Contractor Emissions and Scope 3: What Can a City Actually Measure?

Two accounting lenses, two different answers

Before counting contractor emissions, choose the frame. In a community inventory built to the Global Protocol for Community-Scale Greenhouse Gas Inventories (GPC), a contractor burning fuel on a site inside the city boundary contributes to the city's territorial Scope 1, not to its Scope 3. Scope 3 there is narrower and transboundary: waste exported across the boundary, and, at the expanded BASIC+ level, transboundary journeys.

The picture changes when the municipality treats itself as an organization under the GHG Protocol Corporate Value Chain (Scope 3) Standard. Everything the city buys from contractors — construction, cleaning, waste haulage, machinery and services — then falls into the indirect emissions of procurement. This organizational lens is where contractor emissions become genuinely measurable and manageable.

Keeping the two frames separate prevents the most common error: transferring numbers from one inventory into the other, which causes double counting and misleading conclusions about progress.

• GPC describes a territory and its residents; its Scope 3 covers transboundary flows.
• Organizational Scope 3 describes the city as a buyer: goods, services, capital works, waste, business travel and commuting.
• A contractor's fuel burned inside the boundary is usually territorial Scope 1, not procurement Scope 3.

Which contractor emissions land where

In practice, most of a city's contractor emissions sit in a few corporate Scope 3 categories: purchased goods and services (category 1), capital goods and construction (category 2), fuel- and energy-related activities (category 3), upstream transportation (category 4), waste generated in operations (category 5), and business travel or commuting.

The heaviest contracts are usually roads and civil engineering, building operations, waste collection and treatment, transport services and social care. CE Delft's work for Zeist found that waste processing, civil engineering and the social domain drive the largest share of procurement climate impact, while civil works dominate resource depletion because of the materials in roads, sewers and street furniture.

A practical first step is to map every contract line to a Scope 3 category. The map instantly shows which contractors are visible in the system and which fall out of it, such as small subcontractors or hired machinery.

• Category 1 — goods and services, from office supplies to consultancy.
• Category 2 — capital works and materials such as concrete, steel and asphalt.
• Categories 4 and 5 — inbound freight and waste leaving the boundary.
• Business travel and staff commuting are separate, usually smaller, line items.

Methods: from spend to supplier data

There are three tiers of accuracy. The spend-based method multiplies a contract value by an average sector intensity drawn from environmentally extended input-output tables such as Exiobase or national databases. It is cheap, covers the whole portfolio and works well for a first-year screening, but it carries large uncertainty and cannot tell what a specific contractor actually does.

The activity-based method uses physical units: tonnes of concrete, kilometres driven, litres of fuel, tonnes of waste by disposal route. It is more accurate and lets you compare contractors, but it requires data the city must request in advance through the contract.

The third tier is supplier-specific data: primary reports from the contractor, environmental product declarations (EPDs) for construction materials, or the contractor's own Scope 1 and 2 figures. The City of London is moving exactly this way, from spend-based calculations to contract-specific carbon data: in one year 40 suppliers provided contract-specific data covering nearly half of spend on purchased goods and services.

The GHG Protocol methodology allows a hybrid approach, combining primary data where available with average factors to fill gaps. The essential rule is never to add supplier-specific data on top of the spend-based estimate for the same volume, because that counts the same emissions twice.

• Year one — spend-based estimate across the whole portfolio for ranking.
• For construction materials, ask for EPDs and physical quantities, not invoiced money.
• For waste, collect tonnes by disposal route: landfill, incineration, recycling.
• For machinery and vehicles, collect fuel type, volume or hours run.

What cities actually find

European studies converge on the same picture: most of a municipality's procurement climate impact is indirect and occurs beyond its borders. For Zeist the annual impact of procurement was estimated at roughly 37 kilotonnes of CO2-equivalent, with more than 95% indirect Scope 3 emissions arising during production and processing of materials and services. Nijmegen reached similar conclusions and recommended focusing on extending the service life of infrastructure and using recycled and biobased materials.

The City of London example shows how to shift from average factors to contract-level data and to embed requirements in procurement policy and contract management. The UK goes further at national level: Procurement Policy Note 06/21 requires suppliers bidding for major central government contracts to publish a Carbon Reduction Plan and commit to Net Zero, or risk exclusion from the process.

These are national rules in specific jurisdictions, not a universal standard. A city must reconcile any requirement with its own public-procurement law and avoid copying foreign thresholds blindly. This is general information, not legal advice for any particular jurisdiction.

• Zeist: roughly 37 kt CO2-eq per year, with over 95% in procurement Scope 3.
• Largest sources — civil engineering works, waste processing and the social domain.
• City of London: 40 suppliers provided contract data, about 50% of goods and services spend.
• UK PPN 06/21: carbon reduction plans required from major government suppliers.

Honest limits: risks and constraints

The honest answer to what a city can really count is a set of categories with differing accuracy, not one precise number. Spend-based estimates against sector averages can deviate substantially for an individual contract, especially when a contractor has already cut emissions or, conversely, uses carbon-intensive materials.

The main technical risk is double counting. A contractor's fuel burned on a site inside the city boundary can enter both the territorial Scope 1 and the procurement Scope 3 unless the accounting boundary is agreed in advance. A second risk is data from small and medium contractors, who often have no reporting system, forcing the city back onto average factors and clearly labelled estimates.

A third risk is mixing methodologies and base years. Switching from spend-based to supplier-specific data changes the numbers by itself, with no change in real emissions. Cities should therefore document the methodology version, emission-factor sources, base year and data quality, and review materiality each year as spend and grid factors shift.

• Set the accounting boundary so a contractor's fuel is not counted twice.
• Tag every value: primary data or an estimate from averages.
• Do not compare years directly across a methodology change without recalculating.
• Small contractors are the biggest data gap; decide explicitly where average estimates are acceptable.

A pragmatic measurement ladder

The workable approach is not to cover everything at once but to build a ladder of accuracy. In year one the city runs a spend-based estimate across the portfolio, ranks contracts and picks the material categories. Then, for the top few dozen contractors by emissions or spend, it inserts data and EPD requirements directly into tender documents and contracts.

This staged path delivers priorities in the first cycle and gradually raises the share of primary data without rebuilding the whole system. The key is to place data requirements at the tender stage, not after the contract is signed, when leverage is already limited.