Estimating upstream and downstream supply chain emissions with input-output data: A two-way street

The twenty-eighth Conference of the Parties (COP28) was the first occasion since the Paris Agreement was signed to include a global stocktake report evaluating the progress toward the Paris targets. One key finding of the report was that “Credible, accountable and transparent actions by non-Party stakeholders are needed to strengthen efforts for systems transformations.” More and more companies, including financial institutions, are setting targets to align with the Paris Agreement, and holding them accountable requires monitoring their alignment with their stated emissions targets.

The Greenhouse Gas Protocol assigns companies full responsibility for their direct emissions (also known as scope 1 emissions) – for example, the on-site emissions from burning natural gas for heating. In addition, firms are also responsible for their indirect supply chain emissions. These can be divided into upstream emissions that, for example, result from the production or transport of inputs (e.g. emissions from the production of wheat that is used in a bakery) and into downstream emissions that occur from the processing, transportation, use, and disposal of a product. Both upstream and downstream emissions are generally referred to as scope 3 emissions, with the exception of the upstream emissions from power or heat production, which is referred to as scope 2.

Measuring and reporting a financial institution’s scope 1 and 2 emissions is fairly easy, but these emissions are typically very small compared to its scope 3 emissions. The most important category of a financial institution’s scope 3 emissions is what the Greenhouse Gas Protocol classifies as the “investment” category (see figure 1), which includes all types of financial products and whose associated emissions are also referred to as financed emissions. Financed emissions consist of the client’s scope 1, 2, and 3 emissions. It is especially challenging to attain information on a client’s scope 3 emissions . A recent survey by the Carbon Disclosure Project shows that less than half of the companies that do report emissions also report scope 3 emissions, leaving financial institutions[1] dependent on other options to get an understanding of their indirect impact via their clients’ supply chains.

Where information is lacking, the Greenhouse Gas Protocol acknowledges that environmentally extended multiregional input-output (EEMRIO) models can be used to bridge the gap until more reliable bottom-up data is available. Also, the Partnership for Carbon Accounting Financials (PCAF) in principle accepts the use of data from EEMRIOs. To support its members in their reporting activities, PCAF also provides upstream scope 3 emissions multipliers. The Greenhouse Gas Protocol acknowledges the use of EEMRIO data for upstream scope 3 emissions of investments (category 15 emissions). However, in principle, EEMRIO tables can also be used to explore downstream links. Via these links, emissions associated with “downstream responsibility” can be estimated following a principle comparable to that employed in upstream emissions estimations.[2]

In this article, we want to highlight the combined use of upstream emissions estimates and downstream emissions estimates from EEMRIO tables to gain a broader picture of scope 3 emissions when more precise bottom-up data is unavailable.

[1] PCAF requires reporting of financed scope 3 emissions for every sector from 2025 onward.

[2] See, for example, Lenzen and Murray (2010) and Huidobro et al. (2023).

Read the complete publication 'Estimating upstream and downstream supply chain emissions with input-output data: A two-way street' as a PDF file.