Scope 1, 2 and 3: definitions, examples and how they map to French reporting

The question almost always lands in the first quarter of an hour of a scoping meeting: what goes into scope 3? Behind it sits another question, less openly asked, about what can legitimately be left out.

The language of scopes comes from the GHG Protocol, a carbon accounting standard of American origin. French regulation structures the statutory assessment into six categories instead. Both frameworks circulate in the same files, sometimes in the same spreadsheet, which explains much of the confusion.

Scope 1: what your company burns and emits itself

Scope 1 covers direct emissions from sources you own or control. In practice: gas for the boiler, fuel oil for the generator, gasoline and diesel for the company fleet, refrigerant leaks from chillers and air conditioning, and process emissions for manufacturers, such as limestone calcination in a cement plant.

It is the easiest boundary to collect, because the data already exists somewhere: gas invoices, tank readings, refrigeration maintenance contracts. In service companies it is also the smallest of the three.

The usual trap is refrigerants. They are often forgotten, yet a single kilogram of leaked R404A weighs several metric tons of CO2 equivalent. On a cold storage site, that line can exceed the entire gas consumption.

Scope 2: the energy you buy

Scope 2 covers indirect emissions from purchased energy that you then consume: electricity, steam, heat and cooling delivered by a network.

Two calculation methods coexist. The location-based approach applies the average emission factor of the national grid. The market-based approach uses the carbon content of the supply contract actually signed. The two diverge sharply as soon as a company buys renewable electricity, and you have to state which one you are presenting.

France has a low carbon electricity mix, which mechanically flattens scope 2 compared with similar companies in Germany or Poland. That is good news for the total. It is also why an action plan limited to electricity moves the needle very little for a French company.

Scope 3: everything else, upstream and downstream

Scope 3 gathers the indirect emissions that do not come from purchased energy. Upstream: purchased goods and services, inbound freight, capital goods (buildings, machinery, IT equipment), business travel, employee commuting, waste. Downstream: distribution, use of sold products, their end of life, investments, franchises.

In the assessments we run, this boundary carries most of the total, and purchased goods and services alone frequently exceeds half. A trading company, a consultancy, a software publisher: in all three cases, what matters happens at the suppliers, not inside the walls.

It is also the most expensive boundary to collect and the most uncertain, since a large share rests on spend-based ratios applied to accounting lines. We cover this in detail in our article on emission factors.

How this maps to the six BEGES categories

The French statutory assessment, the BEGES, does not use the word scope. It sorts emissions into six categories:

1. Direct emissions
2. Indirect emissions associated with energy
3. Indirect emissions associated with transport
4. Indirect emissions associated with purchased products
5. Indirect emissions associated with sold products
6. Other indirect emissions

The mapping is mechanical: category 1 corresponds to scope 1, category 2 to scope 2, and categories 3 to 6 share out scope 3. An assessment run under the Bilan Carbone method of ADEME and the Association Bilan Carbone produces both readings from the same dataset, without double entry.

One point is worth keeping: since its 2022 regulatory overhaul, the BEGES requires significant indirect emissions to be covered. The days of publishing a statutory assessment limited to scopes 1 and 2 are over. Thresholds, frequency and penalties are set out in our article on mandatory carbon footprint assessment.

The five boundary mistakes we see most often

The first is putting fuel extraction and transport emissions into scope 2. Those upstream emissions are real, but they belong in scope 3.

The second concerns commuting, regularly excluded on the grounds that the company does not control it. It belongs to scope 3, and the company does hold real levers: remote work, sustainable mobility allowances, site location.

The third comes from the consolidation choice. Operational control or financial control: the rule you adopt changes what enters the scope 1 of subsidiaries and joint venture sites. Fix it during scoping, write it down, and stick to it year after year.

The fourth is double counting between entities of the same group, when a subsidiary's scope 1 also appears in another's upstream scope 3. That is not an error at entity level, but group consolidation becomes wrong if nobody nets out internal flows.

The fifth is the base year. Changing boundary, emission factors or method without recalculating the base year makes any comparison worthless. It is the single point that derails the most reduction trajectories.

What the split is actually for

Spreading tons across three scopes only matters for what comes next: deciding where to act. A dominant scope 1 points toward processes, heat and the fleet. A scope 3 dominated by purchasing points toward specifications, supplier selection and product eco-design.

That shift from figure to decision is what separates an assessment filed in a drawer from an action plan that holds. Our consultants build that plan with your teams in a dedicated workshop, during the carbon footprint assessment.