Physical Trading Houses and How They Arbitrage

Ask a physical trading house like Vitol or Trafigura whether oil is going up or down and you will often get a shrug — that is not really their bet. Their business is buying a specific grade of crude from a specific seller in West Africa, chartering a tanker, moving it to a refinery in Asia that specifically needs that grade, and pocketing the difference. If the outright price of oil moves 10% while the cargo is at sea, that risk is usually hedged away in the futures market the same day the physical deal is struck. What is left over — the thing they are actually paid for — is arbitrage across space, time, and quality.

Three kinds of spread

Location arbitrage exploits Delivery Points and Location Basis: buy crude where it's cheap relative to transport cost, ship it to where it's dear, pocket the gap net of freight. Time arbitrage exploits contango — when future delivery is priced above spot by more than the cost of storing the commodity — by buying now, storing it, and selling forward, a trade covered in Cost of Carry and Storage. Quality arbitrage exploits the fact that a barrel of light sweet crude and a barrel of heavy sour crude are not interchangeable: a refinery configured for one grade will pay up for it and discount the other, so a trading house that understands refinery configurations can buy the grade nobody local wants and find the refinery across an ocean that does.

The hedge that makes it a spread trade, not a bet

The mechanism that turns "buy oil, ship it, sell oil" from a directional bet into a pure spread trade is the futures hedge. The moment a trading house commits to buy a physical cargo, it sells an equivalent quantity of futures. If the price of oil then falls before the cargo is delivered, the physical inventory loses value — but the short futures position gains almost exactly the same amount, because both are pricing the same underlying commodity. What is left exposed is only the difference between the physical price paid and the futures price received, which is the basis the trade was designed to capture in the first place.

Worked example

A trading house buys 1 million barrels of a specific West African crude grade at a discount of $3.00 per barrel to Brent futures, because a temporary local oversupply has widened that grade's basis. Freight to an Asian refinery that wants exactly this grade costs $1.20 per barrel; the refinery is willing to pay Brent flat, no discount, because it has no other easy source of this grade.

1. Purchase price. Brent futures - $3.00, hedged immediately by selling 1 million barrels of Brent futures.
2. Freight cost. $1.20 per barrel.
3. Sale price to the refinery. Brent flat (no basis).
4. Gross margin per barrel. $3.00 - 1.20 = 1.80$, i.e. $1.80 per barrel, regardless of what Brent itself does in the meantime — the futures hedge offsets that leg entirely.
5. Total margin. $1{,}000{,}000 \times 1.80 = 1{,}800{,}000$, or $1.8 million on the cargo.

If Brent rallies $10 while the tanker is at sea, the physical cargo gains $10 million in value but the short futures position loses almost exactly $10 million — the $1.8 million spread survives untouched either way, which is the entire point of hedging the directional leg out.

What this means in practice

This is why physical trading houses can be enormously profitable in years when outright commodity prices are flat or falling: their edge is in logistics, quality knowledge, storage access, and counterparty relationships, not in forecasting prices. It also explains why they invest heavily in owned infrastructure — storage tanks, pipelines, even refineries — because owning the physical assets that create location and quality spreads is more durable than trying to spot temporary mispricings from a desk.