Product Carbon Footprint: What the Number Means

A product carbon footprint looks like a fact and behaves like an opinion with a methodology attached. Two competent teams can assess the same product and produce figures differing by a factor of three, with neither having made an error.

That is not a flaw in the discipline. It is a consequence of choices that have to be made and should therefore be published alongside the result, which is exactly what most published figures omit.

The boundary decides almost everything

The single largest driver of the number is where the assessment starts and stops. The two boundaries in common use answer different questions and are routinely quoted interchangeably.

For an energy-consuming product the use phase frequently dominates the total, so a cradle-to-gate figure can represent a small fraction of the lifetime footprint. Publishing that figure without its boundary is not merely incomplete; it can understate the real impact by an order of magnitude.

Allocation, and why it is contested

When one process produces several outputs, its emissions have to be divided between them. A dairy produces milk and cream; a refinery produces many streams; a slaughterhouse produces meat, hide and rendered products.

How that division is made changes each product’s footprint substantially, and there is no single correct answer. Allocation by mass, by economic value, or by a physical relationship each produce defensible and different results.

• Mass allocation — simple and often unrepresentative, since a low-mass high-value co-product carries little burden.
• Economic allocation — reflects why the process runs, but makes the footprint move with market prices rather than with physics.
• Physical causality — preferred where a genuine physical relationship exists, and frequently not applicable.
• System expansion — avoids allocation by crediting displaced production, and is the hardest to apply consistently.

The practical implication is that a footprint should state its allocation method. Where a supplier will not disclose it, the figure cannot be meaningfully aggregated into your own assessment.

Primary versus secondary data

Every assessment mixes measured data from actual operations with database averages standing in for what could not be measured. The ratio between the two is a quality indicator that is rarely published.

The difference is not marginal. A background database average for a material may reflect a global production mix, while your specific supplier may use a substantially cleaner or dirtier process. Replacing an average with a supplier-specific figure routinely moves a product footprint by tens of per cent.

This is where passport data and carbon accounting converge. The supplier-specific inputs that improve a footprint’s accuracy are the same attributes a passport programme is already collecting, which makes the two efforts substantially the same work.

What has to accompany the number

A footprint published as a bare figure is not usable by anyone downstream. Five items turn it into something another party can act on.

Where regulation is heading

The direction of travel is towards prescribed methodology rather than free choice. The Battery Regulation already requires a carbon footprint declaration calculated to a defined method and independently verified, which removes most of the discretion described above.

That is a significant change in what the number is for. A figure calculated to a prescribed method is comparable between manufacturers, which makes it usable as a purchasing criterion and eventually as a threshold. A figure calculated to a method of your choosing is a disclosure and nothing more.

For any product group heading in that direction, the useful preparation is to establish primary data collection now. Prescribed methods reward companies that measure their supply chain and penalise those relying on database averages, and the collection lead time is measured in quarters.