Crush Spread

A crush spread cannot be computed without deciding how much of each product a unit of seed becomes. That decision is an assumption, and it is embedded in the figure permanently and invisibly. Two crush spreads computed from identical quotations with different yield assumptions are different numbers, and neither is more correct than the other in the abstract — they are answers to slightly different questions.

The assumption is usually a conventional one, standing for a typical processing outcome. Real plants deviate from it in both directions, because yields depend on the seed’s own composition, the plant’s technology and condition, and how the plant is being operated. A processor whose actual yields differ from the convention is looking at a figure computed for someone else’s plant.

Yield assumption

How much meal and how much oil a unit of seed is taken to produce. A convention, not a measurement, and not a property of any particular plant or any particular seed lot.

Unit reconciliation

The seed and its products are frequently quoted in different units. Converting them onto a common basis is a necessary step that introduces its own arithmetic, and an error here is silent.

Contract alignment

The quotations used should refer to compatible delivery periods. Combining an input price for one period with product prices for another embeds a time relationship in what purports to be a processing relationship.

Quotation basis

The seed and product quotations may sit at different points in the chain and on different delivery terms. Where they do, the spread contains a locational difference that has nothing to do with processing.

The gap between a crush spread and an actual processing margin is not a matter of precision. They are different quantities, and the list of what separates them is long, structural, and mostly invisible in the spread figure.

• Conversion costs — energy, labour, maintenance, chemicals, and waste handling are entirely absent from the spread and are a substantial part of any real processing economics
• Capital — depreciation, financing, and the cost of the plant existing at all appear nowhere in a difference between quotations
• Plant-specific yields — the spread uses a convention; a plant earns on its own actual recovery, which differs by technology, condition, and operation
• Seed composition — the oil content of the actual seed purchased determines actual yields, and varies by variety, season, and origin
• Contract reality — a processor buys and sells on their own contracts at their own times, not at a set of simultaneous quotations
• Basis — the processor transacts physical seed and physical products at locations with their own differentials, none of which the quoted spread contains
• Capacity and operating decisions — a plant that is not running earns nothing regardless of what the spread shows

Having established what a crush spread is not, it is worth being precise about what it is. It expresses the market’s current pricing of a transformation: what the products are quoted at, relative to the input, given a conventional conversion. That is a genuine and useful piece of information about how a market is valuing an input against its outputs, and it is observable in a way that any individual plant’s economics are not.

The relationship exists because the transformation is real. Processing capacity connects the seed market to the meal and oil markets physically, and that physical connection is why the prices do not wander independently. When the products are quoted well above the seed, processing is being encouraged and more seed is drawn into crushing; when they are not, the reverse. This is a description of a mechanism operating in the market, and it is the reason the spread is worth computing at all.

What it will not do is tell you what any processor earns, whether any plant will run, or where any of the three prices is going. Those questions require the plant’s own data, the operator’s own contracts, and information that no relationship between quotations contains.