|
| API2 | CIF ARA (Amsterdam–Rotterdam–Antwerp) | Northwest Europe | Main Atlantic import benchmark |
| API4 | FOB Richards Bay | South Africa | Export benchmark — Atlantic and Indian Ocean |
| NEWC (GlobalCoal) | FOB Newcastle | Australia | Main Pacific benchmark |
| API 3, 5, 6, 8, 10, 12 | Various | S. Africa, Australia, S. China, Colombia, India | Quality or regional variants |
The source calls API4 the Pacific benchmark. API4 (Richards Bay) serves the Atlantic/Indian Ocean; the Pacific benchmark is Newcastle (NEWC). Also, API2 is CIF ARA, not Rotterdam alone.
Indicative implied volatilities: API2 25–35%, API4 20–25%, NEWC 18–25%, crude 25–50%, US gas 40–120%, European gas 40–90%.
European utilities, facing less predictable load after liberalisation, moved from one-year fixed-price supply to index pricing plus swaps.
Exchange futures: ICE Rotterdam coal (API2)
| Term | Specification |
|---|---|
| Unit | 1,000 t |
| Quote | USD and cents per tonne |
| Periods | Months, quarters, seasons, years |
| Tick | USD 0.05/t = USD 50 per contract |
| Last trading | Last business day of the delivery period |
| Settlement | Cash against the monthly average API2 |
Coal swaps
An Australian producer sells most output on long-term fixed contracts, but some at the floating API2 price (it has cheap freight to Europe). Fearing falling prices, it receives fixed / pays floating:
| Term | Value |
|---|---|
| Notional | 15,000 t (5,000 t a month for 3 months, starting next June) |
| Fixed payer | Bank, USD 60.00/t |
| Floating payer | Client, monthly average API2 |
| Settlement | Monthly in arrears, paid 5 business days later |
Worked example: June API2 averages 52.
- Physical sale: 52 × 5,000 = 260,000
- Swap: (60 − 52) × 5,000 = +40,000
- Total: 300,000 = USD 60/t
Any mismatch between the physical and swap pricing dates is basis risk. To regain upside on its fixed-price contracts, the producer can do the opposite swap: pay fixed, receive floating. Swaps can also settle as one three-month period (a cash-settled forward), or pay in GBP/EUR at an average FX rate.
Coal swaptions
A European option to enter that swap, expiring in December. The client buys from the bank at USD 1.85/t × 15,000 = USD 27,750. The bank would pay fixed 60, so the client has the right to receive fixed — a floor for a producer who expects prices to rise but cannot afford to be wrong.
| Swap level at expiry | Action | Net price after premium |
|---|---|---|
| 50 | Exercise → receive 60 | 60 − 1.85 = 58.15 |
| 70 | Let lapse → sell at market | 70 − 1.85 = 68.15 |
Covered calls
A producer sells a 3-month USD 55 call (spot 50, vol 40%) for about USD 2.
| Price at expiry | Outcome | Producer receives |
|---|---|---|
| 45 | Not exercised | 45 + 2 = 47 |
| 52 | Not exercised | 54 |
| 60 | Exercised | 55 + 2 = 57 — capped |
Extra income when prices are flat or falling, but the upside is capped — unsuitable for a strongly bullish view.
Checkpoint 2: coal derivatives
Checkpoint — section 10.4
1 / 5A producer receives fixed 60 on 5,000 t; API2 averages 65. Swap cash flow?
Pick an answer to see if you are right.
10.5 Iron ore
Iron (Fe) is abundant but never found as usable metal; iron ore is rock rich enough in iron to be worth processing.
- Fines: sand-like grains — the most traded form. Lump: larger pieces, charged directly into furnaces at a premium.
- Ore + coke in a blast furnace → pig iron → steel (iron plus a little carbon; see lesson 5).
- Iron and steel are about 95% of all metal tonnage and the cheapest metals.
Largest pig iron producers (USGS 2020): China 820 Mt, India 75, Japan 75, Russia 50, Korea 48.
Grades and pricing
Prices are quoted by Fe content; the commonest grades are 58–65%, and 62% Fe is the main benchmark. Iron ore moved from annual benchmark contracts to index pricing around 2010 (lesson 1).
Simple Fe adjustment (pro-rata, ignoring impurity penalties): a 58% cargo against a 62% index at USD 100 → 100 × 58 / 62 ≈ USD 93.55.
Iron ore derivatives (e.g. SGX)
- Futures, swaps, and options (on futures or on swaps)
- Mainly fines, plus a lump contract; grades 58–65%
- Quoted CFR (cost and freight) to a Chinese port such as Qingdao: the seller pays freight and loads the ship, but risk passes to the buyer once loaded, and the seller doesn’t insure
- Cash-settled against a published index
10.6 Freight fundamentals
Vessels
Wet freight: crude, products (plus containers, liners, tugs). Dry freight: coal, iron ore, bauxite, grain. Ships are sized by deadweight tonnage (dwt) — carrying capacity including fuel, water, crew, and stores.
| Dry bulk class | Approx. dwt | Notes |
|---|---|---|
| Handysize | 20,000–35,000 | Own cranes |
| Handymax | 36,000–49,000 | Grain, steel, fertiliser |
| Supramax / Ultramax | 50,000–65,000 | |
| Panamax / Kamsarmax | 65,000–83,000 | Fit the old Panama locks; coal, grain, forest products |
| Post-Panamax / Mini-Cape | 87,000–120,000 | |
| Capesize | ~100,000–200,000 (index ship 180,000) | Too big for Panama/Suez → round the Capes; mostly iron ore |
| VLOC (Valemax, Chinamax) | 220,000–400,000 |
| Tanker class | Approx. dwt |
|---|---|
| Product tanker | 10,000–60,000 |
| Panamax | 60,000–80,000 |
| Aframax | 80,000–120,000 |
| Suezmax | 120,000–160,000 |
| VLCC (very large crude carrier) | 200,000–320,000 |
| ULCC (ultra large) | 320,000–550,000 |
The source gives Capesize as 80,000–175,000 dwt, overlapping Panamax and Mini-Cape. Modern Capesizes are mostly ~150,000–210,000 dwt, and the Baltic Capesize index uses a 180,000 dwt ship.
Voyage charter vs time charter
| Voyage charter (VC) | Time charter (TC) | |
|---|---|---|
| Analogy | Taxi: pay for the trip | Rental car: you drive, fuel, tolls, and delays are yours |
| Priced in | USD per tonne | USD per day |
| Fuel and ports | Owner pays | Charterer pays |
Converting TC → VC per tonne: (round-trip days × daily hire + fuel + port costs) ÷ tonnes carried.
Worked example: 50-day round trip at USD 15,000/day = 750,000; fuel 30 steaming days × 40 t × USD 500 = 600,000; cargo 170,000 t → (750,000 + 600,000) / 170,000 ≈ USD 7.94/t.
Participants
Shipbuilders, owners, brokers (match owners with cargo), charterers (miners, utilities, steel mills, cement makers, grain traders), and operators (crewing and running ships; may charter in long and charter out short at a higher rate).
The Baltic Exchange (London, origins 1744, owned by SGX since 2016) publishes independent freight benchmarks used to settle physical and derivative contracts.
Checkpoint 3: iron ore and shipping
Checkpoint — sections 10.5 and 10.6
1 / 5What is the main benchmark iron ore grade?
Pick an answer to see if you are right.
10.7 Freight indices
Why they are hard to compile
Shipping is varied and opaque; contracts are private and non-standard; deals are infrequent and may carry conditions; nobody has to report trades; and ships vary in size, age, upkeep, and owner credit. Panels of independent shipbrokers submit judgement-based assessments — there is no verifiable “right” rate. Volatility is extreme: the Baltic Dry Index fell from 11,793 (20 May 2008) to 663 (5 December 2008), a 94% drop.
The Baltic Dry Index (BDI)
An index value, not a price — a weighted blend of three sub-indices:
flowchart TB
BDI[Baltic Dry Index]
BCI[Capesize 2014 Index<br/>40%] --> BDI
BPI[Panamax Index<br/>30%] --> BDI
BSI[Supramax Index<br/>30%] --> BDI
Each sub-index defines a notional ship (Capesize: 180,000 dwt, at most 10 years old) and set routes, e.g. C2 Tubarão (Brazil) → Rotterdam (iron ore), C4 Richards Bay → Rotterdam (coal).
Worked example: building an index
Four routes, each to contribute 25% of a starting value of 1,000 → 250 each. Multiplier = 250 ÷ assessed rate.
| Route | Day 1 rate (USD/day) | Multiplier | Day 1 contribution | Day 2 rate | Day 2 contribution |
|---|---|---|---|---|---|
| 1 | 10,000 | 0.02500 | 250 | 10,000 | 250 |
| 2 | 11,000 | 0.02273 | 250 | 11,000 | 250 |
| 3 | 12,000 | 0.02083 | 250 | 12,000 | 250 |
| 4 | 13,000 | 0.01923 | 250 | 14,000 | 269.23 |
| Index | avg USD 11,500 | 1,000 | avg USD 11,750 | 1,019.23 |
Multipliers stay fixed until the index composition changes. Each multiplier is also a sensitivity: a USD 1,000 rise on route 4 adds 1,000 × 0.01923 = 19.23 points.
Worldscale (tankers)
A yearly table of flat rates (WS100, USD/t) for 300,000+ voyages. Deals are struck in points of scale:
Rate = WS points ÷ 100 × flat rate. Example: WS 287 on a route with flat USD 9.13/t → 9.13 × 2.87 = USD 26.2031/t.
Flat rates are set so a notional tanker (75,000 t, 14.5 knots, 55 t VLSFO a day, 4 port days, USD 12,000/day fixed hire) earns the same net daily revenue on every route. Larger ships agree adjustments bilaterally.
10.8 Freight price drivers
| Group | Driver | Effect |
|---|---|---|
| Fleet | Ship availability (~9,000 dry bulkers), lay-ups | More ships → lower rates |
| Newbuild deliveries (~18 months to build) | Watch the orderbook and deliveries | |
| Piracy (223 incidents in 2018), rerouting | Longer voyages absorb ships → higher rates; cargo owners finance stock longer | |
| Choke points (Hormuz, Suez, Panama) | Disruption ties up tonnage | |
| Environmental rules (sulfur caps, scrubbers) | Higher costs | |
| Economy | Bulk commodity demand | Same drivers as coal, ore, grain |
| Credit | Less lending → fewer newbuilds (2008: buyers forfeited deposits to cancel orders) | |
| Seasonality | Weather, harvests, holiday demand | |
| Fuel | ¼ to ⅓ of operating cost | |
| Port congestion | Ships waiting = lost supply |
Short-dated rates react to deliveries and demand; long-dated rates to structural themes (e.g. coal’s decline). Freight is often called a barometer of world trade — a useful but simplistic view.
Rerouting example: Gulf → Rotterdam takes 75 days; via the Cape of Good Hope, 95. At USD 20,000/day hire the extra 20 days cost USD 400,000.
Since 2023, Red Sea attacks have rerouted many ships around the Cape, and drought has restricted Panama Canal transits — real-world examples of choke-point risk lifting rates.
10.9 Freight derivatives
Forward freight agreements (FFAs)
Cash-settled swaps on a Baltic route or index — “a ULCC is not going to turn up at your office”.
- Buy (pay fixed) → protects against rising rates (charterers, cargo owners)
- Sell (receive fixed) → protects against falling rates (shipowners)
The FFA should trade near the physical forward charter rate; gaps from speculation can create hedging opportunities.
Voyage-charter FFA (USD/t)
| Term | Value |
|---|---|
| Start | 3.5 years forward; 12 monthly periods |
| Notional | 10,000 t/month (120,000 t) |
| Fixed payer (buyer) | Client, USD 11.00/t → 110,000 a month |
| Floating | Bank pays the monthly average of daily Baltic C4 rates |
Month averages 11.50 → floating 115,000 vs fixed 110,000 → bank pays 5,000.
Who buys? E.g. an operator that has agreed a fixed voyage rate with a miner but will hire ships later at market rates — it fears rising freight, so pays fixed.
Time-charter FFA (USD/day)
| Term | Value |
|---|---|
| Start | 1.25 years forward; 12 monthly periods |
| Notional | Days in the month × multiplier 0.5 (half a ship) |
| Fixed payer | Bank, USD 15,000/day |
| Floating payer (seller) | Client, average Baltic Supramax S2 |
31-day month at 14,500: fixed 15,000 × 15.5 = 232,500; floating 14,500 × 15.5 = 224,750 → bank pays 7,750 to the client. The client sold the FFA (receives fixed) — consistent with an owner chartering ships out who fears falling rates.
Worldscale FFA (tankers)
| Term | Value |
|---|---|
| Notional | 5,000 t/month for 3 months |
| Fixed payer | Bank, WS178 = USD 24.7954/t → flat rate 24.7954 ÷ 1.78 = USD 13.93/t |
| Floating | Client pays average TD8 (Kuwait → Singapore) in WS points × flat ÷ 100 |
Fixed: 24.7954 × 5,000 = 123,977. Month averages WS170: 13.93 × 1.70 × 5,000 = 118,405 → bank pays 5,572.
Flexibility: size and maturity needn’t match the exposure; positions can be added or offset. Trading views: calendar spreads (Q3 vs Q4), index spreads (Capesize vs Panamax), wet vs dry (BDI vs Baltic Dirty Tanker Index).
LNG freight futures
As LNG moves from oil-linked long-term contracts (freight included) to spot (~20% of trades), freight becomes a separate cost. NYMEX LNG freight futures (Baltic-settled, out to 2 years) cover Gladstone → Tokyo, Sabine Pass → Isle of Grain, Sabine Pass → Tokyo. A trader buying cheap gas can buy freight futures to fix shipping before chartering spot.
Target payout structure
A charterer wants cheap protection against rising Panamax rates:
| Term | Value |
|---|---|
| Index | Panamax TC4 average |
| Strike | USD 10,000/day, monthly for 2 years |
| Client receives (call) | days × max(0, index − strike) |
| Client pays (put) | days × max(0, strike − index) × 2 |
| Target | Trade ends once cumulative call payments reach USD 350,000 |
The client is long a synthetic forward at a better-than-market strike, paid for by 2× downside and a capped upside.
- 31-day month at 12,000: receives 31 × 2,000 = 62,000 (counts toward the 350,000 target).
- 31-day month at 8,000: pays 31 × 2,000 × 2 = 124,000 — offset partly by cheaper physical charters.
Options: calls, puts, zero-cost collars, three-ways — as in lesson 5.
Checkpoint 4: indices and FFAs
Checkpoint — sections 10.7 to 10.9
1 / 6Route assessed at USD 20,000/day must contribute 250 points. Multiplier?
Pick an answer to see if you are right.
Rank: bulk commodities practice
Bulk commodities practice
1 / 3Rank the steel chain from mine to product.
OrderDrag the handle to rearrange — top should be first.
- 1Smelt pig iron in a blast furnace
- 2Ship by Capesize
- 3Convert to steel
- 4Make coke in coke ovens
- 5Mine iron ore and coking coal
Drag items into the correct order (or use ↑↓), then check.
Check: bulk commodities
Bulk commodities
1 / 6What links coal, iron ore, and dry freight as “bulk commodities”?
Pick an answer to see if you are right.
Calculate: numeric practice
Type the number (no units; minus sign for payments).
Formula sheet
| Item | Formula |
|---|---|
| R/P ratio | reserves ÷ annual production |
| Market value | volume × price |
| Swap (fixed receiver) | (fixed − floating) × tonnes |
| Covered call income | min(price, strike) + premium |
| Fe adjustment (pro-rata) | index × grade ÷ 62 |
| TC → VC | (days × hire + fuel + port) ÷ tonnes |
| Index multiplier | target contribution ÷ assessed rate |
| Index change | Δrate × multiplier |
| Worldscale rate | flat × WS ÷ 100 |
| TC FFA cash flow | rate × days × multiplier |
| Target payout put | days × max(0, K − index) × 2 |
Calculate: bulk commodities
1 / 16Coal production 8.1 bn t at USD 60/t. Market value in USD billions?
Enter a number (at most 0 decimal places).
Randomized practice
Practice: bulk drills
1 / 6Round trip 55 days at USD 23000/day, fuel USD 900000, cargo 75000 t. Equivalent voyage rate in USD/t? Round to 2 decimals.
Enter a number (at most 2 decimal places).
Takeaways
- Bulk commodities = the steel chain: iron ore, coking coal, and the ships that carry them
- Coal rank: anthracite > bituminous (thermal / coking) > sub-bituminous > lignite; coking trades at a premium
- Benchmarks: API2 (CIF ARA), API4 (FOB Richards Bay), NEWC (Pacific); GAR > NAR
- Producers receive fixed on swaps, buy receiver swaptions for a floor, or sell covered calls for income
- Iron ore trades by Fe grade (62% benchmark), CFR China, cash-settled
- VC = USD/t (taxi); TC = USD/day (rental car); BDI = 40% Capesize, 30% Panamax, 30% Supramax
- Index multiplier = contribution ÷ rate, also the per-dollar sensitivity; Worldscale = flat × points ÷ 100
- FFAs: buy = pay fixed (fear rising rates), sell = receive fixed (fear falling rates)
Further Reading
- Next: 11. Climate and weather
- 9. Plastics
- 5. Base metals — steel and option structures
- 8. Electricity — dark spreads
- 1. Fundamentals of commodities and derivatives — iron ore’s move to index pricing
- Baltic Exchange route definitions; Worldscale Association; SGX and ICE contract specifications
- Background: N. Schofield, Commodity Derivatives: Markets and Applications (Wiley) — used as a study outline